ML20140F235

Monitoring Marine Environ of Long Island Sound at Millstone Power Station,Waterford,Ct Annual Rept 1996

ML20140F235
Person / Time
Site:	Millstone
Issue date:	12/31/1996
From:	Colby D, Decker G, Keser M NORTHEAST UTILITIES SERVICE CO.
To:
Shared Package
ML20140F232	List: B16435, Forwards 1996 Annual Environ Protection Plan Operating Rept for 1996 & Monitoring Marine Environ of Long Island Sound at Millstone Nuclear Power Station,Waterford,Ct Annual Rept 1996 ML20140F228 ML20140F235
References
NUDOCS 9705020178
Download: ML20140F235 (250)

Text

-- - . - . - . - .- . - . - _ . - - . -.

l 4

Monitoring the Marine Environment l of Long Island Sound at ,

)

Millstone Nuclear Power Station

. i I

1996 Annual Report l

i i

Prepared by: Staffof j Northeast Utilities Service Company Environmental, Health & Safety Services NU Environmental Laboratory Approved by: #E* #" " l l

1 Dr. Milan Keser

)

l April 1997 9705020178 970428 PDR ADOCK 05000423 R PDR

WASTE l

Wly i

l O

i Northeast Utilities has been a Charter member of the WasteWi$e Program since 1994 l i

l l l

l l

l

@ Printed on recycled paper (50% recycled,10% post-consumer waste)

Use of the WasteWi$e logo does not imply EPA endorsement ii Monitoring Studies,1996

Acknowledgements This report was prepared by the staff of Northeast Utilities Service Company (NUSCO),

Environmental, Health & Safety Services, located at Northeast Utilities Environmental Laboratory (NUEL), Millstone Nuclear Power Station, PO Box 128, Waterford, CT 06385. Staff members include:

Dr. Milan Keser, Manager  :

David P. Colby Donald J. Danila Gregory C. Decker David G. Dodge

, James F. Foertch Raymond O. Heller

Donald F. Landers, Jr. Dr. Ernest Lorda J. Dale Miller Douglas F Morgan John T. Swenarton Christitie A. Temichek Joseph M. Vozarik t

Special appreciation is extended to summer staff for their untiring efTorts in field and laboratory support: Bethany Dickert, Cara Endyke, Jennifer McCain, Heather Lisitano, Julie Kristoff, Yvonne Rinehart, and Wendy Sminkey. Additional thanks are extended to Paul A. Brindamour, Norman

W. Sorensen, and Robert J. Stira of Environmental, Health & Safety Services, Berlin, CT, for their contributions to the monitoring programs. Dr. Michael D. Scherer of Marine Research, Inc. and Mark R. Gibson of Rhode Island Fish & Game kindly supplied reports and data on winter flounder.

Critical reviews of this report were provided by the following members of the Millstone Ecological

! Advisory Committee: Dr. John Tietjen (City University of New York), Dr. Nelson Marshall (emeritus, University of Rhode Island), Dr. Saul Saila (emeritus, University of Rhode Island), Dr.

William Pearcy (Oregon State University), Dr. Robert Wilce (emeritus, University of Massachusetts), and Dr. Robert Whittatch (University of Connecticut).

J J

Acknowledgements iii

l l

i i

I i

i l

l t

i l

l '

i

! s  !

a i

'e e

u w e a tej3b :

t

's l%J. ,

.,p ,yff^DF'~;

_ Q -g'y)j T "'N N

_

.N a e' a2; & & .. *, ci ;j Dr. Nelson Marshall, Professor Emeritus of Oceanography and Marine Affairs, University of Rhode Island.

iv Monitoring Studies,1996

Dedication Nelson Marshall has spent a lifetime in the study of the Niantic River and its environs. That study began as a boy when Nelson and his family summered on the River, but for the past 40 years ,

Nelson's study of the River has been as a professional scientist. In the early 1960s Nelson, several colleagues from Connecticut College and graduate students from the University of Rhode Island began an ecological examination of the Niantic River that focused on primary production and scallop ecology. In the mid 1960s when Northeast Utilities selected Millstone Point for location of three nuclear power plants it naturally tumed to Nelson for advice on marine ecology. Thus began a relationship between Nelson and NU that has lasted for more than 30 years. Nelson was the Chairman of the Millstone Ecological Advisory Committee until 1984 when, following his retirement as Professor of Oceanography and Marine Affairs at URI, he and his wife Grace moved to the Maryland eastem shore. Fortunately for NU, Nelson has remained a vital member of the Ecological Advisory Committee to this day.

s Following receipt of his Ph.D. from the University of Florida in 1941, Nelson's career took him to j such institutions as the University of Miami, UCONN, William and Mary, and Alfred University.

Ile was at URI from 1959 to 1984. Honors include Fellowship status in the American Association for the Advancement of Science and lionorary Membership in the Atlantic Estuarine Research Society. IIis research interests, which include coral reef and mangrove ecology and marine resource development, in addition to estuarine ecology, have taken him to Fiji, Malaysia, the Marshall Islands, the Caribbean and elsewhere. lie has produced many graduate students who have l

gone on to excellent careers in marine science thanks to him.

In recognition for the 30 years that Nelson Marshall has served NU as ecological advisor and friend, the scientific staff of the Northeast Utilities Environmental Laboratory and the members of the Millstone Ecological Advisory Committee dedicate this 1996 Annual Report to him.

Dedication v

i i r

1 i

L i

l l

I I

I l

1 1

I I

i i

i i

i

)

1 l

i

(

s l

vi Monitoring Studies,1996  !

I 1

4 i

Executive Summary Fish Ecology Studies coast, effects of MNPS on sand lance abundance is difficult to ascertain, but is probably small.

Studies of fish assemblages in the vicinity of The bay anchovy is typically the most abundant  ;

MNPS were conducted to determine the efTects of ichthyoplankton species collected in estuaries within station operation. %ese effects have been defined as its range and was the dominant larval taxon entrained station-related changes in the occurrence, distribution, at MNPS. Recent abundance has been relatively low and abundance of fishes, which can alter community in comparison to the mid-1980s, but this decline structure. Potential effects include the entrainment of occurred prior to three-unit operation and no early life history stages through the cooling-water significant trends were found for either two- or three-system (probably the most important effect), impinge. unit operation. He egg and larval densities and the ment ofjuvenile and adult fish on the intake screens, entrainment estimate for 1995 were among the lowest which was mitigated by the installation of fish retum of the past 20 years. The numbers of eggs and larvae  !

sluiceways at Units I and 3, and changes in distribu. entrained each year were not significantly correlated l tion in Jordan Cove as a result of the thermal with densities found the following year, implying no l discharge, direct efTect of MNPS on the spawning stock of this Trawl, seine, and ichthyoplankton (fish eggs and short-lived species.

larvae) monitoring programs were established in 1976 Atlantic and inland silversides are among the to provide information for the assessment of impacts most common shore-zone species along the from MNPS operation. These programs provided the Connecticut coast. These species fluctuate in relative basis for identifying taxa potentially alTected, as well abundance from year to year. Typical of short-lived as information on long-term abundance trends used to Species, the abundance of silsersides is highly variable i measure changes in the local populations. About 130 and annual catches by trawl and seine have ranged different fish taxa have been collected in these over two orders of magnitude. Recent catches of monitoring programs. Of thase, six taxa, including silversides by trawl and inland silverside by seine were l American sand lance, anchovies, silversides, gruoby, within historic ranges. Ilowever, the Atlantic .

cunner, and tautog, were identified as having the silverside has significantly decreased in abundance  !

potential to be impacted by MNPS, either by during the three-unit operational period at the Jordan j entrainment or exposure to elevated water tempera. Cove seine site. This decrease was probably not tures from the plant discharge. Abundance data were related to thermal effects, as only a minimal (0.8 C) analyzed separately for the two-unit (1976-85) and increase in water temperature is found at the seine three-unit (1986 through 1995 or 1996, depending station, which is less than typical summer diurnal upon the sampling program) operational periods and variation on the shallow sand flats. Because catches of for the entire 20-year data series (both periods adults by trawl during winter did not show similar combined) to determine if changes in abundance have changes in abundance, MNPS likely has not affected occurred. the local Atlantic silverside population American sand lance larvae ranked third among The grubby is unique because unlike other entrained fish larvae and densities in entrainment potentially impacted species is experiences no fishing

]

samples have decreased after peaking in the late 1970s pressure and has little forage value. Both larval and '

and early 1980s. Declines in sand lance abundance adult grubby abundance indices have been relatively were also apparent in other areas of the Northwest stable throughout the 20 years of monitoring, Atlantic Ocean, with abundance found to be inversely suggesting little plant effect.

correlated with that of Atlantic herring and Atlantic The most abundant of the fish eggs entrained mackerel, both of which prey upon larval sand lance. were cunner eggs, which accounted for more than 50%

llowever, abundance has increased again in recent of all eggs collected since 1979. During three-unit years, although the mean larval density during 1995 operation, cunner eggs increased in abundance, with was the lowest of the past 4 years. Given the large the density of cunner eggs in 1995 the third largest changes in abundance of this fish along the Atlantic recorded, as was the annual entrainment estimate.

Executive Summary vii l

1 I

J

llowever, the densities of larvae found decreased by if egg losses due to entrainment affected l about 50% from 1994. Young-of-the-year cunner recruitment of tautog, then juvenile abundance accounted for a higher proportion (about 70%) of fish should also decrease and the relative abundance of l caught by trawl since three-unit operation began. older fish would appear to increase in the short term.

l Trawl catch at a station near the MNPS intakes showed Based on length-frequency distribution from trawl a significant decline during two-unit operation. This catches, the percentage of juvenile tautog increased decrease was most likely related to the mid-1983 during the three-unit operational period. Therefore, removal of a rock cofferdam at the Umt 3 intake changes in the relative proportion of juveniles and structure that provided habitat for cunner. Aflerwards, adults were probably unrelated to entrainment losses, catches became similar to another station in nearby in addition, the decline in juvenile and adult tautog mid-Niantic Bay. The entrainment of eggs is the abundance in Long Island Sound that began in the greatest potential impact of MNPS on the cunner mid-1980s coincided with the decreasing numbers of population. liowever, if egg losses affected eggs collected at MNPS. If the decrease in adult recruitment, then juvenile abundance should decrease numbers was caused by entrainment losses, then the in relation to older fish. This decrease was not reduction in egg abundance should have lagged the ,

apparent in the length-frequency distributions, and decline ofjuveniles by several years because females '

relative abundance of juveniles actually increased do not mature until age-3 or 4. Therefore, the lower during the three-unit operational period. abundance of tautog eggs was probably due a decline The tautog was the second-most abundant egg in the abundance of spawning adults from fishing j taxon entrained, accounting for over 27% of the total rather than the operation of MNPS. At present, eggs collected since 1979, Wth mean density in 1995 tautog stocks are consikred overfished and because  ;

the largest seen since 1990. Tautog larvae, however, of the long life and slow growth of this species,  !

were not as prominent, ranking eighth in abundance. abundance should remain depressed until fishing  :

No correlation was found between eggs and larvae and mortality rates are substantially reduced. I no significant trends in abundance were found during either operational period. Catches of tautog by trawl Winter Flounder Studies were dominated by young-of-the-year. In contrast to 1994-95, when total trawl catch was the lowest in 20 The local Niantic River population of winter years, the catch during 1995-96 was the largest. Dounder (Meuronectes americanus) is potentially Tautog, particularly fish in size-classes that correspond affected by the operation of MNPS, particularly by to ages-3 through 5, were also taken in lobster pots. entrainment of larvae through the cooling-water The 1996 catch at the Jordan Cove station was six systems of the three operating units. As a result, times the previous high. The reasons for this increase extensive studies of the life history and population are unclear, but may have been related to changes in dynamics of this valuable sport and commercial prey availability. species have been undertaken since 1976.

Special studies on tautog eggs showed that in contrast to the previous 2 years (1994-95),

large (65-80%) decreases in egg abundance occur when monthly mean seawater temperatures were following early evening spawning through the warmer than average, temperatures recorded at the following morning, most probably a result of high MNPS intakes during 1996 were among the coolest l natural mortality. Pelagic tautog eggs disperse of the past 21 years, particularly during spring and rapidly from spawning sites by tidal transport and summer. The cold weather produced heavy ice cover densities in nearshore areas are relatively uniform. in the Niantic River, delaying the start of the adult Based on hydrodynamics, a conservative measure of winter flounder survey until February 27. The ,

the source area for eggs entrained at MNPS includes 3.mean trawl catch-per-unit-effort (CPUE) of fish a radius of about 5 nautical miles. Two daily larger than 15 cm during the spawning season was estimates of the instantaneous standing stock of 1.6, the lowest of the series. Larger females have  ;

tautog eggs within this area equaled or exceeded made up a greater proportion of the spawning stock annual entrainment estimates at MNPS and, in fact, in recent years as abundance declined to low levels.

would have been even larger if high egg mortality The Jolly stochastic model was applied to mark and rates had been taken into account. This implies that recapture data to estimate the absolute abundance of MNPS entrainment effects may be relatively small.

viii Monitoring Studies,1996

the adult spawning population. The abundance Densities of newly metamorphosed demersal estimate for 1995 was 5,574 winter Counder, lower young were relatively low in 1996. Young winter than the estimates of about 10-16 thousand for 1992 Counder were particularly scarce during late summer 94 and considerably less than estimated population and the median beam trawl CPUEs were among the sizes during 1984-91 that ranged between 33 and 80 lowest recorded since this sampling began in 1983. I thousand spawners. One-third to almost two-thirds The A-mean CPUE calculated for young winter of the winter Counder found in the Niantic River Dounder taken during the late fall and early winter at during the spawning period each year were mature the trawl monitoring program stations was 4.8 in females. Female spawner abundance estimates 1995-96, the lowest value since 1976-77. This low ranged from 2,427 (1996) to 68,899 (1982), with abundance was unexpected, given the relatively h:gh corresponding total egg production from about 2.1 to numbers of young produced in the 1995 year-class.

39.9 billion each year. This abundance index was signincantly correlated The low abundance of newly-hatched larvae in with that of young Osh taken in the Niantic River Niantic Bay compared to the Niantic River suggested during 1994, and also indicated that the 1988 and that most local spawning occurred within the river. 1992 year-classes were relatively abundant, whereas in addition, abundance indices of Stage I larvae in the 1993 year-class was weak. Few juveniles have the river were signincantly correlated with been taken within the Niantic River during the adult independent estimates of female spawner egg spawning population surveys in recent years.

production. Densities of Stage I and 2 larvae in the Young-of-the-year abundance indices were either not Niantic River during 1996 were about average, correlated or were negatively correlated with the except for Stage 2 larvae at a station in the upper abundances of age-3,4, and 5 female adult spawners.

river, which was the second highest of the series. Thus, none of the early life stages was a reliable flowever, abundances of Stage 3 and 4 larvae this index of year-class strength for Niantic River winter '

year were at or below average at all stations. Since Counder stock.

1976, annual larval abundances in Niantic Bay Egg production estimates from annual spawning appeared to reflect region-wide trends as they were surveys were scaled to numbers of spawning females l highly correlated with abundance indices for Mount and used as recruitment indices. These indices i ilope Bay, MA and RI. together with adult female spawning stock estimates Smaller size-classes of larvae were dominant in and mean annual February water temperatures were the river and larger size-classes were more prevalent used to St a three-parameter Ricker stock-recruitment in the bay. The reduced cooling-water Oow in 1996 relationship (SRR). Additionally, an indirect esti-resulted in larger catches of smaller larvae in MNps mate of the winter nounder theoretical rate of entrainment samples, which could have been the increase (the SRR a parameter) was used for result of reduced net extrusion under low Dow and modeling winter nounder population dynamics for )

slower water "elocity conditions, impact assessment. Tha value of a in biomass units  !

In Niantic Bay, growth and development were was estimated as 5.87. The estimate of p (the second correlated with water temperature. In the river, SRR parameter), which describes the annual rate of l growth appeared to be related to both water compensatory mortality as a function of stock size,  ;

temperature (positively) and larval density has shown little annual variation since 1988. The ]

(negatively). Growth and development were slower third parameter in the SRR described a negative l than average in 1996, likely due to cooler water relationship between winter Dounder recruitment and I temperatures. Estimated mortality of larvae in the wr.ter tempera:ures in February, the month when  !

Niantic River for 1984-95 ranged from about 82 to most sn* .ning, egg incubation, and hatching occur.

98% and was 94.8% in 1996. Density-dependence fhe number of larvae entrained through the was examined using a function comparing mortality condenser cooling-water system at MNPS is the most t

with egg production estimates (a measure of early direct measure of potential impact on winter stage larmi abundance) and various monthly and flounder. Annual estimates of entrainment were seasonal water temperatures. The best model related to both larval densities in Niantic Bay and indicated that larval mortality increased as egg plant operation. The entrainment estimate in 1996  !

production increased and spring (April-June) water was $3.9 million larvae, the second lowest since temperatures decreased. three-unit operation began in 1986. This was largely Executive Summary ix

i l

attributed to plant operation as cooling-water volume rapidly recovered. New series of stock size during 1996 was the lowest in the three-unit projections were then simulated by adding the effect operationc.) period, with Units 1 and 2 shut down for oflarval entrainment at MNPS. The lowest projected all and Unit 3 for most of the larval winter flounder stock biomass under simultaneous fishing and MNPS season. The decrease in cooling water use resulted in impact again occurred in 1993 (10,604 lbs),whereas a calculated reduction in entrainment of about 72% the greatest absolute decline relative to the baseline (138 million larvae) from that expected if all three occurred in 2000 (a difTerence of 18,682 lbs).

units had operated fully during the season. Generally, greater reductions in stock biomass The impact of larval entrainment on the Niantic resulted from fishing than from larval ectrainment.

River stock depends upon the fraction of the winter The simulated spawning stock returned to within Counder production entrained each year. Empirical about 1,700 lbs of baseline levels in 2030, only 5 mass-balance calculations for 1984-96 showed that a years after the scheduled termination of Unit 3 large number of entrained larvae come from areas of operation in 2025, and became virtually identical to Long Island Sound other than the Niantic River. In the baseline in 2033.

previous years, an estimated 14 to 38% of entrained The probabilities that the Niantic River female larvae originated from the Niantic River, but the spawning stock biomass would fall below selected estimate for 1996 was 59% On the other ; and, the reference sizes (25, 30, and 40% of MSP) were fraction of the annual riser production etrained, determined to help assess the long-term effects of which has ranged from 5.4 to 42.3% in e revious MNPS operation. A stock smaller than 25% of MSP years, was a moderate 25.7% in 1996. is considered overnshed, whereas one that is at 40%

A stochastic computer simulation model of MSP can maximize yield to the fisheries while l (SPDM) was used for long-term assessments of remaining stable. For both baseline and MNPS-  ;

MNPS impact over a 100-year period (1960-2060). impact simulations, stocks were likely (p = 0.92) j Tne winter flounder stock simulated was female greater than 40% of MSP in 1970. At the lowest I spawner biomass (Ibs), which is more directly related point of both stock projections in the mid-1990s, all j to reproductive potential than fish numbers. replicates of the stock projections were less than 25% l Conditional mortality rates corresponding to larval of MSP. Simulated reductions in fishing allowed for i entrainment from mass-balance calculations and a rapid increase in spawner biomass in 2000. By l juvenile and adult impingement at MNPS were 2010, spawner biomass of the impacted stock was simulated according to historical information and likely (p = 0.91) greater than 30% of MSp and had a projections; natural and fishing mortality rates (F) probability of 0.42 of being larger than 40% of MSP.

were provided by CT DEP. For simulation purposes, This simulated recovery _ however, assumed that F was initially set at 0.40 in 1960 and reached a changes in fishing regulatie were implemented as .

maximum of 1.33 in 1990. Based on proposed scheduled and that they achieved reductions in I changes in fishing regulation, F was projected to 'ishing mortality rates as expected. Even with decrease substantially over the next decade to 0.60 by substantial reductions in fishing mortality and 2006 and remain unchanged thereafler. termination of MNPS operation, the probabilistic in the SPDM simulation, an initial stock size of analysis indicated a one in three chance that the new I13,415 lbs was used to represent the theoretical (no equilibrium stock biomass would still be smaller than fishing effects) maximum spawning potential (MSP) 40% of MSP after 2040. To date, however, the of the Niantic River female spawning stock. When- Niantic River winter flounder stock has not shown fishing was added, the annual projections of the evidence of a rebound in abundance as suggested by initially unfished stock become the baseline time- the model. Even though fishing remains high, this series of annual spawning biomass in the absence of population has remained resilient and very small any plant impact. Under the exploitation rates adult spawning stocks in recent years have produced simulated, as provided by the CT DEP, the stochastic relatively large year-classes of young fish.

mean stock size of the baseline declined to 56,243 lbs Nevertheless, continued efforts in reducing fishing by 1970 and to its lowest point of 12,880 lbs in 1993, are necessary to ensure a recovery and avoid a stock The latter value was less than one-half of a critical collapse.

stock size, defined as 25% of MSP. Following simulated reductions in fishing, however, the stock x Monitoring Studies,1996

Lobster Studies years of growth to reach legal size, continued monitoring of lobsters will demonstrate the effects, if The total number of lobsters caught (all sizes) any, of 3-unit operations on the local lobster and total CPUE in 1996 was within ranges of population.

previous years; however, CPUE of legal lobsters landed in 1996 was the lowest observed in the 3-unit Rocky Intertidal Studies period. This decline was expected because total CPUE in 1995 was the lowest observed in the nearly DifTerences among rocky intertidal stations in 20 year study period, so fewer sublegal-size lobsters community composition were attributed to site-were available to molt to the legal size class in 1996. specific environmental conditions created by the There has been an overall decline in legal lobster inDuence of many interacting factors. At three of abundance since 1978, primarily due to increased four study stations, major differences among Hshing rates, which have more than doubled since communities (e.g., based on abundances of the 1978, and to increases in minimum legal size dominant taxa such as barnacles, Fucus and implemented in 1989 and 1990. Chondrus) were attributed to natural variability in Lobster catches and molting peaked earlier factors that afTect the degree of wave disturbance at during the overall 3-unit period (1986-95) than each site. These factors include site orientation to during the 2-unit period (1978-85), probably owing prevailing wind-generated waves, the ability of to the regionally warmer May to August water exposed substratum (slope) to dissipate the horizontal temperatures observed in recent years. Consistent force of those waves, and the character of that with this Gnding, cooler than normal water substratum (e.g., boulders, bedrock ledge).

temperatures in 1996 delayed lobster catch and in addition to these natural factors, impacts molting peaks. Other changes in local lobster related to the MNPS thermal plume have created a population characteristics during 3-unit operation distinctive intertidal community on the shore area were related to implementation of new Gshery immediately adjacent to the discharge outfall to the regulations, rather than to pour plant impacts. The east (Fox island). The unique algal Dora at Fox increased proportion of berried females is as.;ociated Island-Exposed (FE), developed under elevated with the increases in minimum legal size, and should temperature conditions caused by the 3-unit thermal j increase larval production as a larger proportion of plume, continued to be evident in 1995-96 based on females are able to spawn before reaching legal size. qualitative sampling. There was little change to the Similarly, implementation of the escape vent FE Dora resulting from the extended 3-unit outage in regulation in 1984 has led to lower percentage of 1996. The most notable shifts in species occurrence lobsters missing one or both claws (culls) during 3- at FE during 3-unit operation, relative to unimpacted

! unit operation. sites, were the presence of warm water-tolerant The total estimated number of lobster larvae species not typical of other sites (Agardhiella l entrained through the MNPS cooling water systems subulata. Gracilaria tikvahiac and Sargassum i during 1996 was the lowest reported since filipendula), absence of common cold water species entrainment studies began in 1984. This reduction in (Afastocarpus stellatus. Dumontia contorta, and entrainment of larvae was the result of the shutdown Polysiphonia lanosa) and extended or reduced i of MNPS during 1996. For the most part, periods of occurrence of seasonal species with warm entrainment levels have been considerably higher water or cold water affinities, respectively.

during 3-unit operational years relative to 2-unit Dominant species abundance patterns were years, due the additional cooling water demand of altered by 2-cut water circulation patterns and by 3-Unit 3. The long term impact of larval entrainment unit operations only at FE. These changes were most at MNPS is difficult to quantify because the source of notable in the low intertidal zone at FE, where larvae entrained at MNPS is not known, and larval temperature conditions were most severe. The low

( survival, settlement and ultimate recruitment to the intertidal community at FE, which prior to 1983 had fishery are not fully understood. Since lobsters been unimpacted and characterized by perennial require 4-5 years of growth before they become populations of Fucus, Chondrus, and Ascophyllum vulnerable to capture in our traps and an additional 2 and predictable seasonal peaks in barnacle and Executive Summary xi

Afonostroma abundance, has been replaced by a and was eliminated after only two years. Causes for persistent community dominated by Codium, Ulva. the instability of celgrass in the Niantic River are Enteromorpha, and Polysiphonia. These populations unclear, but not related to power plant operation maintained dominance within the FE intertidal because this site is well beyond the zone of influence community during 1996, and small population = of of MNPS. No indications of decline have been noted Sargarsum and Graci/c b, found only in FE study at the current celgrass bed monitored in the Niantic transects, also persisted. River (NR #4), sampled in 1995 and 1996. It Elevated temperatures (2-4*C above ambient) at remains to be seen whether factors causing declines our Ascephyllum station nearest the discharge (FN), elsewhere in the river (perhaps watcr quality, disease coupled with higher than normal ambient or waterfowl grazing) will eventually impact the temperatures, may have created unfavorable eelgrass bed at this cite.

conditions for Ascophyllum growth in 1995-96. Eelgrass beds at the other sites, Wp and JC, Ascophr/ lum growth was signincantly reduced at FN have persisted over the entire study period; however, in 1995-96, relative to stations farther away. Owing analyses of some population parameters indicated to the high degree of variability associated with r' oderate decline at both sites. The WP population is Ascophyllum growth, it is not certain whether this on the fringe of the predicted areal extent of the pattern of reduced growth is related to the thermal plume, but temperature monitoring has never temperature regime. In contrast, thermel incursion in indicated water temperatures above ambient at that most previous years caused growth enhancement at site. Furthermore, low standing stock at WP in 1996 FN. Consistent with previous years, Ascophyllum was observed when no thermal plume was produced mortality, or loss of tagged plants and tips, at our at MNPS due to plant shutdown. Therefore, recent present sampling sites was not related to proximity to indications of decline at WP were attributed to the power plant but rather to degree of exposure to natural variability rather than a power plant impact.

storm forces. While declines are apparent for the JC ec! grass population, some improvement in shoot density and Ecigrass standing stock has been obsen'ed in the most recent sampling years (1995 and 1996). The study Eelgrass (Zostera marina) monitoring studies population at JC is within the predicted thermal during 1996 revealed relatively healthy study plume area, and because of shallow water depths populations at current sampling sites. The study sites there (s;l m), this populatiot; is also susceptible to nearest to MNPS, Jordan Cove (JC) and White Point additional stress from solar warming in summer, (WP), have supported stable populations since the sediment freezing in winter, and waterfowl grazing.

study began in 1985. The present study site in the Sediment transport and sand shoaling may also Niantic River (NR #4) has only been monitored since represent environmental stress. Elevated temper-1995; general decline of the overall population in the atures relative to those at the MNPS intakes have Niantic River has necessitated relocation of the study been measured directly at JC, and may have caused site three times over the 12-year study period. Periodic eelgrass population declines observed there.

Variability in population parameters has been Elevated summer temperatures (by 4-5*C) were observed to some extent at all study sites and was measured in 1996 at JC, and were attributed entirely primarily attributed to factors unrelated to MNPS to solar warming and water circulation patterns in opuation. J rdan Cove since MNPS was shut down. Based on Variability in eelgrass abundance and standing 1996 findings, these natural factors were largely stock has been greatest in the Niantic River. Since responsible for temperature Ductuations observed 1985, this population has been characterized by historically at JC, with the MNPS themial plume only isolated, often transient patches. The number and a mmor mDuence at most.

extent of these patches has declined since early study years, as celgrass at three previous study sites has Benthic Infauna been eliminated. Eelgrass recolonization through seed germination has only been observed once, at the Benthic infaunal studies during 1996 continued original study site (NR #1); this new bed declined to document ongoing community changes related to xii Monitoring Studies,1996

?

1 i

1 impacts from construction and operation of MNPS. Discharge scour directly and continuously

] This was accomplished by monitoring nearby impacts the sedimentary environment and the subtidal soft-bottom habitats for changes in infaunal community at EF. Sediments in 1996 were sedimentary characteristics and infaunal community characterized by increased sediment grain size and structure (total abundance, species number and decreased silt / clay levels relative to 2-unit ,

, species composition). Results of these studies operational years. The infaunal community at EF has J

l through the 1996 sampling year indicated that MNPS developed under the new, relatively stable, high operation-related community changes continued to be curr:nt conditions in the discharge area. Populations

observed at the three study sites nearest to MNPS. of species common during 2-unit operation (e.g., A.

l The only study site not inauenced by MNPS catherinae and P. eximius) have returned to EF, was the GN reference site, located well beyond the while other species (including Tharyx spp.) have 4

area of possible impacts. Data collected at this site declined or do not occur even during periods of area-continued to redect long-term physical and biological wide increase (e.g., M. ambiseta). Little change in stability; sediments collected at GN in 1996 were the impacted infaunal community at EF was observed i similar to previous years, and overall community during 1996, when MNPS was shut down. This l composition was consistent over the study period. indicates that this community will likely persist for l Speci0cally, the same four taxa (oligochaetes, Tharyx some time after sediment scour produced by the I i spp., Aricidea catherinae and Mediomastus MNPS discharge ceases.

j ambiseta) have been numerically dominant at similar j relative abundance levels over both 2-unit and 3-unit t operational periods.

. Two study sites (IN and JC) continued to reDect changes related to past isolated physical disturbances from MNPS that occurred over a relatively short

, duration. In recent years, sediment silt / clay content has declined to near pre-impact ievels and indications of community recovery are evident through 1996.

4 Species richness and abundances of oligochaetes and

, Aricidca catherinae (common taxa prior to 1983) i have increased, while abundances of more

} opportunistic species (e.g., Nucula proxima) have

! declined. Ilowever, continued dominance of post-impact species, such as Tharyx spp., indicates the recovery is still ongoing at IN.

j liabitat and community changes resulting from another past disturbance event attributed to MNPS operation were still evident at JC in 1996.

Speci0cally, the effects of siltation at JC after the start-up of Unit 3 in 1986, associated with increased I cooling water Cow and sediment scour in the immediate area of the discharge, continued to be observed in 1996. Abundances of the previously dominant oligochaetes, and the polychaetes Polycirrus eximius and Aricidea catherinae quickly decreased. The impact of this siltation event has apparently lessened since 1986, as populations of some 2-unit perie,d dominants (e.g., A. catherinae) have recovered. However, long-term persistence of some of the depouted silt / clay was still evident after 1995 sampling, and community recovery is evident but slow at JC.

Executive Summary xiii

t i

l l

I I

b i +

l i

i i

i I

f 8

I I

l 1

l l

I xiv Monitoring Studies,1996 1

l l

Table of Contents A c kn ow l ed ge m e n t s ......... ........ ............ ..................... .............. ........................................................ iii Dedication...................................................................................................................................v Execut.iveSummary......................................................................................................................vn..

Introduction.......................................................................................................................................1 F i s h E c o l o gy S t u d i es ....................................... .................................................................... ............... 9 Wi n t e r FI o u n d e r S t u d i es ..................... ......................................................................................... 59 Lo b s t e r S t u d i es .......... .... ............................. .................................... ... ......................... .................. 153 Roc ky I n t ert i d a l S t u d i es ............................................................................................................... 17 9 Eelgrass.......................................................................................................................................203 B e n t h i c I n fa u n a... ....... ................... .................................. ............ ..... ................................... ......... 2 19 Table of Contents xv I

i

1 I

5 i

i 1

I i

l I I l

f

]

1 4

xvi Monitoring Studies,1996

Introduction Reporting Require:nents " Docket No. 4, Certificate of Environmental Compatibility and Public Need for an Electric nis report summarizes results of ongoing environ- Generating Facility Identified as ' Millstone Nuclear mental monitoring programs conducted by Northeast Power Station, Unit 3,' located in the Town of Utilities Service Company (NUSCO) in relation to the Waterford, Connecticut" and dated March 22,1976).

operation of the three-unit Millstone Nuclear Power his report satisfies the requirements of the NPDES Station (MNPS). MNPS can affect local marine biota Pennit and of the CSC by updating and summarizing in several ways: large organisms may be impinged on various studies conducted at MNPS that were the traveling screens that protect the condenser cooling Presented most recently m NUSCO (1996).

l and service water pumps; smaller ones may be i entrained through the condenser cooling water system, Study Area c

which subjects them to various mechanical, thermal, i and chemical effects; and marme communities in the MNPS is situated on Millstone Point, about 8 km  !

discharge area may be subjected to thermal, chemical, west-southwest of New London on the Connecticut  ;

and mechanical effects resulting from the outflow of shore of LIS (Fig.1). De property, covering an area j the cooling water. In addition, occasional maintenance of al-aut 200 ha, is bounded to the west by Niantic  ;

dredging is done in the vicinity of the intake structures Bay, to the east by Jordan Cove, and to the south by he basis for the studies is the National Pollutant Twotree Island Channel. De MNPS monitoring .

Discharge Elimination System (NPDES) permit programs sample a study area of approximately 50 km  ;

f (CT0003263) issusd by the Connecticut Department of that extends from the northern portions of the Niantic Environmental Protection on December 14,1992 to River and Joitlan Cove to Giants Neck,2 km south of  ;

Northeast Nuclear Energy Company (NNECO), on Twotree Island, and 2 km east of White Point. Work i whose behalf NUSCO has under;aken this work. De takes place from the shoreline into areas as deep as 20 I regulations in the permit allow the MNPS c)oling m southwest ofTwotree Island. 7 water to be discharged into Long Island Sound (LIS) in Strong tidal currents predominate in the vicinity of  :

accordance with Section 22a-430 of Chapter 440k of Minstone Point and influence the physical character- i the Connecticut General Statutes and Section 301 of istics of the area. Average tidal flow through Twotree ,

the Federal Clean Water Act, as amended. P--,.$ 5 Island Channel is approximately 3,400 m'sec and at j

of the MNPS NPDES permit states that: maximum is about 8,500 m'sec~' (NUSCO 1983). .

The permittee shall conduct or continue to conduct Current velocities are about 1 to 1.8 knots in the biological studies of the supplying and receiving channel, slightly less (1 to 1.5 knots) near the plant and l mrters, entrainment studies, andintake impingement in Niantic Bay, and relatively weak in Jordan Cove and ,

monitoring. The studies shall include studies of in the upper Niantic River. De currents are driven by intertidal and subsidal benthic communities, finfish semi-diumal tides that have a mean and maximum >

communities and entrained plankton and shall range of 0.8 and 1.0 m, respectively. nermal and i include detailed studies oflobster populations and salinity induced stratification may occur in regions winterflounderpopulationr. unaffected by strong tidal currents. De greatest in addition, paragraph 7 of the permit requires that: temperature variation has been observed in nearshore  !

On or before April 30,1993 andannually thereq$er, areas where water temperature can vary from -3 to l submitfor review and approval ofthe Commissioner 25'C; salinity varies much less and ranges from 26 to a detailed report of the ongoing biological studies 30 . De bottom is generally composed of fine to required by paragrqph 5 and as gpprowd under medium sand throughout the area, but also includes l i

paragraph 6. some rock outcrops and muddy sand, especially near Furthermore, a decision and order of the CnaMy shore. Strong winds, particularly from the southwest,  !

Siting Council (CSC) requires that NNECO inform the can at times result in locally heavy seas (up to 1.5 m or  !

Council of results of MNPS environmental impact greater) near Mi'Istone Point. Additional information monitoring studies and any modifications made to on local hydrography and meteorology can be found in 1 these studies (paragraph 6 of the proceeding entitled NUSCO (1983).  !

Introduction 1 4

- , y - . _ - - . , m.. .__ ..-- .- --

_ _ _ . -. . . _._m._.._ . . - __ . _ . - . ,.______.__.._._.mmm.__.m.

J I

MNPs l NY CT l

\

) ,

4 N  !

River ihm

{'

0 1 mi i

Niente any NPs e Jo'8ea

% cow wNee ont sessa88 Twoese falend Channel Twosse lesend as,esa neer Long Island Sound -

Fig.1. The area in which biological monitoring studies are being conducted to assess the effects of the operation of MNPS.

Millstone Nuclear Power Station Fish retum systems (sluiceways) were installed at Unit 1 in December 1983 ~and at Unit 3 during its The MNPS complex consists of three operating construction to retum aquatic organisms washed off nuclear power units; a detailed description of the the traveling screens back to LIS. The installation and station was given in NUSCO (1983). Unit 1, a operation of sluiceways have minimized the impact of 660-MWe bo3ing water reactor, began commercial impingement at MNPS (NUSCO 1986,1988a,1994).

operation on November 29,1970; Unit 2 is an A chronology of significant events associated'with 870-MWe pressurized water reactor that began MNPS construction and operation, including commercial operation in December 1975; and Unit 3 installation of devices designed to mitigate environ-(ll50-MWe pressunzed water reactor) commenced mental effects and unit operational shutdowns commercial operation on April 23, 1986. All three exceeding 2 weeks, are found in Table 1. Capacity I units use once-through condenser cooling water factors (the electricity produced as a percentage of systems with rated circulating water flows of 26.5, maximum possible production) during 1996 were the i 34.6, and 56.6 m'sec" for Units I through 3, lowes* /;nce Unit 3 went online: 0% _ for Unit I- 1 respectively. Cooling water is drawn from depths of . (shutdown on November 4,1995),13.7% for Unit 2 l about I m below mean sea level by separate shoreline (February 20,1996), and 24.9% for Unit 3 (March 30, intakes located on Niantic Bay (Fig. 2). The intake 1996). All units remained shut down for all or most of I structures, typical of many coastal power plants, have 1996 for reasons that are beyond the scope of this coarse bar racks (6.4 cm on center, 5.1 cm gap) report.

preceding vertical traveling screens to protect the MNPS cooling water is nominally heated in Units 1, plants from debris. Units I and 2 have always had 9.5 2, and 3 from ambient temperature to a maximum of mm mesh screens Unit 3 originally had 4.8-mm mesh 13.9,12.7, and 9.5*C, respectively. Each unit has l screens, a combination of 9.5- and 4.8-mm mesh separate discharge structures that release the heated l screens from early 1990 through summer 1992, and effluent into an abandoned granite quarry (ca. 3.5 ha ,

only 9.5-mm mesh screens as of August 15, 1992, surface area, maximum depth of approximately 30 m).

2 Monitoring Studies,1996 ,

I

I i

TABLE 1. Chronology of mWor construction and operation events at MNPS through 1994.

Date Activity Reference' December 1965 Construction initiated for Unit 1 NUSCO(1973)

November 1969 Construction initiated for Unit 2 began NUSCO(1973)

October 26,1970 Unit I initial criticality; produced first thermal emuent DNGL November 29,1970 Unit I initial phase to grid DNGL December 28,1970 Unit I began commercial operation DNGL January 15,1971 to February 22,1971 Unit I shutdown DNGL August-December 1972 Surface boom at Unit 1 NUSCO (1978)

November 1972 Fish barrier installed at quarry cut NUEL September 3,1972 to March 20,1973 Unit I shutdown DNGL November 1972 Unit 2 colTer dam removed NUSCO(1973)

April 18 to July 28,1973 Unit I shutdown DNGL August-December 1973 Surface boom at Unit 1 NUSCO(1978)

July-December 1974 Surface boom at Unit 1 NUSCO (1978)

September I to November 5,1974 Unit I shutdown DNGL July-October 1975 Surface boom at Unit 1 NUSCO(1978)

July 1975 Bottom boom installed at Unit i NUSCO(1978)

August 5,1975 Unit 3 cofTer dam construction began NUEL September 10 to October 20,1975 Unit I shutdown DNGL October 7,1975 Unit 2 produced first emuent EDAN November 7,1975 Unit 2 initial criticality; produced first thermal effluent EDAN November 13,1975 Unit 2 initial phase to grid DNGL December 1975 Unit 2 began commercial operation NUEL March 19,1976 Unit 3 coffer dam construction finished NUEL June-October 1976 Surface boom at Unit 2 NUSCO(1978)

October I to December 2,1976 Unit I shutdown DNGL December 20,1976 to January 20,1977 Unit 2 shutdown DNGL May 6 to June 25,1977 Unit 2 shutdown DNGL June-October 1977 Surface boom at Unit 2 NUSCO (1978)

November 20,1977 to May 1,1978 Unit 2 shutdown DNGL March 10 to April 15,1978 Unit I shutdown DNGL March 10 to May 21,1979 Unit 2 shutdown DNGL April 28 to June 27,1979 Unit I shutdown DNGL August 10 to 25,1979 Unit 2 shutdown DNGL November 1 to December 5,1979 Unit 2 shutdown DNGL May 7 to June 19,1980 Unit 2 shutdown DNGL June I to June 18,1980 Unit I shutdown DNGL August 15 to October 19,1980 Unit 2 shutdown DNGL October 3,1980 to June 16,1981 Unit I shutdown DNGL January 2 to 19,1981 Unit 2 shutdown DNGL l

December 5,1981 to March 15,1982 Unit 2 shutdown DNGL March 1981 Bottom boom removed at Unit i NUEL September 10 to November 18,1982 Unit I shutdown DNGL Much 2 to 18,1983 Unit 2 shutdown DNGL April-September 1983 Unit 3 cofTer dam removed, intake maintenance dredging NUEL May 28,1983 to January 12,1984 Unit 2 shutdown DNGL December 1983 Fish return system installed at the Unit 1 intake NUEL August 1983 Second quarry cut opened NUEL April 13 to June 29,1984 Unit I shutdown DNGL February 15 to July 4,1985 Unit 2 shutdown DNGL June 1985 Intake maintenance dredging NUEL September 28 to November 7,1985 Unit 2 shutdown DNGL October 25 to December 22,1985 Unit I shutdown DNGL November 1985 Unit 3 produced first emuent EDAN February 12,1986 Unit 3 produced first thermal emuent EDAN April 23,1986 Unit 3 began commercial operation DNGL July 25 to August 17,1986 Unit 3 shutdown DNGL Introductiott 3

.. --~ - - - - - - ~ ~ ~

TABLE 1. (cont.).

September 20 to December 18,1986 Unit 2 shutdown December i to 15,1986 DNGL Unit I shutdown January 30 to February 16,1987 DNGL Unit 2 shutdown March 14 to April 10,1987 DNGL Unit 3 shutdown June 5 to Au8ust 17,1987 DNGL Unit I shutdown November 1,1987 to February 17,1988 DNGL Unit 3 shutdown December 31,1987 to February 20,1988 Unit 2 shutdown DNGL April 14 to May 1,1988

• DNGL Unit 3 shutdown May 7 22,1988 DNGL' Unit 2 shutdown October 23 to November 8,1988 DNGL Unit 3 shutdown February 4 to April 29,1989 DNGL Unit 2 shutdowr.

April 8 to June 4,1989 DNGL Unit I shutdown May 12 to June 12,1989 DNGL Unit 3 shutdown October 21 to November 24,1989 DNGL Unit 2 shutdown March 30 to April 20,1990 DNGL May 8 to June 15,1990 Unit 3 shutdown; lastallation of some 9.54nm intake screen panels DNGL; NUEL Unit 2 shutdown September 14 to November 9,1990 DNGL Unit 2 shutdown February 2 to April 17,1991 DNGL Unit 3 shutdown; installation of new fish buckets and sprayers DNGL; NUEL April 7 to September 2,1991 Unit I shutdown April 23 to May 11,1991 DNGL Unit 2 shutdown May 26 to July 7,1991 DNGL Unit 2 shutdown July 25,1991 to February 6,1992 DNGL Unit 3 shutdown; installation of new fish buckets and sprayers DNGL; NUEL Au8ust 7 to September 11,1991 Unit 2 shutdown October 1,1991 to March 3,1992 DNGL Unit i shutdown November 6 to December 27,1991 MOSR Unit 2 shutdown January 28 to February 14,1992 MOSR Unit 2 shutdown March 22 to April 6,1992 MOSR Unit I shutdown May 16 to June 4,1992 MOSR Unit 3 shutdown; installation of new fish buckets and sprayers - MOSR; NUEL May 29,1992 to January 13,1993 Unit 2 shutdown July 4 to Au8ust 15,1992 MOSR Unit I shutdown Au8ust 15,1992 MOSR Completed installation of new fish buckets and sprayers at Unit 3 NUEL September 30 to November 4,1992 Unit 3 shutdown July 31 to November 10,1993 MOSR Unit 3 shutdown September 15 to October 10,1993 MOSR-Unit 2 shutdown January 17 to May 1,1994 MOSR Unit I shutdown MOSR April 22 to June is,1994 Unit 2 shutdown July 27 to September 3,1994 MOSR Unit 2 shutdown MOSR September 8 22,1994 Unit 3 shutdown October 1,1994 to Au8ust 4,1995 MOSR Unit 2 shutdown MOSR April 14 to June 7,1995 Unit 3 shutdown November 30 to December 15,1995 MOSR Unit 3 shutdown MOSR November 4,1995 to undetermined Unit I shutdown February 20,1996 to undetennined MOSR Unit 2 shutdown MOSR March 30,1996 to undetermined Unit 3 shutdown MOSR

• DNGL refers to the daily net seneration lo8, NUEL to NUSCO Environmental Laboratory records, EDAN to the environmental data acquisition network, and MOSR to the monthly nuclear plant operating status report.

, i The thermal discharge (about II*C warmer than the cuts and within about 1,100 m of the quarry the ambient under typical three-unit ooeration) exits the surface oriented plume cools to within 2.2*C above quarry through two channels (cuts), whereupon it ambient. Beyond this distance the plume is highly mixes with LIS water (Fig. 2). The cuts are equipped dynamic and varies mostly with tidal currents (Fig. 3).

with fish barriers made up of 19-mm metal grates, All hydrothermal surveys conducted at MNPS were which serve to keep larger fish out of the quarry, The described in detail in NUSCO (1988b).

thermal plume is warmest in the immediate vicinity of 4 Monitoring Studies,1996 i

Millstone Nuejear Power unn 3 i Station Jordan Cove 3 Und2 1

escharges 13 200 m o quemr 0 0,

N m ,,,

orieuw cut Niantic Bay Twotree Island Channel Fig. 2 The MNPS site, showing the intake and discharge of each unit, the quarry, and the two quarry discharge cuts.

Monitoring Prograrns NUSCO. 1978. Impingement studies. Millstone Units 1 and 2,1977. Pages 11 to 4-2 its Annual his report contains a separate section for each major report, ecological and hydrographic studies,1977.

monitoring program, some of which have been Millstone Nuclear Power Station.

ongoing since 1968. Rese long-term studies have NUSCO.1983. Millstone Nuclear Power Station Unit provided the representative data and scientific bases 3 environmental report. Operating license stage.

necessary to assess potential biological impacts as a Vol.1-4.

result of MNPS construction and operation. He NUSCO. 1986. De effectiveness of the Millstone significance of changes found for various communities Unit I sluiceway in retuming impinged organisms and populations beyond those that were expected to to Long Island Sound. Enclosure to Letter D0ll85 occur naturally were evaluated using best available dated May 27,1986 from R.A. Reckert, NUSCO, to methodologies. Programs discussed below include SJ. Pac, Commissioner, CT DEP.18 pp.

Winter Flounder Studies, Fish Ecology Studies, NUSCO. 1988a. He effectiveness of the Millstone Lobster Studies, Eelgrass, Rocky Intertidal Studies, Unit 3 fish retum system. Appendix 1 to Enclosure and Benthic Infauna. Reporting periods for each 3 to Letter D01830 dated January 29,1988 from '

section vary and were predicated on biological EJ. Mroczka, NUSCO, to L. Carothers, considerations and processing time necessary for Commissioner, CT DEP. 21 pp.

samples, as well as on regulatory requirements. In NUSCO.1988b. Hydrotherm J studies. Pages 323-cases where the seasonal ahndance of organisms 355 in Monitoring the marine environment of Long differed from arbitrary annual reporting periods, the Island Sound at Millstone Nuclear Power Station, periods chosen were adjusted to bert define the season Waterford, Connecticut, nree unit operational cf interest for a particular species or community. studies, 1986-87.

NUSCO. 1994. Progress report on the MNPS fish References Cited retum systems. Enclosure I to Letter D08071 dated October 20,1994 from D. Miller, NNECO, to T.

NUSCO. (Northeast Utilities Service Company). Keeney, Commissioner, CT DEP. I1 pp.

1973. Environmental effects of site preparation and NUSCO.1996. Monitoring the marine environment construction. Pages 4.4-1 to 4.51 in Millstone of Long Island Sound at Millstone Nuclear Power Nuclear Power Station, Unit 3, Environmental Station, Waterford, Connecticut. Annual report report. Construction permit stage. 1995. 231 pp.

Introduction 5

1 1

l i

7 i

l  : : :  : .

0 6000 ft N I wps Jordan I Niantic Bay E cp

} d s I

~s aF

% 6F White gg 4F Point - J l

~1.5F g%

N ,

Q Twotree

\ ,,Z N. .

sad "

Point Maximum Ebb Tide V

0 50$0 ft N Niantic Bay WPS Jordan  ;

g cove  !

8F A

/ White

\t ;gQ // Point Seaside lg F j

/  %.4p,../.)

t

'... .. - 1.s F -

Twotree E**

Island Point Low Slad Tide Fig. 3. Locations of selected three-unit thennal plume isotherms (1.5'F,4'F,6*F, and 8'F) under various tidal conditions.

6 Monitoring Studies,1996

0 5000 ft N 84 S Jordan -

Niantic Bay

/ E Cove w White I , , - - - - - - ~ ~ /

4F Point Seaside 1.5 F

(.  ;

• Twotree '

Blas Island Point Maximum Flood Tide l

0 s000 et N j Niantic Bay &#PS Jordan l 3 Cove  !

l 8F, CS'N White h\ pg heide UW,sJ/)

1.5Fg,,,,'k Twottee se* island Point High Slack Tide j

Fig.3. (continued). l l

Introduction 7 l

I 1

l l

b i

l i

1 l

l l

l l

l l

l l

l 8 Monitoring Studies,1996

- - .- . . - - - .- -= .. -. --- . - - . - .

4 a

4 Fish Ecology Studies 4

Introduction..................................................................................................................................11

!, Materials and M e thods . .. .. . . . . .. . . ....... . .. .. . . .. . .. .... .. ..... .. ... ... .. . . . . . .. . . .. . .. ... . . ... . . . . .... . ... . ... . .. ... . .. . .. . . . .

Ichthyoplankton Program . . . ... . . ..... .. . . . .. ...... .... . .. .... . . . . . . . . . . . .. .... . . . . . ... . .. . .. . . . .. . . . . .... .. .. . . . . .. . . .. .. . 12 Trawl Program . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ... . . . . 1 S e ine Pro gram . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

DataAnalyses.....................................................................................................................I4 Abundance Estimates ... .. .. .. . . . .. . . . . .. ... .. .. ... . . . . .. ... . .. .. ... ..... . .. .... .. .. .. . .. ..... .. .. . . . . . ... . ... . ...

Entrainment Estimates . .. .. .. . . .. . . . . . .. . .... . . .. .. .. ... .. .. .. . . .. . . . .. .. ... . .. . .. ... .. .... . . ... . . . ... . .. ... ... .

Results and Discussion . .. ... ... . . . . . .. .. . ... .. .. . . .. ... . .. . .... . ...... . . ... . . . . . ..... . . . . . . .. . .. . . .. ..... .. . . . .. ... .. . . .

Species Composi tion . .. . . . ... ... ....... .... .. . . . ..... . .. . .... . . . . . .. ... . . .. . .. .. . .. ..... .. .. . ... . . .... . .. . . . . . . .. .. . .. .. . . . . . I 5 Entrainment Estimates . . . ..... . .. . . .... .... . .. . . . . . .. . .. .. ... . . . . . ... . . . . . . . .. .. . . . . . . .. . .. . . . . . .. . . . . .. .....

Impingement ................................................................................................................I7 S election o f Potentially Impacted Taxa ............................................................................. 17 American sand lance . . .. . ... .. . . . .. ... .. ... . . .. . .. . .. . .. ... .. . . ... . ... . . . . .... . .... . ..... .. ... ... . .. . ... . ...... . . 1' Anchovies..............................................................................................................20 i Silversides..............................................................................................................21 Gmbby ...............................................................................................................2.4 Cunner ..............................................................................................................25 Tautog ..................................................................................................................28 Conclusions............................................................................................................................42 Re fere nc e s C i te d . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Appendices.................................................................................................................................49 1

i i

l l

i i

l Fish Ecology 9

1 l

s I

l l

I 1

l l

l l

1 I

l l

l l

l l

10 Monitoring Studies,1996

Fish Ecology Studies Introduction Trawi, seine, and ichthyoplankton (fish eggs and larvae) monitoring programs were established to Fish are important members of the aquatic com. provior information for the assessment of impacts munity of eastern Long Island Sound (LIS). LIS frara 'v1NPS operation en local fish populations.

provides a variety of habitats for fishes that in These pmgrams provided a basis for identifying taxa concert with a temperate climate results in a diverse Potentially afTected, as well as information on long-assemblage of species, including residents present all term abundance trends used to measure changes in year that exhibit little movement, highly migratory local populations and have changed over time fish present only in certain seasons, and rarely following evaluations (NUSCO 1987,1994a,1995).

collected transients with ; enters of distribution Potentially impacted species were selected on the (Isewhere. Fish also support ccmmercial and sport basis of life history characteristics, such as fishing activities in Connecticut worth millions of susceptibility of early developmental stages to dollars each year (Sampson 1981; Blake and Smith entrainment at MNPS, stock structure (i.e., localized 1984). or coast-wide populations), and local distribution in The objective of the fish ecology monitoring relation to the thermal plume. In this report, data I programs at Millstone Nuclear Power Station from June 1995 through May 1996 are summarized (MNPS) is to determine whether operation of the and compared to data previous collected from trawl,  ;

three electrical generating units adversely affects the seine, and Ichthyoplankton monitoring programs.

occurrence, distribution, and abundance of local Tautog (Tautoga onitis) support one of the J

fishes. Potential MNPS impacts include entrainment principal sport and commercial fisheries of LIS ,

of early life history stages through the cooling-water (Smith et al.1989; ASMFC 1996), but abundance of l system, impingement of juvenile and adult fish on juveniles and adults has declined since 1984 j intake screens, and changes in distribution as a result (Simpson et al.1995). As a result of this decline, of the thermal discharge. Numbers of fish eggs and likely from overfishing on this slow-growing and larvae entrained have been reliably estimated, but long-lived fish, an interstate management plan is quanufying long-term effects of this impact is more under development to provide for the conservation, l difficult. Effects cf entrainment mortr.lity are restoration, and enhancement of the tautog stock l influenced by biological procer.a. mch as (ASMFC 1996). Because of relatively high '

compensatory mortality, density-depenaent growth, entrainment of tautog eggs at MNPS, concem has fecundity of individual species, population age been raised by the Connecticut Department of structure, and life history strategies. Impingement Environmental Protection (CT DEP) regarding the can also be readily measured but, as in the case of tautog population in the vicinity MNPS. From this eggs and larvae, the implications of fish removal are concem, special studies of tautog early life history difficult to ascertain as adult populations are also were conducted in 1996. This work focused on effected by natural and fishing mortality. In any tautog egg distribution and was based, in pan, on the event, the impact of fish impingement at MNPS has methodology and results of previous studies, which largely been mitigated by the installation and are summarized below in the Results and Discussion operation of fish return sluiceways at MNPS Units I section. In addition, a preliminary assessment of and 3 (NUSCO 1986,1988c,1994b). Changes in the tautog egg entrainment was made in terms of thermal regime of local waters due to MNPS equivalent female spawners remmTd by MNPS operation have been well-documented (NUSCO Operation. Tautog studies in 1996 and those that will 1988b; see also the Introduction section to this be completed during subsequent years are being report). If water temperatures exceed tolerance performed in lieu of sampling at three offshore trawl 1:vels, fish may be forced to move from the area, stations (BR, TT, NB; NUSCO 1996) in accordance vacating potentially important spawning or nursery with an agreement with the CT DEP (NNECO grounds. 1995a,1995b,1996).

Fish Ecology 11

Materials and Methods  :

Ichthyoplankton Program  !

Annual results are presented for a 12-month period that extends from June of one year through May of The sampling frequency for ichthyoplankton the following year. Because of occasional overlap in entrained through the MNPS cooling-water system the occurrence of a species during the May-June varied seasonally during 1995-%. Both day and transitional period, species-specific . analyses are ' night samples were collected twice a week during based on actual periods of occurrence instead of. June through August, once a week in September and being constrained to a May 31 endpoint. When a February; and three times a week during March q

. species season of occurrence crossed a calendar year, through May. Only one day sample per week was the annual period was termed a report year (e.g., collected during October through - January. 1 1995-%). When a species was collected only within Generally, samples were collected each week at only a calendar year, the annual period was presented as a one of the three plant discharges (station EN, Fig.1), j specific year (e.g.,1996).~ Although methods of with the site of collection usually alternating weekly j collection for the 1995-% report year were between Units 1 and 2, if plant operations permitted. q essentially the same as those used in previous years, Following the shutdown of Unit I in November  !

the number of stations used in the trawl monitoring 1995 (no circulating water pumps in operation), l program was reduced from six to three in January samples were taken exclusively at Unit 2 until mid- '

1996 with the deletion of stations NB, TT, and BR April of 1996, when sampling was also done at Unit (NNECO 1995a). The materials and methods that 3. Beginning in early May, circulating water pumps  ;

follow correspond to the 1995-% report year. at Units 2 and 3 were operated for only a few hours l

l 1

f 1KM N . - - TRAWL.S l 0l. . s s e SE!NES 1 M1 . PLANKTON  !

I NIANTIC RIVER NR i

, 1 o' -

NIANTIC JORDAN BAY - COVE  ;

EN

',,, , .J.C. . .

IN 9 '\ o Nez Fig.1. Location of current trawl, seine, and ichthyoplankton sampling stations.

12 Monitoring Studies,1996

= ._

each week and entrainment samples were taken spawning) for a total of 12 paired surface-bottom during the infrequent occasions when this occurred. collections. Sampling was conducted near the time To collect samples from the plant discharge, a 1.0 x of slack tidal currents on July 10 and 1I and 3.6-m conical plankton net with 333-pm mesh was maximum tidal currents in both July 16 collections.

deployed with a gantry system. Four General The sampling was conducted at a point midway Oceanic flowmeters (Model 2030) were mounted in between Millstone Point and Black Point having a the mouth of the net and positioned to account for water depth of about 10 m. Temperature and salinity horizontal and vertical flow variations. Sample were determined at surface, midwater, and bottom volume (about 200 m', except during periods of high derghs using a YS1 Model 30 meter.

plankton or detritus concentrations when volume was To determine spatial distribution offshore of reduced) was determir.ed from the average readings MNPS, sampling was conducted within a 5 nautical of the four flowmeters. mile (n mi) radius of MNPS, which was determined Ichthyoplankton samples were separated using a as the most likely potential source area for tautog NOAA-Bourne splitter (Botelho and Donnelly eggs entrained at MNPS. Five sampling points were 1978); fish eggs and larvae were removed from the located at I n mi intervals along three separate samples with the aid of a dissecting microscope. transects, with the point of origin for each transect at Successive splits were completely sorted until at least the mid-point between Millstone Point and Black 50 larvae (and 50 eggs for samples processed for Point. From the origin, the transects extended eggs) were found, or until one-half of the sample had approximately southeast (SE), south (S), and been examined. Larvae were identified to the lowest southwest (SW). The rationale for selecting these practical taxon and enumerated in all samples, except stations and the precise location of each will be for June samples, for which only two (one day and detailed in the Results and Discussion section.

one night) samples per week were examined. Vertical tows were taken at each station with the 60-Tautog, cunner (Tautogolabrus adspersus), and cm bongo sampler described above. The sampler aachovy (bay anchovy, Anchoa mitchilli and striped was lowered from the surface to the bottom and anchovy, A. hepsetus) eggs were identified and immediately retrieved to the surface while the boat i enumerated in all samples collected from April remained at idle; water was filtered during both through September. Tautog and cunner are both in descent and ascent. Replicate tows were taken at the Family Labridae and their eggs are very similar shallower sampling sites until approximately 30 m' in appearance. They were distinguished on the basis of seawater had been filtered (both nets combined).

of a weekly bimodal distribution of egg diameters Material retained by both nets was combined as one (Williams 1967). Allichthyoplankton densities were sample for each station. Sampling was conducted in reported as a number per 500 m' of water filtered. the morning after sunrise on July 2 and July 9.

During 1996, special studies were conducted to Sampling simultaneously with two boats required examine the spatial distribution of tautog eggs about 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br />, starting about 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> prior to slack potentially entrained at MNPS. Tautog egg entrain- current, and was conducted during the period ment is likely a function of egg distribution, that is encompassing slack after ebb (July 2) and slack after related to spawning aggregations of adults and flood (July 9). On both dates, each site was sampled hydrodynamic transport during the approxi.nate 2 to once, except for the station at the transect origin, 3-day span between spawning and hatching. He which was sampled at the beginning, near the l vertical distribution of tautog eggs was examined middle, and at the end of each sampling period.

using paired surface and near-bottom tows by simultaneously sampling with two 60-cm bongo Trawl Program samplers having 333-pm mesh nets and 22.7 kg depressors. Volume filtered was determined from a single GO flowmeter mounted in the center of each Triplicate bottom tows were made using a 9.1-m bongo opening. Sampling was conducted with one otter trawl with a 0.6-cm codend liner. As of January bongo system depkyed just below the surface and 1996, demersal fishes were collected every other week throughout the year at three stations: Niantic ancther near the bottem. Three replicate paired tows River (station NR), Jordan Cove (JC), and Intake were taken during the evening peak spawning period (lN)(Fig.1). A standard tow was 0.69 km, but if the on both July 10 and 15 and during the mornings of July 11 and 16 (apprt ximately 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> after peak trawl net became loaded with macroalgae and Fish Ecology 13

detritus, tow distances were shortened and catches data used for calculating the A-mean were restricted '

standardized to 0.69 km by proportionally adjusting by date to reduce the number of zero values in the '

the catch. His change in sampling procedure -

distribution tails, which extend beyond the season  ;

occurred frequently in 1995-% due to abundant boundaries. Two-unit operational period A-means concentrations of macroalgae and detritus at NR and were calculated from the beginning of two-unit JC. Catch was expressed as the number of fish per operation (1976) to the beginning of three-unit ,

standardized tow (CPUE). Up to 50 mndomly operation (1986). A nonparametric, distribution-free chosen individuals of certain selected species per Mann-Kendall test (Hollander and Wolfe 1973) was station were measured (total length) to the nearest usgi to determine whether the direction of change of mm. Catch of tautog in lobster pots (see Lobster an annual A-mean time-series represented a '

Studies section for sampling methods) were used to significant (p s 0.05) trend. Sen's (1%8) nonparn-supplement the trawl abundance data for this species. metric estimator of the slope was used to describe the rate of change of significant trends. This approach to ,

Seine Program trend analysis was suggested by Gilbert (1989) as particularly well-suited for analysis of environmental Shore-zone fish were sampled using a 9.1 x 1.2-m monitoring data because no distributional assump-knotless nylon seine net of 0.6-cm mesh. Triplicate tions are required and because small sample sizes are shore-zone hauls (standard distance of 30 m) were acceptable. Wilcoxon's signed-ranks test (Sokal and  ;

made parallel to the shoreline at Jordan Cove (JC) Rohlf 1969) was used to compare the catch of tautog biweekly from April through November (Fig.1). among lobster pot stations. Spearman's rank-order Collections were made during a period 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> correlati n (Sokal and Rohlf.1969) was used to ,

before and I hour after high tide. Fish from each describe associations among various abundance l

haul were identified to the lowest possible taxon, indices.

counted, and the total length of up to 50 randomly ,

selected individuals of each species from each Entrainment Estimates replicate were measured to the nearest mm total length. Catch was expressed as number of fish per Entrainment estimates of dominant ichthyoplankton t haul. were calculated from daily density estimates at station EN. These estimates were based on the  :

Data Analyses parameters of a Gompertz function fitted to the j entrainment data. De distribution of egg and larval abundance over time is usually skewed because their Abundance Estimates densities mcrease rapidly to a maximum and then decline slowly, ne cumulative density over time The A-mean was used as an m. dex of abundance of Juvenile and adult fish collected m, the trawl and from this type of distribution resembles a sigmoid-shaped curve, for which the inflection point occurs at seme programs, and of fish eggs and larvae m the ichthyoplankton program. the time of peak abundance. The Gompertz function Begm, nmg with this report, trawl catch data will focus on the three

, (Draper and Smith 1981) was used to describe the cumulative egg and larval abundance distribution.

stations currently bem, g sampled; catch data from the Thus, the inflection point was not constrained to be six stations previously sampled may be found in the mid-point of the sigmoid curve as is the case in NUSCO (1996). The A-mean was selected because it the frequently used logistic and probit curves. De is the best estimator of the mean for population particular form of the Gompertz function . used abundance data that approximately follows the lognormal distribution and contains numerous zeros (Gendron 1989)was:

(Pennington 1983, 1986). Calculation of this index C, = a exp(-exp[-K {t -p)]) (1) and its variance estimate was described m detail m where NUSCO (1988a). Because of varying sampling C, = cumulative density at time t frequencies, the A-mean indices of ichthyoplankton t = time in days from the date when the eggs or taxa were weighted by the largest number of samples larvae first occur collected in a week to standardize data across weeks a = total or asymptotic cumulative density and years. With species that occurred seasonally, the 14 Monitoring Studies,1996 l

l p - inflection point in days since first occurrence americanus) comprised about two-thirds of the

, date larvae collected with 13 other taxa making up most  !

x = shape parameter. of the remainder (Table 1). Cunner, tautog, ar:d anchovies accounted for over 86% of the eggs ne origin of the time scale was set to the date when collected. Silversides (Atlantic silverside, Menidia the eggs or larvae generally first appeared in the menidia and inland silverside, M beryllina) waters off MNPS. Least-squares estimates, standard dominated (80%) the seine catch at station JC errors, and asymptotic 95% confidence intervals of (Appendix II); another 16% were killifishes (striped the a,p, and x parameters were obtained by fitting kilgfish, fundulus majalis and mummichog, F.

the above equation to the cumulative abundance data heteroclitus) and fourspine stickleback (Apeltes using nonlinear regression methods (Proc NLIN; quadracus). Seven taxa accounted for about 82% of SAS Institute Inc.1985). The cumulative data were the total catch at the three trawl stations (Appendices obtained as the running sums of the weekly geomet- lil-V). These were the winter flounder, scup ric means of the abundance data per unit volume. A (Stenotomus chrysops), silversides (mostly Atlantic geometric mean of weekly densities was used in silverside), windowpane (Scophthalmus aquosus),

analyses because the data generally followed a log- grubby (Myoxocephalus aenaeus), skates (mostly normal distribution (McConnaughey and Conquest little skate, Raja erinacea; also the winter skate, R.

1993) and weekly sampling frequencies varied. ocellata and cleamose skate, R. eglanteria), and A " density" function was derived algebraically by anchovies (mostly bay anchovy). Total catch of fish calculating the first derivative of the Gompertz over the 20-year period was very similar between IN function (Eq.1) with respect to time. This density (105,065 specimens) and NR (104,097), each of function, which directly describes abundance over which was about 1.8 times the total catch of 57,766 time (abundance curve), has the form: fish at JC.

d, = a'.x.exp(--exp[-x {t -p)] - x[t -p]) (2) where a, equals 7.a because the cumulative densttles TABLE 1. Taxonomic composition ofichthyoplankton collected at EN (as a percentage of the total) from June 1976 through May were based on weekly (7-day period) geometric 1996 as larvae and April 1979 through september 1995 as eggs.

means, d, is density on day i and all the other w u,y,, pm parameters are as described in Equation 1. Daily entrainment was estimated by multiplying these daily "^

f,fl,l",,,',P,P

,, ,,,,,,,c,,,,, yj "

densities d, by the daily volume of cooling water that Ammodytes americanus 7.9 passed through MNPS. Annual entrainment ((f'g,%^d"'f,,'""'"' j8 estimates were determined by summing all daily thohs gunnet/us 24 estimates during the period of occurrence. ("gl",$,s , adspusus y y6 Enchetyopus combruus  ! .6 Results and Discussion $,',',",'p"p**'/"""'" d Syngnathusfuscus D .0

. .. Scophthalmus aquosus 08 Spectes Composttsos1 repr,ius triacanzhus 0s Clupea harengus 0.6 At least 130 fish taxa were collected as eggs, larvae, juveniles, and adults in the trawl, seine, and ne temporal changes during the 20-year period in ichthyoplankton programs as part of the Fish the composition of the above dominant taxa collected Ecology monitoring studies for MNPS from June in trawl and ichthyoplankton programs were com-1976 through May 1996. This includes fishes pared using A-means. Changes in the composition of collected at present and former stations during this seine catches were not examined because silversides 20-year period, with 116 taxa taken by trawl,51 by have always dominated the catch. In trawl sampling, seine, and 67 collected in ichthyoplankton samples winter flounder and scup had the largest annual (Appendix I). For ichthyoplankton sampling at the A.mean CPUE during each report year, with MNPS discharges (station EN), anchovies (mostly silversides, windowpane, and skates also relatively bay anchovy) and winter 71ounder (Pleuronecte3 numerous (Table 2). Because of the elimination of Fish Ecology 15

TABLE 2. The annual a-mean' CPUE (nol0.69 km) of the most abundant fish collected by trawl at JC, tN and NR for each report year from June 1976 through May 1996 (two-unit operational period; 1976-85; three-unit operational period: 1986-96).

Taxnn 76-77 77-78 78-79 79-80 80-81 81-82 82-83 83-84 84-85 85-86 86-87 87-88 88-89 89 90 90-91 91-92 92-93 93-94 94-95 95-96 )

P. americanus 23.9 15.6 16.7 26.6 34.8 28.9 49.4 30.6 313 23.5 273 273 41.0 23.1 28.4 26.7 253 16.9 22.6 11.4 5 chrysops 14.8 13.0 5.6 6.2 9.2 7.9 25.1 25.9 143 83 24.1 17.4 11.4 11.0 25.8 176.0 563 2.7 26.4 5.1 <

Menidea spp. 18.2 8.5 10.1 7.1 33 2.5 33 2.8 2.0 3.8 23.1 4.1 5.0 2.4 2.9 8.6 18.4 2.5 2.2 1.6 l 5 oguosus 1.7 1.8 0.9 1.8 1.6 1.5 2.2 3.0 2.4 2.5 3.0 43 3.6 4.9 33 2.4 3.7 6.0 3.4 34 M aancus 0.6 0.9 0.9 1.9 1.8 2.5 33 2.1 1.8 1.2 23 1.6 3.5 1.7 2.2 1.4 2.7 13 23 0.9 Rgospp. 0.7 0.6 0.4 0.4 0.8 0.6 1.0 2.6 0.7 1.8 1.8 2.2 2.6 2.4 3.4 3.2 2.2 3.2 2.1 3.5 Data were seasonally restricted to June-October for K chrysops. October-FebruIry for Menidia spp., but unrestricted (June May) for the remaining taxa.

the three offshore trawl stations as of January 1996, taxa fell within the range of abundance values for skates and anchovies became less numerically previous years. The A-mean density for cunner eggs dominant, whereas the grubby increased in relative of 7,126 was the largest value since 1980-81, when it proportion of the catch from the totals for six trawl was 8,223. Anchovy egg abundance remained stations reported in NUSCO (1996). Winter depressed, with the A-mean of 153 the lowest annual

)

flounder, windowpane, grubby, and skates are density index since 1992-93.

, collected throughout most of the year by trawl and Winter flounder larvae ranked second or third each their respective abundances have remained relatively year, except for 1992-93 and 1995-96, when la val l stable since the early to mid-1980s. In contrast, grubby had their third highest abundance in 20 years l scup, anchovies, and silversides were collected (A-mean density of 85). American sand lance '

seasonally (mostly summer and fall for scup and (Ammodytes americanus) larvae were abundant from anchevies and winter for silversides). The annual 1976-77 through 1980-81 and decreased abundances of these fishes fluctuated to a greater considerably until the past few years when A-mean degree because most of the catch was young-of-the- densities began to increase. Hewever, the 1995-96 l year with abundances related to variable reproductive A-mean of 18 represented a substantial decrease from I success. Also, anchovies and silversides school and last year (63). Larvae of the Atlantic menhaden occasional large catches affected the magnitude of (Brevoortia tyrannus) have increased in abundance annual A-mean CPUE. In fact, A-means could not be in recent years and are becoming a dominant summer computed for trawl catches of anchovies because of species. Larvae of other species, particularly those infrequent catches interspersed with a large number of cunner, tautog, rock gunnel (Pholis gunnellus), i of zeroes in the time-series. The 1995-96 A-mean fourbeard rockling (Enchelyopus cimbrius), radiated CPUE for both winter flounder (11.4) and silversides shanny (Ulvaria subbifurcata), and snailfishes (1.6) were the lowest and for scup (5.1) the second (Liparis spp.), occasionally were relatively abundant.

lowest in 20 years. In contrast, the A-mean of 3.5 for skates was the largest annual CPUE for that species Entraittment Estimates group, although annual means were less than when all six stations were included (NUSCO 1996). Trawl Entrainment of fish eggs and larvae in the catch of grubby (A-mean CPUE of 0.9) at the three condenser-cooling water system represents a direct inshore stations during 1995-96 was the lowest since impact from the operation of MNPS. The annual 1978-79. numbers of eggs and larvae entrained were related to All dominant ichthyoplankton taxa were collected their abundance at station EN and plant operations seasonally at EN. Therefore, A-mean densities (i.e., cooling water usage). Due to the start-up of (no/500 m') were computed from data collected Unit 3 in 1986, cooling-water :,sp increased nearly during standardized periods of occurrence for each twofold upon full three-unit operations. However, taxon (Table 3). Cunner eggs were always the most this did not necessarily result in co nparable abundant of the fish eggs collected. Tautog eggs increases in entrrmment estimates, de to were second-most abundant after a large decline in fluctuations in annual abundances of the dondriant anchovy egg abundances that occurred during the ichthyoplankton available to entrainment as indicated mid-1980s. For 1995-96, abundances of tautog and by abundances at station EN (Table 3). Most fish anchovy eggs and most of the 14 dominant larval eggs collected at ststion EN were from cunner, 16 Monitoring Studies,1996

TABLE 3. The annual 4-mean* density (nol500 m') of the most abundant fish eggs and larvae collected at EN for each report year from June 1976 through May 1996 (two-unit operational period: 1976-85; three-unit operational period: 1986-96).

l ] axon 76-77 77-78 78-79 79-80 80-81 81-82 82-83 83-84 84-85 85-86 86-87 R7 88 88-89 89-90 90-9191-92 92-93 93-94 94-95 95-96 EGGS

• I adspersus 5,870 8,223 5,171 5,501 7,068 5,719 7,484 2,969 5,002 5,395 6,904 4,998 6,954 4,416 5,436 7,057 7,126 T anars 1,364 2.842 2,647 2,244 2.114 2,157 3,237 2,756 3,011 2.269 2.887 2.060 1,878 1,449 1,596 1,650 2,074 4

Anchoa spp. 1,447 1,245 1,080 765 2,257 4,880 145 910 89 38 54 127 476 107 542 423 153 LARVAE" Anchoa spp. 1,152 931 433 2,168 2,430 5,768 816 1,421 302 1,102 1,244 126 359 619 1,122 799 178 2 33 475 181 4 P. americanus 106 143 114 285 129 233 297 210 180 BL 109 116 203 106 99 388 21 142 224 81 A. americanus 94 318 119 Ill 136 21 27 18 9 3 13 41 31 24 7 18 28 43 63 18 M acnaeus 41 38 36 38 107 72 68 50 68 34 29 95 63 30 24 58 34 48 43 85 B ryrannus 5 4 4 0 3 I 11 23 2 41 3 2 6 72 18 97 41 9 54 66 P. gunnellus 13 13 16 13 58 27 13 14 14 22 4 26 9 6 3 15 8 28 17 41 I adspersus 29 58 1 13 58 78 31 49 4 12 4 5 9 14 68 209 8 10 25 12 T onnis 37 36 1 11 46 83 44 33 3 15 3 7 17 15 33 99 13 6 12 8 E cimberus 2 8 6 8 6 1 6 13 5 8 8 12 45 31 37 98 5 18 9 8 U. subbiyurcata 5 9 14 14 16 17 6 4 60 7 9 23 41 51 34 28 2 18 8 <1 Liparas spp. 27 30 10 16 22 5 13 8 36 1 4 42 18 12 3 23 14 12 5 <1 Sfuscus 4 7 4 9 8 13 7 9 9 5 4 6 7 3 6 4 5 5 3 6 S aquosus 10 II I 5 5 5 2 13 3 1 4 3 5 3 4 12 2 2 3 i P. trracanthus 14 3 1 2 11 17 9 9 1 2 3 0 9 5 29 10 2 2 5 3 i

i O Data seasonally restricted to May 22 July 23 for I adspersus, May 23-August 25 for T onitis, and June 15-August 5 for Anchoa spp.

Data seasonally restricted to July-September for Anchoa spp., March-June for P. americanus, December May for A. americanus, ,

February May for M aenaeus, January-May for P. gunnellus, July December for B. tyrannus, June-August for T adspersus, June August l for T omiss, March-May for Liparts spp., April-September for S fuscus, April-June for U. subbifurcata, April-July for E. cimbrius, l May-October for S aquosus, and June-September for P. triacanthus.

i 1

tautog, and anchovies and the most numerous larvae was discontinued in 1987 (NUSCO 1988c). This were anchovies, winter flounder, American sand monitoring reduction was implemented because '

i lance, grubby, and Atlantic menhaden (Table 1). impingement losses were well documented and most The estimated number of cunner (4.9 billion) and impingement was mitigated by the operation of tautog (2,7 billion) eggs entrained in 1995 fell within retum sluiceways at Units 1 and 3 (NUSCO 1986, the range of previous years for three-unit operation 1988c,1994b). Impingement at Unit 2, which does (Table 4). Entrainment estimates for anchovy eggs not have a return sluiceway, is routinely monitored were much lower in recent years than during a period by plant operational personnel and impingement of greater abundance that occurred during the early counts are only made when a large impingement 1980s. In 1995, only an estimated 16 million were event occurs; none occurred during 1995-96. The entrained. Larval entrainment estimates in 1995 for Unit I sluiceway may be taken out of service during anchovies (186 million) and in 1996 for American periods of high debris loading that would adversely sand lance (23 million) were less than long-term affect plant operations, The number of days that averages (about 579 and 65 million, respectively) Unit I sluiceway was out of service each month from that included the period prior to three-unit operations 1985 through 1995 was determined from plant (Table 5). Entrainment estimates for gmbby have operational records (Appendix VI). During 1995, the been relatively consistent, with the estimate of 43 latest period for which the information was available, million determined for 1996 near the long term the slaiceway was out of service for 54 days, mostly l average of 49 million, Entrainment of winter in May, August, and September.

flounder larvae totaled nearly 54 million and this loss is evaluated in the Winter Flounder Studies section Selection of Potentially Impacted Taxa of this report,

, Additional data analyses were completed for Imp,mgement selected taxa that were identified as potentially impacted, either because of their prevalence in Although impingement of organisms on the intake entrainment samples or because of possible influence traveling screens is a potential direct impact at by the thermal discharge, Taxa potentially vulner-MNPS, impingement monitoring on a regular basis able to entrainment include American sand lance, Fish Ecology 17

_. ._ _ . . _ _ . . _ - - . _ . _.-_. .~~ - - - - - -

i t

i l 4

I i* TABLE 4. Estimated number of cunner, tautos, and anchovy e8gs entrained each year from 1979 through 1995 at MNPS and the volume of cooling water on which the entrainment estimates were based (two-unit operational period: 1979-85; three-unit operational period: 1986-95).

Cunner Tautog . Anchoves Year Volume (m')' No. entrained j No.(x10 entrg)ined (x 10 ) energ)ined Volume No.(x10 (x 10 )

(m')'

(x10')

Volume (m')'

(x 10 )

4 2

~ 1979 1,534 728 705 728 215 711 1980 2,302 806 1,273 8% 91 795 1981 1,736 816 1,735 816 172 799 1982 2,726 853- 1,486 853 234 843  :

, 1983 2,631 798 1,180 g 798 618 786 4 1984 2,031 827 1,369 827 652 812 1985 2,802 831 1,784 831 20 825 1986 2.932 1,870 3,907 1.870 $17 1,846 4

1987 4,533 1,784 3,740 1,784 17 1,752

! 1988 4,366 1,953 2,813 1,953 16 1.920 1989 3,885 1,643 3,094- 1,643 5 1,611 4 r 1990 3,651 1,823 2,185 1,823 28 1,795 1991 4,758 1,265 1,589 1,265 [

147 1,247 -

, 1992 2,754 1.565 1,390 1,565 17 I,537 j i- 1993 5,746 1,748 2,168 1,748 237 1,728 4 1994 5,982 1,726 2,162 I,726 170 1,693 1 1995 4.876 1,633 2,671 I,633 16 1,600

)

• z l Volume was determined from the condenser cooling wster flow at MNPS during the season of occurrence for each taxa. 3 7

' TABLE 5. Estimated number of anchovy, winter flounder, American sand lance and grubby larvac entrained each car from 1976 through 1996 at MNPS and the volume of cooling water on which the entrainment estimates were based (two-unit operationa period: 1976-85; three-  !

unit operational period: 1986-96).

Anchovies Winter Flounder Amer:can sand lance

• Grubby ,

Year No.entrpined Volume (m')* No. entrained Volume (m')* No. entrained Volumej)m')*

No.entrg)ined Volume (m')*

1 (x107 (x 10 ) (x10 (x 10 ) (x10') (x 10 ) (x10') - (x10

/ l y

1976 419 616 108 663 20 839 13 625 i 1977 424 570 31 586 84 983 32 l 653 l 1978 173 657 87 491 190 808 11 446

- 1979 887 552 48 474 154 941 21 534 i 1980 918 505 176 633 124 1,090 34 702

! 1981 1,784 633 48 455 90 713 43 414 .

1982 464 550 170 674 32 1,065 49 629 4

1983 623 482 219 648 di 1,127 57 704 1984 169 602 88 574 20 981 41 643 l 1985 712 601 83 528 10 I,031 37 582

1986 1,328 1,259 131 1.353 5 1,734 56 1,286

1987 124 1,161 172 1,324 48 2.186 $$ 1,370 1988 3% 1,338 193 1,382 126 2.036 124 1.273 1989 546 1,201 174 1,046 55 1,927 72 -1,110 1990 1,025 1.272 139 1,303 61 2.242 49 1,335 1991 478 786 121 934 7 1.330 34 1,024 1992 174 1.018 514 1,199 32 1,672 76 1,132 1993 220 1,098 45 1,412 50 2,261 54 I,374 1994- 536 1,241 182 1,175 77 2,091 58 1,118 1995 186 1,247 223 1,134 114 2,013 61 1,444 1996 . - 54 545 23 1,246 43 723

• Includes data from December of the previous calendar year. )

Volurae was determined from the condenser cooling nier flow at MNPS during the season of occurrence for each taxa. ,

,' No' Lalculated

, because larvae occur after the end of the report period (May 1996).  !

l anchovies, grubby, cunner, tautog, and winter for abundant life stages of these selected taxa.

flounder. The distribution of silversides in Jordan Information on the winter flounder is presented in a l Cove may be affected by the MNPS thermal separate section of this report (see Winter Flounder discharge. Therefore, A-mean densities (no/500m') Studies) and is not included among the fishes for eggs and larvae, and A-mean CPUE for trawl discussed below.

(no/0.69 km) and seine (no/30 m) were calculated J 18 Monitoring Studies,1996

d

, t

i American sandlance i 3@st Suer  !

, remoo remoo  !

. The American sand lance is a schooling fish com- <

mon in estuaries, along the coast, and in inshore wa.

• b l ,.  !

ters from Labrador to Chesapeake Bay (Richards 1982; Nizinski et al.1990). Sand lance have a life l* l l

l t

g200- '

I span of 5 to 9 years, but populations are dominated ,,,_ l j by the first three age groups (Reay 1970). Sexual l l 3 maturation occurs at age-1 or 2 with adults spawning *~ __i.___,,,_____, l 4

'once a year, predominantly between November and so- l l March (Richards 1963,1982; Scott and Scott 1988; {N:l A' ' I Westin et al.1979; Grosslein and Azarovitch 1982). , ddddedder'dddsides'dede/dddd '

Eggs are demersal and adhesive.(Fritzsche 1978; "7" Smiglelski et al.1984). Embryonic and larval devel.

,, Fig. 2. Annual a-mean densities (noJ500 m') of American -

1 opment is lengthy (Smigielski et al.1984).

sand lance larvae at station EN during MNPS two-unit t American sand lance were taken in all three Fish (1976-M and three-unit (1986 %) operational periods. A Ecology programs, although relatively few juveniles A-mean density calculated for the entire two-unit period is

. and . adults were collected by seine or trawl represented by a horizontal solid line that is extended as a (Appendhs I V). Most sand lance were found as dashed line through the three-unit period to serve as a ref- ,

j larvae in winter and spring at station EN. Larval crence level for abundance.

{

abundance peaked in the late 1970s and early 1980s, E the median value of the last 15 years. A significant followed by a rapid decline during the mid 1980s

! (Table 6; Fig. 2). Because sand lance larvae were so decline (p = 0.022; Mann-Kendall test) occurred i

! abundant from' 1976-77 through 1980-81, larval during the two-unit period (slope -19.3); a similar densities in entrainment samples during the three unit negative trend was not apparent during the three-unit period to date have been lower than the two-unit av. Period. The large change in abundance was reflected erage. The A-mean density oflarvae at EN during in a wide range (5 to 190 million) in annual entrain-1995-96 was 18, less than one-third of the abundance ment estimates, which were also dependent upon index of the previous year. However, following the cooling water flow during the larval season (Table large decline in 1981-82, this A-mean isjust below 5). Plant cooling-water flow in 1996 during the pe- i riod of occurrence oflarval sand lance was the lowest TABLE 6. The annual A-mean' density (no/500 m') and 95% of the three-unit period and, coupled with relatively conndence interval of American sand lance larvac conected at EN low abundance, resulted in an entrainment estimate from June 1976 through May 1996.

g g year EN Declines in sand lance abundance during the 1980s 1976-77 94*17 were also apparent in other areas of the Northwest i 1977 78 31st 117 Atlantic Ocean. Larval densities in LIS over a 32- l lN 19so-si

!!N!6 136

• 32 year period (1951-83) were highest in 1%5-66 and {

1978 79; the latter peak was also evident throughout l Ujj NN the entire range of American sand lance (Monteleone 1983-s4 Is

• 4 et al.1987). This high abundance persisted through.

l fjj y out the Northwest Atlantic until 1981 and the decline

{

19s6-s7 13

• 4 that followed appeared to be inversely correlated '

l$yj8; jj *lj with that of Atlantic herring and Atlantic mackerel 1989 90 24* 7 (Nizinski et al.1990). Sand lance likely increased in j,N, ,j 3,7

• 2 abundance _ due to a decrease in their' predators, 1992-93 2s e10 herring and mackerel, the abundance of which had y

l99 '4 jj*y been reduced by overfishing during the 1970s 1995-96 is*7 (Sherman et al.1981). In more recent years, Atlantic mackerel, which prey heavily upon sand lance

• Data scasonaHy restricted to December-May. (Monteleone et al. 1987), have become more Fish Ecology 19 l

I

abundant as sand lance abundance decreased. Given adults are important forage for many recreationally  !

the large abundance changes of American sand lance and commercially important fishes (Vouglitois et al.

along the Atlantic coast, effects of MNPS operation 1987) and also have high mortality rates (Newberger on this species are difficult to ascertain, but are likely and Houde 1995).

small in comparison to the large-scale natural Anchovies were collected in all three programs, but fluctuations typically associated with this fish. rarely by seine and only sporadically by trawl (Appendices 1-V). Juvenile anchovies resulting from Anchovies the summer spawn were typically captured by trawl frogt August through October. Most anchovies were The bay anchovy is one of the most abundant collected in only I or 2 years of sampling at the three fishes found along the Atlantic Coast (McHugh inshore trawl stations, including 1978 79 (23% of 1967) and usually the dominant summer ichthy- total catch at the station) and 1985-86 (52%) for IN, oplankton species found within its range (Leak and 1991-92 (42 %) for JC, and 1989-90 (62%) for NR Houde 1987). This species ranges from Mexico to (Appendices Ill-V).

Cape Cod and occasionally into the Gulf of Maine Anchovies dommated larval collections and their (Hildebrand 1943; Bigelow and Schroeder 1953; eggs ranked third in abundance (Table 1). Annual Grosslein and Azarovitch 1982). Bay anchovy are egg and larval abundances were correlated common in nearshore and estuarine waters during (SPearman's rank-order correlation coefficient r -

warmer months, but move offshore in winter 0.55; p = 0.022). The A-mean densities in 1995 for (Vouglitois et al.1987). Chesapeake Bay bay anchovy eggs (153) and larvae (181) fell within anchovy were found to have little genetic variation, historical ranges, but were much lower than found in indicating a lack of stock structure, likely due to the early to mid 1980s (Table 7). Large annual enormous population size and considerable changes in bay anchovy egg abundance were also ,

movements and mixing (Morgan et al.1995). bserved in LIS during 1952-35 (Richards 1959) and Although the striped anchovy also occurs from Nova in Bamegat Bay, NJ during 1976-81 (Vouglitois et ,

Scotia to Uruguay, its occurrence north of Chesa- al.1987). All egg and larval densities during three- '

peake Bay is variable and the striped anchovy is unit operational period were below the two-unit usually found further offshore than the bay anchovy average because of a decline that occurred in the (Hoese and Moore 1977; Smith 1985). The eggs of mid-1980s, prior to three-unit operation (Fig. 3).

the two species can be readily_ distinguished and TABLE 7. The annual A-mean' density (no1500 m') and 95% '

since 1979, when eggs were first identified to confidence interval of anchovy eggs and larvae collected at EN species, about 96% of the eggs collected at station from June 1976 through May 1996.

EN were those of the bay anchovy. Therefore, most EGGS LARVAE of the anchovies collected in the Fish Ecology Year W i programs were likely bay anchovy, even if only 1976 1.152

• 419  !

identified to genus. 1977 931

• 40s i The bay anchovy can mature at 2.5 to 3 months in l9N 2,NNN!

1.447

• 336 age and individuals spawn repeatedly during the 1980 1,245

• 597 2.430

• 1.249 summer (Luo and Musick 1991). In LIS, spawning l$j 'isN2$s 2 $'s No takes place at depths of 20 m or less from May 1983 2.257

• i,076 t,421* s30 through September, with a peak during June and July ll8j 98N 3'680 3 ,, ej6j j (Wheatland 1956; Richards 1959).' Spawning 1986 91o

• 547 1,244

• 893 appears to be correlated with high zooplankton abundances (Castro and Cowen 1991; Peebles et al.

ll8, jyjj j2 1989 54

• 47 619

• 416 1996) and warm water temperatures (Zastrow et al.

1991). Eggs are pelagic and at 27*C hatch in about Q

1992 jy*j 107* 112 l@*@

17s

• so 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> (Kuntz 1914). Mortality rates of eggs and ly jjj*jy jj*$

larvae are relatively high (Leak and Houde 1987; 19 % 153

• 295 isi*117 Houde et al.1994; Dorsey et al.1996), particularly from predation by etenophores and jellyfish (Govoni

• Data seasonally restricted to June 15-August 5 for eggs and and Olney 1991; Purcell et al.1994). Juveniles and July-september forlarvae.

20 Monitoring Studies,1996 1

i i

l l

i l

i

,8,gi, ,Mo should be readily evident if the production of eggs I e

• toos , and larvae declines because of decreased adult l abundance. Neither the number of eggs nor larvae

, entrained at MNPS were significantly correlated with he. l the densities of eggs or larvae found the following g,

4 year, implying no direct effect on the anchovy

\l spawning stock. Although both egg and larval j 2e i l abundances declined in the mid-1980s, these  !

deqeases occurred prior to three-unit operation in o , ,,,,,,....

J Q ....,,..

1986. In addition, the lack of negative trends during three-unit operational period suggests that MNPS has i

re 7e e e2 e4 a a e e2 o' ,

had minimal or no impact on anchovy adult j abundance. '

• wvn ,

\

Silversides je.  :

[ l The Atlantic and inland silversides are sympatric E 4mo- l along most of the east coast of North America and h i l reside in bays, estuaries, and salt marshes. The 8m ( k l Atlantic silverside ranges from the Gulf of St.

y Lawrence to northern Florida (Conover 1992) and

\ Lm the inland silverside from Cape Cod to South

, /s ' /e ' k ' e'2 'k'k'k'k'k'L' Carolina (Johnson 1975). Both species are abundant,

" l but in general the Atlantic silverside is more i Fig. 3. Annual A-mean densities (noJ500 m') of anchovy numerous than the inland silverside, except in low i eggs and larvae at station EN during MNPS two-unit salinity waters (Bengtson 1984, 1985). Most l (1976-85) and three-unit (1986-96) operational periods. A silversides mature as yearlings and only live I to 2 4-mean density calculated for the entire two-unit period is years. Spawning begins at water temperatures of 9-represented by a horizontal solid line that is extended as a 12*C and occurs during the day at high tide on a dashed line through the three-unit period to serve as a ref- semilunar cycle (Middaugh 1981; Conoser and Ross l crence level for abundance.

1982; Jessop 1983; Conover and Kynard 1984).

Adhesive eggs are laid in shallow water on j Since then, A-mean densities decreased by an order vegetation (Conover and Kynard 1984). Larvae are '

of magnitude and have remained relatively low, planktonic, but remain near spawning areas. Sex is except for larvae in 1990. For both eggs and larvae, indeterminate until fish reach 8 to 21 mm in length no significant trends were detected with the Mann- ,

and sex ratio is affected by prevailing water l Kendall test during either the two- or three-unit temperature during development (Conover and operational periods. {

Kynard 1981; Conover and Fleisher 1986). Growth '

He entrainment of eggs and larvae by MNPS of young is fast and mean lengths can exceed 90 mm probably represents the greatest direct operational by November, with fish from early spawns larger at

. impact on anchovies that spawn in the Millstone any given age than late-spawned fish (Conover 1979; area. In 1995, the entrainment estimates for eggs (16 Bengtson et al.1987). Atlantic silverside migrate million) and larvae (186 million) veere within historic offshore during winter, but remain in waters within ranges, but were among the lowest estimates of the 40 km of shore and in depths of less than 50 m past 20 years (Tables 4 and 5). Anchovies mature (Conover and Murawski 1982). In contrast, inland within a year and have a maximum life span of not silverside have small home ranges (Hoff 1972).

more than 2 or 3 years (Stevenson 1958; Newberger High (>99%) overwintering mortality typically and Houde 1995). Estimates of the abundance of occurs (Conover and Ross 1982; Conover 1992).

(ggs collected at station EN are representative of Both species are important trophic links between annual egg production. Derefore, a reduction in the zooplankton and larger fishes and birds.

adult spawning stock size due to plant operation Fish Ecology 21

in the trawl monitoring program nearly all silver- 474 sides collected have been the Atlantic silverside, with 474 12o -

less than 0.1% identified as the inland silverside during the past 20 years. Since 1981, when the two f

l l species were first differentiated in the MNPS sampling, more than 80% of the silversides collected {"_-

g so ll lj 4

i' c by seine were Atlantic silverside, although relative proportions varied from year to year. Silversides g "-

3 dominated symmer shore zone seine catches at JC g- \

(Appendix II) and were taken by trawl in winter ,

lf--------\---

l ,.- \

(Appendices , lil V), but were rarely found in

• rd7e 7s' 4 o'ai42 ede4 edes'e7'4e'ee'ao'es'-e2 ode 4 ele 4

ichthyoplankton collections. Potential impingement i

effects have been mitigated by the installation of *~

• return sluiceways. Therefore, any effect of the 100 -

l operation of MNPS is likely related to the influence of the thermal plume in Jordan Cove, which serves as {.

e l a locally important spawning and nursery area. E "-

.' l Atlantic silverside were collected by trawl @ 4o. l primarily at the three inshore stations (IN, JC, NR) 5 ,, ,  ;

N  ;

currently sampled (Fig.1). Most were taken from t ,

October through February after individuals withdrew a ,, . 7, T, C . , , j,,bi from shoreline areas to overwinter in deeper waters. 77 7e tu sia: e244 e er4s w4 ew2 e344 eus

• ~

Annual trawl abundances fluctuated considerably #

(Table 8), with the A-mean CPUE for the past few years below the. two-unit average at each of the T

• g ,, _ l stations (Fig. 4). Although recent trawl catches have e l

been low, the only significant trend (slope of -3.2) E l

was found for station NR (p = 0.025; Mann-Kendall 6 to-l i

test). During the two-unit operational period, a 5 A :A ,

significantly (p = 0.031) negatis e slope of-2.0 was *

/

l TABLE 8. The annual A-mean' CPUE (noJ0.69 km) and 95% 77.re 7o4o ei42 e244 esas eras ee-so sas.e2 es44 es-se confidence interval of Atlantic silverside collected by trawl at RENT YEAR selected stations imm June 1976 throu8h May 1996.

Year IN JC NR Fig. 4. Annual A mean CPUE (no/0.69 km) of Atlantic silverside taken by trawl at stations NR, IN, and JC during 1976-77 15

• 16 13

• 20 - 77*283 MNPS two-unit (1976-85) and three-unit (1986 %) opera-1977 78 29

• 92 6

• 612 - 10

• 21 tional periods. A 4-mean CPUE calculated for the entire 1 - 4

• f7 two-unit period is represented by a horizontal solid line

! 1980-81 8

• 17 4*5 3*4 that is extended as a dashed line through the three-unit 1981-82 6*9 Is0 6*8 period to serve as a reference level for abundance. (Note j*2 *

, that the vertical scales differ among the graphs).

I 1984-85 2*6 5*!! l*1 1985-86 7*8 6*8 3*6 1986-87 5*3 8*7 110

• 222 found for IN, probably a result of the high initial 1987 88 3*5 2*2 15

• 27 abundance found at that station in 1976-77.

1988-89 2*1 1*0 25

• l4

'Jhe A-mean CPUE of Atlantic and inland l'9 1991 92 lo 12

• 10 2*1

$ l $$

19

• 7 silversides taken by seine during 1995 (87 and 60, respectively) were within historic ranges (Table 9).

l993 .

2 l In 1995, the abt:ndance of Atlantic silverside 1994-95 1*l 4*3 2*2 1995-96 2 *l 2*2 exceeded that of the inland silverside (Fig. 5).

1*l Nearly all annual A-mean CPUE values for inland silverside during three-unit operation were above or

• Data seasonally restricted to November-February at IN and NR

and October-January at JC. near the twc~ unit average, but abundance of Atlantic i

22 Monitoring Studies,1996

, - , . . - - , - .- , - - . - ~ , - n ,- - .+ e , v ,

.. . - - ~ - . ~ - - . . . . - . - .. .- - .-_ - . - . . . - -

4 1

TABLE 9. The annual 6-mean* CPUE (no/30 m) and 95% 2.uut mai

,- confidence interval of Atlantic silverside and inland silverside M40D M4oD j collected by seine at JC from June 1981 through May 1996. 400-s LvensIDE ,

! Year Ar!.ntic silvmide Inland silvmide l 3an_ f 1981 152*251 3*3 .  !

i 1982 il4* l62 6*l6 g

• l 1

i 1983 1984 1985 397

• 598 29

• 24 19*12 88*243

.t .: 2 4*8 h*

5

/

l

/.' gN

1986 172*385 14*21 Igo- - - - - - - - - - - - -

1987 109

• 90 3*2 k /e w' V/

1988 96

• 108 27

• 54 1989 70

• 93 14

• 16 LJ

o A '

7 1990 83

• B0 133

• 234 '

4 f, ' f3 ,, f7 f, ' f, ' f3 $5 1991' ,8*11 74

• 37 l 1992 78655 43*27 1so- ,m j 1993 60

• 73 5*5

. slLvenssDe  !

l994 37

• 28 63

• 64 ' 6 -

j 1995 87

• 73 60

• 114 l b '

0 f ,

1 Data seasonally restricted to June-November. E l g

I silverside has mostly been below the two-unit f3,_ f average (Fig. 6). These abundance trends for both l i j species were supported by the results from a Mann-i l A  !

Kendall test, where the only significant trend (p = ,

- - , ~ -

l 4 '

0.025; slope of-18.6) was for the Atlantic silverside s'i ' e's $s ' $7 ' de ' di ' e3 ' es

} during three-unit operation. YE^"

To determine if a change in size distributions

• Fig. 6. Annual A-mean CPUE (no/30 m) of Atlantic sil-occurred after Unit 3 became operational, the length- verside and inland silverside taken by seine at station JC J frequencies for seine catches (expressed as percent- during MNPS two-unit (1976-85) and three-unit (1986-95) j ages) were examined for the periods before and after operational periods. A A-mean CPUE calculated for the j three-unit operation (Fig. 7). The length-frequency entire two-unit period is represented by a horizontal solid distribution remained similar during these two line that is extended as a dashed line through the three-unit i operational periods, suggesting that no changes in period to serve as a reference level for abundance. (Note l growth resulted from increased thermal discharge that the vertical scales differ between the graphs).

! into Jordan Cove. .

The primary potential impact of the operation of plume mto shore-zone area, causing potential l

MNPS on silversides is the meurs,on i of the thermal disruption in spawning activity and in distribution of l

juveniles and adults. Annual abundance of inland silverside collected by seine has fluctuated over the l ,oo years, but without apparent relation to operational ti li _ _ _

fM periods. Atlantic silverside abundance in the shore-zone area showed a decreasing trend during mree-

{* l$ S oc unit operation. This could be related to MNPS

]

g am- lj operations, although both the predicted and measured f

g , i (based on dye studies) maximum thermal increase is g s '

i A only 0.8'C at the site of station JC (NUSCO 1988b).

ino- $ / ' s, , Fucher, elevated summer temperatures in Jordan N p, - Cove appear to be more directly related to solar 6J i

9

/ heating of the shallow sand flats (such as at the JC

, Ei E2 ds 54 $sdeirdedeEodie2da54 Es seine site) than to the MNPS thermal plume (see the

'E^"

Eelgrass section of this report). Therefore, it appears Fig. 5. Annual 4-mean CPUE (no/30 m) of Atlantic unlikely that a small increase in water temperature silverside (dashed line) and inland silverside (solid line) could affect the reproductive success of a species that taken by seine at station JC from 1981 through 1995. ranges as far south as northern Florida. Also, Fish Ecology 23

ATLAny1C slLVER$ log To- Grubby so-g 8'" a m ,,,,,,,, The grubby is a demersal fish found in shallow waters along the Atlantic coast from the Gulf of St.

4o- 7 m**= Lawrence to New Jersey that tolerates a wide range y 3o-  : _ of temperature and salinity (Bigelow and Schroeder g,

y _

1953). ladividuals reside in protected shallow water f  : t- g on mud or sand bottoms, peat reefs, and in ec! grass

'0-  ?

, 1 g beds (Ennis 1969; Lazzari et al.1989) and are found o , , , , ,M, throughout the year in waters of LIS near MNPS.

< 4o no4o eseo esioo m" ""

Similar to the winter flounder, grubby produce blood m.Am savERsioE Pl asma antifreeze proteins and can remain active in 70- very low water temperatures (Reisman et a!.1987).

,o_ Grubby spawn throughout the winter and have a g demersal, adhesive egg with an incubation time of 40 g*~ g ,, to 44 days at a water temperature of 4.6-6*C (Lund g e- '

and Marc) 1975; Lazzari et al.1989). Richards 3o. , (1959) reported larvae present in L!S from February

,_ j through April and Laroche (1982) noted that they are l more abundant near the bottom than at the surface.

'0- i i A small species, the grubby has no sport or o ,

,nN, , , , commercial value and, given its protective spines and

.e 4aeo sano esim em ""

cryptic coloration, it probably also has limited forage value. The grubby preys upon many small fishes and benthic invertebrates (Lazzari et al.1989; Levin TRAWL CATCH oF SILVERStoES 199 l),

I-The grubby is the fourth-most abundant larval fish 80 -

collected at EN, accounting for 4.8% of the total J l h so- e from June 1976 through May 1996 (Table 1). The l

8, , . l A-mean density of larvae for 1996 of 85 was within l

30 -

g l the range of historic data, but was the largest l

i

.. observed since 1988 (Table 10). Three-unit opera-20 - '; ( l l tional annual A-mean densities of larval grubby

]

,o- . ) fluctuated about the two-unit average (Fig. 8). No

_ _ , _ _ n-g significant temporal trend for larval grubby abun-

< 4o ' 404o ' co.so ' so.1oo ' 100120' 12o ' dance was detected during either two- or three-unit LENoTH operation when examined with the Mann-Kendall test. Despite a relatively high density in 1996, the i Fig."1. Length-frequency distribution (20-mm length in-entrainment estimate of 43 million larvae was less tervals) of Atlantic silverside and inland silverside taken by than the average of 49 million for these annual seine at station JC and Atlantic silverside taken by trawl at stations NR, IN, and JC during MNPS two-unit (1976-85) estimates because of the smallest cooling-water flow and three-unit (1986-95) operational periods and the 1995- that occurred during the larval grubby season for any 96 report year (June-May). year of the three-unit period (Table 5).

l Predominantly a shallow-water fish, the grubby abundance of adult silversides co!!ceted by the trawl was the fifth-most abundant fish taken by trawl at the monitoring program (an index of future spawners) th ce inshore stations; about 60% were caught at NR did not show a similar pattern of abundance, (Appendices ill-V). In 1995-96, catches at all three suggesting that plant operation has not affected stations fell below tSe two-unit A-mean average Atlantic silverside spawning population size near CPUE (Table 11; Fig. 9). The only significant trend MNPS. found using the Mann-Kendall test was at station NR l

24 Monitoring Studies,1996

1 TABLE 10. The annual A-mean* density (no1500 m3) and 95% TABLE 11. The annual A mean' CPUE (no10.69 km) and 95%

confidence interval of grubby larvae collected at EN from June confidence interval of grubby collected oy trawl at selected 1976 through May 1996. stations from June 1976 through May 1996.

~

Year NR JC IN vear r'N I976-77 0.9

• 0.3 0.6

• 0.2 0.6

• 0.I 1977 41

• 9 1977-78 0.5

• 0.1 2.2

• 0.5 1.1

• 0.2 1978 38

• 9 I978-79 1.2

• 0.2 2.0

• 0.6 0.7

• 0.2 1979 36

• 7 1979-80 3.3

• 0.9 0.7

• 0.1 0.9

• 0.2 1980 38

• 7 38*1.1 1981 107

• 27 1980-81 1.1

• 0.2 2.1*06 I981-82 7.5

• 2.5 1.0

• 0.2 2.3*06 1982 72* 13 198k33 l 1.7

• 2.7 1.4

• 0.2 2.2

• 0.5 1983 68

• 19 1983-84 4.1

• 0.8 1.7

• 0.3 1.7

• 0.3 1984 50*I5 1 M 4-85 5.9

• 1.2 1.6

• 0.3 09*0.2 1985 68

• 23 N1-86 2.3

• 0.5 1.4

• 0.3 1986 0.7

• 0. I 34

• 10 t W6-87 7.2

• 2.3 1.1

• 0.2 09*0.2 1987 29

• 7 1987-88 3.7

• 1.2 1.2

• 0.2 1.1

• 0.2 1988 95

• 35 1988-89 10.5

• 2.3 1.0

• 0.1 1.4

• 0.3 J 1989 63

• 18 1989-90 3.6

• 2.0

> 0.4

• 0.1 1.0

• 0.3 1990 30* 8 1990-91 8.0

• 2.0 0.4

• 0.1 0.8

• 0.2 1991 24

• 6 1991-92 3.4

• 0.5 0.5

• 0.1 1.0

• 0.2 1992 58

• I7 1992-93 6.2

• 2.0 1.4

• 0.3 1.9

• 0.3 1993 34

• 9 1993-94 2.2

• 3.0 0.7

• 0.5 1.9

• 3.8 1994 48

• 16 1994-93 3.7

• I.6 1995 2.9

• 1.1 1.6*06 43* 15 1995-96 1.9

• 1.0 0.7

• 0.3 0.7

• 0.2 1996 85

• 37

' Data seasonally restricted to December June at IN. but year-Data seasonally restricted to February-May. round (June May) at JC and NR.

(p = 0.007) during two-unit operation, which was vicinity of the MNPS intakes and subsequent increasing with a slope of 0.78. hatching may directly affect the number of larvae The percent length-frequency distributions of collected at station EN. Entrainment oflarvae is the grubby taken by trawl were similar before and after primary direct plant impact on the resident grubby three-unit operation, although in 1995-96 more population. However, abundance of grubby has not smaller fish were taken in comparison to the declined during the three-unit operational period and composite of either of the two operational periods this species has been among the most stable of the (Fig.10). fishes residing near MNPS.

Because grubby eggs are adhesive, spawning in the Cunner

2. UNIT 3. UNIT pea o PEm o tas- The cunner, found from Newfoundland to Chesa-peake Bay (Scott and Scott 1988) is closely

,- soo- f l associated with structural habitats, such as rocks, fg / l pilings, celgrass or mussel beds, and macroalgae.

7s-

[\ \

A' l Cunner are inactive at night and when water temperatures fall below 5-8'C, they become to/pid i5- 'J \l l

/ (Green and Farwell 1971; Olla et al.1975; Dew 1976). Individuals maintain highly localized home l

ranges (Green 1975; Olla et al.1975; Gleason and a ' ' ' Recksiek 1988), may establish defended territories h ' h ' Ei ' 53 es 57' 5e ' di 53 Es ' (Poltle and Green 1979b), and most, but not all YEAR individuals, do not undertake extensive movements Fig. 8. Annual 4-mean densities (nol500 m') of grubby (Green and Farwell 1971; Olla et al.1979; Lawton et larvae at station EN during MNPS two-unit (1976-85) and al.1996). Most cunner live only 5 to 6 years, with three-unit (1986-96) operational periods. A A-mean den- maximum age likely about age-10, less than one-sity calculated for the entire two-unit period is represented third of the life span of the closely related tautog by a horizontal solid line that is extended as a dashed line (Dew 1976; Regan 1982),

through the three-unit period to serve as a reference level for abundance.

Fish Ecology 25

2 Ur#7 3-UM T PEmoD 40-

, PEmoo 12 - wg s c) 3.unn es ine

\  ;

10- j ,

D30-. O ** s '=e I e

,, _ E SHe

, i a 20- r

~

6 ._ h I\-r \. ,h r l e10-

i.

u. , _

2-

~ g g i 0- ,,,,,,,,',i,,,,,,,,i ,

O 60-70 7140

, , , , i el e0 e1 100 101 110 *110 77-7e 7e40 e142 e344 e54f e748 e9-e0 e142 e3 e4 e5-e6 LENGTH 4

Fig. 10. Length-frequency distribution (10-inm length

,,e  ;

intervals) of grubby taken by trawl at stations NR, IN, and JC during MNPS two-unit (1976-85) and three-unit (1986-y , ,, l

94) operational period.s and the 1995-96 report year (June-g l e 6 May).

a h

E8~ <

! (Williams 1967; Dew 1976). Williams et al. (1973) h 9 ----- --.i-..

((}W *

^

/ \) \

noted that only about 5% of cunner eggs survive to hatching. Newly-hatched larvac are 2 to 3 mm in length, metamorphose by 10 mm, and settle into o ,,,,,..i

, ,,,,ii,,,

n.7e 7e40 si42 em es as e7-ae es-e0 ei42 em em preferred habitats (Miller 1958; Levin 1991).

The cunner has no commercial value and is generally not sought n'ter by sport fisherman, although numerous individuals are caught while fishing for other species (MacLeod 1995). Region-l

e ..  ;

ally, declining trends in adult cunner abundance have been observed in LIS (Smith et al.1989), Cape Cod E 2- Bay (Lawton et al.1994), and Mount Hope Bay k / '

(MRI 1994).

5 in the MNPS area, cunner eggs and larvae are pres-1-

(/ V

\ ent primarily from June through August. In these O '

early developmental stages collected at station EN, i

,,,,,,i,,,,,,,,,,i n.7e re40 e142 e s e74e as-e0 si.e2 em es so cunner eggs have been most abundant of all egg taxa, whereas larvae have been les; common, ranking only Fig. 9. seventh overall (Table 1). Following a relatively Araual A-mean CPUE (no/0 69 km) of grubby large A-mean density of 7,057 for eggs in 1994, the taken by trawl at stations NR, IN, and JC during MNPS two-unit (1976-85) and three-unit (1986-96) operational A.mean density of 7,126 for 1995 was the third larg-periods. A A-mean CPUE calculated for the entire tw* est recorded (Table 12), as was the entrainment esti-unit period is represented by a horizontal solid line that is mate of about 4.9 billion for 1995 (Table 4). During extended as a dashed line through the three-umt period to three unit operation, eggs have increased in abun-serve as a reference level for abundance. (Note that the dance from a low m. 1986 and the A-mean m. 1995 vertical scales differ among the graphs), exceeded the two-unit average (Fig.1I). However, the 1995 larval A-mean density decreased about 50%

Cunner mature at age-1 to 2 and spawn during May from the previous year (Table 12). Except for a peak through September from afternoon into the evening in 1991, annual larval abundances during three-unit (Johansen 1925; Dew 1976; Pottle and Green 1979a; Operation remained below the two-m it average (Fig.

Green et al.1985). Lawton et al. (1996) reported all 11). For larvae, no significant trends were detected cunner larger than 65 mm observed in western Cape using the Mann-Kendall test durin<,; either two- or Cod Bay to be mature. The pelagic eggs hatch in 2 three-unit operation. Cunner egg asundance varied to 6 days, depending upon water temperature without trend during the two-unit period, but 26 Monitoring Studies,1996 i

1 TABLE 12. The annual a-mean' density (no1500 m') and 95% zwt swt confidence interval of cunner eggs and larvac collected at EN PEmoD PEmoD from June 1976 through May 1996. EGGS ,

Year EGGS LARVAE e eo w l FN FN lq, ,

fl , , ,

1976 29

• 14 1977 58

• 28 { 6000-i

\j y , j g j 1978 1979 1*0 5,870

• IJ01

,oon.

l/ V l 1980 13

• 5 g 8.223

• 1.645 58* 19

• 1981 5,171

• 882 78

• 36 ,'

4000-1982 5,501

• I,377 31

• 14 s 1983 7,068

• 2,679 49

• 26 1984 5,719

• l.246 4*2 0 i iie iiiie i iiiiiiiii 1985 7.484

• 2,659 12

• l0 rs re so e2 ed as as so e2 e4 1986 2,969

• 1,082 5*1 1987 5,002

• 1,644 5*3 1988 5.395

• 1,756 l 9*4 300-1989 6,904

• 3,077 14

• 12 m '

1990 4.998

• 2.250 68

• 6l 1991 6,954

• 3.228 250-209

• 157 a 1992 4,416

• 2.238 8*4

'f 1993 5,436

• 2,364 10*6 200- ,

1994 7,057

• 3,315 l j 25

• I8 $ a 1995 7,126

• 4,307 12

• 9 150- i

[ 100- i

/

• Data seasonally restricted to May 22-July 23 for eggs and June- 0 A

August for larvac. l /

so- A [\, i j

showed a sigmficant (p = 0.025) positive (slope =

0

\ / \ j ~ ~7 ~ ~t_ ~#' '

i

,iiiiiiiiriii,,iiii 277.0) trend during three-unit operation, possibly " w s2 e4 caused by the lowest density of the data series that

d as i

was recorded in 1986 at the beginning of the three. Fig.11. Annual 6-mean densities (no/500 m') of cunner  !

unit period. eggs and larvae at station EN during MNPS two-unit l

Juvemles and adult cunner are caught by trawl, (1976-85) and three-unit (1986-96) operational periods. A l mostly from spring through summer. Twice as many A-mean density calculated f r the entire two-unit period is j represented by a horizontal solid line that is extended as a cunner were taken at IN (4.126) than at JC (2,063), i dashed line through the three-unit period to serve as a ref-with relatively few (416) found at NR (Appendices Ill-V). Relatively high A-mean CPUE values were crence level for abundance. (Note that the vertical scales differ between the graphs).

recorded for IN from 1976 through 1981, followed by moderate abundance from 1982 through 1984, was very similar to the overall three-unit average size j with annual A-mean CPUE becoming more similar at composition, IN and JC (Table 13). Annual A-mean CPUE at both Although the abundance of trawl caught cunner in stations during three-unit operation remained l the MNPS area decreased during the 1980s, '

considerably below the two-unit reference level (Fig. particularly at IN, most of the decline occurred prior 12). The only significant decreasing trend (p = to three-unit operation. Contributing to the decrease 0.002; Mann-Kendall test) occurred at station IN, at IN was the mid-1983 removal of a cofferdam that with a slope of-3.0 found during two-unit operation. was in place during the construction of the Unit 3 To determine an age-frequency distribution of intake structure. This rock cofferdam provided good l cunner collected by trawl, ages were assigned based habitat for cunner and increased their availability to {

on an age-length key provided by Serchuk (1972). sampling by trawl at the nearby IN station. After its

' Percent length-frequency distributions were deter- removal, CPUE at IN decreased considerably and mined for both the two- and three-unit periods and became more similar in magnitude to that of former i for 1995-96 (Fig.13). The size distributions differed trawl station NB (Fig.14), which was located only greatly between the two operational periods. Nearly about 500 m to the west in mid-Niantic Bay. The 70% of the cunner caught during three-unit operation annual A-mean CPUEs at both stations from 1983 l wire young-of-the-year, but relatively high through 1994 were significantly correlated I frequencies of older fish were taken prior to 1986. (Spearman's rank-order co relation coefficient r = '

The length-frequency distribution during 1995-96 0.78; p = 0.0042).

Fish Ecology 27

._____e .-._._..__.~___..-______.___-.m .. _ .. _ . _ _ _ _ .-.

l  !

I i'

TABLE 13. The annual a-mean* CPUE (no10.69 km) and 95% s.tssT s.taer -!

confidence interval of cunner collected by trawl at selected - Pa#oD . Paeoo stations from June 1976 through May 1996. #~ W

~

Year IN JC 20-1976 I977 26.0

• 19.0 24.0

• 23.0 4.0

• 2.0 f ,

3.0

• 1.0 g

• I978 6.0

• 3.7 3.0

• 1.4 e,16-1979 29.0

• 23.0 9.0

• 5.0 h -r---------.  !

1980 1981 23.0

• 16.0 12.0

• 10.0 6.0

• 2.0 5.0

• 2.2 5 V

\

1982 5.0

• 3.0 4.0

• 2.0 g 5-3.0* 13 i 1983 4.0

• 2.0 N

• 1984 2.0

• I .0 - 2.0

• l.0 0 . ........ .., . .i, 1985 1.0

• 0.6 1.0

• 0.5 77 7e ai s3 as s7 es e1 e3 e5 )

1986 0.I

• 0.2 0.5

• 0.4 1987 0.2

• 0.2 0.4

• 0.2 (

e 1988 03 *0.1 3.0

• 3.4 i 1989 0.9

• 0.4 0.8

• 0.4 1990 0.4

• 0.1 0.9

• 0.2 to- x l ,  !

1991 0.4

• 0.1 23

• 0.7-

• a '

1992 1.0

• 0.7 1.4

• 0.5 e-  ;

1993 0.1 e 1.1 1.4

• 0.7  ;

! 1994 0.4

• 0.1 0.8

• 0.5 e- '

1995 0.8

• 0.4 13 *l.1 l S, l. [

l

\

b s- \ ,

5

• Data seasonally restricted to May-August at IN and May- b l I September at JC. 2-The entrainment of eggs is the greatest potential o , , ,.,,...f,,....,,.

impact of MNPS on the cunner population. ** '7 " '5 '

( However, if egg losses affected recruitment, then L l juvenile abundance should decrease in relation to Fig.12. Annual A-mean CPUE (no./0.69 km) of cunner older fish. This decrease was not apparent in the taken by trawl at stations IN and JC during MNPS two-unit comparison oflength-frequency distributions, which, (1976-85) and three-unit (1986 96) operational periods. A conversely, indicated a relative increase ofjuveniles A mean CPUE calculated for the entire two unit period is represented by a horizontal solid line that is extended as a

occurring during three-un.it operation. The greatest dashed line through the three-unit period to serve as a ref-l effect of MNPS on cunner may have been the loss of habitat formerly provided by the Unit 3 cofferdam. erence level for abundance. (Note that the vertical scales difTer among the graphs).

Tautog 80-General biology. The tautog ranges from New ro. a w ,,,,,,

Brunswick to South Carolina, but is most common from Cape Cod to the Delaware Capes (Cooper D se 5'

i 1965). Adult tautog prefer rocky areas and similar y reef-like habitats near shore from spring through fall; 4- a em l juveniles are also found in eelgrass beds and among "- '

I

~~

macroalgae in coves and estuaries (Tracy 1910; 20- " l l Bigelow and Schroeder 1953; Wheatland 1956; io-  :  : ,

-,L l Cooper 1965; Briggs and O' Cone.er 1971; Hostetter o , , inE, ,rt-_ ,

and Munroe 1993). Tautog are active during the gTH 70 71-110 111 140 14 170 17 200 m l day, but are quiescent during night (Olla et al.1974). -

! During winter, adults move to deeper (25-55 m)

Fig.13. Length-frequency distribution by length (mm) and l water whilejuvemies remain inshore to overwinter in age (determined from age-length key of Serchuk 1972) of l a torpid state (Cooper 1965; Olla et al.1974). cunner taken by trawl at stations IN and JC during MNPS Tautog are long lived with maximum age reported two-unit (1976-85) and three-unit (1986-94) operational '

for males of 34 years and 22 years for females periods and the 1995-96 report year (June-May).

(Chenoweth 1%3; Cooper 1965). Adult growth 28 Monitoring Studies,1996 l

1 i

i

g previous 16-year period, but was the highest 30 ] g abundance since 1990 (Table 14). Since the early

^ 25 f j; \ \ U "'3* * **

NB 1990s, annual abundances have remained below the two-unit operational average (Fig.15). During three-R 3 unit operation a significant (p = 0.040; Mann-E 15j f Kendall test) decreasing (slope -138.3) trend was 5 105 V detected for t utog eggs, despite increases in p i y '

abundance occurring during recent years. No trend

'3 5j .

s _

watpresent during the two-unit operational period.

0 , , ,

? In contrast to eggs, tautog larvae were not a 76 78 80 82 84 86 88 90 92 94 predominant laXon, ranking eighth since 1976 (Table

" I). Larval abundance appeared to rise and fall Fig.14. Comparison of annual 6-mean CPUE (no10.69 rapidly in relation to two peaks observed in 1981 and km) of cunner taken by trawl at stations IN and NB from 1991. No correlation was found between annual 1976 through 1994. The vertical line indicates the removal abundances of tautog eggs and larvae (Spearman's l of the MNPS Unit 3 cofferdam in late summer of 1983. rank-order correlation coefficient r = 0.101; p =

0.701) and no trends in larval abundance were found rates have been estimated for several regions ranging using the Mann-Kendall test during either two- or from Narragansett Bay to Virginia (Cooper 1965; three-unit operational periods. However, abundances Simpson 1989; Hostetter and Munroe 1993), of tautog and cunner larvae were highly correlated Male tautog mature when 2 to 3 years old and (SPearman's rank-order correlation coefficient r =

females at age-3 to 4; fecundity at size and age was 0.891; p = 0.0001). Relative annual survival indices i reported by Chenoweth (1963). Adults return to for tautog and cunner were calculated by dividing A- l nearshore waters in spring prior to spawning, with a mean densities of larvae by those of eggs. The high proportion of fish returning to the same survival indices for tautog and cunner were also spawning area cach year (Cooper 1965). Spawning highly correlated (Spearman's rank order correlation occurs during afternoon or early evening hours from coefficient r = 0.853; p = 0.0001), indicating mid-May until mid-August in LIS (Wheatland 1956; common processes that affected the recruitment of Chenoweth 1963; Olla and Samet 1977,1978). The young for both of these wrasses.

pelagic eggs hatch in 42 to 45 hourt at 22*C (Williams 1967; Fritzsche 1978). The pelagic larval TABLE confidence 14.,

m terval of tautog eggs and larvac collected at EN f stage lasts about 3 weeks and individuals settle on June 1976 through May 1996.

the bottom when they reach a size of about 17 mm LGGS LARVAL (Sogard et al.1992; Dorf 1994). Estimated growth year rN FN rate during pre-settlement is about 0.75 mm per day and durmg post-settlement is about 0.5 mm per day 1977 -

36

• 17 (Sogard et al.1992; Dorf 1994). Size at the end of 1978 -

I*1 the first growth season in Narragansett Bay (about 50 mm total length; Dorf 1994) was less than found in a l$

1981

$$$lll3 2.647

• 434 N ig5 83

• 36 southern New Jersey estuary (75 mm standard length; Sogard et al.1992) and this was attributed to l$$

1984

$$l$$

2.157

• 440 lllll 3*2 a longer growing season in southern waters.

l$ l$lj'0[3 9 33 * $2 Entrainment. Tautog were collected primarily as 1987 3.011

• 823 7*3 eggs in the ichthyoplankton entrainment program. l$8 lj$l$ lNjo Since 1979, eggs have ranked second in abundance 1990 2.060

• 933 33

• 28 from collections at station EN (Table 1). Tautog ard Q9j l 878 g *g

• ji cunner eggs are very similar in appearance, but their 1993 1.5 %

• 567 6*3 annual A-mean densities were not correlated l994 9 j%* y8 lj

• 8 (Spearman's rank-order correlation coefficient r =

0.108; p = 0.680). The 1995 A-mean density for

• Data seasonally restricted to May 23 August 20 for eggs and tautog eggs of 2,074 fell viithin the range for the June-August for larvac.

Fish Ecology 29

8 T T pgfgo [gggg TABLE 15., Total annual catch of tautog collected by trawl at 3500- gogg selected stations from June 1976 through May 1996.

300o- , l Year NR JC IN Total 2500- 1976-77 39 71 63 173

%f , 1977-78 16 106 70 192 o 2000- [ .

/ 1978 79 30 59 86 175 E *

,,1 1979-80 45 57 68 170 1500- l 1980-81 25 47 22 94 e

1981-82 129 20 27 176 1* '

90 1982 83 37 50 177 l 198&B4 16 18 41 75 800- i 1984-85 11 15 46 72 0 1985-86 22 31 47 100

. i i1 i ie i iiiiiie i i i i T- 1986-87 110 76 7e er 82 57 23 190 64 86 88 90 92 94 l 1987-88 15 30 17 62 1988-89 57 26 42 135 l 1989-90 28 20 18 66 l 125- uwAE 1990-91 105 40

, 16 161 l

. 1991-92 51 35 14 100

'* 1992-93 24 91 9 124 l 1933-94 13 50 24 87

. 1994-95 14 20 17 51 75- li '

1995-96 128 74 28 230 2 l

/\ l 5- / \ '

/ \ l tautog were caught at NR, the catches in the Niantic

~ \ '

River also varied to a greater degree (CV = 84%)

than at the other two stations (both had a CV = $8%). I o ' '

\O'l ' '

No significant trends were found for the combined 7e'7s'do'52 ' 54 e6 ' s'8 ' do ' J2 ' d4 catch of tautog at the three inshore trawl stations l

• " during either the two- or three unit operational l Fig.15. Annual A mean densities (noJ500 m') of tautog periods. By station, the only significant trend was for JC during the two-unit operational period, when eggs and larvae at station EN during MNPS two-unit (1976-85) and three-unit (1986-96) operational periods. A catch decreased over time (slope = -6.7; p = 0.006; A-mean density calculated for the entire two-unit period is Mann-Kendall test).

represented by a horizontal solid line that is extended as a Length-frequency distributions of tautog caught W dashed line through the three-unit period to serve as a ref- trawl, before and after three-unit operation and Dr erence level for abundance. (Note that the vertical scales the current year, are shown in Figure 16. Ages were differ between the graphs).

Trawl monitoring program. Tautog are caught infrequently by trawl because they prefer rocky or 90-reef habitats and are less vulnerable to this sampling so; D *~~"a gear; annual A-mean CPUE cannot be calculated gro; g because of too many zero values. As an alternative, g soi the annual (June-May report year) sum of catches at gso- u em the trawl stations may be used as index of abun- g4o-dance. in contrast to 1994-95, when the total catch gso- ,

was the smallest recorded in the 20-year data series t 20- -

- i.

(NUSCO 1996), the combined catch at the three io-inshore stations during 1995-96 was the largest and was due to a relatively strong year-class of young Act

( se

, ],

u m iv v

(" ,

e vi (Table 15). The 128 tautog collected at NR was only one fish less than the series high in 198182; the 74 Fig.16. Length-frequency distribution by length (mm) and fish at JC also was the second highest catch for that age (determined from age-length key of Simpson 1989) of station. However, since the mid-1980s, tautog have tautog taken by trawl at stations NR, IN, and JC during become less abundant at IN, which may also be MNPS two-unit (1976-85) and three-unit (1986-94) opera-related to removal of habitat (i.e., the rock tional periods and the 1995-96 report year (June-May).

cofferdam) as noted for the cunner. Although more 30 Monitoring Studies,1996

i l

l assigned to length categories that were based on age- were reported to be significantly greater than at the length data from Lis (Simpson 1989). Young-of- other two stations. However, because of the changes the-year tautog accounted for a higher proportion of in abundance noted above, no significant differences the fish caught after three-unit operation began were found among the stations for tautog abundance.  !

(64%) and in 1995-96 (83%) than during the two- Also, no abundance trends were detected using the unit period (28%), when proportionately more fish Mann Kendall test during the 1988-96 period at any I were seen in older age classes. As noted previously, particular station or for total lobster pot catch.

young tautog were relatively common in 1995-96. {

The reasons for the increased catch of tautog in the Lobster pot sampling. Tautog have been JC )bster pots in 1996 are unknown. A large blue routinely found in pots used in the lobster monitoring mussel (Mytilus edulis) bed located on a ledge in the J

program (see the Lobster Studies section for details). MNPS discharge area may have been greatly reduced '

Since 1988, these fish have been counted and following the shutdown of all three MNPS units as of measured to provide another index of tautog March 30 that continued throughout 1996. Reduced l abundance. Total annual (May October) catches at prey availability as well as a lack of a thermal plume j each of the three lobster monitoring program stations in the shallow Jordan Cove may have increased the (Jordan Cove, designated herein as JC; Intake, IN; number of tautog foraging in the cove, particularly 1 and Twotree, TT) were examined. From 1988 near the rock outcrops where the lobster pots were through 1993, annual catch was usually greatest at set. In addition to greater catches during 1996, lobster pot station IN, followed by TT and JC (Fig. tautog entering the JC pots apparently attacked and 17). However, beginning in 1994, catches at IN killed or damaged a considerable number of trapped decreased relative to the other two stations. In 1996, lobsters, which had unusually high injury rates this  ;

catch at TT more than doubled from 1995 and at JC year (see Lobster Studies section for details). This j the number of tautog was about six times the was likely an indication of increased tautog attacks  !

previous high. In NUSCO (1996), catches at IN cn lobster, perhaps in lieu of other prey. '

Lobster pots should select for certain size-classes 120-JC of tautog because the 2.5 cm2 wire mesh should not

)

, gj g retain smaller individuals when pots are hauled.

Also, the 15-cm diameter of the funnel entrances i

[

S 80 -

restricts the entry of most larger individuals. The S60j length-frequency distribution of tautog caught in 8 lobster pots was dominated by fish from 200 to 349 d 40 :; .

. ,E mm, particularly at IN (Fig.18). This size range h 20i: h ..y. "

, primarily includes 3 to 5-year-old fish, ages during which both males and females become mature.

88 89 90 91 92 93 94 95 96 Therefore, the lobster pot catches provide a reliable l

index of newly recruited adults. Relatively similar

- 1803 age structure was found at the three station locations,

$ although fish in both smaller and larger size-clases i adjacent to modal size-classes were taken more d)2

[ frequently at JC and TT than at IN.

S 80 j Previous tautog early life history studies. l 5 so j Previous special studies on the hydrodynamics of LIS

$ 40 j near MNPS and of tautog early life history (focusing S 20 j predominantly on the egg stage) are summarized o i below with some additional commente. Thew 88 89 90 91 92 93 94 95 96 studies were important to the design of the W96 YEAR special field study that attempted to identify potential Fig. 17. Total catch of tautog in lobster pot sampling s urce areas for the tautog eggs entrained at MNPS.

(May-October) by station (JC, IN, TT) and at all stations Tautog eggs are pelagic and their dispersal from combined from 1988 through 1996. (Note that the vertical spawning sites in LIS is primarily by tidal transport.

scales differ between the graphs). The number of eggs entrained by MNPS should be related to egg abundance in Niantic Bay, the source Fish Ecology 31

l I

100 -

~

!I I l 90 -

l "l IN ' 1 80 ; , E JC i .

g 1 70 l D

z 60]  : km4 i +

m s o3 50 i i

) s w l n

g 40 EV 9

30 )

20 >

,a7 N

y fgj  ! -3e r? I h i E 10 1 g . g- ' ' a q' ~a l 0' --

~'

<100 100-149150-199 200-249 250-299 300-349 350-399 400-.449 450-499 >=500 LENGTH (mm)

Fig.18. Length-frequency distribution of tautog taken by lobster pot sampling at stations JC, IN, and TT from 1988 through 1996.

1 of condenser cooling water. Results from drogue Thames River. Based on this information, farfield studies in 1991 (see NUSCO 1992a for details) studies of tautog egg abundance conducted during indicated that during ebb tide water from LIS enters 1996 extended about 5 n mi from the mouth of Niantic Bay from the west, and conversely, during Niantic Bay.

flood tide water enters from the east (Figs.19 and Based on MNPS monitoring data collected from 20). The distance a planktonic tautog egg is trans- 1979 through 1994, the annual temporal occurrence ported during a tidal stage should be directly related of tautog eggs in eastern LIS is generally from about to tidal current velocity. Average current velocities early May through mid- to late September. This for ebb and flood tidal stages in the area from the seasonal occurrence of tautog eggs is similar to that Connecticut River to the Thames River were esti- reported by Monteleone (1992) for Great South Bay, mated from information provided in tidal current NY. The annual timing of peak spawning, as indi-tables (NOAA 1993) at five locations west of Niantic cated by egg abundance:, can be estimated from the Bay and five locations to the east (Fig. 21). For inflection point of the Gompenz function (Eq.1).

simplicity, the duration of both ebb and flood stages From 1979 through 1994, peak spawning occurred was assumed to be 6 h. Estimated average tidal during mid- to late June and appeared to be related to velocity was 0.94 knots during an ebb tide and 0.88 annual spring water temperatures. A significant (p =

knots during a flood. Based on these velocities, a 0.002) negative relationship was found between May lautog -egg could be transported 5.7 nautical miles (n water temperatures and the estimated date of peak mi) during an ebb tide and 53 n mi during a flood spawning (Fig. 22). The 1996 spring water tempera-tide. Therefore, the potential source area for a tautog tures were abnormally cool, with an average May eggs entering Niantic Bay during a full tidal cycle water temperature at the MNPS intakes of 9.4'C, lies within a radius of about 5 n mi, with a center at a resulting in an estimated peak egg abundance during mid-point between Black Point and Millstone Point.

the latter portion of June to mid-July. This informa-This would encompass a shoreline extending from tion was used as the basis for the timing of tautog egg about 2 n mi cast of the Connecticut River to the studies conducted during 1996.

32 Monitoring Studies,1996

. . ~ . . - ... - . ~ - . . . - . _ . ~ . . . .. . . _ _ . . . . ._ -_ - .

I i

J i i Niantic River a

S I

I t

i

.m een cee I

l I

hsNPS N f ff

/ whet.

P8' O '* .

g, ' a 3 s.:s b

a r

g i na c

.% g i/ Twi.

ll Black Pt.

, Vowy ,

10 fuenc (03 m)

......... i n um, em n,,,,  ;

mammann ete current i

i 1

Fig.19. Esumated tids! current direction and velocity in Niaritic Bay for an ebbing tide during the first hour after high slack tide and at the time of maximum ebb current (based on results of drogue studies conducted in 1991; NUSCO 1992a). Note that the relative length of the arrows corresponds to the estimated average current velocities. ,

Fish Ecology 33

. . . . _ . _ . ._. . . . . . . - - _ . . . . . ~ . _ . . . . . _ . _ _ . - _ . . _ . . . _ _ _ _ _ . .

.i I

I i

1

/

)

I Nient6c 1 River \

-)

\

3

( i i

)

i 4

cove i

1 i

S N

9 4... T* White

/ o *::::;

4 -- ,

a... r j

eeny y%eg W

i  %

v wee m.

)

' /

/

V stock Pt.

vooav .,

I to e/ese (0 3 minec)

.......... i n ener m m.

me= man no ea mrw Fig. 20. Estimated tidal current direction and velocity in Niantic Bay for a flooding tide during the first hour after low slack tide s and at the time of maximum flood current (based on results of drogue studies conducted in 1991; NUSCO 1992a). Note that the - '

relative length of the arrows corresponds to the estimated average current velocities.

i l

34 MonitoringStudies,1996 i

. . .._-.m... .. ._.._ _ . . . _ .__ ._ _ _ - . . _ . . _ . _ _ - . _ _ _ . . _ . . _ _ _ . _ .

I i

4

i 1

[

d N s ]

i

! l S n mi l

} MNPS i

i NB i #27 1 O 9 #2736

) g #2816 #2791 e g

}

j

1. 2821 e #2776 #2766 Fishers 1.

1. 2814 i g #2711

1. 2796 M

Plum I.

3 Onent l [ Pt.

i Gardiners 1.

..i i

i Montauk Pt. ,

Fig. 21. Approximate location of sites used to estimate average current velocities during ebb and flood tides, including the

' NOAA tide current tables reference numbers (NOAA 1993).

Fish Ecology 35

l 6/30- Ns 3 , 6/8-9/93 J -i y e, , 1.0 -

N* A 6/20- NN e' O.84 ... . 6/15-16/93

.'/ . '\.

e 2

• \

q *

0. 6

-*~~

7/19-20/93 $ '5 'g

,2.o.5o' j- '

. e 2

3 6/10 J l p = 0.002 e

O as d

1 9

i 10 11 12 13 0.2 - ',*,/ A. w . ., . .\,[

j ' ,

May water temperature (C) 0.01 1 2 4 6 8 10 12 14 16 18 20 22 24 Fig. 22. Relationship between annual mean May water TIME (24hr) temperature ("C) at the MNPS intake, and the date of peak abundance of tautog eggs frm 1919 thrcugh 1994.

8000]

5000 - e Comparison of results from routine day and night entrainment collections at MNPS indicated that E ,g, _ /

tautog eggs were more abundant at night. Therefore, EN- j 24-hour entrainment studies were conducted during E 1993 to examine the diel change in abundance 2000 --l * \.,

t (NUSCO 1994a). Three 24-hour periods (June 8-9, June 15-16, and July 19-20 in 1993) were sampled at 0i

] ' + .s, 8

• i i i i i i i i i i i i 2-hour intervals using the entrainment sampling 2 4 e e to 12 14 is to 20 22 24 methodology described previously. TIME (24hr)

The pattern of change in tautog egg abundance every 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> over a 24-hour period showed very Fig. 23. Daily proportional abundance (top; expressed as sample density / maximum sample density for each date) of consistent results on the three study dates (Fig. 23)-

tautog eggs for three 24-hour studies conducted in 1993, in general, egg abundance peaked at about 1800 h, and the geometric mean density of the three studies com-decreased through the night, remained relatively bined (bonom).

constant until late afternoon, and increased rapidly during the evening. This pattern indicated a short, The nearfield spatial distribution of tautog eggs was early evening spawning period, consistent with labo-examined from Black Point to Twotree Island Chan-ratory observations of Olla and Samet (1977). The nel and in Niantic Bay in 1994 (NUSCO 1995) with timing of peak abundance was not related to tidal five stations sampled (Fig. 24). The offshore stations stage because sampling in 1993 for the June 8-9 and (BP, L1, NB, and SS) were sampled with the 60-cm June 15-16 studies occurred during opposing tidal bongo system (333-km mesh nets) using a stepwise stages. The rapid decline in abundance from 1800 to oblique tow pattern for a 6-minute duration with 2200 h cannot be attributed to hatching, as egg incu- equal sampling time at surface, mid, and near-bottom bation takes longer than i day. Therefore, this de- depths. Station EN was sampled using the previously cline was probably due to high natural egg mortality, described entrainment gantry system. The water likely from predation, as was suggested for eggs of depth at all offshore stations ranged from about 6 to s the cunner, a sympatric species (Williams et al. 10 m. Station BP was sampled during an ebb tide 1973). Natural mortality, which likely accounts for and station SS during a flood tide, so that collection the rapid decline in tautog egg abundance, from peak densities of tautog eggs would represent those poten-spawning at 1800 h through 0200 h was about 70% tially imported into Niantic Bay from the west and and through 0600 h was 80%. This information, east, respectively. The remaining three stations (EN, suggesting high natural egg mortality during the first NB, and LI) were sampled during both tidal stages.

12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> after spawning, followed by a reasonably Samples were collected during the period of 0500 to stable abundance, was used in designing the 1996 1100 h. This time period was selected because 24- l sampling program.

hour studies conducted in 1993 showed that tautog )

36 Monitoring Studies,1996 l

1

0l .

1 Ml NIANTIC RIER NIANTIC g - COVE NB k

)

l S' & ss o . t ko

• SP Fig. 24. Location ofichthyolankton stations (') sampled for nearfield spatial distribution of tautog eggs during 1994.

egg densities remain relatively stable at this time of The results of this nearfield study of tautog egg day (Fig. 23). Stations EN and NB were sampled at abundance indicated that the geometric mean approximately at the same time. The collection se- densities of tautog eggs at each station were similar quence of stations sampled was LI, NB, and SS dur-and had overlapping 95% confidence intervals (Fig, ing a flood tide and L1, NB, and BP during an ebb 25). The lack of localized egg concentrations was tide. These sequences facilitated paired comparisons confirmed by the results of paired comparisons of stations BP, L1, and SS with EN. By sampling NB between station EN and the other stations (NB, LI, second in the sequence (with EN sampled almost SS, and BP) when tested with the Wilcoxon's signed-simultaneously), the sampling interval between EN rank test. Although the number of paired and the other three stations was minimized. Sam- comparisons was rather low (12 pairs for NB and Ll pling dates in 1994 were June 23 and 24 during a and 6 for SS and BP), no significant (p < 0.05) flood tide and June 29 and 30 during an ebb tide. differences were detected between station EN and the These dates occuned during peak density of tautog other four stations. These nearfield data indicated eggs. On each sampling date, three sequences of that eggs were not concentrated near MNPS and samples were taken (L1, NB, EN, BP during an ebb entrainment densities of tautog eggs were tide and LI, NB, EN, SS during a flood tide), with the representative of a more homogenous distribution, first sequence starting about i hour before maximum including areas outside of Niantic Bay.

tidal current, the second starting at near maximum cunent, and the third immediately after the second wr.s completed.

Fish Ecology 37

l l

the surface than near-bottom for all 12 paired E 5000 - C mParisons, even though vertical temperature and

.g salinity measurements showed no apparent water g4000- column stratification (Table 16). Examination of the geometric mean densities of the three replicate pairs

. 3000 -

indicated that differences between surface and near-E 2000 - bottom were much greater for collections made in the g - -

evening just after spawning than for collections 1000d l dunng the morning, approximately 12 hour1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> after l *

} 1 spawning (Fig. 26). Tidal currents did not appear to 0 i i - i i i affect this pattern. Relatively similar egg densities EN NB Ll ss BP stations were found during the evening from near-bottom collections and in the morning at both surface and Fig. 25. Comparison of nearfield spatial distribution of near-bottom relative to densities at the surface during tautog eggs in the vicinity of MNPS, based on geometric the evening. This pattern of densities, in addition to mean densities with 95% confidence intervals for cach the large difference in tautog egg abundance between i station sampled (see Fig. 24) during June 1994. surface and near-bottom collections during evening i and the reduction in density at the surface from l Tautog early life history studies in 1996. The evening to morning, suggested selectively higher Draft Fisheries Management Plan for Tautog mortality for eggs near the surface during the 12-hour I (ASMFC 1996) stated that tautog eggs are primarily period following spawning.

found near the surface, the basis for which apparently was the results of a study conducted in Narragansett Bay by Bourne and Govoni(1988). A portion of that "E 12000 ,

bay is a two-layered estuary and pelagic tautog eggs g Evering may concentrate near the halocline, in contrast, the

[ 10*0 j g j ] uomng )

water column in eastern LIS is relatively soooy  !

homogenous for salinity (and temperature), with no 1 halocline evident. However, even under these ,J l g l conditions (i.e., eggs tending to concentrate near the E dNo d l surface), the effect of winds on their transport would :f 2000 J need to be considered along with tidal currents. E -

rj Therefore, to examine the vertical distribution of 0i ami n r. ,

J i > i i i i i tautog eggs, paired surface and near-bottom tows s e s e s e s a were taken in 1996 at a point midway between I'dai current slack u nximum Millstone and Black Points. This location also was near the point of origin for the transects used in the Fig. 26. Comparison of geometric mean densities of tautog farfield spatial distribution study discussed below. eggs from paired surface and near-bottom samples The abundance data for the vertical distribution study collected during the evening and morning at the time of were also used to estimate tautog egg mortality given slack and maximum tidal currents from a site midway below. The densities of tautog eggs were greater at between Millstone and Black Points during July 1996.

TABLE 16. Time of sampling. tidal current stage, water temperature, and salinity when paired surface and near-bottom collections were taken to compare the vertical distribution of tautog eggs at a site midway between Millstone Point and Black Point during July 1996.

Samphng Tidal T emperature (*C) Salmity (ppt) time (h) current Surface Mid Bottom Surface Mid Bottom 2005-2105 Slack 17.6 17.6 17.6 29.2 293 293 0832-0939 Slack 17.6 173 16.9 29.5 29.5 29.6 2005-2046 Maximum 19.1 17.5 17.0 28 0 29.0 29.4 0822 o914 Maximum 18.2 17.5 17.1 27.7 28.8 29.2 38 Monitoring Studies,1996

__ -.. .__ . _ _ _ _ . . _ . . _ _ . _ _ . _ . . . . _ . . . . _ . . _ ..._.___m._____.__.__._m_.___..

4 i

d N

\f

\ .

I l l l 1 3 I

{

S n mi .

MNPS 1 I

a one l

l 2 2 il2 l 3 3 4 ,,3 4 Fiahors L

s. ,,4 .s <

SW SE

.s )

S i l

Plum L

{

Orient Pt.

Gardeners I.

i Montauk Pt.

Fig. 27. Sampling sites for the farfield (within about 5 n mi of MNPS) spatial distribution study of tautog egg abundance conducted in 1996. Three transects (SW, S SE) extended from a common origin (station ORI), with stations spaced at I n mi intervals.

Tautog egg mortality during about the first 12 nearfield study completed in 1994. In 1996, sam.

hours after spawning may be estimated if the pling was extended to the 5 n mi boundary discussed combination of surface and bottom densities above to encompass a potential source area of tautog

. (geometric mean of 6 samples) is assumed to be eggs entrained at MNPS. Sampling sites were at I n representative of abundance at the time of mi intervals along three separate transects (Fig. 27).

collections. Under this assumption, them was about The point of origin (station ORI) of each transect was 65% mortality (68.9% for slack current collections at the mid-point between Millstone Point and Black and 64.4 % for maximum current collections) during Point, which was also used in the vertical distribution the 12-hour period. "Ihis mortality estimate, although study. For this farfield study, sampling took place in large for a 12-hour period, was less than the the moming after sunrise because results of the 1993 approximate 80% mortality estimated from the 24- 24-hour studies showed that tautog egg densities hour studies summarized above, remained relatively stable during this time of day No information was available for tautog egg (Fig. 23).

thundance further offshore of Niantic Bay than the Fish Ecology 39

Abundances of tautog eggs, expressed as density from 24-hour studies and evening-moming per 500 m', for both sampling dates at station ORI abundance comparisons from vertical distribution were generally similar (geometric mean of three studies, the standing stock estimates only accounted replicates), with no consistent trends evident between for about 20 to 35% of the spawn from the previous dates or along transects (Fig. 28). On July 2, the evening because of natural mortality.

greatest abundances of eggs were collected along The daily standing stock estimates were also transect SE at n mi 2 through 4, but the remaining compaied to an average lifetime fecundity estimate two transects had relatively similar densities. On for female tautog (Table 19). Parameters used to July 9, abundances were more similar among estimate average lifetime fecundity under 1996 transects with slightly higher densities at n mi 1 and 2 conditions were length at age for LIS (Simpson for transects SE and SW. For both dates no clear 1989), fecundity at age (Chenoweth 1963), fraction nearshore to offshore gradient of tautog egg densities of mature females at age (Chenoweth 1963), and was found.

natural (M = 0.15) and fishing (F = 0.54) mortality The depths of the sampling sites varied greatly, rates from ASMFC (1996). In addition, present ranging from 8 to 58 m (Tables 17 and 18). Water fishing regulations were used, including a 14 in (356 temperature and salinity measurements at surface, mm) legal size limit with a natural mortality of M =

mid-depth, and bottom were similar at each site, 0.15 and a discard mortality of F = 0.04 (D. Simpson, indicating a relatively well-mixed water column CT DEP, Old Lyme, CT, pers. comm.) for fish less throughout the 5 n mi radius from Niantic Bay. Due than the legal size limit. Lifetime fecundity was to the large variation in water depths among estimated to be 142,655 eggs per female. The daily sampling sites, tautog abundance indices were egg standing stock was adjusted for mortality that recomputed to give the number of eggs under I m2 of occurred during the 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> after spawning by using sea surface at each sampling site (Fig. 29). This l

abundance index suggested greater similarity among transects than the volumetric density estimates, with - SE no obvious nearshore to offshore trends evident.

s This lack of trends and no indication of preferred a,3,3,j /.

spawning areas may have been due to the time of /

g 1 - sw sampling. Results of previous 24-hour studies showed that tautog adults spawn primarily during the 900,j /

g  ; ,

evening. The time period of sampling for the farfield spatial distribution was during the morning, which j ,j ' ',

1 e occurred a full tidal cycle after spawning and allowed 1 C- ~ ~ ~ ~ - ' ~

a j j for relatively complete mixing and distribution of eggs by tidal currents.

f ( $ $

Nauticai mnes An instantaneous standing stock within the 5 n mi radius of Niantic Bay was calculated to estimate the 4000 ; ayg - SE number of tautog eggs that could be potentially -

entrained by MNPS from tidal transport. The "E sooo ; s geometric mean density of all 16 stations combined g ,_

3, was calculated and extrapolated to a total number of E2000 eggs based on the average depth of the stations #

samplei The estimated number of tautog eggs k sooo v .j ; _. ,-

within a 5 n mi radius of Niantic Bay during the time 1 period of sampling was 4.9 x 10' on July 2 and 3.1 x . " " y~^

od 10' on July 9. These daily egg standing stock o 1 2 3 4 5 estimates equaled or exceeded the estimated annual Nautical mdes total number of tautog eggs entrained at MNPS since 1979, which ranged from 0.7 to 3.9 x 10' (NUSCO Fig. 28. Comparison of tautog egg densities along threc 1996). In addition, these standing egg stock esti' transects sampled at I n mi intervals on two dates m July 1996. Nautical mile o is station ORl (see Fig. 27), the mates represented the number of eggs approximately origin of the three transects.

12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> after peak spawning. Based on the results 40 MonitoringStudies,1996 l

TABLE 17. Station depth, water temperature and salmity for the July 2,1996 collections to examine farfield spatial distribution of tautog

'l emperature (*C)

Salimty (ppt)

Station Depth (m) Surface Mad Bottom surface Mid Bottom ORl* 8 163 16.5 163 28.7 28.5 28.6 SEl 11 16.1 16.0 16.0 28.7 28.7 SE2 13 28.7 16 4 16 4 16.1 28 4 SE3 28.5 28.7 18 17.1 16.1 16.1 27.2 28.8 28.8 SE4 22 16.5 16.1 16 0 283 288.8 SE5 20 28.8 16.0 15.9 15.8 28 8 28.8 29.0 Si 20 163 16.2 16.1 28.4 28.5 28.7 S2 24 16.1 15.9 16.0 28.7 28.9 S3 28.8 45 16.2 15.9 -' 28.5 28.8 -

$4 58 16.6 16.1 -

27.6 283 -

SS 52 16.7 16.4 -

27.6 27.9 -

SWI 28 16.5 163 16 4 28.6 28 6 28.6 SW2 29 16.5 16.2 16.2 27.7 28 3 28.4 SW3 34 16.6 16.2 16.2 27.4 28.2 283 SW4 37 16.6 16.2 16.1 27.4 283 28.5 SW5 49 16.8 163 -

273 27.9 -

• Parameters for station ORI are a mean of three collections during the sampling date.

• No temperature and salinity measurements because the bottom depth was greater than the length of the probe cable.

TABLE 18. Station depth, wster temperature and salinity for the July 9,1996 collections to examine farfield spatial distribution of tautog

'lemperature (*C) Salmity tppt)

Station Depth (m) Surface Mid Bottom Surt' ace Mid Bottom ORl* 8 17.5 17.4 17.4 29 4 29.5 29.5 SEl 12 17.5 173 16.9 294 29.5 29.7 SE2 12 17.2 17.1

' 16.8 29 5 29.5 29.7 SE3 17 17.2 16.4 16.5 29.0 29.9 29.9 SE4 23 17.2 16 4 16.0 29.0 29.8 30.2 SES 19 17.7 16.6 163 28.6 29.7 29.9 Si 21 17.1 16.8 16.8 293 29 6 29.6 S2 24 17.2 16.8 16 4 29.2 29.6 29.9 S3 47 16.8 16.1 -' 29.6 30.1 -

S4 $$ 16 9 16.4 -

29.4 29.8 -

$5 51 17.9 163 -

28.2 28.9 -

,SWI 29 17.4 16.9 16.7 293 29.7 29.7

$W2 30 17.1 17.0 16.9 293 29 6 29.7 SW3 34 17.0 16.5 163 29.1 29.8 30.0 SW4 37 17.8 16.8 16 4 28.2 29.4 29.8 SW5 50 18.0 16 3 -

28.1 30.0 -

• Parameters for station ORI are a mean of three collections during the sampling date.

No temperature and salinity measurements because the bottom depth was greater than the length of the probe cable.

Fish Ecology 41

i I

l l

'* M2 -

this area equaled or exceeded annual entrainment

_ j i

• ~

estimates at MNPS and, in fact, would have been  !

even larger if high egg mortality rates had been taken 8 ,, . -

into account. This implies that MNPS entrainment I ~

effects may be relatively small, f so- If egg losses due to entrainment affected recruit-

~1 ment, then juvenile abundance should decrease and 20 ~

)

the, relative abundance of older fish would appear to  ;

, increase in the short term. Based on length-  ;

u mi oi a4 a oia 4 s e1 : 4 a frequency distribution from trawl catches, the l

'=M 6E 6 SW percentage of juvenile tautog increased during the j

'* three-unit operational period. Therefore, changes in i M' the relative proportion of juveniles and adults were

, probably unrelated to entrainment losses. In addition, the decline in juvenile and adult tautog

[ em abundance in Long Island Sound that began in the g _

mid-1980s (Simpson et al.1995) coincided with the ,

}do _

decreasing trend in eggs collected at EN. If the  !

_ decrease in adults was caused by entrainment losses, 9 ~ ~

~~

then the reduction in egg abundance should have

,) T lagged the decline of juveniles by several years w mi oia34 s oia 4 a oi 4 s because females do not mature until age-3 or 4. l T= = se s sw Therefore, the lower abundance of tautog eggs was Fig. 29. Comparison of tauto egg abundances based on Probably due a decline in the abundance of spawning j the number of eggs under 1 m of sea surface along three adults from fishmg rather than the operation of transects sampled at I n mi intervals on two dates in July MNPS. During the 1990s, the instantaneous fishing j 1996. Nautical mile 0 is station ORI (see Fig. 27), the mortality rate for tautog was estimated at about 0.54 '

origin of the three transects. (annual fishing mortality of 42%) and various survey biomass indices declined by more than half from the previous decade (ASMFC 1996). At present, tautog 65% fro n the vertical distribution study and 80% stocks are overfished and because of the long life and from the 24-hour study. Equivalent-female spawners slow growth of this species, abundance should for the July 2 study ranged from 98,139 to 171,743 remain depressed until fishing mortality is reduced to and for the July 9 study ranged from 62,388 to less than half of current levels.

108,654. These numbers of equivalent-female spawners in the 5 n mi radius are conservative (i.e., Conclusions low) because tautog are serial spawners and may spawn over an extended period (Oila and Samet Evaluation of data coll .cted in the tr. wl, seine, and I9773-ichthyoplankton programs in the fish eco?ogy studies Summary. The greatest direct impact of MNPS in 1995-96 did not change conclusions made in on tautog stocks is most likely the entrainment of recent years regarding the potential impact of the eggs. Large (65 80%) decreases in egg abundance operation of MNPS. Examination of species occur following early evening spawning through the composition in each program indicated similar following moming, most probably from high natural dominant taxa as found in previous years. In trawl mortality. Pelagic tautog eggs disperse rapidly from catches, annual abundances were more stable for spawning sites by tidal transport and densities in resident species '(e.g., grubby, skates) than for those nearshore areas are relatively uniform. Based on fishes caught seasonally (e.g., anchovies, scup),

hydrodynamics, a conservative measure of the source probably because individuals of the latter were area for eggs entrained at MNPS includes a radius of primarily young-of-the-year and annual reproductive about 5 n mi. Two daily estimates of the success was variable. For all species, tne estimated instantaneous standing stock of tautog eggs within 42 Monitoring Studies,1996

1 1

TABLE 19. Average lifetime egg production of an age-3 female tautog spawner. l Length Surmal fracuon Egg d

Age (mmf Fecundity" probability' mature production l

3 223 16,069 1.000000 0.8 12,855 4 273 26,476 0.826280 1.0 21,876 l 5 317 39,000 0.682738 1.0 26,627 1 6 354 53,519 0.564133 1.0 30,192 7 386 69,938 0.282955 1.0 19,789 ,

8 414 88,181 l .0 12.515 0.14192( l 9 439 108,184 0.071186 1.0 7,701 10 460 129,894 0.035705 1.0 4,638 1 1I 478 153,264 0.017909 1.0 2,745 l 12 493 178,251 0.008983 1.0 1,601 l 13 507 204,820 0.004505 1.0 923 14 519 232,937 0.002260 1.0 526 15 529 262,571 0.001133 1.0 298 16 538 293,697 0.000569 1.0 167 17 545 326,288 0.000285 1.0 93 18 552 360.321 0.000143 1.0 52 19 558 395,774 0.000072 1.0 28 20 562 432.628 0.000036 1.0 16 21 567 470,864 0 000018 1.0 8 22 570 510,463 0.000009 1.0 5 Total 142,655

• Length at age from Simpson (1989).

• Fecundity at age from Chenoweth (1%3).

Instantaneous mortality rates (Z) were:

Natural (M) = 0.15 from ASMFC (1996)

Discard through age-6 (F) = 0.04 from Simpson (CT DEP, Old Lyme, CT, pers. comm.)

Fishing age 97 through 22 (F)= 0.54 from ASMFC (1996).

d Female maturity from Chenoweth (1963).

number of eggs and larvae entrained was a function two-unit operation for larval American sand lance of both annual abundance and the volume of and juvenile and adult tautog and cunner were condenser cooling-water used at MNPS, with the probably related to regional declines, likely resulting frequency of large entrainment estimates greater from increased predation or overfishing in the case during most years of three-unit operation. of tautog. Also, removal of specific habitat (i.e., the Entrainment estimates for several taxa and life stages Unit 3 cofferdam) contributed to fewer cunner and (e.g., anchovy and cunner eggs; anchovy, American tautog being collected by trawl at station IN, The sand lance, and grubby larvae) decreased in 1995 or large numbers of tautog and cunner eggs entrained at 1996 relative to the previous year for these estimates. MNPS did not appear to affect the future spawning Detailed analyses were conducted on six taxa that stocks of these two fishes because the proportion of were most susceptible to MNPS operational impact juvenile recruits relative to adults has increased due to entrainment or effects of the thermal during three-unit operation. Finally, early life discharge. These analyses generally focused on history stages of both tautog and cunner appear to be comparing temporal trends during two- and three- affected similarly by environmental and biological unit operations. The reasons for the decline in processes, which seem to have greater effects than Atlantic silverside taken by seine that occurred MNPS operation.

during three-unit operation remains unknown, but probably was not related to the MNPS thermal discharge, which affects the JC seine site only minimally. Trends of declining abundance during Fish Ecology 43

References Cited Conover, D.O. 1992. Seasonality and the scheduling of life history at different latitudes. J. Fish. Biol.

ASMFC (Atlantic States Marine Fisheries 41:161 178.

Commission). 1996. Draft fisheries management Conover, D.O., and M.H. Fleisher. 1986. Tempera-plan for tautog. Public hearing summary. ture-sensitive period of sex determination in the Washington, DC.10 pp. Atlantic silverside, Afenidia menidia. Can. J. Fish.

Bengtson, D.A. 1984. Resource partitioning by Aquat. Sci. 43:514-520.

Afenidia menidia and Afenidia beryllina Conover, D.O., and B.E. Kynard. 1981. Environ.

(Osteichthyes: Atherinidae). Mar. Ecol. Prog. Ser. gnental sex determination: interaction of 18:21 30. temperature and genotype in a fish. Science Bengtson, D.A. 1985. Laboratory experiments on 213:577-579.

mechanisms of competition and resource Conover, D.O., and B.E. Kynard. 1984. Field and partitioning between Afenidia menidia and Afenidia laboratory observations of spawning periodicity beryllina (Cope) (Osteichthyes: Atherinidae). J. and behavior of a northern population of the Exp. Mar. Biol. Ecol. 92:1-18. Atlantic silverside, Afenidia menidia (Pisces:

Bengtson, D.A., R.C. Barkman, and W.J. Berry. 1987. Atherinidae). Envir. Biol. Fish. I1:161 171.

Relationships between matemal size, egg diameter, Conover, D.O., and S.A. Murawski. 1982. Offshore time of spawning season, temperature, and length winter migration of the Atlantic silverside, Afenidia ofhatch of Atlantic silverside, Afenidia menidia. J. menidia. Fish. Bull., U.S. 80:145-150.

Fish Biol. 31:697-704. Conover, D.O., and M.R. Ross.- 1982. Patterns in Bigelow, H.B., and W.C. Schroeder.1953. Fishes of seasonal abundance, growth and biomass of the the Gulf of Maine. U.S. Fish Wildt. Serv. Bull. Atlantic silverside, Afenidia menidia, in a New

$3:1-577. England estuary. Estuaries 5:275-286.

Blake, M.M., and E.M. Smith. 1984. A marine Cooper, R.A. 1965. Life history of the tautog, resources management plan for the State of Tautoga onitis (Linnaeus). Ph.D. Thesis. Univ, of Connecticut. CT Dept. Envir. Prot., Mar. Fish. Rhode Island, Narragansett, RI.153 pp.

244 pp. Dew, C.B.1976. A contribution of the life history of Botelho, V.M., and G.T. Donnelly.1978. A statistical the cunner, Tautogolabrus adtpersus, in Fishers analysis of the performance of the Boume plankton Island Sound, Connecticut. Chesapeake Sci. l splitter, based on test observations. NMFS unpub. 14:101 113. l ms. Dorf, B.A. 1994. Ecology of juvenile tautog )

Bourne, D.W., and J.J. Govoni. 1988. Distribution (Tautoga onitis, Family Labridae) in Narragansett I of fish eggs and larvae and patterns of water Bay, Rhode Island. Ph.D. Thesis, Univ, of Rhode circulation in Narragansett Bay, 1972-1973. Am. Island, Narragansett, RI. 213 pp.

Fish. Soc. Symp. 3:132-148. Dorsey, S.E., E.D. Houde, and J.C. Gamble. 1996. l Briggs, P.T., and J.S. O' Conner.1971. Comparison of Cohort abundances and daily variability in shore-zone fishes over natural vegetated and mortality of eggs and yolk-sac larvae of bay sand-filled bottoms in Great South Bay. N.Y. Fish anchovy, Anchoa mitchilli, in Chesapeake Bay.

Game J.18:15-41. Fish. Bull., U.S. 94:257 267.

Castro, L.R., and R.K. Cowen. 1991. Environmental Draper, N., and H. Smith. 1981. Applied regression factors affecting the early life history of bay analysis. John Wiley and Sons, New York. 709 anchovy Anchoa mitchilliin Great South Bay, New PP.

York. Mar. Ecol. Prog. Ser. 76:235-247. Ennis, G.P. 1969. Occurrences of the little sculpin, Chenoweth, S.B. 1963. Spawning and fecundity of AfroxoccPhalus aeneus, in Newfoundland waters.

the tautog, Tautoga onitis (Linnaeus). M.S. Thesis. J. Fish. Res. Board Can. 26:1689-1694.

Univ. of Rhode Island, Narragansett, RI. 60 pp. Fritzsche, R.A.1978. Development of fishes of the Conover, D.O. 1979. Density, growth, production Mid-Atlantic Bight. An atlas of egg, larval and and fecundity of the Atlantic silverside, Menidia juvenile stages. Vol. V. Chaetodontidae through menidia (Linnaeus), in a central New England Ophidiidae. Power Plant Project, Off. Biol. Serv.,

estuary. M.S. Thesis. Univ. of Massachusetts, U.S. Fish Wildl. Serv., U.S. Dept. of the Interior, Amherst, MA. 59 pp. FWS/OBS-78/12. 340 pp.

44 Monitoring Studies,1996

l Gendron, L.1989. Seasonal growth of the kelp, Jessop, B.M. 1983. Aspects of the life history of the Laminaria longicrurls in Baie des Chaleurs, Atlantic silverside (Afenidia menidia) of the Quebec, in relation to nutrient and light availability. Annapolis River, Nova Scotia. Can. Ms. Rep. Fish.

Bot. Mar. 32:345-354. Aquat. Sci. 1694. 41 pp.

Gilbert, R.O. 1989. Statistical methods for Johansen,F. 1925. Natural history of the cunner environmental pollution monitoring. Van Nostrand- (Tautogolabrus a6persus Walbaum). Contrib.

Reinhold Co., New York. 320 pp. Can. Biol. 2:423-468.

Gleason, T., and C. Recksick. 1988. Synopsis of Johnson, M.S. 1975. Biochemical systematics of the biological data for the cunner Tautogolabrus atherinid genus Afenidia. Copeia 1975:662-691, adspersus (Walbaum). Univ. of Rhode Island. Kuntz, A. 1914. The embryology and larval Contrib. 240 of the R1 Exp. Sta. development of Bairdiella chrysura and Anchovia Govoni, J.J., and J.E. Olney. 1991. Potential mitchilli. U.S. Bur. Fish.. Bull. (1913) 33:1-19, predation on fish eggs by the lobate etenophore Laroche, J.L. 1982. Trophic pattems among larvae of Afnemiopsis lei @l within and outside the five species of sculpins (Family: Cottidae) in a Chesapeake Bay plume. Fish. Bull., U.S. 89:181 Maine estuary. Fish. Bull., U.S. 80:827-840.

186. Lawton, R.P., B.C. Kelly, V.J. Malkoski, J.H.

Green, J.M.1975. Restricted movements and homing Chisholm, P.Nitschke, B. Starr, and E. Casey.

of the cunner Tautogolabrus adpersus. Can. J. 1994. Semi-annual report on monitoring to assess 2001.53:1427-1431. impact of Pilgrim Nuclear Power Station on marine Green, J.M., and M. Farwell. 1971. Winter habits of fisheries resources of Western Cape Cod Bay in the cunner, Tautogolabrus adspersus (Walbaum Marine Ecology studies related to the operation of 1792), in Newfoundland. Can. J. Zool. 49:1497 Pilgrim Station. Semi-annual Rep. No. 44. Boston 1499. Edison Company.

Green, J.M., G. Martel, and E.A. Kingsland. 1985. Lawton, R., B. Kelly, V. Malkoski, J. Chisholm, Foraging time allocation in a territorial fish: P.Nitschke, and J. Boardman.1996. Annual report influence of reproductive activities. Mar. Ecol. on assessment and mitigation of impact of the Prog. Ser. 24:23-26. Pilgrim Nuclear Power Station on finfish Grosslein, M.D., and T.R. Azarovitch. 1982. Fish populations in Western Cape Cod Bay in Marine distribution. MESA New York Bight Atlas Ecology studies related to the operation of Pilgrim Monogr. 15. New York Sea Grant Institute, Station. Semi-annual Rep. No. 47. Boston Edison Albany,NY 182 pp. Company. 86 pp.

Hildebrand, S.F. 1943. A review of the American I amn, M.A., K.W. Able, and M.P. Fahay. 1989.

anchovies (Family Engraulidae). Bull. Bingham Life history and food habits of the grubby, Oceanogr. Coll. 8:1 165.

Affoxocephalus aenaeus (Cottidae), in a Cape Cod Hoese, H.D., and R.H. Moore. 1977. Fishes of the estuary. Copeia 1989:7-12.

Gulf of Mexico. Texas A&M Univ. Press, Coll. Leak, J.C., and E.D. Houde. 1987. Cohort growth and Sta. 327 pp, survival of bay anchovy, Anchoa mitchilli, larvae in Hoff, J.G. 1972. Movements of adult tidewater Biscayne Bay, Florida. Mar. Ecol. Prog. Ser.

silverside, Afenidia beryllina (Cope), tagged in 37.109-122.

New England waters. Am. Midl. Nat. 88:499-502. Levin, P.S.1991. Effects of microhabitat on recruh-Hollander, M. and D.A. Wolfe.1973. Nonparametric ment variation in a Gulf of Maine reef fish. Mar.

statistical methods, John Wiley and Sons, New Eccl. Prog. Ser. 75:183-189.

York. 503 pp. Lund, W.A., and B.C. Marcy, Jr. 1975. Early Hostetter, E.B., and T.A. Munroe. 1993. Age, development of the grubby, Afyoxocephalus growth, and reproduction of tautog, Tautoga onitis acnaeus(Mitchill). Biol. Bull. 149:373-383.

(Labridae: Perciformes) from coastal waters of Luo, J., and J.A. Musick, 1991. Reproductive biology Virginia. Fish. Bull., U.S. 91:45-64. of bay anchovy in Chesapeake Bay. Trans. Am.

Houde, E.D., J.C. Gamble, S.E. Dorsey, and J.H. Fish. Soc. 120:701-710.

Cowan, Jr. 1994. Drifting mesocosms: the MacLeod, R.E.1995. Job 1: marine angler survey. In influence of gelatinous zooplankton on mortality of A study of marine recreational fisheries in bay anchovy, Anchoa mitchilli, eggs and yolk-sac Connecticut. Federal aid to sports fish restoration.

larvae. ICE 9 J. Mar. Sci. 51:383-394. F54R final report. March 1,1989-February 28, Fish Ecology 45

1995. CT Dept. Envir. Prot., Bur. Na:. Res., Fish. Miller, NNECO, to S.J. Holbrook, Commissioner, Div. 25 pp. CT DEP 12 pp.

McConnaughey, R.A., and L.L. Conquest. 1993. NOAA. 1993. Tide current tables,1994 Atlantic Trawl survey estimation using a comparr;ive coast of North America. U.S. Dept. of approach based on lognormal theory. Fish. Bull., Commerce, National Oceanic and Atmospheric U.S. 91:107-118. Administration, National Ocean Service. 201 pp.

McHugh, J.L. 1967. Estuarine nekton. Pages 581 NUSCO (Northeast Utilities Service Company). 1986.

620 in G.H. LaufT, ed. Estuaries. Amer. Assoc. The effectiveness of the Millstone Unit I sluiceway Advan. Sci. Publ. 83, ip returning impinged organisms to Long Island Middaugh, D.P. 1981. Reproductive ecology and Sound.18 pp.

spawning periodicity of the Atlantic silverside, NUSCO. 1987. Fish ecology. In Monitoring the Menidia menidia (Pisces: Atherinidae). Copeia marine environment of Long Island Sound at 1981:766-776. Millstone Nuclear Power Station, Waterford, CT.  !

Miller, D.1958. A key to some of the more common Annual report, 1987.102 pp.  !

larval fishes of the Gu!f of Maine. Woods Hole NUSCO.1988a. Delta distribution. Pages 311-320 in Lab. Ms. Rep. 58-1. 56 pp. Monitoring the marine environment of Long 1; land i

Monteleone, D.M. 1992. Seasonality and abun-Sound at Millstone Nuclear Power Station, (

dance of ichthyoplankton in Great South Bay, Waterford, CT. Annual report,1987.

New York. Estuaries 12:230-238. l NUSCO.1988b. Hydrothermal studies. Pages 323 Monteleone, D.M., and W.T. Peterson. 1987.  ;

355 in Monitoring the marine environment of Long  ;

Interannual fluctuations in the density of sand Island Sound at Millstone Nuclear Power Station, lance, Ammodytes americanus, larvae in Long Waterford, CT. Annual repon,1987.

Island Sound, 1951 1983. Estuaries 15:246-254.

NUSCO. 1988c. The effectiveness of the Millstone Morgan, R.P.11, B.M. Baker, and J.H. Howard. 1995. Unit 3 fish rerum system. Appendix l to Enclosure Genetic stmeture of bay anchovy (Anchoa 3 to Letter D01830 dated January 29,1988 from mitchilli) populations in Chesapeake Bay. E.J. Mroczka, NUSCO, to L. Carothers, Estuaries 18:482-493. Commissioner, CT DEP 21 pp.

MRI (Marine Research, Inc.). 1994. Brayton Point NUSCO. 1992a. Niantic Bay current studies.

Station biological and hydrological repon.

Pages 317 331 in Monitoring the marine l January-December 1993. Submitted to New environment of Long Island Sound at Millstone England Power Co. Nuclear Power Station, Waterford, CT. Annual Newberger, T.A., and E.D. Houde. 1995. Population repon,1991.

biology of bay anchovy Anchoa mitchilliin the mid NUSCO.1992b. Progress repon on the MNPS fish Chesapeake Bay. Mar. Ecol. Prog. Ser. I16:25-37. retum systems. Enclosure 2 to letter D05905 dated Nizinski, M.S., B.B. Collette, and B.B. Washington. October 22,1992 from J.F. Opeka, NUSCO, to T.

1990. Separation of two species of sand lance, Keeney, Commissioner, CT DEP.

Ammodytes amcricanus and A. dubius, in the NUSCO. 1994a. Fish ecology. Pages 113132 in Western Nonh Atlantic. Fish. Bull., U.S. 88:241 Monitoring the marine environment of Long Island 255.

Sound at Millstone Nuclear Power Station, NNECO (Nonheast Nuclear Energy Company). Waterford, CT. Annual report,1993.

1995a. Ecological studies proposed for 1996 at NUSCO.1994b. Progress repon on the MNPS fish Millstone Nuclear Power Station. Enclosure 1 to return systems. Enclosure 1 to letter D08071 dated Letter D08923 dated July 25, 1995 from D.B. October 20,1994 from D. Miller, NNECO, to T.

Miller, NNECO, to S.J. Holbrook, Commissioner, Keeney, Commissioner, CT DEP.

CT DEP. 33 pp. NUSCO. 1995. Fish ecology. Pages93-121 in NNECO.1995b. Letter D09375 dated November 29, Monitoring the marine environment of Long Island 1995 from D.B.' Miller, NNECO, to S.J. Holbrook, Sound at Millstone Nuclear Power Station, Commissioner, CT DEP. Waterford, CT. Annual report,1994.

NNECO.1996. Ecological studies proposed for 1997 NUSCO. 1996. Fish ecology. Pages 199-231 in at Millstone Nuclear Power Station. Enclosure 1 to Monitoring the marec environment of Long Island Letter D10084 dated July 15, 1996 from D.B. Sound at Millstone Nuclear Power Station, Waterford, CT. Annual report,1995.

46 Monitoring Studies,1996

Olla, B.L, A.J. Bejda, and A.D. Manin.1974. Daily marine fishes having different longevities. Mech.

activity, movements, feeding, and seasonal Ageing Devel. 18:59-66.

occurrence in the tautog, Tautoga onitis. Fish. Reisman, H.M., G.L. Fletcher, M.H. Kao, and M.A.

Bull., U.S. 72:27-35.

Shears. 1987. Antifreeze proteins in the grubby Olla, B.L., A.J. Bejda, and A.D. Manin. 1975. sculpin, Affoxocephalus aenaeus and the tomcod, Activity, movements, and feeding behavior of the Aficrogadus tomcod; comparisons of seasonal cunner, Tautogolabrus aspersus, and comparison cycles. Envir. Biol. Fish. 18:295 301.

of food habits with young tautog, Tautoga onitis, Richards, S.W.1959. Pelagic fish eggs and larvae of off Long Is'and, New York. Fish. Bull., U.S. kong Island Sound. Bull. Bingham Oceanogr.

73:895-900. Coll.17:95-124.

Olla, B.L., A.J. Bejda, and A.D. Manin 1979. Richards, S.W.1963. The demersal fish population of Seasonal dispersal and habitat selection of cunner, Long Island Sound. Bull. Bingham Oceanogr.

Tautogolabrus adspersus, and young tautog, Coll.18:1 101.

Tautoga onitis, of Long Island, New York. Fish. Richards, S.W. 1982. Aspects of the biology of Bull., U.S. 77:255-262.

Ammo @tes americanus from the St. Lawrence Olla, B.L., and C. Samet. 1977. Courtship and River to Chesapeake Bay, 1972-75, including a spawning behavior of the tautog, Tautoga onitis comparison of the Long Island Sound postlarvae (Pisces: Labridae), under laboratory conditions. i with Ammo @tes dubius. J. Northw. Atl. Fish. Sci. '

Fish. Bull., U.S. 75:585-599. 3:93-104.

Olla, B.L., and C. Samet. 1978. Effects of elevated  !

Sampson, R. 1981. Connecticut marine recreational temperature on early embryonic development of l fisheries survey 1979-1980. CT Dept. Envir. Prot., j the tautog, Tautoga onitis. Trans. Am. Fish. Soc. Mar. Fish. 49 pp. i 107:820-824.

SAS Institute, Inc.1985. SAS user's guide: statistics.

Peebles, E.B., J.R. Hall, and S.G. Tolley. 1996. Egg l Version 5 ed. SAS Institute Inc. Cary, NC. 956 production by the bay anchovy Anchoa mitchilli pp.

in relation to adult and larval prey fields. Mar. Scott, W.B., and M.G. Scott. 1988. Atlantic fishes of Ecol. Prog. Ser. 131:61-73.

Caaada. Can. Bull. Fish. Aquat. Sci. 219. 731 pp.

Pennington, M. 1983. Efficient estimators of Sen, P.K. 1968. Estimator of the regression abundance for fish plankton surveys. Biometrics ,

coefficient based on Kendall's tau. Amer. Stat. '

39:281-256.

Assoc. 63:1379-1389. I Pennington, M.1986. Some statistical techniques for Serchuk, F.M. 1972. The ecology of the cunner, estimating abundance indices from trawl surveys. Tautogolabrus aspersus (Walbaum) (Pisces:

Fish. Bull., U.S. 84:519-525.

Labridae), in the Weweantic River Estuary, l Pottle, R.A., and J.M. Green. 1979a. Field Wareham, Massachusetts. M.S. Thesis, Univ. of observations on the reproductive behaviour of the Massachusetts Amherst, MA.111 pp.

cunner, Tautogolabrus adpersus (Walbaum), in Sherman, K., C. Jones, L. Sullivan, W. Smith, P. i Newfoundland. Can.J.Zool.57:247-256. Berrien, and L. Ejsymont.1981. Congruent shifts Pot;le, R.A., and J.M. Green. 1979b. Territorial in sand eel abundance in westem and eastem North behaviour of the north temperate labrid, Atlantic ecosystems. Nature (London) 291:486-Tautogolabrus aspersus. Can. J. Zool. 57:2337- 489.

2347. Simpson, D.G. 1989. Population dynamics of the Purcell., J.E., D.A. Nemazie, S.E. Dorsey, E.D.

tautog, Tautoga onitis, in Long Island Sound. M.S.

Houde, and J.C. Gamble. 1994. Predation 7hesis. Southem Connecticut State Univ., New monality of bay anchovy Anchoa mitchillieggs and Haven, CT. 65 pp.

larvae due to scyphomedusae and etenophores it; Simpson, D.G., M.W. Johnson, and K. Gottschal.

Chesapeake Bay. Mar. Ecol. Prog. Ser. 114:47-58. 1995. Job 2: Marine finfish survey. Pages 27-53 Reay, PJ.1970. Synopsis of biological data on North in A study of marine recreational fisheries in Atlantic sand eels of the genus Ammodytes. (A. Connecticut. CT Dept. Envir. Prot., Mar. Fish.

tobianus, A. dubius, A. americanus and A. Smigielski, A.S., T.A. Halavik, LJ. Buckley, S.M.

marinus). FAO Fish. Synop. No. 82. 28 pp. Drew, and G.C. Laurence. 1984. Spawning, Regan, J.D., W.L Carrier, C. Samet, and B.L. Olla. embryo development and growth of the American 1982. Photoreactivation in two closely related Fish Ecology 47

sand lance Ammodytes americanus in the laboratory. Mar. Ecol. Prog. Ser. 14:287-292.

Smith, C.L. 1985. Toe inland fishes of New York State. NY State Dept Envir. Conserv. Albany, NY. 522 pp.

Smith, E.M., E.C. Mariani, A.P. Petrillo, L.A. Gunn, and M.S. Alexander.1989. Principal fisheries of Long Island Sound, 1961-1985. CT Dept. Envir.

Prot., Mar. Fish. 47 pp. .

Sogard, S.M., K.W. Able, and M.P. Fahay. 1992.

Early life history of the tautog Tautoga onitis in the Mid-Atlantic Bight. Fish. Bull., U.S. 90:529-539.

Sokal, R.R. and FJ. RoMf.1969. Biometry. W.H.

Freeman and Company, San Francisco. 775 pp.

Stevenson, R.A.1958. The biology of the anchovies Anchoa mischilli and Anchoa hepsetus in Delaware Bay. M.S. Thesis. Univ. Delaware, Newark, DE.

56 pp.

Tracy, H.C.1910. Annotated list of the fishes known to inhabit the waters of Rhode Island. R.I. Ann.

Rep. Comm. Inland Fish. 40:35-176.

Vouglitois, J.J., K.W. Able, RJ. Kurtz, and K.A.

Tighe. 1987. Life history and population dynamics of the bay anchovy in New Jersey.

Trans. Am. Fish. Soc. I16:141-153.

Westin, D.T., K J. Abernethy,1.E. Meller, and B.A.

Rogers. 1979. Some aspects of biology of the American sand lance, Ammodytes americanus.

Trans. Am. Fish. Soc. 108:328-331.

Wheatland, S.B.1956. Oceanography of Long Island Sound.1952-1954.11. Pelagic fish eggs and larvae.

Bull. Bingham Oceanogr. Coll. 15:234-314.

Williams, G.C. 1967. Identification and seasonal size changes of eggs of the labrid fishes, Tautogolabrus adrpersus and Tautog onitis, of Long Island Sound.

Copeia 1967:452-453.

Williams, G.C., D.C. Williams, and R.J. Miller. 1973.

Mortality rates of planktonic eggs of the cunner, Tautogolabrus adspersus (Walbaum), in Long Island Sound. Pages 181-195 in A. Pacheco, ed.

Proceedings of a workshop on egg, larval and juvenile stages of fish in Atlantic coast estuaries.

Nat. Mar. Fish. Serv., Mid. Atl. Coast. Fish. Ctr.

Tech. Pub. No.1.

Zastrow, C.E., E.D. Houde, and L.G. Morin. 1991.

Spawning, fecundity, hatch-date frequency and '

young-of-the-year growth of bay anchovy Anchoa  !

mitchilliin mid-Chesapeake Bay Mar. Ecol. Prog.

Ser. 73:161-171.

48 Monitoring Studies,1996 i

l

l APPENDIX L List of fishes collected in the Fish Ecology sampling programs (June 1979-May 1995; all stations).

b Scientific name Common name Trawl Seine Ichthyoplankton Acapenser osyrhynchus Atlantic sturgeon

• Alosa aestivalis blueback herring *

• Alosa medwcris hickory shad

• Alosapseudoharengus alewife * *

• Alma sapidissima American shad *

• Alara spp. rivet herring * *

• Aluterus schorpfl orange fliefish

• Ammodytes americanus . American sand lance * *

• Anchoa hepsetus striped anchovy

• Anchoa mitchilli bay anchovy * *

• Anguilla rostrata American cel * *

• Apeltes quadracus fourspine stickleback * *

• Batrdeella chrysoura silver perch

• Bothus acellatus eyed flounder

• Brevoorria tyrannus Atlantic menhaden * *

• Brosme brosme cusk

• l Carans crysos blue runnet *

• Caranxhippos . crevalle Jack Centroprostss striata black sea bass *

• Chaetondon ocellatus spotfin butterflyfish *

• l' Clupcidae herrings *

• Clupea harengus Atlantic herring * *

• Conser oceanicus congereel * *

Cyclopteruslumpus lumpfish *

• r C)*noscion regalss weakfish * *

• Cypronodon variegatus sheepshead minnow *

• l Dactylopterus voittans flying gurnard *

!- Dasyatts centroura roughtail stingray

• Decapterus macarellus mackerei scad * *

, Decapteruspunctatus round scad

• l Enchelyopus cimbrtus fourbeard rockling *

• Etropus microstomus smallmouth flounder *

• Eucinostomus lefroyt mottled mo}arra

• l Fistularia cabacaria bluespotted cornettish *

Fundulus diaphanus banded killifish

• l Fundulus heterochrus mummichog *

• l Fundulus luciar spotfin killifish

• l Fundulus majahs striped killif1sh

• Gadidae codfishes *

• Gadus morhua Atlantic cod *

• Gasterosseus aculeatus threespine stlckleback * * * -

Gasterosteus wheatlandi blackspotted stickleback * * * '

Gobiidae - gobics *

• Gobiosoma ginsburgi seaboard goby

• l liemetrtpterus americanus sen raven *

• l Hippocampus erectus iined seahorse *

• Labridae wresses

• Lactophrys spp. boxfish

• Leiostomus samhurus spot

• Liparis spp, seasnail *

• Lophius americanus goosefish *

• Lucaniaparva rainwater killifish *

• Macrosoarces ameracanus ocean pout

• I Melanogrammes aeglefmus haddock

• l Mentacirrhus sasatilis northern kingfish * * *

( Menidu; heryllmo inland silverside *

• Menadsa menida Atlantic silverside * *

• Merluccous bilinearts silver hake * *

• Mscrogadus somcod Atlantic tomcod *

• i Monacanthus hispidus planchend filefish

• Fish Ecology 49

=- . . ~ ~. . . _ - . ., - . - , . . - . - . - -

l I

APPENDIX 1. (continued).

Scientific name Common name Trawl Se;ne Ichthyoplankton Monocanthus spp. filefish

• Morone americana uhite perch

• Morone saxatdis striped bass *

• Mugilcephalus striped mullet * *

• Murilcurema white mullet

• Mullus auratus red gostfish

• Mustehs canis smooth dogfish

• Myhobardfremmvillet bullnose ray

• Myoxocephalus aenaeus grubby * *

• Myoxocephalus octodecemspinosus longhorn sculpin * *

• Myoxocephalus spp, sculpin

• Ophidiidae cusk-ecis

• Ophidson margmatum striped cusk ect * *

• Ophidion nelshi crested cusk ecl

• Opsanus tau oyster toadfish

• Osmerus mordaz rainbow smelt * *

• Parahchthys dentatus summer flounder

• Parahchthys oblongus fourspot flounder

• Peprilus triacanthus butteriish * *

• Petromyzon marmas sca iamprey

• Phohs gunnellus rock gunnel * *

• Pleuronectes americanus winter flounder * *

• Pleuronectesferrugmeus yellowtail floundet

• Pollachius varens pollock *

• Pomatomus saltatrix bluefish *

• Priacanthus arenatus bigrye

• Priacanthus cruenta.us glasseye snappet

• Pristigenys alta short bigeye

• Prionotus carohnus northern scarobin * *

• Proonotus evolans striped scarobin * *

• Pungitius pungitius ninespine stick lcback * *

• Raja egiantersa cicarnose skate *

• Raja evenacea little skate

• Raja ocellata winter skate

• Salmo trutta brown trout

• Sciaenidae drums

• Scophthalmus aquosus windospane * *

• Scombee scombrus Atlantic mackerel

• Scyltorkmus rettfer chain dogiish

• Selar crumessopthalmus bigeye scad

• Selene setapinnd Atlantic moonfish

• Selene vomer lookdown *

• Synodusfoetens inshore lizardfish

• Sphyraena borechs northern sennet

• Sphoeroides maculatus northern puffer * *

• Squalus acanthias spiny dogiish

• Stenotomus chrysops scup

• Strongylura marina Atlantic needlefish

• Syngnathusfuscus northern pipefish * *

• Tautogolabrus adspersus cunner * * *

• Tautoga omtis tautog * *

• Trachmotusfalcatus permit *

• Trachurus fathami rough scad

• Trachinocephalus myops snakefish

• Trinectes maculatus hogchoker

• Ulvarsa subbtfurcata radiated shanny *

• Upeneusparvus dwarfgoatfish

• Urophycis chuss red hake *

• Urophycss tenuis white hake *

• Urophyris spp. hake * *

• 50 Monitoring Studies,1996

- . - . . . - . . _ . _ _ . _ . . _ _ _ _ . . . . . - . - . . . . _ . . . _ . ~ _

I 1

i aG$*S.g~R CoC ~~ S30032 R=22-2 2**-**-----  %

}N

,. . s'd d d d n

g 7

• e s5.---ee-.ee-e.ere. -.-...........

1 f ~w a 6 e i

e gggg=.._=.g-. .......-~.-...._..-

w-hU h

,- g

• .J. O gg.;g.............-......--..... g e a e e E.g......

4,-

.---.o.....--......~...  !

g v E .

a

. g g g g ;- - w - . .

s y e . . .--.............

p . -J44 A ' ,

j e __,2.,-............--.. ..--..-...

g._ 3 ... e .

)

J -

Jg g . g -

g:.-~=-...._-----......--...

. a .....

[g 8 .

ge--=== g-e.._n....--.......... ,

n - _.

+

ggg-----....=............. ......

j g9 S .

.r -

• sag--- ---- -a----- ---- --- -

J w

]  ;- 3 e j

+ * ,g3,..--...---._...., ........... ,

-l. . -

a

. ,e==.--.......................-..

g'P.w __-t- I.

~

1 -

Eg 4 ~

. .a: s

.:gn-.g,.._.-~................... g-g g - El:= _ _

=~~c-  :---- - ---- --- - -- *

- - w .

5 - W s -e -. gg,2._----.=................_.... -

.: s

.1

- a l g._ ~. .xe,-..--

.g .... ............... .

1, s

93.._- --................_...... g i t -

qg ggg=...-. ......-=..._ .. .....-. g g n -

. ~ ,a

..... g g2 -$~2,g...--..................

'i g -

, ,t 4:

628'.-~.....--...................

5  :

l

.t i

E 3 s- q' 2

= l E 3

<>  ! :: c , u = cl o w a xlo wa[, .iv cau,ww,,=*c.cfk .

!b ; Ill !R -

Fish Ecology Si i

i l

N g APPENDIX IU Tasal number d samples ccJtected and mwnt er d fish caught by trawl at stanon IN during each report +=ar from June 1976 tfweveh Mn 1996 O Year D 7677 77 7s 7s.79 79 30 eas t 81-82 82-43 33-s4 34-95 e5.s6 96-87 tv-se ts-se 89-90 90-9t 91-92 92-93 93-94 94-95 95-96 7asmi

c Nuneber of saunples 7s 78 7s 7s 7s 7e el 7e 73 78 7s 78 78 73 73 73 7 St 7s 7e 35ee Q

g Taman*

00 p empren, , 924 1336 5161 2079 2143 1908 2331 21:2 1452 1907 1499 1844 24 to 1259 153t loos 1621 1532 2160 1975 33,271 3rAry=ye 693 1465 398 421 903 000 14M 1943 1264 398 152B 13t$ 996 747 IM9 7903 3749 277 1948 282 29,622 I a 3 squemme 203 294 107 IM ISO let 2t6 294 199 3t2 322 768 591 682 357 308 616 t000 845 558 t.132 E A6meds syp 287 718 005 728 190 146 75 115 to 144 123 $3 60 26 24 261 19t1 107 112 82 6.0%

JImpe spp 97 90 44 99 144 152 t91 354 IM 312 344 385 473 448 469 543 299 484 408 156 5.977 T adyresws 632 666 227 1922 $96 342 207 76 68 27 9 9 12 35 38 I? 95 8 38 21 4,126 i g Aa:Ans syp 161 $$ $06 0 44 354 1 20 13 1799 99 41 It i 3 16 2 2 3 12 3,446 Gediese il 69 63 82 ell 3t$ 140 194 94 135 35 373 21 20 17 24 36 23 26 50 2,138 M asumros 45 37 $6 72 362 I?6 200 242 76 59 126 35 Ill le 62 47 122 90 147 38 2.001 P ;. L . $ 2 12 3 4 9 4 7 0 3 10 5 929 to 128 60 24 16 I 7 1.439 P essenrs 75 40 to 9 24 49 37 53 to 39 107 125 6I le 63 75 115 los 94 69 I,245 ,

himaasms upp 42 30 30 46 66 72 33 67 38 31 los 27 36 215 76 21 19 42 98 Ile 1.238  !

C senser 8 2 0 3 5 39 13 24 25 43 241 3 32 46 49 35 47 5 277 34 9M M Mismarvs 108 13 2 M tot et 33 $2 26 38 44 4 23 51 47 47 73 $ 38 23 til thyArcis app 2 $ 7 5 21 23 182 45 19 29 11 26 49 25 $9 13 43 59 93 44 760 T esiWs 63 10 86 68 47 27 50 48 46 47 23 17 42 It 16 14 9 24 37 23 7$3 E smeressesses 6 0 0 0 t I? 4 35 14 34 107 39 $9 12 85 32 06 31 96 36 724 t P geme=Nhes 54 28 il 7 35 25 36 27 23 12 21 15 24 22 6 7 12 12 27 to dit N eurrdasses  ? 5 Il 19 62 96 Its 60 16 7 I I O 2 2 1 0 0 0 $ 410 ,

O s=wdsr i 6 le 0 9 19 29 2 4 16 9 4 4 4 16 6 334 I II  ? 296 A p. _m i 2t6 9 3 0 4 0 l 0 3 4 I 1 0 0 2 7 17 1 e 2 72 K jlemus 0 14 9 12 24 18 12 12 25 ll  ? 15 13 7 5 9 to It S 7 229 Clupeeene 1 0 0 0 0 0 0 0 0 110 0 0 0 0 0 0 0 0 2 0 It3 ,

5 a==wassen 2 1 0 e t 0 0 2 0 0 1 0 4 3 3 22 16 8 7 5 73 C@ 0 I 6 le 4 0 4 0 4 0 0 7 2 0 4 0 I O O O $7  ;

C egro6s I 8 0 1 0 3 4 0 0 1 4 2 le 0 4 8 I _O I 2 *37 8 eprennes t 0 t t 0 0 0 0 0 0 0 0 0 0 0 5 24 0 3 0 31 P seAwierws t 0 t 6 3 I I 2 5 3 7 1 2 0 0 0 1 2 5 1 32 L re-.s app I 3 4 2 2 2 5 4 3 0 0 0 4 0 0 0 0 0 0 0 27 M amerkmee 2 7 i O 6 5 0 0 0 0 0 0 1 2 0 2 0 t i I 25 A. syndishes 15 2 0 0 4 0 0 t 0 0 0 0 0 0 0 0 0 0 t t 24 M ary=A,s 2 0 2 0 0 0 1 0 3 a 0 0 1 2 2 0 0 3 0 is a Anne 1 0 0 0 0 0 0 1 0 0 1 4 e 2 3 0 1 0 0 2 46 N. rerrves 0 0 0 0 0 0 0 0 0 2 1 0 1 0 2 6 0 0 0 4 16 M al. 0 0 1 2 1 6 3 1 0 0 0 I O O O O O O O O 16 A seimises 0 0 0 0 0 0 0 9 0 0 2 I e 0 0 0 1 0 0 0 14 G arwArarms 1 0 I I O 2 I I I O O O O O O 4 2 1 0 I 13 O are 1 0 0 4 4 I I O 2 0 I I O O O O O l 0 0 13 M emmenas 0 0 0 0 0 I O O O O I O I O O O O $ I 2 18 4

M ar2 rJIse=s 0 0 0 0 0 0 2 0 0 0 0 0 3 0 0 0 0 $ 0 0 It  ;

Am .pp 9 0 0 0 0 0 0 0 0 0 0 0 t I O 7 4 0 0 0 IO L A emernisses e 2 0 1 1 0 2 0 0 2 1 0 0 0 0 0 0 0 0 0 9 A rouman, I e 1 0 0 0 I O 2 I O t 0 0 0 1 0 0 1 0 9 F. asAmwas I t 0 0 t 0 0 0 0 0 0 0 0 1 3 I e t 0 0 9 A scherpft 0 0 0 0 0 0 0 0 1 1 0 0 0 2 1 0 1 0 0 I e P eenensspes 0 0 0 0 0 0 0 0 0 0 0 3 2 2 l 0 0 0 0 0 8 f

1 I

I i

L

. , . , , , . _ . _ .,-,.e_ - -. , , , . ~ ,

T' i

.J i

2 4

(

k 1

i n .

e 3 s

d E i

1 2

......................... = ,

i 4

t,

. 4 3

1 , , .......................... ,  ;

4 -

5 3 g

l 1

i  ; l 3

1

......................... =

e i 3 t =

, ......................... g j , , ......................... g 5

, y e

d =

, , ......................... g 1

, A a

g

5

. p i 3

~

e

>. x l h!

. c a w n. c z , v u w v a z .1

. e I

I )

www=  !

I I 1 j

]

I 4

1 '

i l

Fish Ecology 53 l k <

a i

I I

a

1 ja

.a .s s_ss -- ~x z 5==Rr_ssenas:::=mnessess---~~

. - s.

m . aaaa _s. _._- _e .I

-~ _ _

I e ggresg Essa-esam-s~-~-----***----*----**-**=

g a gg: =g-nsgg =na --n-- z-- - ------ ----------  !

t o s e s.es ===vas_= =ns:-

-~__ _ =s-~~=----~-- --- ~~~------

s e g .n g a _r n _e _= e s _e s r = - - s e s = * = - - - - - - - - - - - * - - - ~ ~ * - - -

g _

g e gg~=g

sse:-ca: --=- -s---a-- -----~~- ---- -

.=

g,gr e =gn=33-==un

- sess-- ---

-- = ---~ --- ---~~- ,

ge ggag: gar sa-a=a=g----------------- ------ - -

,1,s e ges,=_===_.e

- ~_ s=_,an ----- =----- ---=-~~~-~ ---

.__:=_ss=:s:s_s n =nar--

.l  : e

, = ===-~---- --------------

w -

g e e-a. _g n ssssze: ~_ m:n---=s- :=----~-----------~~-~-

= e i=sna_sn=: sex =a==--n:---- s----- --- ---------

.s

. - _-_ _ _ o

e. :U*#~:':-*********-~**********"

,, E8::: . - -- 5:8::88_8 - _

,a e s.x:::

_-~ __ es_a=gnase:-s----- == ---- - -------- --

• :=m

_- ns-sggrass== g-- --- -- ---

t g .

a2 e =:n :=-a-sg=

- - a ma===------------------- - ----

i l-e gsg gs:ssgan nn--s- s-------~- -------- ----- '

a e t

.asI-rea_s=:s-

-.~ :ce-a--- --- ----------------- - )

e e

_gggas==::::=t=se ----m-----

e -

)

e e -s=

s_. ss

_ _me:=__s==sa-- _ -~e--------~=-------- ----

t - )

n.-e.

3 2 as gsa= s a- ~= s= =- ---------- -~~-----

1 E

I

< >I I S=.4 r l e = w o. a e s =k !wu==w,ucesc=,w_I I I 3 c,2=,,amses,eJ

$4 Monitoring Studies,1996

5 5

i

. ... .,em

.. ee n n n e. n - - - - - - - - - - - - . - -

] ~8 I.

> & t O >

g g p

g ; . . . ...........-.............-..

g e . . . ............................ g

$ H e . . . ............................ g

.z a g , .. . ............................ g

y - . . n..e...e-..-..ee..e......... y 8 l 2

e g g a )

a . eena.. .......-............... g j

i 2

e.

s a g on--........................... g

s. .

1 A 3

a g , ......-.-......-............... g

. ................-.............. 3 A

L e y

......n........................ -

i 3 b P=

............................... . m N

^

A 0 y a

t s

~

g  ! i ll  ! $

>: o= vw.-2a.seuew&26 , , c c c . a a : . 2 .I.c.f

. E .

l Fish Ecology 55

a g2" IUS;Z"2*: a 3.2IR

,J s I

g.- . .4,E 3 -

w-a .: ---#.3 as- *R==38330%U20:0:23:022:

nn4----

f , 2 gagg3 -a=a*R

- Rz:*"g***=2**-**a-~**-*******

• I ,s R  ::gs : s a =c:a 42=-*2'5:=----*---**---*-****

'h, -

, l a,

-I.e,E-=-g=.e=,. .....

- 2~.=. ..--.~.-~..-.-**

S, p 2 2 A43R:n-=A"25*22*a-aa"- * * *

-g s n4 wen

=

- - 2"*-*"****"***

A g R 2 .nn- r cR823 e =g=rR j5

- - "a *"R:Q':***3*******a *-**-*

h ; R gtaa Kd .. 32 tgt:3aag220 '**:23:*2-*********-****-**

g g ggggg-t U.a m

aa g.Eggegg=.

-E .a-..aeoona-.....--..o*e 8 2 c;Mn323*3 1,5 W"R

...--- *CR*2*2"*":2-8 -"*"***--**-"*

-  : a d n.q R 8,mge*nRR2 2 2 a: " ;R **R***N******a ******-*****

I; R

.l d .BSII22n*3"222M2

.s

. n- - R'*23*--"**"2*'***-*****-**

! 3 R REE*Rt:3a 2=gga2= 2*h:*-2**"--a-***--****-*

3 -- -- *

2

• 14 5. : ~E~~ ; 3 R g g a * * = 2 " A g 2 a * :

-*aa

• • • a-**"**********

• 2 2 3, sages **-a-2:==ns-*a***-~ :-----*******-***

5 -

q ; 3 ):,m enRE #Rg:'RKASARR*2"*=**22**-****a-*"**"*

." a T O R j - 2R5082J13*

0-a --- $g2223:*M"*="*"*"*-**----**-****

E g g - g a

g 3 2-Rm393'*-=p*22****"**-'**"*"-*--*********

$ R IM

,2 g 3%2242*"*;**':-***-****"""*******-*****":

2 2 R

• ;- =2=nsa*-a==2-*a*******R*****--**a*******

R e

-.a:ast: amas=2*2 a - m'3*==***************-******

4 : 2 ca

-atasa"2s2*m-R** -22*=*a -******a k aa aa****-**

• i l

s Y

Z k wt )- s.]a%(=,ec k .w

$ 3

.sewe I L

,v,wce,===lwwa,e 4 wans sv 56 Monitoring Studies,1996

.... . .-~~~~~~~----------- g

~~

s E g n ....~..................-... y t -

, e ..-....--.................. g 3 --

g ........................... g c a g , ................... ....... .

i  ?

g n ~.....-.... .~............. -

i 0 y n .--.-.......-....-......... y i

g e ...........................  :

i  ;

e ......-...--............... g a  ;

e ~~-........................ g

, n .....-...............-..... g b ~~

g e g

..............~............ .

, e -....--...... ..............  ;

i I g e g

-.~....-................... .'

i 2 g , .........-................. g E '

g g e ....-...................... g i -

e , -...... ............-...... ,

d O e e ..-.-..-................... g i

g

" ls

e .........................-. a* -

e -

8 2

s t i

3 w <t lil

< > ~! hl]ld :ce:INd.Il{Ia;1t}

       - vocw vuma         46m   cviaiss . .!. .a Fish Ecology 57

i APPENDIX VI. Total number of days the MNPS Unit I sluiceway was not in service by month from 1985 through 1995. l Month 1985 1986 1987 1988 1989 1990 l 1991 1992 1993 1994 1995 January 0 0 0 0 2 0 ' 0* 4 4' 3 February 0 0 0 0 6 0 ' 0' 2 16' 5 March 0 0 1 0 31 * ] 2 10 7 0' April 0 0 0  ! 21 0 15* 3' O O 12 5 May 3 8 2 0 31' i 18 O O Jww O 10 4 1 6' 8 6 l' * O July O 6 1 2 6 31' 10 3 0 0' 0' 2 3 0 August 8 21 10' 3 6 4 15' 6' 4 4 September 28 3 4 9 18 11 3* 22 9 5 2 17 October 7 5 0 15 0 18' 4' 4 2 2 4 November 30' 0 3 3 2 6 4' 6 4 20 0' December 21' O O O O 8 0' 7 3 5 0'

• Unit I shutdown.
• No information available (NUSCO 1992b).
• No information for June 1 15,1991; sluiceway in service from June 16-30,1991.

l i 1 i 58 MonitormgStudies,1996

Winter Flounder Studies Introduction..................................................................................................................................61 Materials and Methods . .. .. .. .. ... ... .. . . .... .. .. .. ...... . . . ... ..... . . . . .... . ... .. . ... . .. ... ... . .. . ... ... .. .. ... . .. . .. ... Sampling Programs . . . . . . . . . . . . . . . . . ... . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . .. . . Adult Winter Flounder Sampling ............... ........................................................... 63 Larval Winter Flounder Sampling .......................................................................... 63 Juvenile Winter Flounder Sampling ....................................................................... 65 Indices of Abundance .. ..... . ..... .. ... ..... .. ..... .... ... . .... ... . .. ...... . .... .. . . .. ... ..... . . . . .. .. .. . .. . .. .

Relative Annual Abundance of Adults ................................................................... 66 Absolute Abundance Estimates of Adults ............................................................. 66 Adult Spawning Stock Size and Egg Production ......................................... .......... 66 Development and Growth, Abundance, and Mortality ofLarvae ........................... 67 Abundance, Growth, and Mortality of Juveniles in Summer ................................. 68 Abundance of Juveniles during Fall and Winter ...... .............................................. 69 Stock and Recruitment Relationship and Biological Reference Points .................. 69 Assessment of MNPS Operation on Niantic River Winter Flounder ................................ 71 Estimates of Larval Entrainment at MNPS ............................................................. 72 Mass-Balance Calcula tions ... . .. . ........... .. .. . . ... . . .. ... .. . .. .. .. . ..... ... ... ... . . . .. . ... ..... .

Stochastic Simulation of Winter Flounder Stock D Resul ts and Di sc ussion . .. . .. .. ...... ... ..... ........ .... ..... ....... . .. .... ........ . .ynamics . .

                                                                                                       .................................................82 Seawater Temperature .... ... . . .. ............ ..... .. .. .. ....... . .. . ..... . ... . ... ........ .. ...... .. .. .. .. ..... .. ...... . . .. . . .. 8 2 Ad ul t Winter Flo under ... ........... ..... .... ....... ... .... ... .... . . ..... .. .. . . ... .. . .. .. ... .. ... ..... . .. . .

Relati ve Annual Abundance . ........ ... . ... .. . .. ...... ... . . . .. . ... ... . . . .. . ..... ... ....... .. . ...... .. Absol ute A bundance Estimates .......... ............ . . .... . . ... .. .. .... . . . . . .... . ... .... .... ...... . . . Spawning Stock Size and Egg Production ........................................................... Larval Winter Flounder .. ... . .... .... .... .. ....... ...... .. ....... .. 92 ....... .... .. Abundance and Distri bution ........... ........ ....... . ... . . ..... . . ....... ........... . ... .. .. ... . . ... .. . .. . Development and Growth ..... .... ......................... ... .. . . .. ........ ... .. . .. .... .. ... . ....... . .. ..... . .. 98 Mortality............................................................................................................. J u venile Winter Flounder . . . ................... ... .... ........ .106.. .. .... ..... . Age-0 Juveniles during Summer ... .................................................... ............ ........ 1 % Age-0 Juveniles during Late Fall and Early Winter .............................................113 Age- 1 Juveniles during Late Winter ...................................................................... 1 15 Comparisons among Life-Stages of Wm' ter Flounder Year-Classes ...............................117 Stock-Recruitment Relationship (SRR) ................................ .........................................121 MNPS Impact Assessment ... . .... . . ..... .. ........... .. .. . . .. . . .. . . .. . .. . . . .. . . . . .. ... . .. . . . .. .... .. . . . .... ... . . ... .. . . Larval Entrainment ...... ............ . . .... . . .. .. .... ..... .. ... ....... . ... .. .. . ... .. .. . ..... .. . .. .. .. . . Stochastic Simulation of the Niantic River Winter Flounder Co ncl us i ons . . . . . . . .. . . . . . . . . . . . . . . . .. . . . .. . . .. . . . . . . .... . . . ....... References Cited .. . . . . . . . . . . . . . .. . . ... . . . . . . . . .. . ... ... . .. . .... . . . . ... ... . .

                                                                                                                                                                      . . ..1 44 WinterFlounder 59 1

1 I

s 4 4 60 Monitoring Studies,1996

Winter Flounder Studies i introduction temperature range is 1215'C (McCracken 1963), although a few remain in estuaries, apparently i The winter flounder (Pleuronectes americanus) has avoiding temperatures above 22.5'C by burying been the object of environmental impact studies by themselves in cooler bottom sediments (Olla et al. Northeast Utilities Service Company (NUSCO) at the 1969 Millstone Nuclear Power Station (MNPS) since 1973, have). Other aspects of winter flounder life history been summarized by Klein-MacPhee (1978). It is an important sport and commercial fish in Because the early life history of the congeneric Connecticut (Smith et al.1989) and an abundant European plaice (Pleuronectes platessa) has many member of the local demersal fish community. The similarities to that of the winter flounder, relevant i winter flounder has been reported from Labrador to literature for this species was also reviewed for this G;orgia, but is most numerous in the central part of report to gain further insights into winter flounder its range (Scott and Scott 1988), which includes Long Population dynamics. Island Sound (LIS). Its seasonal movement pattems MNPS operation results in the impingement of and reproductive activity are well-documented (e.g., juvenile and adult winter flounder on the traveling Klein-MacPhee 1978). Most adult fish enter screens of the cooling-water intakes and the entrain-cstuaries in late fall and early winter and spawn in ment of larvae through the condenser cooling-water upper portions of estuaries during late winter and system. The impact of fish impingement at MNPS carly spring at temperatures between I and 10*C has been largely mitigated by the installation and (peaking at 2 5*C) and salinities of 10 to 35 Operation of fish return sluiceways at MNPS Units 1 (Bigelow and Schroeder 1953; Pearcy 1962; Scarlett and 3 (NUSCO 1986b,1988a,1994b). Unlike many and Allen 1992). Three years are required for oocyte marine fishes, mortality of entrained winter flounder mituration (Dunn and Tyler 1969; Dunn 1970; larvae potentially has greater significance as it is a Bunon and idler 1984), in eastern LIS, females Product of local spawning with geographically begin to mature at age 3 and 4 and males at age 2 isolated stocks associated with specific estuaries or (NUSCO 1987). Average fecundity of Niantic River coastal areas (Lobell 1939; Perlmutter 1947; Saila females is about $61,000 eggs per fish. Eggs are 1961). In panicular, the population of winter demersal and hatch in about 15 days, and larval flounder spawning in the nearby Niantic River has development takes about 2 months; both processes been studied in detail to assess the long-term effect of are temperature-dependent. Small larvae are plank. larval entrainment through the MNPS cooling-water tonic and although many remain near the estuarine system. Although the 1996 spawning season was the spawning grounds, others are carried into coastal eleventh year in which winter flounder could have waters by tidal currents (Smith et al.1975; NUSCO been impacted by the operation of all three MNPS 1989; Crawford 1990). Some of the displaced larvae units, the plant was shut down during most of this are returned to the estuary on subsequent incoming period, resulting in the smallest cooling water flow tides, but many of them are swept away from the area since 1985, before Unit 3 went on-line. into coastal waters, where their survival may be Devel0Pment of a long-term assessment capability reduced. Larger larvae maintain some control over was the ultimate goal of NUSCO winter flounder their position by vertical movements and may spend studies. Presently, a combination of various sampl-considerable time on the bottom. Following met. ing programs and analytical methods are used to amorphosis, demersal young-of-the-year winter examine current abundance of the Niantic River fkunder predominantly settle or move into shallow Population and obtain annual estimates of the spawn-inshore waters. Yearlings (age-1 fish) become ing stock. This report summarizes data collected dur-photonegative and most are usually found in deeper ing 1996 and updates results reported previously in witers (Pearcy 1962; McCracken 1963). Some adult NUSCO (1996). A computer population simulation fish remain in estuaries following spawning, while model, the NUSCO winter flounder stochastic others disperse offshore. By summer, most adults Population dynamics model (SPDM), is used for leave warmer shallow waters as their preferred assessing tong-term effects of MNPS operation. The Winter Flounder 61

~ _ __ _ _ - _ _ __ __ __ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __ _. SPDM can simulate the long-term effects of ing program at MNPS. Additionalinformation used I historical and projected rates of fishing mortality and in various assessments was presented in NUSCO I simultaneous plant operation, resulting in annual (1987), which summarized various life history mortalities from impingement ofjuveniles and adults studies of the winter flounder prior to the operation , and the entrainment of larvae through the MNPS of Unit 3. Ongoing sampling programs that contrib-cooling-water system. Results of SPDM simulations uted data to the Niantic River winter flounder studies and a probabilistic risk analysis help to assess the ) are shown in Figure 1, which includes the seasonal s effects of MNPS operation on the Niantic River duration of sampling and timing relative to the I winter flounder population through the year 2060, annqpt life cycle of Niantic River winter flounder, well after the scheduled shutdown of Unit 3 in 2026. Brief descriptions of field methodologies used in these programs are given below. Materials and Methods Information on water temperature (*C) was obtained from continuous temperature recorders at Sampling Programs the intakes of MNPS Units I and 2; daily mean I temperatures were determined from available records Data needed to assess MNPS s,enpact on the winter of 15 min average temperatures. Monthly, seasonal, flounder come from several biological sampling or annual means were calculated using daily means. programs. Some programs (e.g., Niantic River adult Water temperature and salinity measurements at adult I and larval surveys, age-O survey) were designed to (surface and bottom) and larval (surface, midwater, investigate specific life history stages of winter and bottom) sampling stations were recorded using a flounder. Other programs provide information from Rosemont RSS-3 Portable Salinometer or a YSI  ; year-round sampling of the entire local fish Model 30 Salinity / Temperature / Conductivity meter. Temperature at juvenile winter flounder stations in community, such as the trawl monitoring pmgram (TMP) and the entrainment ichthyoplankton monitor- the Niantic River was taken with a mercury ) thermometer. l l l

                                                                                                                                       \

I 5 { 4 - A08

• O Juvenile Survey 3

2 y Winter Flounder Larval Survey 1 Nianbc River Adult Winter Flounder Survey lF lM lAlM lJ lJ lA lSlOlN lDlJ lF lMlA l L, Year- 0 m, Year- 1 yw r

1. February-April sampling (spawning season) for adults and juveniles throughout the Niantic River.

2. February-June larval sampling at three sations in the Niantic River and one in Niatic Bay.

3. Year-round monitoring of allichthyoplankton at the MNPS discharges.

4. Late May-September sampling of age-O juveniles at two stations in the Niantic River.

5. Year-round monitonng of all benthic fishes at six (1976 95) or three (1996) stations near MNPS  ;

(juvenile catch data come from two stations in November, three to four in December, and three to six in January).  ! Figure 1. Current sampling programs contributing data for computation of winter flounder abundance indices (hatched area show months from which data wett med in this report). 62 Monitoring Studies,1996

i

                                                                                                                          +

Adult Winter Flounder Sampling

   ' SamplLng nethodology for the adult winter flounder spawning surveys in the Niantic River has remained essentially . unchanged since 1982. Surveys usually                                            Niantic 54l begin between mid-February and mid-March, after                                                River most ice cover disappears from the river, and                              A continues into April. Sampling ceases when the               ,

proportion of reproductively active females decmases  ; to less than 10% of all females examined for 2 I consecutive weeks, an indication of completion of N c most spawning. In each survey, the Niantic River was divided into a number of sampling areas, s2 referred to as stations (Fig. 2). Since 1979 no p 1m s1 samples have been taken cutside of the navigational , channel in the lower portion of the river because of i an agreement made with the East Lyme-Waterford Shellfish Commission to protect habitat of the bay /y -d e l scallop (Argopecten irradians). Winter flounder sman  ; were collected on at least 2 days of each survey week cow 5 , using a 9.1-m otter trawl with a 6.4 mm bar mesh l l codend liner. Fish caught in each tow were held in i

                                                                                                           -2 water-filled containers aboard the survey vessel before processing. Since 1983, each. fish larger than 20 cm was measured to the nearest mm in total length and its gender ascertained. Before 1983, at least 200                                                                    -

randomly selected winter flounder were measured during each week of sampling. Fish not measured were classified into various length and gender (  ; groupings; at minimum, all winter flounder examined were classified as smaller or larger than 15 cm. Fig. 2. Location of stations sampled in the Niantic River Gender and reproductive condition of larger winter during 1996 for adult winter flounder from February 27 flounder was determined by either observing eggs or  ;

    .                          ..                       through April 3 (numbers) and age-0 winter flounder from milt, or as suggested by Smigielski (1975), noting the May 22 through September 25 (letters).                            I presence (males) or absence (females) of ctenii on                                                                       ,

left-side caudal peduncle scales. Before release, discharges of Units I and 2, depending upon plant  ! healthy fish larger than 15 cm (1977-82) or 20 cm operation and the resulting water flow from the (1983 and after) were marked in a specific location condenser cooling water pumps. This year, because i with a number or letter made by a brass brand cooled of the shutdown of Unit 1 (no pumps in operation)in in liquid nitrogen. Marks and brand location were November 1995, most collections early in the larval ' varied in a manner auch that the year of marking winter flounder season were made at Unit 2 and later > would be apparent in future collections. attemated between Units 2 and 3. Most of these > samples were collected with only one or two Larval Winter Flounder Sampling circulating water pumps in operation. Larvae were collected in a 1.0 x 3.6-m plankton net of 333- m j Winter flounder larvae entrained through the MNPS mesh deployed from a gantry system. Four General t cooling-water system have been sampled at the Oceanic (GO) Model 2030 flowmeters were MNPS discharges (station EN, Fig. 3) since 1976. In positioned in the net mouth to account for horizontal most years, collections usually alternated between the and vertical flow variation; sample volume was WinterFlounder 63 ,

d determined by the average of four volume estimates

.                                                                         during February and March and 333 m for the from the flowmeters. The net was usually deployed for 3 to 4 min (filtering about 200 m'), with variation             remainder of the season. Volume of water filtered in sampling time dependent upon the number of                       was determined frons a single GO flowmeter circulating water pumps in operation and tidal stage.               mounted in the center of each bongo opening. The Sampling frequencies and volume filtered have                       sampler was towed at approximately 2 knots using a l

varied since 1976 (NUSCO 1987,1994a). In 1996, stepwise oblique tow pattern, with equal sampiing time at surface, mid-depth, and near bottom. The sampling was conducted during both day and night length of tow line necessary to samp'.e the mid-water once per week in February and 3 days and nights per week during March through early May and the last 3 and tgttom strata was determined by vtater depth and tow line angle measured with an inclinometer. Nets weeks of June. During the last 4 weeks of May and  ! the first week of June, only one day sample was were towed for 6 min (filtering about 120 m'). One of the duplicate samples from the bongo sampler was { collected each week, as unlike previous outages, the i circulating water pumps were shut off for most of the retained for laboratory processing. When present,  ! jellyfish medusae at the river stations were removed time. All ichthyoplanktop samples, including those described below, were preserved with 10% formalin. from the samples using a 1-cm mesh sieve and their mass estimated volumetrically to the nearest 100 mL. Winter flounder larvae have been collected in Niantic Bay at station NB since 1979 and in the The larval winter flounder sampling schedule for Niantic River at stations A, B, and C since 1983 (Fig. Niantic River and Bay was based on knowledge 3). A 60-cm bongo plankton sampler was weighted gained during previous years and was designed to with a 22.7-kg oceanographic depressor and fitted increase data collection efficiency while minimizing j sampling (NUSCO 1987). Larval sampling at the with 3.3-m long nets with mesh size of 202 pm i Niantic River stations usually begins in early to mid-I a

                                                                                "Y        cr     "g""8   ,

2

                                                       /B N

Niantic 1km 1 ml lc NB EN < ' s I %ntie Minston, p ,n, Bay Blace O Point ( Long Island Sound R B3. Location of stationsidenoted Y b iCHers) sampled for larval wimer flounder during J 996.

]

Monitoring Studies,1996 s

e l 4 i i

 !                       February, in 1996, collections were only taken at          used during each sampling trip; nets were deployed
)                        stations NB and C during most of February because          in a random order. A change to the next larger mesh i                       ice conditions prevented sampling in the upper river       in the four-net sequence was made when fish had i                       until February 23. Daytime tows were made within I
 '                                                                                  grown enough to become retained by it, as use of hour of low slack tide through the end of March.          larger meshes reduced the amount of detritus and During the remainder of the season, until the             algae collected. At each station, four replicate tows disappearance of larvae at each station, tows were        were made, two each with the two nets in use.

i made at night during the second half of a flood tide. Rarely, only three tows were taken at a station From 1983 through 1990, sampling was conducted 2 becage of bad weather or damage to the net. Tow } days a week. Starting in 1991, sampling was reduced distance was estimated by letting out a measured line to 1 day a week (NUSCO 199la). Through 1992, attached to a lead weight as the net was hauled at i station NB was sampled during day and night every d approximately 25 m. min . The length of each tow 1 two weeks during February and at least once a week ] from March through the end of the larval winter was increased from 40 to 100 m in 20- or 40-m j increments at a station as fish abundance decreased flounder season. Beginning in 1993, station NB was over time. However, in years when densities of j sampled weekly only during the day from the start of young were high, maximum tow length at a station

the larval season through March and at night from was 60 or 80 m.

1 April throu, h the remainder of the larval season. Catches from the TMP (see the Fish Ecology i section of this report for methods) were used to { Juvenile Winter Flounder Sampling follow the abundance of age-0 winter flounder during } fall and winter. In addition to the TMP, juvenile ~ information on juvenile (age-0 and age 1) winter winter flounder smaller than 15 cm in length (mostly flounder was obtained from three sources (Fig.1). A age-1) were caught along with adults in the annual special sampling program specifically targeted post. Niantic River spawning stock surveys. Dese fish j larval young-of-the year. A second source of data is were processed similarly as adults, although gender ! the trawl monitoring program (TMP), and the third was usually not specified, and the fish were not i data source is the Niantic River adult spawning branded. When small winter flounder were abun-j abundance surveys, during which winte. flounder dant, a subsample of at least 200 fish was measured i juveniles are collected incidentally. Data on juvenile each survey week; otherwise, all specimens were j fish abundance were available from about May of measured. j their birth year into April of the following year.

Juvenile indices were referred to as age-O when fish j

Indices of Abundance were collected as post-larval young in summer and during the subsequent fall and winter by the TMP. Data from the field sampling programs described i These fish are recorded as age 1 when taken during above were used to calculate annual and seasonal the February April adult spawning surveys. indices of relative abundance. Indices, calculated , ne abundance of post-larval age-0 winter flounder j using various sampling statistics, were computed for has been monitored at two stations (Fig. 2) in the

various life-stages of winter flounder, from newly Niantic River since 1983 (LR) or late 1984 (WA). hatched larvae to adult spawners and also included i

nrough 1992, collections were made weekly, but estimates of egg production. Specifics of each abun. beginning in 1993, sampling frequency was reduced dance index depended upon the particular stage of

to biweekly. Stations were sampled during daylight life, sampling effort, and suitability of the data; a

} from about 2 hours before to I hour after high tide. detailed description of each follows. He indices 1 { Monitoring began in late May and continued through ' enabled timely assessments to be made regarding the the end of September.

current status of the Niantic River winter flounder Young winter flounder were sampled using a 1 m population and many of these data were used with the beom trawl having two tickler chains and nets of 0.8 ,

j SPDM for long-term predictions of MNPS impact. 1.6 ,3.2 , and 6.4-mm mesh. In 1983, triplicate tows For some indices, a long term mean was calculated ) were made at LR using nets ofincreasing larger mesh with variability described by the coefficient of j as the season progressed. Beginning in 1984, two variation (CV; standard deviation divided by the frames with nets of successively larger mesh were mean). e i 3 i WinterFlounder 65 9

_ __ _______ _ __.___ __ _ _ _ -.m _ _ _ ~ _ __ Relative Annual Abundance ofAdults was standardized within each year by replicating as necessary the median CPUE value for a given week The relative annual abundance of winter flounder such that the number of tows used in calculating in the Niantic River during the late February-carly CPUE was the same for each week sampled. April spawning season is determined by trawl catch- A second relative index of abundance was based on per unit-effort (CPUE). An annual relative the size distribution of female fish from adult abundance index with 95% confidence interval (CI) spawning survey catches standardized by variable was calculated by using the A-mean CPUE (NUSCO weekly and yearly effort (i.e., number of tows). 1938c) following data standardization. This repre. Catches were adjusted by sampling effort to insure sents a departure from the median CPUE abundance that each size group of fish'was given equal weight index previously used and will be discussed in more within each week of work, among weeks in each detail below. Components of standardization for survey year, and to adjust for varying effort among CPUE calculation included tow length, tow duration, years (see NUSCO 1989 for more details). To avoid weekly effort, and fish length and gender categories. confusion with the CPUE index, this adjusted catch is Tow distance had been measured using radar, but referred to as " annual standardized catch" throughout more recently with LORAN or differential Global the remainder of this report. The annual standardized Positioning System. Distance was fixed in 1983 catch of females was the basis for the calculation of (with exceptions noted below) because using the annual recruitment and egg production described same tow length at all stations was expected to below, reduce variability in CPUE; previously, tows of variable length had been taken at all stations and Absolute Abundance Estimates ofAdults  ! catch was standardized by time of tow. A distance of 0.55 km was selected as the standard because it Absolute abundance estimates of winter flounder . represented the maximum length of a tow that was spawning in the Niantic River were obtained using formerly possible at station 1. In most years, but mark-and-recapture methodology and the Jolly especially during 1987 and 1989 91, tows one-half or (1%5) stochastic model. This model is among the two-thirds of this length were occasionally taken in most appropriate ones for open populations as long the upper river to avoid overloading the trawl with as basic assumptions are approximately met macroalgae and detritus. Because catch data from (Cormack 1%8; Southwood 1978; Begon 1979; station 2 were used also in the TMP, tows there were Pollock et al.1990). Annual absolute abundance made over 0.69 km, the standard for that particular estimates for Niantic River winter flounder larger sampling program. In 1990, tow distance at station I than 20 cm were calculated by pooling together all was reduced to 0.46 km because of the construction fish marked and released during each annual survey of a new bridge at the mouth of the river, { and by observing the recaptures made in subsequent Catches of winter flounder larger than 15 cm in years. Absolute abundance estimates could not be  ! tows made throughout the spawning surveys were generated for years prior to 1984 because of standardized to either 15-min tows at stations I and 2 uncertainty in data records and ambiguity caused by or 12 min tows at all other stations; a standard tow brands used during early surveys. Estimates of distance was not set prior to 1983. Duration of tows annual population size (N) and other model varied and was usually greater in the lower river than Parameters, including survival ($), recruitment (B), in the upper river because of differences in tidal and sampling intensity (p), were made using the currents and amounts of extraneous material computer program ' JOLLY'(Pollock et al.1990). collected in the trawl, even though distance was similar. To lessen error in the calculation of CPUE, Adult Spawning StockSize data from either exceptionally long or brief tows andEgg Production made prior to 1983 were excluded from the analyses. The minimum fish length of 15 cm used for CPUE The proportion of mature female winter flounder in calculation was smaller than the 20 cm used for mark and recapture estimates described below because of each 0.5-cm length increment beginning at 20 cm data limitations from the 1977 82 surveys. Effort was estimated from qualitative observations of reproductive condition (percent maturity by 0.5-mm  ; 66 Monitoring Studies,1996 i l l l

size-classes) made from 1981 through the present. Stage 5. Transformation tojuvenile stage Pooled estimates were adjusted to give continuously complete and intense pigmentation increasing fractions of mature fish through 34 cm; all present near the base of the caudal fin. females this length or larger were considered to be Larval data analyses were based on standardized mature. De fecundity (annual egg production per d densities (number 500m of water sampled). A female) was estimated for each 0.5-cm size-class by geometric mean of weekly densities was used in using the follov/ing ietationship determined for analyses because the data generally followed a Niantic River winter flounder (NUSCO 1987): lognormal distribution (McConnaughey and Conquest 1993) and weekly sampling frequencies fecundity = 0.0824.(length in cm)* (1) varied among some stations and years. Because older larvae apparently remained near the bottom nis relationship was used with the annual standard- during the day and were not as susceptible to ized catch of mature females and their length com- entrainment or the bongo sampler, data from daylight position to calculate egg production. Annual mean samples collected after March at stations EN and NB fecundity was determined by dividing the sum of all were excluded from abundance calculations, except individual egg production estimates by the standard- for estimating entrainment at MNPS. During May ized catch of females spawning per year. Absolute and June of 1996, when night collections could not estimates of spawning females and corresponding be taken at station EN due to insufficient water flow, annual egg production estimates for 1977 through data from night collections at station NB were used 1996 were determined by assuming that the relative for abundance and entrainment estimates. values represented 4.0% of the absolute values (see The distribution oflarval abundance data over time Absolute Abundance Estimates in Results and is usually skewed because densities increase rapidly  ! Discussion for how this fraction was determined). to a maximum and then decline slowly. A Annual estimates of the number of female spawners cumulative density over time from this type of were also used in the derivation of a relationship distribution results in a sigmoid-shaped curve, where between stock and recruitmlnt for Niantic River the time of peak abundance coincides with the winter flounder. inflection point. De Gompertz function (Draper and Smith 1981; Gendron 1989) was used to describe this Development and Growth, Abundance, cumulative abundance distribution because the andMortality ofLarvae inflection point of this function is not constrained to the mid-point of the sigmoid curve. He form of the Ichthyoplankton samples were split to at least one- mPednh used war half volume in the laboratory. Sample material was viewed through a dissecting microscope and winter C, - a exp(-exp[-x-{t-p))) (2) flounder larvae were removed and counted. Up to 50 randomly selected larvae were measured to the where C, = cumulative density at time t nearest 0.1 mm in standard length (snout tip t t = time in days from February 15 notochord tip). He developmental stage of each a = total or asymptotic cumulative density measured larva was recorded using the following p = inflection point scaled in days since February 15 identification criteria: x = shape parameter Stage 1. Yolk-sac present or eyes not pigmented (yolk-sac larvae); he time of peak abundance was estimated by the Stage 2. Eyes pigmented, no yolk-sac pr:sent, parameter p. The origin of the time scale was set to no fin ray development, and no flexion February 15, which is the approximate date when of the notochord; w nter flounder larvae first appear in the Niantic Stage 3. Fin rays present and flexion of the River. Least-squares estimates, standard errors, and notochord began, but left eye not yet asymptotic 95% Cls for these parameters were migrated to the midline; obtained by fitting the above equation to the Stage 4. Left eye reached the midline, but cumulative abundance data using nonlinear juvenile charactenstics not present; regression methods (SAS Institute Inc. 1985). Cumulative data were obtained as the running sums WinterFlounder 67

b of the weekly geometric means of the abundance The presence of density-dependent mortality was data, he a parameter of the cumulative curve was investigated by relating annual larval abundance in used as an index to compare annual abundances. the 7-mm and larger size-classes from station EN to A " density" function was derived from the first the annual egg production estimate for the Niantic derivative of the Gompertz function (Eq. 2) with River using the following relationship (Ricker 1975): respect to time. His density function, which directly describes the larval abundance over time (abundance log,(L / E) = a + b E curve), has the form: (4) wherg L = annual larval abundance oflarvae 7 mm i d, = u* k.exp(-exp[-x.{t-p)]- x-[t-p]) (3) and larger at EN as estimated by a (see  ! Equation 2) where d, = density at time i and all the other para. E = annual estimate of egg production in the meters are as described for Equation 2, except for a*, Niantic River which was re-scaled by a factor of 7 (i.e., n' = 7a) a = intercept because the cumulative densities were based on b = slope or index of mortality dependence upon j weekly geometric means and, thus, accounted for a annual egg abundance ' 7-day period. Since the ratio L divided by E represents the fraction Larval mortality rates were estimated from data of larvae surviving from eggs to 7 mm, density-collected at the three Niantic River stations. Data dependent mortality may be assumed when the slope i from 1983 were excluded as smaller larvae were (b)is significantly different fro _m zero_. His mortal-

• undersampled then because of net extrusion (NUSCO ity is compensatory when the slope b is negative and 1987). - The abundance of 3 mm and smaller larvae depensatory if positive. l was used to calculate an index of newly-hatched Regression analyses were used to examine possible larvae because 3 mm was the approximate length at relationships between variables and, at times, to hatching. De decline in the frequency of larvae in make predictions. Ordinary least-squares linear
• progressively larger size-classes (in I mm groups) regression was used when the independent variable was attributed to both natural mortality and as a was arsumed to be measured without error (e.g.,

result of tidal flushing from the river. Hess et al. water temperature). The test of a relationship was (1975) estimated the loss of larvae from the entire based on the slope being significantly (p s 0.05) river as 4% per tidal cycle and also determined that different from zero. Functional regression methods the loss from the lower portion of the river was about developed by Ricker (1973,1984) were used in the 28% per tidal cycle. Hus, the weekly abundance cases where the independent variable was measured estimates of larvae 3 mm and smaller at station C in with error (e.g., abundance indices). For functional the lower portion of the river were re-scaled by a regressions, the probability that the correlation factor of 1.93 to compensate for the 28% decline per coefficient r was significantly (p 5 0.05) different tidal cycle (two cycles per day). The abundance of from zero was the criterion used to decide whether a larvae in the 7 mm size-class was used to calculate an valid relationship existed prior to determining the- i index of larval abundance just prior to slope and its 95% Cl. ' metamorphosis. Because previous studies (NUSCO t 1987,1989) showed a net import oflarger larvae into Abundance, Growth, and Mortality -; the Niantic River, the weekly abundance oflarvae in the 7-mm size-class at station C was not adjusted for ofJuveniles in Summer  ! tidal flushing. To calculate each annual rate of mortality, sums were made of weekly mean catch ohoung-of-thegear winwr Coundu in . abundance indices (three stations combined) of each of the three or four replicated 1-m beam trawl newly-hatched larvae (after adjusting for tidal tows was standardized to a 100-m tow distance ' flushing) and larvae in the 7-mm size-class. Survival re C mPuting mean CPUE for each day and j rates from hatching through larval development were station; density was expressed as the number per 100 ' estimated as the ratio of the abundance index of the m of bottom. A median CPUE abundance mdex was larger larvae (7 mm size-class) to that of the smaller determmed for each half-season, with late May larvae (3-mm and smaller size-classes). through July denoting the early season and August. { 68 Monitoring Studies,1996 i a

September the late season. A 95% Cl was calculated tions (NR and JC), December through February for for each median CPUE using a distribution-free nearshore Niantic Bay stations (IN and NB), and method based on order statistics (Snedecor and January and February at offshore stations (TT and Cochran 1967). BR). This selection resulted in a uniform sample size Nearly all of the age-0 winter flounder collected of 42 collections per season. These catches were were measured fresh in either the field or laboratory pooled and used to calculate year-class abundance to the nearest 0.5 mm in total length (TL). During described by a A-mean CPUE (NUSCO 1988c). the first few weeks of study, standard length (SL) Beginning in January 1996, stations BR, TT, and NB was also measured because many of the smaller were, deleted (see Fish Ecology section for details), specimens had damaged caudal fin rays and total resulting in a sample size of 28 tows for the 1995-96 length could not be ascertained. A relationship A-mean. between the two lengths determined by a functional The annual A-mean CPUE ofjuveniles smaller than regression was used to convert SL to TL whenever necessary: 15 cm (mostly age-1 fish) taken during the adult winter flounder spawning surveys was determined as described previously for fish larger than 15 cm. Two TL in mm = -0.2 + 1.212-(SL in mm) (5) A-mean indices were calculated, one for stations in the lower Niantic River navigational channel (1 and Growth of age-O winter flounder at each station was

2) and, when sufficient data were available, one for examined by following weekly mean lengths all river stations combined. For comparative pur-throughout the sampling season. Mean lengths of poses, an annual A-mean abundance index ofjuvenile young taken at the Niantic River stations LR and WA fish of similar size was also determined using catch from late July through September were compared data from the five (or, in 1996, the two) trawl using an analysis of variance; significant differences monitoring program stations outside of the Niantic among means were determined with Duncan's River during the period of January through April, multiple-range test (SAS Institute Inc.1985).

with an annual sample size of 45 collections (18 in To calculate a total instantaneous mortality rate (Z), 1996), which temporally overlapped the adult all young were assumed to comprise a single cohort spawning surveys. with a common birthdate. A catch curve was constructed such that the natural logarithm of density was plotted against age (time in weeks); the slope of Stock andRecruitment Relationship the descending portion of the curve provided an and Biological Reference Points estimate of the weekly rate for Z. Once this rate was determmed, the monthly mortality rate (Z ) was A stock-recruitment relationship (SRR) described calculated as Z-(30.4 / 7). by Ricker (1954,1975) is the basis of the life-cycle The relationship between growth and abundance of algorithm that drives the population dynamics sim-young and water temperature was examined using ulation model of Niantic River winter flounder. multiple linear regression (SAS Institute Inc.1985) Application of this SRR to MNPS winter flounder and functional regression methods that were stock assessment was described in detail in NUSCO desenbed above for larval wm, ter flounder. (1989, 1990). The stock and recruitment data for determining the SRR were derived from the catch-at-age of female winter flounder during the Niantic Abundance ofJuvenileS River spawning survey. Because the spawning stock during Falland Winter is made up or many year-classes, the true recruitment consists of the total reproductive contribution over in fall and early winter, age-0 winter flounder the life of each individual in a given year-class gr'adually disperse from areas near the shoreline to (Garrod and Jones 1974; Cushing and Horwood deeper waters. Catch of these fish during this time 1977). Therefore, the index of annual parental stock period at the TMP stations (see the Fish Ecology size was based on derived egg production and the section elsewhere in this report for methods) was also index of recruits or year-class size was based on used as an index of relative abundance. Data used calculated egg production accumulated over the life-included November through February for inshore sta-time of the recruits. This method accounted for Winter Flounder 69

_ _ _ . . - - _ _ .. _ ._ _ _ . _ _ . _ _ _ _ _ _ _ _ _..___.m.. I 3 variations in year-class strength and in fecundity by age-3 fish were thought to be unreliable, this sim and age. The assumptions and methods used to i estimation process was only carried through the 1992 age Niantic River winter flounder and to calculate a 1 year-class (i.e., age-4 females taken in 1996). De I { recruitment index expressed as equivalent numbers I adjusted numbers ofmature fish provided an index of

• of female spawners were described in detail in the fully recruited year-class expressed as the i

NUSCO (1989,1990) and are summarimd below. i aggregated number of female spawners passing Stock and recruitment indices. Methods used to through each age-class. An implied assumption was 1 calculate the annual standardiud catch index and that catches in the Niantic River were representative l total egg production of the parental stock were given l j of tly population, with the exception of immature previously (see Adult Spawning Stock Sim and Egg { I fish that did not enter the river until fully recruited. Production section above). The recruitment index Although this recruitment index could be used j was determined by applying an age-length key to the j together with the annual number of female spawners annual standardimd catches of females partitioned j to derive an SRR, this would ignore sim composition into length categories. Based on a re-examination of l 1 differences that affected annual egg production. ' data, the age-length key used this year differed from i herefore, the above index was adjusted for the one used previously (described in NUSCO 1989) i differences in fecundity among fish using the i and will be discussed in more detail in Results and length-fecundity relationship of Niantic River winter

Discussion. A common age length key was used  ;

j flounder given above (Eq 1). Finally, annual egg l over all years because Witherell and Burnett (1993) production was summed up over the lifetime of each 1 reported that no trends were observed in mean l year-class to determine the recruitment index as eggs length-st-age during 1983-91 for Massachusetts j

and, then, converted to equivalent female spawners at '

winter flounder despite a 50% reduction in biomass , the rate of one female spawner for each 561,000 eggs over that period. Aging females allowed for the (i.e., the current mean fecundity). determination of their numbers by year-class present Stock and recruitment parameters. The Ricker at ages 3,4,5, and 6* during successive spawning SRR appeared best suited for use with the Niantic seasons. De age-6* group was further subdivided River winter flounder stock because the relationship 1 ' into the numbers of fish expected to survive to a l between recruitment and spawning stock indices was terminal age of 15 by assuming various annual a dome shaped curve with substantial decline in 4 instantaneous mortality rates as fishing pressure recruitment when the stock was larger than average i increased from the 1970s into the 1990s. To follow (NUSCO 1989). Furthennore, this particular form of each year-class from 1977 through 1992 to its l a SRR has been applied to other New England terminal age (e.g., 2007 for the 1992 year-class), flounder stocks (Gibson 1989). De mathematical values of Z (= F + M) were used that represented form of this SRR is: estimates of current and anticipated annual instantaneous rate of fishing (F) as provided by the R, = a P,.exp(- P,) Connecticut Department of Environmental Protection (6) (CT DEP). These were the same mortality rates used where R, is the recruitment index for the progeny of  ! in the stochastic population dynamics model, the spawning stock P, in year I and a and are para-discussed below. An instantaneous natural mortality meters estimated from the data.' he a parameter rate (M) for winter flounder was assumed constant at describes the growth potential of the stock and 0.25 over all years. From observations made of log,(a), the slope of the SRR at the origin, is abundance and age over the years, a large fraction of  ! equivalent to the intrinsic natural rate of increase age 3 females, considerable numbers of age 4 fish, (Roughgarden 1979) when the stock is not exploited. and even some age-5 females were apparently The p parameter is the instantaneous rate at which immature and not present in the N! antic River during recruitment declines at large stock sims due to some the spawning season (NUSCO 1989). Thus, the total form of density-dependent mortality. The natural number c,f females was reduced to spawning females logarithm of winter flounder recruitment was found using length specific proportions of mature fish correlated with mean water temperature during estimated from annual catches in the Niantic River February at the intakes of MNPS, which is when for fish age-3 to 5; all females age-6 and older were most spawning and early larval development occurs assumed to be mature. Because the estimates of (NUSCO 1988b,1989). Therefore, the parameters a 70 Monitoring Studies,1996

  . ~ . _ _ _ _ _ _ . . . _ _ . .                           - ._... _ _ _ _ _. _ ..                                __                   ._. _ _ .

4 i i f i and p were estimated initially by fitting Equation 6 to Reananging tenns and solving for the rate of fishing i the data and then re-estimated under the assumption that would achieve a given equilibrium stock sin ! that there was a significant temperature effect; this results in: was accomplished by adding a temperature-effect I component to Equation 6. Following Lorda and F = log,(a)- p-(Par >) (10) l Crecco (1987) and Gibson (1987), annual mean water temperatures .were used as an explanatory When F = 0, Equation 9 becomes the equilibrium or i variable to adjust the two-parameter SRR for replacement ievel of the unfished stock: ! temperature effects, which served to reduce , i recruitment variability and obtain more reliable P,,, = (log,[a}) / p (1!) { parameter estimates for the SRR. The temp-ersture-dependent SRR had the form: He fishing rate for " recruitment overfishing" has - been recently defined for winter flounder stocks as , R,= a P, exp(- P,) exp($ T,4) i (7) the rate of fishing that reduces the spawning stock j ' biomass to less than 25% of the stock for maximum where the second exponential describes the effect of spawning potential (Howell et al.1992). 1 February water temperature on recruitment and the  ; Although Equations 9 through 11 can be used to i j new parameter & represents the strength of that effect. calculate equilibrium stock sins and fishing rates for This effect either decreases or increases the number the winter flounder, the results are only deterministic ' j of recruits-per spawner produced each year because i approximations that ignore age structure effects. L trmperature was defined as the deviation (Tr4) of Therefore, these equations are primarily useful to cal-

each particular mean February temperature from a i

culate initial values of the corresponding biological long-term (1977 92) average of February water ! reference points. 7hese are better estimated through temperatures. When the February mean water temp-simulations using the SPDM or other similar pop- ) erature is equal to the long-term average, the ulation or production models that include age struc-i deviation (Tra)in Equation 7 becomes zero and the j ture and both natural and fishing mortality, exponential term equals unity (i.e., no temperature

effect). Rus, Equation 7 reduces to its initial form

(Eq. 6) under average temperature conditions. Assessment of MNPS Operation on l i Nonlinear regression methods (SAS Institute Inc. Niantic River Winter Flounder j 1985) were used for estimating the parameters in the j above equations. Several well established methods available for 4 . Biological reference points. Fishing mortality (F) stock assessment are based on stock recruitment l

! ss an important factor affecting the growth potential theory (Smith 1988). These methods assume I , of the stock (Goodyear 1977) and, thus, as relevant constant fishing rates and populations with stable , for assessing other impacts. Because fishing and age-structure, which result in equilibrium or steady. l j n:tural mortality of winter flounder take place state stocks that replace themselves year after year.  ! j concurrently through the year, the actual fraction of Some analytical methods are based on equilibrium i the stock removed by the fishery each year (i.e., the equations, such as Equations 9 through 11, which exploitation rate)is obtained as: have been modified to incorporate effects of 1 mortality caused by activities other than fishing. I u = (F / Z)(1 - exp[-Z]) Several problems may exist with an SRR-based (8) approach to impact assessment at MNPS. Because i Stock-recruitment theory and the interpretation of stock-recruitment theory (Ricker 1954) was several biological reference points denved from developed for semelparous fish (i.e., those which Ricker's SRR model were discussed in detail in spawn only once in their lifetime), Equation 11 may . NUSCO (1989). The equilibrium or sustainable provide unreliable estimates of equilibrium stock stock sin of an exploited stock (i.e., when F > 0) is sims for iteroparous fish (multi-aged spawning given by: stocks), such as the winter flour. der. Although the parameter a in Equation 9 could be adjusted for the Pur) = (lo8.[a] - F) / S (9) WinterFlounder 71 i

f e that no fishing mortality occurs prior to maturation. lations to estimate the fraction of Niantic River 1 This assumption cannot be met in the case of winter l I annual flounder production lost through larval ' flounder because many immature fish (ages-2 and 3) entrainment at MNPS; estimation of the equivalent are vulnerable to fishing gear. Wigley and Gabriel instantaneous mortality rates of females that were (1991) noted that concentrations of immature winter attributed to impingement; stochastic simulation of flounder found off Rhode Island may be subjected to the winter flounder stock dynamics to predict stock i significant mortality from fishing. Howell and biomass at selected levels of entrainment and fishing , ~ Langan (1987,1992) found that discard mortality rates; and an analyses of simulation results leading to rates of trawl-caught fish in New England waters estingtes of the probability that the stock would fall may be substantial. Simpson (1989) reported that below selected reference sizes. 4 about 72% of LIS winter flounder landed by the } commercial fishery were between 28 and 32 cm; many of these fish would have been age-3. Estimates ofLarval Entrainment at MNPS j Additional problems are found when ap*ying

,                 determmistic mcdels (i.e., assuming steady-state                                     he estimated number of larvae entrained in the j                  conditions) to fish stocks whose exploitation rates are                           MNPS condenser cooling water system each year is a a

not stable, especially when such stocks merease m direct measure ofimpact on the local winter flounder l abundance, as m the case of the winter flounder rtock. Annual estimates were detemnined using j during the late 1970s and early 1980s (Smith et al. larval densities at station EN (Fig. 3) and the

1989). Environmental variability also results in year- measured volume of cooling water used by the three to-year variation of natural mortality rates, which MNPS units. The Gompertz density function (Eq. 3) .

further weakens the results of determmistic was fitted to larval data and daily densities asussmWs. (number 500m) were calculated. Daily entrainment An approach to stock assessment mcorporstmg estimates were determined after adjusting for the environmental variability and all types of mortality, daily condenser cooling-water volume and an annual both constant and variable, involves the computer estimate was calculated by summing all daily simulation cf fish populatiou using a simple model

                                                                                                  ,,,;,,,g.         ggg of population renewal with spawning stock feed-back                                  The reduction in larval entrainment as a result of (e.g., a functional stock-recruitment relationship)..                            the 1996 shutdowns at all MNPS generating units This sppivacu has two advantages: assumptions of                                 was estimated by simulating full cooling water flows population equilibrium are not necessary, and much                               at each unit with weekly winter flounder larval detail can be incorporated into the conditions or                                entrainment densities.      The difference between scenarios used to simulate changes in fish                                       estimates based on the actual and simulated flows                ,

populations through time. An additional advantage is was the avoided larval entrainment attributed to the l that Monte Carlo methods readily provide the shutdowns' stochastic (as opposed to deterministic) framework needed for probabilistic risk assessment and for Mass-Balance Calculations testing hypotheses about the probable size of the stock at some future point. His simulation approach ne number of winter flounder larvae entrained de-was applied in NUSCO (1990) to assess the impact of Pends upon larval densities in Niantic Bay. Potential larval entrainment under a simple scenario. In impact to the Niantic River stock from larval entrain-NUSCO (1991b), the same approach used various ment is related to the number oflarvae in Niantic Bay combinations of historic and projected fishing and originating from the river. Mass-balance calculations larval entrainment rates to assess more realistically were used to investigate whether the number of the impact of MNPS operations on local winter winter flounder larvae entering Niantic Bay from the flounder. In NUSCO (1992a), the impact resulting Niantic River could sustain the number of larvae  ;

             - from the impingement of juvenile and adult winter                                 observed in the bay during the winter flounder larval             '

flounder was also simulated, ne basic steps leading season each year from 1984 through 1996, nree to the final impact assessment using this simulation Potential larval inputs to Niantic Bay include eggs , approach are: direct estimation of annual larval hatching in the bay, larvae flushed from the Niantic ' entramment rates at MNPS; mass-balance calcu. River, and larvae entering the bay from LIS across

                                                                                                                                                                    =

l l 72 Monitoring Studies,1996 1 l

                 - . - - - - .              .       . . . - - -          - - - - . ~ . - . - . - - .-.-                           .-
                                                                                                                                     !t the boundary between Millstone Point and Black                      Because these mass-balance calculations were based Point (Fig. 3) The few yolk sac larvae collected                    on the change in the number oflarvae in Niantic Bay annually in Niantic Bay suggested that minimal                      over a 5-day period:

spawning and subsequent hatching occurred in the bay, which was therefore considered a negligible 5-day change = NB, . 3 - NB, (14) source of larvae. Larvae were known to be flushed from the river into the bay and this input to the bay Thus: was 'imated from available data. The number of Source or Sink = 5-day change + Ent + Mort - larvae entering Niantic Bay from LIS was unknown. FromNR + ToNR

                                                                          .                                                  (15)

Four ways in which larvae may leave Niantic Bay include natural mortality, advection into the Niantic Daily abundance estimates were demed from the River during a flood tide, entrainment at MNPS, and Gompertz density equation (Eq. 3) and the daily flushing from the bay into LIS. Estimates could be densities for Niantic Bay at two points in time (NB, made for the number of larvae lost through natural and NB, . 3) for each 5-day period were calculated mortality, advected into the Niantic River, and from data collected at stations NB and EN combined. entrained at MNPS, but little was known about the These densities, adjusted for the volume of Niantic number oflarvae flushed into LIS. The numbers of Bay (about 50 x 10' m'; E. Adams, Massachusetts larvae flushed to and from LIS were combined as an Institute of Technology, Cambridge, MA., pers. unknown tenned Source or Sink in the mass-balance comm.), provided an estimate of the instantaneous cilculations. Thus, the form of the mass-balance daily r.anding stock. The difference between these  ; equation was: two estimates (NB, and NB, . 3) was the term 5-day  ; change in Equation 15. The selection of 5 days as 1 NB, . 3 = NB, - Ent - Mort + FromNR - ToNR the period of change was arbitray and a cursory {

                * (Source orSink)                            (12)  examination of results based on 10-day periods                      !

showed that the same conclusions were reached with where t = time in days either 5- or 10-day periods. NB,.s = number oflarvac in Niantic Bay 5 days Daily entrainment estimates were based on data after day I(instantaneous daily estimate) collected at station EN and the actual daily volume of Nb, = initial number oflarvae in Niantic Bay on condenser cooling water used at MNPS. The daily i day I (instantaneous daily estimate) entrainment estimates were summed over each 5-day l Ent = number oflarvae lost from Niantic Bay by period (Er.r). Annual stage-specific mortality rates entrainment in the condenser cooling water for 1984-89 were determined by Crecco and Howell system (over a 5-day period) (1990), for 1990 by V. Crecco (CT DEP, Old Lyme, Mort = number oflarvae lost from Niantic Bay CT, pers. comm.), and for 1991 through 1996 by due to natural mortality (over a 5-day NUSCO staff. Mortality was partitioned among period) developmental stages by comparing the rates of FromNR = number oflarvae flushed from the decline of predominant size-classes of each stage. Niantic River (over a 5-day period) Each developmental stage was assigned a portion of ToNR = number oflarvae entering the Niantic the total annual larval mortality rate (Z); similar River (over a 5-day period) mortality rates were assumed for Stages 3 and 4. Source or Sink = unknown number oflarvae in Although estimating stage-specific mortality in this Niantic Bay that flush out to LIS manner was imprecise, sensitivity analysis on the or enter the bay from LIS (over mass-balance calculations (NUSCO 1991b) indicated a 5-day period) that larval mortality was the least sensitive parameter Solving for the unknown Source or Sink term, the in Equation 15 above. These annual rates were equation was rearranged as: modified to daily stage-specific mortality rates by assuming 10-day stage durations for Stages 1,3, and Source or Sink = NB, . s - NB, + Ent + Mort - 4 larvae, and 20 days for Stage 2 larvae. The fromNR + ToNR (13) proportion of each stage collected at station EN during each 5-day period was applied to the daily standing stock for Niantic Bay (NB,) to estimate the I WimerFlounder 73 i

     - _    . .             - _.--               --        -. _-. - - . - - - -                                 - -         _ ~ - . - ~ _ -          . . - .

I number of larvae in each developmental stage for daily densities for Niantic. Bay. ^ Daily density stage specific mortality calculations. De daily loss i estimates for 1991-93 were combined and functional j due to natural mortality (Mort) was summed for each regression was used to determine the relationship  ; 5-day period. between abundance at stations NB and RM. The _ The 5 day input oflarvae to Niantic Bay from the average density oflarvae flushed from Niantic Bay river (FromNR) was based on daily density estimates into the river was estimated by the functional for station C in the river after adjusting for the rate of regression equation: l flushing between station C and the mouth of the river. To determine the relationship between the esti. i @NR = 128.149 + 2.073 NB, (17) mated daily density at station C and the average density oflarvae leaving the river on an ebb tide, the

• ne 95% CI for the slope (r = 0.705; p = 0.001) was geometric mean density of samples collected during 1.827 - 2.351. After being adjusted for the average an ebb tide for ten import-export studies conducted at tidal prism and the number of tidal prisms per day,  !

the mouth of the Niantic River during 1984, 1985, i these daily estimates of the number oflarvae entering and 1988 (NUSCO 1985, 1986a, 1989) was the river during a flood tide were summed over each } compared to the estimated daily densities at station C. 5 day period to calculate the term ToNR in the mass-The average density of larvae flushed from the balance equation. Because of the large intercept in  ; Niantic River was estimated from the functional the above regression line when no larvae were ! regression equation: j present in Niantic Bay (NB, = 0), the term ToNR was , i conservatively set to zero. The term Source or Sink fromNR = 9.751 + 0.473-(Daily density at in Equation 15 represents the 5-day net loss or gain { station C) (16) oflarvae to Niantic Bay from LIS required to balance l

                                                                                                                                                             )

The 95% Cl for the slope (r = 0.%9; p = 0.001) was the calculation. For a net loss oflarvae (flushed to LIS), the Source or Sink term would be negative and 0.387 - 0.579. The estimated average density, the j for a net gain of larvae (imported from LIS), the average tidal prism of 2.7 x 10' m' (Kollmeyer Source or Sink term would be positive. Results from i 1972), and about 1.9 tidal prisms per day were used j i mass-balance calculations by developmental stage  ; to estimate the daily flushing oflarvae from the river i were used to estimate the number oflarvae entrained i into Niantic Bay. This daily input to the bay was at MNPS each year from the Niantic River. If ' summed for each 5-day period to calculate the term fromNR can support the number of larvae entrained  ! ' FromNR in the mass-balance equation. by MNPS, then the Source or Sink tenn is negative I Stepwise oblique tows were collected during 1991 (i.e., no import) to balance the equation. Dese larval in the channel south of the Niantic River railroad bridge (station RM) during a flood tide to estimate an losses were then used to calculate conditional j mortality rates for Niantic River larvae for under average density to compute ToNR (NUSCO 1992a).

both actual operating conditions and projected full in 1992 and 1993, sampling was conducted again at MNPS three-unit operation. Deir derivation will be RM during a flood tide, but the collections were provided in greater detail in the following section and made by mooring the research vessel to the railroad

! later in the Results and Discussion section. bridge and taking continuous oblique tows (NUSCO 1994a). Comparison of densities fro:n the paired

stations of NB and RM showed a poor relationship. Stochastic Simulation ofWinter j Therefore, daily densities at the two stations were Flounder Stock Dynamics estimated using the Gompertz density curve (Eq. 3).

For station RM in 1992, the equation could only be Modeling strategy and background. De { adequately fit by smoothing the data using a 3-week stochastic population dynamics model (SPDM) deve-i running average prior to calculatmg a weekly loped for the Niantic River winter flounder stock was 1 cumulative density. ne Gompertz function could based on the Ricker SRR (Eq. 7) fitted to the data, i not be fit to data collected at station NB during 1993, even though the SRR equation does not explicitly j nerefore, catches from stations NB and EN were appear in the model fonnulation. The mechanisms combmed to calculate the weekly geometric means underlying the Ricker form of recruitment are prior to fitting the Gompertz function and estimating ncorporated in the set of equations that the model i 1 74 Monitoring Studies,1996 4 ) 1

uses to calculate mortality through the first year of fish population either as biomass (allowing for size ' life. Beyond that point (i.e., age.1) in the life. cycle .:riation within each age-class) or numbers of fish. simulation, the population model simply describes A similar implementation of an adult fish population the annual reduction of each year. class through dynamics simulation was used by Crecco and Savoy natural mortality and fishing together with growth (1987) in their model of Connecticut River American and reproduction. These processes occur at the shad (Alosa sapidissima). beginning of each model time. step oflength equal to Model components. Figure 4 illustrates 1 year. The projection of adult fish populations over components of the computer program used for the time has been implemented in many models by SPDM. Components depicted by solid.line boxes

,                 means of Leslie matrix equations (e.g., Hess et al.                               constitute the model presently in use, while the box 1975; Saila and Lorda 1977; Vaughan 1981;                                        with dashed lines corresponds to the mass. balance Spaulding et al.1983; Reed et al.1984; Goodyear                                   calculations dealing with spatial larval distribution and Christensen 1984). In the SPDM, adult winter and entrainment loss estimates, which are not an flounder were projected over time by grouping fish integral part of the model. The functionality of most into distinct age. classes and by carrying out the                               model components should be clear from the flow computations needd (mostly additions and                                          chart and no further details will be provided. Some multiplications) iteratively over the age index (I critical components, such as the one labeled age.1 through 15) and over the number of years specified                                cohort and the two random input boxes, are described

' for each simulation. This approach was algebraically below. A list of the actual input data used in the identical to the Leslie matrix formulation, facilitated application of the model to the Niantic River winter the understanding of how the model works, and flounder stock is also given. simplified the computer code when describing the w a==

                                                        .no s.muw.n                                                                                                          {

r.m.ws . v Ranoom w

                                                                                              =mper.iur.

rM&;

                                                          ...q             i,.inaitr&r 1_

ui.n p.....

U"".'? .'""*""'"'l . . J  :

v _  :

4 E* ' N ""

                                                             'an      Q . . . . . . . . . . . . . . . . . . . . . . .!
                                                                       <                  l Randomnome l v

Adun ctMort cecutshons with Annual edun nnpmosment. Procataliste r.atural, and fishmg mortaley y popuisuon

                                                                                                                       .g 4 nsa estrnates                        assessment V                                                V                      V sp. ning sioca su and Egg production estrnsees OUTPUT Fig. 4. Diagram of NUSCO stochastic population dynamics computer model for assessing the long term effect oflarval win flounder the   model. entrainment at MNPS. Dashed boxes and arrows refer to components and calculations which are not an WinterFlounder 75

_ _ _ _ _ _ - . _ _ _ . - ~ _ _ _ _ - _ . . _ . _ _ . . _ _ _ . _ _ _ _ _ . The most critical aspects in the formulation of a class via the mortality from egg to age-1 (Eq.18). stock-recruitment based population model are the The random noise term n,in Equation 18 is simulated specific equation and parameters used to calculate as independent random variates from a normal total mortality during the first year of life (i.e., from distribution with zero mean and variance equal to o', egg through age-1). He equation used for.this The value of a is estimated during the model purpose in the SPDM was derived from Ricker'r calibration runs as the amount of variance required to equilibrium equation for 2. (total instantaneous generate a values within the 95% Cl of the estimate mortality from egg through maturation age). This of a used in the model (NUSCO 1990). Similarly, involved the extension of stock-recruitment theory, the term ( WT, in Equation 18 represents the effect which was developed for fish that spawn only once, of annual environmental variability of February to iteroperous fish with multi-age spawning stocks. water temperatures on larval survival. This effect The form of the equation used in the present model

                     . was:

becomes random when the February water temperatures are themselves simulated as Ze., = log,(FEC) + log,(ASF) - log,(a) + n, - independent random variates from a normal

                                  $ WT,- Z .2 + p P,                            (18)                  distribution with mean and variance equal to the mean and variance oiFebruary water temperatures at where the subscript t denotes the time-step (each                              the MNPS intakes for 1977-92.                               !

time-step represents a year) and non-subscripted _ The stochastic simulation of fish population terms remain constant from year to year; a, p, and $ dynamics provides a framework for probabilistic risk are the parameters of the SR function (see Eq. 7), but assessment methodology. This type of assessment is with a estimated independently of the stock and based on Monte Carlo methods (Rubinstein 1981), recruitment data; FEC is the mean fecundity of the where many independent random replicates of the stock expressed as the number of female eggs time series are generated so that the mean of the produced per female spawner; ASF is a scaling factor series and its standard error can be estimated. Monte to adjust a for the effect of a multi-age spawning Carlo replications can be used to derive the sample stock; n, and WT, are independent random variates from two specified normal distributions described distribution function (Stuart and Ord 1987) without assuming any particular statistical distribution His below; Z ,3 is the instantaneous mortality through the methodology was used to assess the risk of stock immature age-classes; and the last term (p P,) is the reduction resulting from the effects of entrainment  ; feed-back mechanism simulating stock-dependent and impingement at MNPS. De probabilities of compensatory mortality, which varies according to stock reductions were empirically derived from 100 the size of the annual spawning stock P,.  ; The Monte Carlo replicates of winter flounder annual complete derivation of the above equation was given abundances in the time-series of impacted stocks. in NUSCO (1990: appendix to the winter flounder Briefly, the probability that a stock will be smaller section). De scaling factor ASF is a multiplier that than some postulated size is given by the proportion , converts age 3 female recruits into their spawning of replicates that are smaller than the reference size  ! potential throughout their lifetimes. This spawning in a given year. Additionally, the 5th and 95th potential is defined as the cumulative number of percentiles of the frequency distribution of stock mature females from the same year-class that survive sizes for specific years were calculated. These l to spawn year after year during the lifetime of the percentiles help describe the uncertainty associated ) fish. The algebraic form of this multiplier is identical with point estimates of annual stock sizes in the i to the numerator of Equation A-4 in Christensen and SPDM projections. Goodyear (1988). SPDM assumptions and limitations. Major Stochasticity in the winter flounder model (Fig. 4) assumptions of the SPDM relate to the underlying has two annual components: a random term that re-form of the SRR used and the reliability of the SRR presents uncertainties associated with the estimate of parameter estimates. Ricker's a parameter (n,) and annual environmental ' Because the SPDM incorporated the Ricker form of SRR, it was assumed variability in the form of random deviations from the that stock-dependent compensation and the long-term mean February water temperature ($.WT,). postulated effect of water temperature on larval These two components of annual variability are survival (Eqs. 7 and 18) applied reasonably well to incorrwed inta the calculation of each new year-the Niantic River winter flounder stock. A second 76 Monitoring Studies,1996 i

assumption was that the three parameters of the SRR power plant operation) estimated from the mass-were correctly estimated and that ct, in panicular, was balance calculations described above; a schedule of a reliable estimate. Although the population was not changes when any of these rates was not assumed assumed to be at steady state, the average fecundity constant; and the length of the time series in years. and survival rates for fish age-1 and older were The combined mortality of F + IMP was used only assumed to remain fairly stable over the period during the simulation period (1971-2025) that ccrresponding to the time-series data used to estimate corresponded to MNPS operation (Table 1). the SRR parameters. Although this last assumption Because the ability of a fish stock to withstand can generally be met in the case of fecundity rates additional stress is reduced by fishing mortality and adult natural monality, fishing mortality rates are (Goodyear 1980), simulations of the long-term much less stable. Changes in exploitation rates from entrainment of winter flounder larvae also included year to year should not cause estimation problems as effects due to the substantial exploitation of the long as the changes are not systematic (i.e., change in stock. He annual schedule of nominal fishing rates the same direction year after year). Because these was determined from recent DEP estimates (P. assumptions are seldom completely met, early Howell, CT DEP, Old Lyme, CT, pers. comm.) and applications of the model (NUSCO 1990) included differed from those given in NUSCO (1995a). These cclibration runs to validate predictions under both exploitation rates took into account length-limit deterministic and stochastic modes by comparing regulations in effect from 1982-96 and from changes model results to recent series of stock abundance in regulations proposed by the DEP to reduce fishing data. Finally, no temperature trend or large-scale monality in Connecticut waters (Tables 2 and 3). environmental changes (e.g., global warming) were Vulnerability factors for age classes I through 5+ assumed to have occurred during the years simulated were calculated for the commercial fishery (60% of in each population projection. the total winter flounder catch) and were based on Model input data. The dynamics of the Niantic actual or proposed changes in length limits and River winter flounder stock were simulated using the minimum commercial trawl fishery codend sizes; the SPDM under a credible real-time scenario running size-at-age of female Niantic River winter flounder at from 1960, well before operation of Unit 1, to 2060, mid-year (age + 0.5) determined using the von long after the projected shutdown date for Unit 3 in Bertalanffy growth equation (NUSCO 1987); 2025 (Table 1). The scenarios used power plant selection curves for ll4-mm (4.5-in) and 140-mm effects based on actual generating units in operation (5.5-in) trawl mesh codends provided by the DEP; each year, concurrently with estimates of F that were and a discard mortality rate of 50% for undersized based on historic and projected rates of commercial fish. The sport fishery was estimated to take 40% of exploitation and spon fishing for winter flounder in the total catch, having a discard monality rate of Connecticut. Parameters used in the SPDM included: 15%. Values of F used in the simulations were F, with an additional mortality equivalent of 0.01 to stepped up from 0.40 in the 1960s to a peak of 133 account for impingement (IMP) losses (NUSCO in 1990 (Fig. 5), which reflected an historical 1992a); larval entrainment conditional monality rates iac ease in fishing and the current high exploitation (i.e., ENT, the fraction of the annual production of of winter flounder. F was subsequently reduced to Niantic River winter flounder removed as a result of meet a targeted values of 0.90 in the late 1990s, TABLE 1. Cooling-water requirements and dates of operation for MNPS Units I through 3. cach with an assumed life span of 40 years. Coohng water flow Fraction of MNPS First year of 4 Projected last year Unit (m'sec ) total flow Start-up date entrainment ofoperation 1 29.18 0.227 November 1970 1971 2010 2 37.62 0.292 December 1975 1976 2015 3 61.91 0 481 April 1986 1986 2025 MNPS total 128.71 1.000 WinterFlounder 77

TABLE 2. Connecticut winter flounder regulations in effect for the commercial and sport fisheries since 1982. Minimum length limit (in) Minimum length limit (mm) Period Commercial fishery Sport fishery Commercial fishery Sport fishery Seasonal closure 1982* 8 8 203 203 1983 (Jan-May) 8 Nonc 8 203 203 None 1983 (Jun-Dec) Ii 8 279 203 Nonc 1984 (Jan-Aug) !I 8 279 203 Nonc 1984 (Sep-Dec) 10 8 254 203 1985 1986 10 None 10 254 254 None 1987 (Jan-Aug) 10 10 254 254 1987 (Sep-Dec) 11 10 279 Dec 1 Mar 31 (within Niantic River) 254 1988 1989 il 10 Dec I Mar 31 (within Niantic River) 6 279 254 1990-93 11 10 Dec ! Mar 31 (within Niantic River) 279 254 1994* 11 11 Dec I Mar 31 (within Niantic River) 305 279 1995 (Jan-Sep) 12 Mar 1 Apr 14 (in all state waters) 11 305 279 Mar 1 Apr 14 (in all state waters)' 1995 (Oct-Dec) 12 12 305 2 1996 305 None 12 12 305 305 None Prior to 1982 there were no size regulations, but it was assumed that fish between 6 inh (152 mm) and 8 inches (2

• subjected to about 50% of the nominal fishing mortality for each year. Fish larger than 8 mehes were full On 14 andJanuary 3 inches for 1,1989, the minimum May 15-November 14. trawl codend mesh size for the commercial fishery was establishe On April 22,1994, the minimum trawl codend mesh was established at 4.5 inches for Novemkr 15 June 3 November 14. On November 15.1994, the minimum trawl codend mesh size was increased to 5.5 inches for fish creel limit was ;stabCshed far the vport fishery.
• Closed season res,inded 9s of September 25,1995, but creet limit of 8 fish remained in effect.

l TABLE 3. Vulnerability facton' for castein LIS winter flounder by age class', adjusted for discard mortality of und the commercial (60% of total landings) and sport (40%) fisheries, according to fishing regulations in effect for the p l Niantic River winter flounder population dynamics simulation model. , { Age-classes: Commercial fishery Sport fishery Period i 2 3 4 Total fishery 5+ 1 2 3 4 5+ 1 2 3 4 5+ s1981 0.03 036 0.60 0.60 0.60 0.06 0.24 0.40 0.40 0 40 0.09 0.60 1.00 1.00 1982 0.00 036 0.60 0.60 0.60 0.06 0.13 0.40 0.40 0.40 1.00 0.06 0 49 1.00 1.00 1983-84 0.00 030 0.60 0.60 0.60 0.06 0.13 0.40 0.40 1.00 1985-87 0.00 030 0.60 0 40 0.06 0.43 1.00 1.00 1.00 0.60 0.60 0.06 0.06 0.40 0.40 0 40 0.06 036 1.00 1.00 1.00 1988-89 0.00 0.21 0.57 0.60 0.60 0.06 0.06 0 40 0.40 1990-93 0.00 0.12 0.57 0.40 0.06 0.27 0.97 1.00 1.00 0.60 0.60 0.06 0.06 0.40 0.40 0 40 0.06 0.18 0 97 1.00 1.00 1994 0.00 0.12 0.57 0.60 0.60 0.06 0 06 030 036 0.40 0.06 0.18 0.87 0.% l.00 1995 0.00 0.01 0.25 0.49 0.60 0.06 0.06 030 036 0.40 2 1996 0.06 0.07 0.55 0.85 1.00 0.00 0.01 0.25 0.49 0.60 0 06 0.06 0.07 030 0.40 t 0.06 0.07 032 0.79 1.00 l ' These factors assume discard mortality at 50% the nominal F rate for fish caught by commercial gear and at 15% o undersized fish caught by anglers (CT DEP estimates; P. Howell, Old Lyme, CT, pers, comm1

• The notation 5+ refers to fish that are age-5 and older, 0.70 during 2000-2005, and 0.60 thereafter, The fishing on ages-1 and 2 has been or will be greatly efTect of the changing fishing rates on partially diminished and many age 3 and 4 fish should be vulnerable fish is seen in Figure 6, As a result of protected as well. The derivation of the equivalent more protective regulations, the efTect of commercial mortality rate IMP was given in NUSCO (1992a) and 78 Monitoring Studies,1996

  -   . .. - . ..-.. -. . - . ~ . - -                                                               .. - . _ ..-.                      - . . . - , .                            . _ _ . - .               - . .    . - ~ .   . , - ,

i 1.4 i  ! l: k0 t

                                                                                                                                                                                                                                         -1 a          1D                                                      !,                                                                                                                                              -

I e . . l il t I 1 i h D  : id g 8 I t

                                          .                                                                  i W                 .

[ :/ '\ -

                       @ 08-                                                          It-                      n\

c ,1 - r I h.) ) 3 4 't i \ 50.6-i j i  : { l 3-unit operation O.4 - i

                                                                   ~

i C J i 02i  ! . i 1  : 3  ! 0-, , , , , . , , 5 1960 1970 1980 1990 2000 2010 2020 2030 2040 2050 206 YEAR Fig. 5. Historic and projected annual instantaneous mortality rate due to fishing (F), as detennined in consultation wi l DEP, lations. plus a small (0.01) component accounting for impingement mortality (IMP) at MNPS as implemented in the SP 1.41 . i 1.2i' 9

v. .

l tu 1

                                    ]
                      $ 'i          j                                                              ,*

b g0.6i  ; 1 0.6 0 / !9

                                                                                                       ~

E

                                   ,1 a                          ;/
                                                                .~
                                                                                                                     \                                              AN                                                                       1 l
                     $ 0.4 ;                                  ~'             '

I E 2 A ge-2 ) 4 A 9e-3 0.2 ; --- - [4 A9e-1 ' s._ e i 0-, , , , , , , 1960 1970 1980 1990 2000 2010 2020 2030 2040 2050 206

                                                                                                                    . YEAR Fig. 6. Estimated reductions in instantaneous fishing mortality rate F (including discard mortality) of age-1 through 4; flounder as a result of actual or planned regulations imposed by the CT DEP on the winter flounder commercial a ies.

I l i  ! I 4 i

- WinterFlounder 79 i

l l i I a

                                             --+,,m-    - -

w, - - - . - - - . , , , - - - ,-g

is an additional small (0.01) component of mortality the simulations were less than the theoretical added to F during the years of MNPS operation, maximum under full three-unit operation. Other data, rates, and inputs to the SPDM are sum- Simulation of MNPS impact. The simulation marized on Table 4 and include the number of age- output consisted of a time-series of annual stock sizes - classes, age-specific rates of maturation, natural generated under a specified set of population l mortality, average weight and fecundity at age, the parameters and conditions (including random three-parameter SRR estimates, February water variability) that constituted a scenario. All model temperature statistics, and other specific factors for runs of the 1 % 0-2060 stock projection series I each simulation. consisted of 100 replicates, which were judged to be Conditional mortality rates for larval entrainment sufficient given the amount of variability present in

. (ENT) from 1984 through 1996 were estimated SPDM simulations (NUSCO 1990). Thus, the Monte directly using the mass-balance calculations under Carlo sample size was set to 100 and the geometric actual MNPS operating conditions for use in the mean of

he replicates was computed for each year in SPDM simulations. Values of ENT determined for the projection. All stock projections are given in other years (1971-83; 1997-2025) were varied units of spawning biomass (Ibs) because overfishing stochastically by randomly choosing one of the criteria often rely on assessments of biomass, which annual values that was adjusted for full MNPS three-tend to be more conservative than those based on fish unit operation; this selection process was done by re- numbers. Furthermore, larval entrain nent effects sampling with replacement using uniform prob- result in long term stock reductions which can be abilities. Estimates of ENT were made by assuming quite different depending on whether the stock is that all three units used cooling water pum  !

expressed as fish numbers or as biomass. Population maximum capacity (11.1 x 10* d mE ay'). The ped at I reproductive capacity is more accurately reflected by I selected value of ENT was then scaled by both the biomass, which takes into account the size of number of units in operation in a particular year individual females (egg production is a function of (Table 1) and the fractions of cooling-water flow length or weight), as well as the number of spawners, actually used during the annual March-May larval A complete simulation of MNPS impact consisted winter flounder season (Table 5). MNPS cooling- of three model runs, which provided a set of time-water use was known for 1976 through 1996 and series generated under the same scenario, but with actual flow values were used to scale the randomly different combinations of F (plus IMP) and ENT. selected value of ENT. Because no data were These model runs were designed to simulate the available during 1971-75 for Unit 1, flow values for natural variability of the theoretical unfished stock these years were estimated from net electrical (i.e., with no fishing or plant operational effects), the generation records. Estimates for 1972 and 1975, reduced stock biomass when subjected to fishing years during which this unit apparently operated near mortality (i.e., the baseline time-series without maximum capacity, were normalized to the value for MNPS effects), and the expected biomass when all 1987, the year of maximum flow for the Unit I time-three types of anthropogenic mortality (F, IMP, and series; other years were scaled accordingly. Since ENT) occurred (i.e., the impacted stock). The first the simulation time-series extended to 2060 time-series with no fishing or plant effects was the (including a recovery period following the end of reference series against which the potential for MNPS operation), historic cooling-water flow rates recruitment failure was evaluated when the largest calculated for 1971-95 were re-used to predict reductions of stock biomass occurred during any of entrainment for 1997-2025 by re-sampling the the other simulations. The second time-series historic flows with replacement using uniform represented the most likely trajectory of the exploited probabilities to randomize the process. This stock without MNPS operation. The third time-series approach assumed that the existing 26-year record of MNPS operation adequately described the was the expected stock trajectory when the conditional mortality rates corresponding to ENT and operational variability expected at the station in the IMP were added to the fishing mortality simulated future. Except for those cases where randomly for the baseline. This last time-series was the basis chosen values for a year had all three units operating for quantitatively assessing MNPS impact on the near 100% capacity, annual values of ENT used in Niantic River winter flounder population. 80 MonitoringStudies,1996

                                                                                              .. _.                   . . _       _         _    2_. . _. _

i l l 1 I TABLE 4. Data, rates, and other inputs used with the Niantic River winter flounder population dynamics simulation model. I i Model input Value used or available Number of age-classes in population 15 l Earliest age at which all females are mature 6 Fraction mature, mean wt (Ibs), and mean fecundity by age: Age I females 0 0 011 0 Age-2 females 0 0 125 0 Age-3 females 0.10 0.554 223,735

}                 Age-4 females t                                                                                       0.38           0.811        378,584 Age-5 females                                                                                                                              l 0.98            1.0lt9      568,243 Age 6 females 1.00           1.377        785,897 Age ? females

• 1.00 1.645 1,004,776 Age-8 females .

1.00 1.873 1,201,125 I Age-9 females 1.00 2.057 1,366,951 Age 10 females 1.00 2.203 1,502,557 Age lI females 1.00 2.304 1,598,597 Age 12 females  ; 1.00 2.390 1.682,208 Age 13 females  ; 1.00 2.461 1,754,800 Age-14 females j 1.00 2.516 1,809.000 Age-15 females { 1.00 2.552 1,845,800 Age after which annual mortality is constant 4 Instantaneous mortality rates M and F at age 1 0.50 0' Instantaneous mortality rates M and F at age-2 and older 0.25 0 Initial number of female spawners 72,239" Biomass of female spawners ) 113,415 lbs Mean fecundity of the stock (eggs per female spawner) 972,205' a from the three-parameter SRR for the virgin (F = 0) stock (numbers of fish) d 5.87 p from the three-parameter SRR 2.450 X 10'8

              $ from the three parameter SRR
                                                                                                     -0.418 Mean February (1977-92) water temperature (*C)                                          2.81 standard deviation 1.22 minimum temperature 0.36 maximum temperature 4.76 Number of spawning cycles (years) to simulate 100 Number of simulation replicates per run 100 Fraction of age-O group entrained at MNPS (i.e., impact)                                kOO'
   Values are entered here only when mortalities remain constant dunng all the spawning cycles or years simulated Zero valu

• model to get a detailed schedule of mortalities from an auxiliary input file set up as a look up table (see Results and Discussion).
• Corresponds to the undished stock at equilibrium (see Table 32 in Results and Discussion).

Calculated for the Niantic River winter flounder female spawning stock at equilibnum in the absence of fishing (see Table 33 in Results a Discussion). d

  '  Indirectly calculated from life history parameters (see Stock-recruitment relationship in Results and Discussion).

A zero simulates a non-impacted stock; otherwise the conditional mortality due to entrainment is used. Winter Flounder 81

TABLE 5. Annual average cooling-water flow and percent of nominal maximum flow at MNPS Units I through 3 during the March-May winter flounder larval entrainment season from 1971 through 1996. Unit i Unit 2 Unit 3 Nominal flow at 100% capacity: 29.18 m' seed d 37.62 m'sec 61.91 m'sec" Fraction of total MNPS flow: 0.227 0.292 0 48i March-May March-May March-May average flow  % of nominal average flow  % of nominal  % of nominal d d average flow Year" in m see maximum in m see maximum in m see d maximum I971 - 67.41 - - - - 1972 - 99.64 - - - - 1973 - 33.81 - - * - 1974 - 83.50 - - - - 1975 - 99 64 - - - - 1976 2539 90.80 29.16 80.83 - - 1977 27.61 98.73 24.61 68.20 - - 1978 17.48 62.53 18.91 52.41 - - 1979 17.18 61.44 21.48 59.53 - - 1980 27.60 98.70 31.75 88.01 - - 1981 1.52 5.43 33.98 94.18 - - 1982 27.60 98.70 3233 89.61 - - 1983 26.79 95.83 30.90 85.63 - - 1984 13.88 49.61 35.83 9931 - - 1985 27.86 99.64 16.40 45 45 - - 1986 27.21 93.25 36.89 98.07 49.82 80.48 1987 29.01 99.40 36.99 9832 47.12 76.12 1988 28.84 98.81 32.83 87.27 55.58 89.78 1989 13.85 47.46 24.72 65.72 5133 82.91 1990 27.55 9439 33.28 88 48 48.71 78.68 1991 10.79 36.98 32.29 85.83 38 65 62.44 1992 25.11 86.06 28.50 75.75 51.10 82.55 1993 27.78 95.21 33.52 89.10 58.82 95.00 1994 433 14.84 3139 83 44 58.20 94.01 1995 29.04 99.52 21.61 57.44 37.35 6033 1996 0.72 2.47 14 44 38.38 31.05 50.15 I

• No records of cooling-water flow were available for 1971 75. Net electrical generation records were used to estimate flow, with val 1972 and 1975 normalized to the value for 1985 (maximum of the Unit I time-series); 1971,1973, and 1974 were adjusted accord!l below average. April (5.38'C) and May (9.36*C) i Results and Discussion had the coolest water temperatures since 1978, June (14.40*C) since 1982, and July (17.93*C) and August Seawater Temperature (19.44*C) since 1983. This trend continued during late summer and fall, with september (18.86*C) and October (15.35'C) about 0.8*C cooler than average.

In contrast to late 1994 and most of 1995, monthly mean seawater temperatures recorded at the MNPS The mean for November of 10.34*C was the third intakes during 1996 were among the coolest since coldest after 1976 and 1980. Although cooler than sampling began in 1976 (Table 6). The winter of average, the December mean of 6.99*C, however, was not exceptionally low. 1995-96 began with the coldest water temperature By season, winter (2.86*C) was cooler than (6.89'C) in December since 1989. January (3.55'C), average, spring (9.71*C) the second coolest after February (2.12*C), and March (2.87'C) were also 1978, and summer (18.74*C) the coolest of the 21-82 Monitoring Studies,1996 l l l l

                                      ~.                 -                         . .

TABLE 6. Monthly and annual mean seawater temperature (*C) from January 1976 through December 1996 as calculated from mean daily water temperatures recorded continuously at the intakes of MNPS Units I and 2. Year Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Annual mean 1976 3.65 331 4.81 7.55 10.75 15.11 18.29 19.60 18.93 15.04 9.28 4.73 10.90 1977 0 67 036 2.85 5.66 10.72 14.92 19.08 2033 19.41 15.58 12.18 6.72 10.84 1978 3.01 i.09 1.67 4.85 9.10 14.24 17.68 19.82 19.24 16.14 12.47 7.74 10.64 1979 4.53 1.48 335 5.93 10.50 15.57 18.84 20.91 20.05 15.99 12 41 8 60 11.57 1980 5.16 238 2.80 638 10.44 14.76 18.44 20.23 20.16 16.07 10.25 5.73 11.10 1981 1.M 2.63 336 6.40 10.19 15.48 19.51 "0.86 19.94 11.07 14.75 6.29  !!.01 1982 2.20 1.56 3.04 5.41 10 06 14.16 17.98 21.10 20.01 15.95 12 47 8.97  !!.13 1983 5.58 3.74 4.55 7.07 10.50 15.05 19.10 19.17 20.57 1737 12.57 7.90 11.98 1984 4.84 4.02 3.98 6.58 10.84 15.53 18.90 20 60 19.52 16 41 13.04 9.07 11.97 1985 436 236 4.17 7.02 10.95 14.99 18.98 21.24 20.44 17.46 13.14 7.95 11.98 1986 4.62 338 4.11 7.25  !!32 15.99 18.83 20 62 18.80 16.53 12 43 8.19 11.89 1987 5.28 3.27 4.53 7.51 11.26 15.91 19.19 20.47 1930 15.70 11.10 7.16 11.78 1988 2.65 2.67 4 49 7.01 10.67 14 69 18 30 2031 18.86 14.91 11.41 7.20 11.12 1989 4.49 3.24 3.67 6.21 10.59 15.25 18.95 2031 19.92 15.83 12.25 4 87 1134 1990 3.60 4.28 4.96 6.84 10.73 14.93 18.65 20.80 20.23 17.74 12.47 9.12 12.08 1991 5.72 4.76 5.61 8.11 12.26 16 61 19.53 20.48 19.99 17.11 12.00 8.17 12.59 1992 5.20 3.68 4.42 6.80 10.72 15 42 1841 19.62 19.20 15.17 11.12 7.28 11.45 1993 5.09 3.10 3.12 6.09 1137 15.64 18 96 20.88 19.88 1535 11.73 8.47 11.69 1994 3.15 1.59 2.81 6.62 9 96 1537 2030 20.78 19.27 16.27 13.21 9.15 11.60 1995 6.60 4.11 5.14 7.82 10.98 15.28 1930 21.06 20.43 1833 13 41 6.89 12.51 1996 3.55 2.12 2.87 538 936 14 40 17.93 1945 18.86 1535 1034 6.99 10.58 Overall mean 4.04 2.82 3.82 6.61 10.63 15.21 18.82 20.41 19 64 16 14 11.91 7.47 11.51 CV(%) 38 41 26 12 6 4 3 3 3 6 9 17 - year series (Table 7). At 10.90*C, the fall of 1996 occasionally during the annual adult winter flounder ranked as the fourth coolest. Because of these water surveys. From 19M through 1996, these means temperatures, the annual mean for 1996 was 10.58'C, differed from those receded at MNPS by 0.6*C or nearly 1*C cooler than the long-term average of less. For 10 of the years the water temperature in 11.51*C, and was the coldest in 21 years. These March was slightly cooler in the river than in Niantic relatively extreme water temperatures in 1996 likely Bay and in 8 of the years, including 1996 (+0.02*C) had important effects on adult spawning, as well as the river was slightly wanner; data were insufficient larval growth, development, and mortality, and the to calculate comparative means during three surveys. settlement, growth, and mortality of demersal young. Monthly mean temperatures were most variable Adult Winter Flounder during January through March (monthly CV = 26-41%; Table 6), the period when winter flounder spawning and early larval development occurs and Relative Annual Abundance most stable (CV = 3-6%) from May through October, when collections of winter flounder were dominated The cold winter temperatures during the winter of by young and other immature fish. 1995-96 produced heavy ice cover in the Niantic The mean temperatures given above reflected water River as far south as Smith Cove (Fig. 2), which temperatures in Niantic Bay, where the MNPS Persisted until late February, when air temperatures intakes are located. Water temperature in the Niantic exceeding 10*C cotr.cided with heavy rains. The River usually has a wider annual range, with adult wmter flounder survey finally began on somewhat colder temperatures in winter and warmer February 27 and sampling continued for 6 weeks in summer. During March, when considerable until April 3 (Table 8). By this time, few fish spawning, egg incubation, and larval development remained in spawning condition, as illustrated by the takes place, mean water temperature in the Niantic percentage f females 26 cm and larger that were River was determined from readings taken gravid (Fig. 7). The pattern in the decline of gravid Winter Flounder 83

t TABLE 7. Seasonal" mean seawater temperature (*C) for 1976 throu8h 1996 as calculated from mean daily water temperatures , continuous recorders at the intakes of MNPS Units I and 2. Year Winter Sprin8 Summer Fall 1976 3.94 11.14 18.94 9.69 1977 132 i 10.72 19.61 11.49 1978 1.95 9.40 18.91 12.11 I979 3.17 10.67 89.93

                                                                                                                               .1233 1980                   3.47                        10.53                                                               l 19.61                   10.69
                                                                                                                                                       ~

1981 234 10.69 20.11 10.70 1982 2.29 9.88 19.69 12.46 1983 4.65 10.87 19.61 12.61 1984 4.29 10 99 19,68 12.84 t 1985 3.67 10.98 20.22 12.85 1986 4.06 11.52 19.43 12.38 1987 4.40 l1.56 1988 19.66 ll32 3.28 10.79 19.16 ' 11.17 1989 3.82 10.68 19.72 10.97 1990 4.28 10.83 19.89 13.16 - 1991 538 1232 20.00 ' 12.48 1992 4.45 10.98 19.08 11.19 1993 3.79 l1.03 19.91 1l.85 ' 1994 2.55 10.64 20.13 12.87 1995 531 1135  ; 20.26 12.87 1996 2.86 9.71 1 18.74 10.90 l Overall mean 3.58 10.82 19.63 11.85 CV (%) 30 6 2 8

• Winter is January through March, spring is April throu8h June, summer is July through September, and fall is Oct females was similar to 1995, but more of the gravid the 21-year time-series (Fig. 9; Table 9). The small i

females were found each week in 1996 until early CPUE values for 1992-96 reflected extremely low April. This likely reflected colder water temperatures adult stock sizes present in recent years. The A-mean this year. Even so, most spawning apparently CPUE was highly correlated (Spearman's rank order j occurred earlier in the season under the ice cover, correlation coefficient r = 0.%75; p = 0.0001) with because more than half of the females were spent at the median CPUE values (Fig.10). He A mean ' the start of the hurvey in late February, index was slightly greater in magnitude than the ' Relative annual abundance of spawning winter median for all years, with largest differences i flounder in the Niantic River was measured by otter occurring during 1976-80. trawl CPUE. More than one-third of the tows made Female winter flounder taken during the 1996 ] s during 1996 had no winter flounder larger than 15 survey were mostly larger than 31 cm, with fish 45 cm (Fig. 8). Because of the increasing frequency of cm or larger relatively common in comparison to all zero catches in recent years, the relative abundance but the smallest (20.0-21.5 cm) fish (Fig.11). A index was changed in this report from a median to a comparison of the annual standardized catch of A-mean CPUE (NUSCO 1988c). De A-mean index females from 1993 through 1996 showed the scarcity of abundance is the best estimator of the population of all sizes of winter flounder this year (Fig.12). De mean when the data come from a distribution that decline in winter flounder abundance was even more contains numerous zero values (as it has for the adult striking when catches from 1981 (largest CPUE since winter flounder surveys during the past few years) 1976; Table 9),1985,1990, and 1996 were compared and the distribution of the non-zero values is (Fig. 13). Large decreases in abundance have approximately lognormal (Pennington 1983, 1986), occurred for all size-classes of female winter he A-mean CPUE of winter flounder larger than 15 flounder, with the exception of the very largest cm in 1996 was 1.6, which was the lowest CPUE of females. Although not abundant, larger fish in the 84 Monitoring Studies,1996 l

  . _ _ .               . .              - - . _ . . _   ____-~__.______..m_                _ _ _ _ _ _ .                                   _ -_. __

t , TABLE 8. Annual Niantic River winter flounder' population c 35-surveys during tiu spawning season from 1976 through 1996. y 30-Number of O 25k l Year Dates sampled weeks sampled 5 202 ' g -

                                                                                                                                         /

1976 March 1 April 13 7 m 1977 March 7. April l2 1978 March 6. April 25 6 g10-8 H $_ \ j 1979 March 12 - April 17 6 # _ ,. 1980 March l7. April l5 5 0_,_,,,,_ ,,,,,,,,,,,,,,,, _ 1981 March 2. April l4 7 76 78 80 82 84 86 88 90 92 94 96 1982 February 22 April 6 7 YEAR  : 3 1983 February 21 - April 6 7 1984 February 14 April 4 8 l

1985 February 27 - April 10 7 Fig. 8. Percentage of tows with no fish larger than 15 cm j 1986 February 24 April 8 7 collected in the Niantic River by year from 1976 through 1

1987 March 9. April 9 1996. 5 1 i 1988 March 1 April 5 6 1989 February 21 April 5 7 1977 and 1978. Despite relatively high abundance of i 1990 February 20 - April 4 7 age-0 fish produced in 1988 and 1992, female winter

1991 Tcbruary l3. March 20 6 1992 February l8. March 31 flounder from 20 to 30 cm in length have been

] 7 1993 February 16 April 7 8' relatively scarce in the spawning survey catches in 1994 March 22. Aprill3 4 recent years. 1995 February 28. April 6 6 Also noted during the 1996 spawning survey was 1996 February 27. April 3 6 the collection of five partially eaten winter flounder carcasses and one wounded fish that, based on canine

• Minimum size for markins was 15 cm during 1976 82 and tooth holes in several specimens, were apparently

                                                                                                                                                     ]

preyed upon by one or more harbor seals (Phoca ' "

• Limiicd s s dunns week 2 because ofice formation. "# ##nco Seals wm nrst notd in h

   ' Atmost no sampling during week 3 and limited sampling during             Niantic River during the 1993 winter flounder weeks 2 and 5 because ofice and weather conditions.

spawning season, with sightings increasing during j cach succeeding year, in some years, seal predation Niantic River now make up i relatively larger may represent an increased mortality risk to Niantic l proportion of female winter flounder than in previous l years. The large numbers of females from 23 to 34 River winter flounder spawners, particularly since population size has become depressed. ' cm collected in 1981 were likely 3. and 4-year old  ! 1 fish from the very strong year. classes produced in 1 m b n

5 6 @ 70 :

w 45- o 60 - - . g 40 . kr 50 j ,, g, , 95 ' 40 w 25 c. 30 i T [', ) 0 j # 15j 20 j

                              '-                                                6 gol '                             3 .

- o 0 Qgj u V, b A N '513 ,,~........ 0 ,,,,,,,,,,,,,,,, N,,,,s ., , tg 0-~ , , , , , i b 76 78 80 82 84 86 88 90 92 94 96 MARCH APRIL <!: YEAR

, Fig. 7 Weekly percentage of Niantic River female winter flounder larger than 26 cm that were gravid during the                     Fig. 9. Annual A-mean CPUE and 95% confidence inter-1994-% adult population abundance surveys.                                 val of Niantic River winter flounder larger than 15 cm from 1976 through 19%.

J

)

j WinterFlounder 85

    -- ~                . . _ _       . . =_ - .                  .          --          . . . -              .      .=         .--               -

TABLE 9. Annual 9.l.m otter trawl adjusted a-mean CPUE' of winter flounder larger than 15 cm' taken throughout the Niantic River during l the 1976 through 1996 adult population abundance surveys. 1 Wc:cks used Tows Adjusted a-mean 95% confidence Survey for CPUE acceptable number of Non-zero CPUE Standard interval for year computation

• for CPUE' tows used* observations estimate error a mean CPUE 1976 7 132 224 223 48.0 2.7 42.7 53.2 l 1977 6 I83 228 226 28.6 1.9 24.9 32.4

, 1978 6 135 162 162 31.2 ' 2.4 26.5-35.9 1979 5 116 140 140 41.5 3.6 34.6 48.5 1980 $ I12 145 144 41.6 2.5 36.6 46.5 1981 7 171 231 231 51.4 2.4 46.7 56.0 1982 5 116 150 150 48.1 3.4 41.4 54.8 i 1983 7 232 238 237 31.4 13 28.8 33.9

1984 7 244 287 286

' 18.4 0.7 17.1 19.7 1985 7 267- 280 277 17.I 07 15.8 18.5 1986 7 310 336 334 " 123 0.5 113 13.4 1987 5 233 239 236 16.8 0.9 15.0 I8.6 1988 6 287 312 310 17.9 0.7 16.6 19.3 i 1989 6 231 271 267 12.5 0.6 11.4 13.7

 ,               1990                    7                   260                 315

' 314 10.7 0.5 9.8 - l1.7 1991 6 2% 330 324 163 0.9 14.5 -17.8 1992 7 377 406 , 395 7.7 03 7.0 83 j 1993 7 287 1 392 344 3.3 0.2

~

1994 3.0 - 3.7 4 184 212 201 6.4 0.5 5.5 73 1995 6 316 342 284 2.6 0.1 2.42.9 } 1996 6 310 342 242 I.6 0.1 1.41.8 i

• Catch per standardized tow (see Materials and Methods); differs from NUSCO (1996) because median CPUE was replaced b the index of abundance.

i

• Mostly age-2 and older fish.
• Effort equalized among weeks; during several years weeks with very low effort were not used for computing CPUE.
• Only tows of standard time or distance were considered.

l l wnnter tsounsa r.15 cm 60 - Absolute Abundance Estimates

                    .i 50 ..               /-                            medan epue                            Adult winter flounder distribution in the Niantic 402.  ,u
                               /"/       -

River during the 6 weeks of sampling was not delta-mean k 30- U, . ', - i consistent in either time or place. Few (n = 36) fish

          "                                                                                       were captured and marked during the week of March 20  2l                                                                               18 in comparison to weeks befora (50-61) or
                                                 %.f% .*

10 immediately after (61). A total of 118 fish was 1 .... marked during the last week of the survey (April 1 0- , , , , , , , , , , , , , , , , , , , ,', and 3), which may have reflected fish moving into 76 78 80 82 84 86 88 90 92 94 96 deeper waters from shallow flats not sampled as YEAR waters warmed from about 3 to 5'C fro.n late March Fig. 10. Comparison between the annual median and through early April. As found during other recent A-mean CPUEs of winter flounder larger than 15 cm from years, most adults were concentrated in relatively 1976 through 1996' small areas, including the upper river arm (stations 52-54), particularly at station 54 (Fig. 2). As noted during 1995, catches at station 51 remained 86 Monitoring Studies,1996

8; 3 94 93 71

                                                                                     },

i

              !d        :                                                           !

8 6; l!i l!!g a w

i-x 59

i!:

95 i ;. : i 4 '

                                                                             ,i l                                                                                                           ?

j 96 . /.

                                                                           ~

A^\ ' n 3q

                                                                                                       \
                                                                                                              ..., /,

z a iill \  ; j\ w a  :::.r!.- , o 2.j i

                                                                                                                                                                        ,t x                                                ; i                                                   .4: . * \.

A i w 4 .

n. 1 s-

                                                                                                                                                             ,-                      /.-
                     ,                                  i            :                                                                     . . ..         s;        -

Y ...?..

                                                                                                                                                                                                 \
                     .    ...,.. ,v x                                                                                                    N ,......
                     .    /                    /

On,,i...,,,, ..........i.....,,,,,,. . . ,,,,,,,.,, l 20 2122 23 24 25 26 27 28 29 30 3132 33 34 35 36 37 38 39 40 4142 43 44 45+ l LENGTH (cm) l Tig. I1. Comparison of percent length. frequency distributions of female winter flounder 20 cm and larger taken in the Niantic River during the spawning season from 1993 through 1996. m 45- 94 m J . g  : i 5 40-  !! 1 g '

                                                                                   .~.

i

u.  !: :!!

w 35-o .

                                                                                 !         ii I                                                                     !        !i 0 30 Ja                                                              i        ii H
           <         a                                                         .

O 25 i l . H 3 /" i. 5 201 / i k i  : [ i i I O 15"4 i ! L. . ..-. ii ' l'it p j .! .! 95 93  :.

. :i m 104 ' ::: .

                                                                                                                                      '~.

ii

                                                       /                   -
                                                                                                                      /

• 5: ,,,,,..,.,,,/ . M.' 96

                                               ,                                                                                                       (3. j.[\  , , g,
                                         ,... 3, /

j- ,.T3 - 0 .).. ..... .

                                      ..ii.. ............ .. ... ,, ..,,,.... .

20 21 22 23 24 25 26 27 28 29 30 3132 33 34 35 36 37 38 39 40 4142 43 44 45* LENGTH (cm) Fig. 12. Comparison of annual standardized catch by length of female winter flounder 20 cm and larger taken in the Niantic ! river during the spawning season from 1993 through 1996. i l WinterFlounder 87

w 1801 5  : 160- 3

                                                         ;                                          81                                                                           !

l u., I f [\ i 2;*t " E120-' d

                                                       '                                  l\. ) U      \

; /

t ,  :

                                                                                        ,       c        .              ,

U l'. ' Il*II ' o 1005a ,.' uJ \! n \ ,: ,i

• I .l p l' ./

i

                   $ 802                                     l Il                                          o 

i

                                                                                                           \.

e

                   %          i            e         'b                   af                                                       \i\

s 85 (l A

                                                                                                   ~*'                              '

m 409 ,

                                                                                                              ..                          V
                                                                                                                                             \

h 2o k K:..'...:..'... ', 1'.~ ' ~^'. . C '..'.s g .r

                           ~
                                                    ... ..    ..l.....'.....                     96 p,,,,..g                         i 0- O m r,,.                                     cri M T,.,,,                                ,,                           C , , >M 20 2122 23 24 25 26 27 28 29 30 3132 33 34 35 36 37 38 39 40 41 42 43 44 45*

LENGTH (cm) l Fig.13, Comparison of annual standardized catch by length of female winter flounder 20 cm and larger taken in the Nian River during the spawning season in 1981,1985,1990. and 1996, particularly poor this year, with the few fish collected recruitment (B), and sampling intensity (p) were also there mostly found near the castern shoreline. generated by this model. Because of continued low Absolute abundance of winter flounder larger than abundance of Niantic River winter flounder popula-20 cm (N) spawning in the Niantic River was tion, this year only 376 fish 20 cm and larger were estimated using mark and recapture data with the Jolly (1%5) model. Estimates of survival (@), marked with a freeze brand and released (Table 10). This was the lowest total of the time-series and only i TABLE 10. Mark and recapture data from 1983 through 1996 used for estimating abundance of winter flounder larger than 2 Nianti: River during the spawning season. i Total Total not Number Total Number of fish marked in a given year Survey number previously marked and number year observed marked recaptured during subsequent armual surveys: released recaptured 83 84 85 86 87 88 89 90 91 92 93 94 95 1983 5,615 5,615 5,615 0 . 1984 4,103 3,973 4.083 130 130 1985 3,491 3.350 3,407 141 47 94 19R6 3,031 2,887 3,010 144 23 45 76 1%7 2.578 2.463 2,5 73 115 2 13 27 73 1988 4.333 4,106 4,309 227 7 22 31 63 104 1989 2,821 2.589 2.752 232 2 11 9 33 32 145 1990 2,297 2,135 2,275 162 1991 4,333 4,067 1 7 4 15 14 38 83 4J24 266 1 5 4 12 27 33 54 130 1992 2.346 2,119 2.336 227 0 0 1993 984 1 2 3 21 20 53 127 830 972 154 0 0 0 t 0 4 4 15 21 109 1994 1,035 959 1,033 76 0 0 0 0 0 4 1995 682 601 681 1 5 14 25 27 81 0 0 0 0 0 1 I 2 8 8 1996 379 18 43 341 376 38 0 0 0 0 0 0 0 2 2 5 5 4 20 88 Monitoring Studies,1996

55% of the 1995 total of 681 fish branded, the with N because of the panicular form of Jolly's previous low, Because so few fish were captured, variance formula. Therefore, the 95% CIs computed only 38 previously-marked fish were collected in are generally considered unreliable as a measure of 1996. About half (n = 20) the recaptures had been sampling error, except at very high sampling marked in 1995, with most others from 1992-94, intensities (Manly 1971; Roff 1973; Pollock et al. The mark-recapture data from 1996 provided an 1990). initial abundance estimate for 1995 of 5,544 winter Sampling intensity (p), or the probability that a fish flounder (Table 11); this value and those of other will be captured, was estimated as 0.122 for 1995, recent years will be subject to change as additional which was the second highest estimate for this marked fish are found during future surveys. The parameter, perhaps indicating relatively intense standard errors of N given in Table 11 are correlated sampling on fish concentrated in relatively few small TABLE 11. Estimated abundance' of winter llounder larger than 20 cm taken during the spawning season in the Niantic River from 1984 through 1995 as determined by the Jolly (1965) mark and recapture model. Abundance Standard Probability Standard estimate error of 95% Cl ofsurvival error 95% Cl Year (N) N for N (@) ofo for $ 1983 0 328 0.040 0.251 - 0 405 1984 57,706 8,370 41,300 74,112 0.558 0.065 0.430 0.686 1985 79,607 10,851 $8,338 100,876 0360 0.041 0.279 0.440 1986 49,057 6,194 36,917- 61,197 0.654 0.M8 0.522 0.786 1987 75,909 9,783 56,733 95,084 0.596 0.062 0.474 0.7I8 1988 66,688 7,244 52,491 80,885 i 0 453 0.048 03 60 - 0.546 1989 41,744 4,730 32,474 -51,014 0391 0.04 1 0310 0.472 1990 32,983 3,778 25.577 40,389 0.844 0.096 0.656 1.032 1991 61.131 7,248 46,925 75J36 0.200 0.025 0.151 - 0.249 1992 16.153 2,057 12,122 20.184 0 445 0.075 0.298- 0.593 1993 10,435 1,830 6,849 14,022 0 481 0 103 0.278 0 683 1994 16.094 3,437 9,357- 22,831 0306 0.093 0.123 - 0.488 1995 5,544 1,668 2,274 - 8,814 Mean 42,754 1,837 39,153 -46,355 0 468 0.013 0.444 - 0 492 r Sampling Standard Annual Standard intensity error of 95% Cl recruitment error 95% Cl Year (p) p forp (B) ofB for B 1984 0 071 0.0103 0.050 0 091 47,428 9,083 29,626 - 65,23l 1985 0.044 0.0060 0 032 0.055 20,454 5,200 10,262 30.647 1986 0 061 0.0078 0.046 0.077 43,850 8,499 27,191 60,509 1987 0 034 0.0044 0.025 - 0.042 21,472 6,379 8,969 - 33,975 1988 0 065 0.0071 0.051 - 0.079 11,524 3.663 4,344 - 18,704 1989 0 067 0 0077 0.052 0.082 16.692 3,074 10.667 - 22,716 1990 0 069 0.0080 0.054 - 0.085 33,311 5,453 j 1991 22.624 43.998 i 0 071 0.0084 0.054 0 087 3,925 1,440 1,102 6,748 1992 0.145 0.0186 0.108 0.181 3.245 1,104 1,082 5,408 { 1993 0.094 00166 0.061 0.126 11,083 j t 2,713 5,765 - 16,401 ' 1994 0.064 0.0137 0 037 - 0.090 621 887 -1,116-2,359 1995 0.122 0.0365 0.050 0.193 j Mean 0.075 0.0043 0.067 0 084 19,419 856 17,741 - 21,096 l I Estimates may vary from those reported in NUSCO (1996) because of mark and recapture data added from the 1996 adult winter flounder population survey. l I Winter Flounder 89 I l l 1

_ _ _ ~ _ _ _ . _ _ _ _ _ . _ - _ . _ _ _ _ _ ___.____ i i t ^ areas of the river. Sampling intensities of about 0.10 Estimated recruitment (B) values were low in 1991 j are recommended to obtain reliable and precise (3,925) and 1992 (3,245), increased to 11,083 for estimates of population sin and survival rates with 1993, and fell to a particularly low estimate of 621 ! ) the Jolly model (Bishop and Sheppard 1973; Nichols fish for 1994. However, estimates for both these

et al.1981), ahhough Hightower and Gilbert (1984) ,

• parameters are considered to be less reliable than found that low sampling effort may give acceptable i those of abundance when using the Jolly model estimates if population sia is relatively large and the (Bishop and Sheppard 1973; Arnason and Mills i

! number of marked animals is also relatively high. i 1981; Hightower and Gilbert 1984). Estimates of B However, Gilbert (1973) and Carothers (1973) were also relatively imprecise. reported that N was underestimated and had low As for other parameter estimates based on only I year of accuracy when sampling intensities were low (5-9%), { recapture information, those for e and B may change regardless of population sin or number of fish considerably with the addition of data from the next marked. Estimates of p only approximated or annual suivey. The low estimates of B in recent exceeded 0.10 this year, in 1992 (0.145) and 1993 years, however, appeared to accurately reflect weak (0.094). Loss of information because brands were recruitment of winter flounder. missed, or due to mortality of fish handled, also Because of a reasonable correspondence between requires increased sampling effort. Other sampling median trawl CPUE and Jolly abundance estimates, errors, model assumptions, and biases inherent in the the annual standardimd catches of all fish larger than j Jolly model that could have affected these estimates 20 cm for 1984-95 were compared to total abundance  ; were discussed in NUSCO (1989) and Pollock et al. estimates from the Jolly model. (1990). The relative numbers of females and eggs produced each year, as Although the Jolly estimates are subject to determined from the standardimd catch estimates, considerable error, annual A-mean CPUE and Jolly were conservatively assumed to represent about 4% abundance estimates were significantly (Spearman's of the absolute values; the range for annual values rank-order correlation coefficient r = 0.9021; p = was 2.7 - 6.3%. Thus, a multiplier of 25 was used to 0.0001) correlated (Fig.14). Thus, based on a A-mean CPUE of 1.6 for 1996, absolute abundance scale standardimd catch indices to absolute numbers of female winter flounder spawning in the Niantic of winter flounder may have been less than 5,000 fish. By extrapolation, abundance in 1981 could River that are given below. In using this scaling i factor it was assumed that ratios of annual have exceeded 150 to 200 thousand winter flounder. standardimd catch to absolute abundance during Estimates of survival (@) have varied considerably 1977 through 1983 would have been similar to those from year to year (0.200 - 0.844; Table 1 !). for 1984-95, had estimates of absolute abundance been available for the earlier period. n l Spawning Stock Size and Egg Production {g 20-- Ab

                              ,A', u  p.- *bu"8*ac' '
                                                                   . so $

tey

                  @ 16 /

ho The sin of the Niantic River winter flounder 5

                                        /               ,          L ,o ,,9  female spawning stock is used in various assessments 5 " "1              S. / ..            "          F w

of MNPS impact. The annual standardimd catch of { $3 k. M female spawners (an index of spawning stock sia) g si e

L3o and the production of eggs were determined from 1:>

k' 2o , available data on sex ratios, sexual maturity, and fish 5 j., 4-mean CPUE . E o', , , , ,', , o , to w5 length frequencies. The sex ratio of winter flounder I N es sa larger than 20 cm during the 1996 spawning season

• so 92 H h YEAR
• in the Niantic River was 1.78 females for each male Fig. 14. Comparison between the esumates of absolute (Table 12). 'Ihis sex ratio is larger than the long-term abundance in thousands of winter flounder larger than 20 average for the 20-year time-series of 1.44, but less cm in the Niantic River during the spawning season than the ratio of 2.70:1 found during 1995, which (dashed line) and the corresponding 4-mean CPUE (solid was the largest one found since 1977. Ratios of 1.50 line) from 1984 through 1995. 133 6 dfd P by M (1%2a,1%2b) and by Howe and Coates (1975) for 90 Monitoring Studies,1996

l TABLE 12. Female to male sex ratios of winter flounder taken relatively cold years (e.g., 1977 and 1978) during the spawning season in the Niantic River from 1977 proportionately fewer females spawned during the thmu8h 1996. carlier portion of the survey, compared to warmer masured years (e.g.,1989 and 1992) when more fish were Year All fish captured fish > 20 cm spent at the beginning of sampling. During each year, the proportion of females 1977 1.03 1.26 1978 2.23 1.95 estimated to be mature in each 0.5-cm length increment was used with the annual standardized ll,7l jaj [2jg catch of females to obtain annual abundance indices 1981 1.42 1.61 for female winter flounder. Annual estimates of ) 1982 1.16 1.50 female spawner abundance and egg production were

       .1983                1.52           1.52 generated by multiplying relative standardized catch 1984                1.07           1.07 3                                   estimates for each by 25 (see Absolute Abundance ll8j g                  92          0.9                    Estimates, above). This multiplier has decreased 1987               0.78           0.78                   from values of 26.316 and 28.571 used in recent 1988               1.50           1.50                   years (NUSCO 1992a,1993, l'994a,1995a,1996).

1989 1.32 132 1990 This has resulted in reductions of approximately 5 to 1.24 1.24 13% for estimates of absolute female stock size and [ 1993 3] $ total egg production previously reported. Estimates

                                                                                                                       ]

1.47 1.47 of female stock size ranged between 2,427 (1996)  ; 1994 1.70 1.70 and 68,899 (1982) fish (Table 13). Mature females 1995 2.70 2.70 1996 generally comprised approximately one-third to one-1.78 1.78 half of each annual total, with 1995 having the

  %,,,,,,,                 g              g                     highest fraction of mature females at 63%, which was j

related to the highly skewed sex ratio and 1 proportionately larger fish found during that year. other winter flounder populations in southern New Despite this preponderance oflarger mature females, England. Witherell and Burnett (1993) also found the number of spawners has been very low in recent 1 greater proportions of female winter flounder in l years because of low overall abundance of winter j Massachusetts waters, particularly in older age- flounder. The total number of female spawners was classes. They believed that males likely have a used as an estimate of parental stock size for the SRR higher natural mortality rate, based on evidence of (see below). earlier ages of senescence reported for males by Annual egg production estimates ranged from Burton and Idler (1984). about 2.1 to 39.9 billion (Table 13). Differences in The rate of spawning was determined by observing percent maturity due to changes in length frequency weekly changes in the percentage of gravid females distributions ' somewhat affected mean fecundity, larger than 26 cm, the size at which about half of all which was low during the late 1970s when smaller observed females were mature (NUSCO 1988b). fish were more abundant, but increased during recent This is comparable to L o3 estimates of size at- years because ofincreasing proportions of older and maturity of 28.3 and 27.6 cm reported for Massachu- larger fish. Total egg production was greatest from setts waters by Witherell and Bumett (1993) and 1981 through 1983 because of peak population O'Brien et al. (1993), respectively. In recent years, abundance and moderate mean fecundity. Estimates spawning in the Niantic River was mostly completed were also relatively high in 1988,1989, and 1991 as by late March or early April as relatively few gravid proportionally older and larger females dominated a females were found afterwards (Fig. 7). During most moderately-sized reproductive stock.' Total fecundity years, ice in the upper river prevented the start of decreased to relatively low values of 2,1 - 8.2 billion field work in January or early February, so during 1993-96 because of very low abundance, approximately one-half to two-thirds of the females Female size and time of spawning affects various examined during late February and early March had reproductive parameters, including egg size, spawned before sampling began. Spawning was fecundity, and viability; embryos deposited earlier in liksly correlated with water temperature, as in the season appear to have better survival than eggs WinterFlounder 91

TABLE 13. Relative and absolute annual standardized catch of female winter flounder spawners and corresponding egg production in the Niantic River from 1977 through 1996'. Relative index Relative index Survey ofspawning  % mature Average oftotal Total female Total egg  ! year females' females

• fecundity' egg production' stock size" production (X 10'f 1977 889 36 446,374 394.2 22,226 9.854 1978 1,415 51 506,220 716 2 35,368 17.904 1979 1,129 38 474,665 535.7 28,217 13.394 1980 916 35 464,104 425.0 22,893 10.625 1981 2,683 45 1,382.3 515.241 67,070 1982 2,756 34.557 49 578,530 1,594 4 68.899 39.860 1983 1,873 46 577,307 I,081.2 46,821 1984 27.299 872 40 574,214 500.7 21,801 12.518 1985 931 43 607,083 564.9 23,264 14.123 1986 654 42 666,312 436.1 16,361 10.902 1987 852 39 623 254 530.9 21295 1988 1,278 13272 53 677,596 865.7 31,939 1989 983 21.642 52 727,934 715 4 24,570 1990 17.885 580 42 637,693 370.1 14,510 9.253 1991 1,060 47 602,499 638.7 26,502 15.968 1992 533 52 732,366 390.7 13,336 9.767 1993 273 54 816,797 223.4 6,837 5.585 1994 507 55 649,622 329.4 12,676 8.234 1995 218 63 775,416 169.3 5,458 4.232 1996 97 52 844,911 82.0 2,427 2.051
• Some estimates differ slightly from those reported in NUSCO (1996) because of changes in the length-age key use
• Based on proportion of the relative annual standardized catches ofwinter flounder that were mature females.

' As a proportion of all winter flounder 20 cm or larEer.
' Total egg production divided by the number of spawning females.

• A relative index for yeardo-year comparisons and not an absolute estimate of production.
• Calculated on the assumption that the relative annual standardized catches were approximately 4.0% of abs by approximately 8% or less from those reported in NUSCO (1996) because of a 0.2% increase in the scaling factor used.

produced by smaller fish late in the season (Buckley anoxia. Skeletonema costatum was one of the most et al,1991). Egg deposition apparently takes place abundant of the phytoplankton collected at MNPS on gravel bars, algal mats, celgrass beds, and near during entrainment sampling from 1977 through freshwater springs in Rhode Island salt ponds 1980 (NUSCO 1981). However, highest densities (Crawford 1990). Viable hatch is greatest at 3*C in occurred in summer, after the winter flounder salinities of 15 to 35 and decreases with increasing spawning season. Based on a comparison of temperature (Rogers 1976). DeBlois and Leggett estimates of egg production and abundance of Stage (1991) found that the amphipod Calliopius I larvae (discussed below). egg mortality may be laeviusculus preyed heavily upon demersal capelin considerable in the Niantic River. i (Mallotus villosus) eggs, removing up to 39% of the production. They suggested that invenebrate predation on demersal fish eggs may be an important Larval Winter Flounder regulatory mechanism for population size in marine fishes having demersal eggs, Morrison et al. (1991) Abundance and Distribution reported high monality of demersal Atlantic herring (Clupea harengus) eggs in the Finh of Clyde, The a parameter of the Gompertz function (Eq. 2) Scotland because of heavy deposition of organic was used as an index for temporal (year to year) and matter resulting from a bloom of the diatom spatial (Niantic River and Bay) abundances of winter Skelefonema costarum. The decomposing material fl under larvae, Based on the a parameter estimates, caused a depletion of oxygen and egg death due to larval abundance during 1996 m both the n,yer 92 Monitoring Studies,1996

4 1 (stations A, B, and C combined) and the bay (stations Annual spatial abundances of the first four larval EN and NB combined) were about average for the developmental stages were based on cumulative 14-year series (Table 14). In general, annual weekly geometric means (Figs.15 and 16). The ebundances in the bay varied less than in the river, abundance distribution of Stage 5 fish (i.e., newly in 1985,1987,1988,1989,1995, and 1996, larval transformed juveniles) was not examined because so abundance was at least four times greater in the river few were collected by ichthyoplankton gear, than in the bay, No consistent relationship was found Cumulative density data (the running sum of the between the indices of annual abundances in the two weekly geometric means) were used to compare areas (Spearman's rank-order correlation coefficient r abundances as a surrogate for the a parameter from

   = 0.455; p = 0.102). This lack of a relationship has                    the Gompertz function (Eq. 2) because in some two possible causes. First, if many of the larvae in                    instances this function could not be fitted. This the bay came from the river, then annual larval                         usually occurred when a developmental stage was mortality rates prior to the period when larvae were                    rarely collected at a station (e.g., Stage 1 at stations flushed from the river to the bay were highly                           EN and NB or Stage 4 at station A). Cumulative variable. Second, the Niantic River may not be the                      weekly geometric means and the corresponding a only source of larvae entering the bay (NUSCO                           parameters were found to be highly correlated 1992a,1992b,1993,1994a,1995a,1996) and this                            (Spearman's rank-order correlation coefficient r =

possibility will be addressed again later in this 0.999; p < 0.001) in a previous study (NUSCO section. Larval abundance in the bay appeared to 1989), indicating that cumulative weekly geometric reflect regional-wide trends as annual abundance (a means could be used as an attemative index oflarval parameter) at EN since 1976 was correlated abundance.

(Spearman's rank-order correlation coefficient r = Stage I abundance during 1996 in the river was

! 0.635; p = 0.002) with annual abundance indices in about average compared to the previous 13-year Mount Hope Bay, MA and R1 (Marine Research, Inc,                        period of sampling at all three stations, (Fig.15). A 1992; M, Scherer, Marine Research, Inc., Falmouth,                     comparison of annual Stage I abundance among MA,, pers. comm.).             As was found for the                     years showed a similar relative ranking at the three             1 comparison between Niantic River and Bay, no                            stations, with 1988 and 1989 ranked the highest and              l i  relationship was found between the abundances in 1983,1986, and 1993 the lowest. Except for a the Niantic River (1983-96) and Mount Hope Bay                                                                                           J slightly greater abundance at station A in some years, (Spearman's rank order correlation coefficient r =

annual abundances at the three river stations have 0.108; p = 0,714) This suggested that Mount Hope been similar. This indicated a somewhat Bay, similar to Niantic Bay, is not a preferred winter homogeneous distribution of Stage I larvae through-flounder spawning area, as discussed below. out the river, Because winter flounder eggs are TABLE 14 Index of annual larval winter flounder abundances and 95% confidence intervals for the Niantic River and Bay, based on the ci parameter from the Gompertz function for 1983 throagh 1996. Year Niantic River Niantic Bay 1983 1,863 (1,798 - 1,929) 3,730 (3,670 -3,791) 1984 5.018 (4,884 5,152) 2,200 (2,088 2,311) 1985 11,924 (11,773 -12,075) 1,801 (l,717 1,886) 1986 1,798 (I,726 - 1,87I) 1,035 (979 - 1,09I) 1987 5,381 (5,172 5,589) 1,301 (1,240-1,363) 1988 l 24,004 (23,644 - 24,364) 1,784 (1,708 - 1,861) l 1989 18,586 (17,965 19.207) ' I,75) (1,6% - 1,8%) 1990 5,544 (5,378 - 5,709) I,532 (I,474-1,589) l 1991 4,083 (3,973 -4,193) 1.444 (1,388 -1.500) 1992 10,646 (10,184 I1.108) 4,415 (4,214 -4,617) j 1993 1,013 (1,470 1,557) 459 (391 - 526) 1994 l 5,685 (5,564 5,805) 2,378 (2,269 2,486) 1995 l 14,075 (13,416 - 14,735) 3,091 (2,966 - 3.216) 1996 9,916 (9,631 10,202) 1,690 (1,535 - 1,844) Winter Flounder 93

1 W g25000 STAGE 1 ' < 4

           @ 20000 -1 3                ,

E 15000 , ~ y 10000 -

                                  -                                                                                          ~
          "b 5000                 i                 1-         !       F,                                         ~-

s - - - 5 01:- -l -U ' _U l l- ' H '* TI O YEAR 83 85 87 89 91 93 95 83 85 87 89 91 93 95 STATON A 83 85 87 89 91 93 95 w B y20000 C

          <                                                                          STAGE 2
                                             ~
          @ 16000-m               4                       ;                   -
          @ 12000e                                                        '

i 1

          $ 8000 1 s               -

q J 5 4000 -l  !).' U 0

                            ''~       "'

Lj'

                                                        ~
                                                                        '      J                M'          ~

L~ ^~~ YEAR 83 85 87 89 91 93 95 83 85 87 89 91 93 95 w STATON A B 83 85 87 89 91 93 95 C

        @ 6000 -                                                                     STAGE 3
        @ 5000 a               1                                   -

g 4000 i lU p 3000h { _ , , l h [ - 5i"! b 2000 -I 5 fg {LU il'( f O l 3' d U o. i - _I " I{ ?- .

                                                                                       ._.         Fib e-l j : 'mi      ,                W!

YEAR 83 85 87 89 91 93 95 '83 85 87 89 91 93 95 STATON A 83 85 87 89 91 93 95 w B C

       @ 3500 5 3000 -

STAGE 4 z 4 ' 1 g 2500g  ! l

• 2000 e l'i  !

i

       $ 15005                                                                                                                                      I g 1000j                                                                                       g i

n soo 4 - m _j i ,I[1 ( u 0 - ~ -

                                                                                                  ~~                                        '

YEAR 83 85 87 89 91 93 95 STATON 83 85 87 89 91 93 95 83 85 87 89 91 93 95 A B C Fig.15. Index of cumulative density for each developmental stage oflarval winter flounder at the Nian from 1983 through 1996. (Note that the vertical scales differ among the graphs). 94 Monitoring Studies,1996 l 1

d w GTAGE 1

                             @ SM ,
                             @ 400 -

e - h

                             < 300 1                                                                                   $

g _~~ ~ { uYEAR 83 85 87 89 91 93 95 i 83 85 87 89 91 93 95 STATION EN NB 1600 , AE2 a i i

                            $ 12001 e           i 800 i                                         i                                      ,-

6,

                            $          J~                                                                 .

4g g i .rg!

                            $ 400u fh! , d_ ~                               F                           f      -      4 9

D . - mumE2Jd O 0! - YMR 83 85 87 89 91 93 95 83 85 87 89 91 93 95 STATION EN I NB w 4000 STAGE 3 3000 - e h 2000 r pl [g _. 7

                                       )                      !!

i  % 2

                                                                                                                   )

7

                                          ! ; i IT l l l i u iU{

g 1000- 1,

                                                                                         %              $    [':)N i

g U 0 Y MR 83 85 87 89 91 93 95 n N4" -l B2~ -- j STATION 83 85 87 89 91 93 95 1 EN NB I. W , @ 600 - STAGE 4

                          @ 500 l     4*

300 ,

                                     -ll,                           r i      s y}

g , 1 d I' b2 I' ' I i b' h ,h qu l '7,  ! l'- kll l;Pi!it W ui!: hGfLm :n YEAR 83 85 87 89 91 93 95 83 85 87 89 91 93 95 STATION EN NB 4 Fig.16. Index of cumulative density for each developmental stage oflarval winter flounder at the Niantic Bay station from 1983 through 1996. (Note that the vertical scales differ among the graphs). s' I WinterFlounder 95

i demersal and adhesive and the duration of Stage 1 is stations B and C with their abundance at the two bay short (about 10 days), the homogenous distribution stations (EN and NB) increasing to levels similar to j suggested either that spawning was not restricted to a or greater than at stations A and B (Figs.15 and 16). specific area of the river or that the river is well- { In comparison to other years, abundances of Stage 3 mixed. Low abundance in 1983 was attributed, in and 4 larvae during 1996 were at or below average at part, to undersampling because of net extrusion all stations. For Stage 3, discrete spatial relationships j (NUSCO 1987). However, this was rectified in 1984 for annual abundances were found between adjacent when a net with smaller mesh (202 pm) was used stations; correlations were significant (p s 0.05) during the early portion of the larval season. between stations A and B; B and C; C and both NB Abundance of Stage 1 larvae at the two Niantic Bay and EN; and NB and EN (Table 15). Similar to Stage stations (Fig.16) was low in comparison to the river 3, Stage 4 larvae were generally more abundant at (Fig.15), indicating that little, if any, spawning station C and the two Niantic Bay stations in occurred in the bay. Except for 1985 and 1996, comparison to stations A and B. abundances at station NB were consistently greater Annual abundance of newly hatched winter than at EN, possibly because NB was located closer flounder larvae should relate to adult spawning (i.e., to the river mouth, the likely source of Stage 1 egg production) and the fraction of eggs that hatch.  ; larvae, or because undersampling occurred at EN as a To examine this relationship, the annual egg produc-result of extrusion through the 333 m mesh net used tion estimates (Table 13) were compared to the there. Additional evidence for possible net extrusion annual abundance of Stage I larvae. The index of at EN is discussed in the Development and Growth Stage I larval abundance was the et parameter from section that follows. At station NB, ranks of annual the Gompertz function (Eq. 2) for the Niantic River abundance indices were similar to those of the river (stations A, B, and C combined). A functional stations; this suggested that most Stage I larvae regression indicated a strong positive relationship (r collected in the bay probably originated from the

                                                                = 0.625; p = 0.022) between egg production and                1 Niantic River. Significant (p s 0.05) positive Stage I abundance (Fig.17). The abundance of correlations were found among Stage 1 annual                 newly hatched larvae was directly related to the adult abundances at all stations, except for station EN with egg production under the assumption that egg stations C and NB (Table 15).                                hatchability was similar among years. However, Stage 2 abundance in 1996 at stations B and C was Stage 1 abundance for both 1995 and 1996 appeared about average, but at station A was the second to be greater than expected from this relationship, greatest (Fig.15). In general, annual ranks of Stage 2       suggesting a greater egg batchability or larval abundance at the three river stations were similar to       survival during the past 2 years, which had the lowest those of Stage 1. This implied a similar annual rate        estimates of egg production.

of larval loss (monality and flushing) during larval Dates of peak abundance, estimated from the development from Stage I to 2. Annual abundances inflection point p of the Gompertz function (Eq. 2), at stations B and C were almost identical. Stage 2 were used to compare the times of occurrence in the larvae occurred predominantly in the river, but were river (station A, B, and C combined) and bay (sta-more prevalent in the bay compared to Stage 1 (Fig. tions EN and NB combined) for each developmental 16). Annual Stage 2 abundances were consistently stage (Table 16). Dates of peak abundance of Stage greater at station NB than at EN, unlike annual Stage I larvae were not estimated for bay stations because 1 abundances in 1985 and 1996, which were similar, during several years this larval stage was rarely Significant (p s 0.05) positive correlations of collected outside of the Niantic River. In 1996, peak abundance were found among all river stations and abundance for both Stages 1 and 2 larvae in the river between stations EN and NB (Table 15). occurred on the third latest dates of the 14-year The later developmental stages (3 and 4) of winter period. Based on water temperatures of 2 to 3*C flounder larvae were usually not homogeneously during February (Table 6) and egg incubation times distributed within the Niantic River. The abundance reported by Buckley (1982), peak spawning decline at the upper river stations (A and B) as generally occurred in early to mid-February. development progressed likely represented a gradual Buckley et al. (1990) reported that egg developmen-flushing to the lower portion of the river and into the tal time was inversely related to water temperature bay. Stage 3 larvae were usually most abundant at during oocyte maturation and egg incubation. Colder

 % Monitoring Studies,1996

TABLE 15. Matrix of Spearman's rank order correlations among staticas for the indices of annual cumulative abundance of each developmental stage oflarval winter flounder from 1983 through 1996. ' Stage Station  ! B C EN NB i A 0.9429' O.8813 0.5919 0.7099 0.0001 " 0.0001 " 00258' O.0045 " 1 B 0.8637 0.5655 0.6396 0.0001 " 0.0351

• 0.0138
• C 0.4158 0.7187 l 0.1392 NS 0.0038 " l EN  !

0.2992 1 0.2987 NS 2 A 0.8725 0.8857 0.1517 0.3011 0.0001 " 0.0001 " 0.6048 NS 0.2955 NS I B 09385 0.3802 0.5648 0.0001 " 0.1799 NS 0.0353

• C 0.3670 0.4725 0.1967 NS 0.0880 NS EN 0.7275 0.0032 "

3 A 0.8374 0.5121 0.5780 0.3539 0.0002 " 0.0612 NS 00304' O.2145 NS B 0.7758 0.6352 0.4857 0.0011 " 0.0147 ' O.0783 NS C 06132 0.7714 0.0197

• 0.0012 "

EN 0.7143 0.004I " 4 A 0.5823 0.4000 0.5956 0.5050 0.0289

• 0.1564 NS 0.0246* 0.0655 NS B 0.7099 0.2791 0.2904 0.0045 " 0.3338 NS 0.3138 NS C

0.1780 0.2772 0.5426 NS 0.3373 NS EN 04621 0 0962 NS

• The two statistics shown in each correlation matrix element are:

correlation coefficient (r), and probability of a larger r (NS - not significant lp > 0.05), * - significant at p 5 0.05, " - significant at p s 0.01). than average winter water temperatures in 19% dates of Stage 1 peak abundance in the river showed (Tables 6 and 7) could have lengthened egg a significant negative relationship (Spearman's rank-development time, A comparison between the order correlation coefficient r = -0.660; p = 0.010), 1983-96 February water temperatures and the annual The later dates of peak abundance in 1996 were also Winter Flounder 97

    . - - -              - - . -                  .    -.-           - - . - - . -             - - - . _ _ . -               - - . _ . . . - ~ ~ _ ~

i ' A ) J estimation of developmental stage from length-

  ;               g 20 i r = 0 625                                                        qm                a.

l

                 '5w i p = 0.022 88 he length-frequency distributions of larvae (all          j
  .                    154 stages combined) collected in the Niantic River               '
}                  M      q                                                            (stations A, B, and C combined) were different from

a. 0 ' ,.

                      ,g q                                    ,                        those obtained for Niantic Bay (stations EN and NB            i combined) in 1996 (Fig.19). Smaller size-classes
}                  {      e

• 8,5 ,g

94.

• 92 f .e5 predominated in the river. In 1996, about 79% of i river larvae were in the 3.5-mm and smaller size-j fe o j 96 a,6,,* " 91
• classes, which was similar to the historical (1983 95 combined) length-frequency distribution, showing

                                                                                                                                                     ?

l 0 4 8 12 16 20 24 81% of larvae in the same size-classes. The size- , EGG PRODUCTION IN BILLIONS class distribution for the bay in 1996 was not f l e nsistent with previous findings (NUSCO.1987, Fig.17. The relationship (functional regmssion) between the index of annual Stage I -t J={ in the Niantic River (a 1988b, 1989, 1990, 1991 b, 1992a, 1993, 1994a, parameter of the Gompertz function) and estimated annual 1995a, 1996). For the bay in 1996, the length- + egg production from 1984 through 1996. frequency distribution was bimodal with the greatest i frequencies in the 2.5 to 3.0-mm size-classes (30%)  ; ! evident for Stages 3 and 4 larvae in the river and and the 6.5-mm size-class (9%). Typically, the i Stages 2,3, and 4 larvae in the bay. The later peaks length frequency distribution in the bay (1983-95 of abundance in 1996 for older developmental stages l combined) was unimodal, with the greatest frequency

                                                                                                                                                      ]

{ could have resulted from a later hatch or possibility (12%) at about the 6.0-mm size-class. Small larvae i j that colder than normal spring water temperatures in accounted for only 12% of the total. The length-  ! the bay (Table 7) may have slowed larval frequency-distribution for the bay' was described i 4 developmental rates. The relationship between water from a combination of samples' collected at stations temperature and larval developmental rate is { EN and NB; a majority of the samples were from EN. l

discussed in more detail below.- Samples at station EN were collected using a 333-pm i mesh net. Previous studies indicated that smaller l Development andGrowth larvae, primarily Stage 1, can be extruded through j

'                                                                                   this size net (NUSCO 1987). Larval extrusion The length frequency distribution of each larval                 through net mesh is related the size of the mesh stage has remained relatively consistent since                         opening and also may be related to the velocity of developmental stage determination began in 1983                        water filtered by the net. Historically (1983-95

! (NUSCO 1987,1988b,1989,1990,1991b,1992a, combined), about 87% of the Stage I larvae from j 1993,1994a,1995a,1996). Stage-specific length- station EN were collected in March and April.

frequency distributions by 0.5-mm size-classes in During this period in 1996, all collections were made i 1996 showed some separation in predominant size- at the discharges of Unit 2 and 3 under reduced

[ classes by developmental stage (Fig.18). Stage 1 circulating water flow conditions (Unit I was not larvae were primarily (73%) in the 2.5 to 3.0-mm operating). The percentage of nominal circulating i size-classes, 91% of Stage 2 were 2.5 to 4.0 mm, water flow at 100% capacity (Table 5) for the dates 91% of Stage 3 were 4.0 to 7.5 mm, and 83% of sampled in 1996 during March and April were 47% Stage 4 were 6.5 to 8.0 mm. These results were f r Unit 2 and 24% for Unit 3. Reduced flow consistent from year to year and indicated that resulted in a lower cross-net water velocity that developmental stage and length of larval winter appeared to increase the number of Stage I larvac ! flounder were closely related. These data agreed retained. Generally, the average flow (as a I with laboratory studies on larval winter flounder, Percentage of nominal) on sampling dates during  ; which showed that there were positive correlations Mach and April from 1983 through 1995 ranged' 1 I between growth and developmental rates (Chambers from about 90 to 100% at all units. k and Leggett 1987; Chambers et al.1988; Bertram et Additional evidence for the greater retention of j i 4 al.1996). This relationship was the basis for the Stage I larvae at station EN was the similar estimated annual abundance at both EN and NB (Fig.16). A l ? I 98 Monitoring Studies,1996 k

TABLE 16. Estimated annual dates of peak abundance oflarval winter flounder for each development stage in the Niantic River and Ba the number of days corresponding to the 95% confidence interval from 1983 through 1996. Year Stage i Stage 2 Stage 3 Stage 4 Nianric River 1983 March 5 (3) March 15 (2) April 18 (1) May 2 (4) 1984 March 7 (5) March 9(5) April 24 (5) May 19 (10) 1985 March 11 (!) March 16 (2) April 25 (3) May16 (7) 1986 February 26 (I) March II (5) April 20 (3) May 12 (10) 1987 March 10 (2) March 17 (3) April 20(2) May 9 (4) 1988 February 29 (1) March 9 (1) April 7(4) May I (5) 1989 March 8 (6) March 12 (5) April 14 (3) May 11 (9) 1990 February 17 (3) February 18 (5) April 21 (2) 1991 May 9 (14) February 27 (3) March 14 (II) April 13 (5) April 29(3) 1992 March 16 (4) April 6 (3) April 16 (2) May 2 (2) 1993 March 9 (2) March 14 (8) April 11 (7) J 1994 March 22 (4) March 31 (5) April 24 (I) 1995 May 10(3) March 2 (3) March 7 (5) April 20 (2) 1996 May 4 (2) March 14 (6) March 21 (4) April 19 (8) May 17 (5) Niantic Bay 1983 - April 7 (2) April 23 (1) 1984 May 10 (4) April 8 (2) May 4 (3) 1985 May 25 (8) April 1 (4) April 29 (6) 1986 May 18 (3) April 5 (30 April 28 (3) 1987 May II (2) April 6(6) April 28 (2) 1988 - May 16 (4) March 24 (3) April 22 (2) May 9 (5) 1989 - April 13 (1) April 23 (2) 1990 May 17 (3) April 3 (8) April 23 (2) 1991 - May 7 (5) March 28 (5) April 11 (3) April 29 (4) 1992 - April 15 (4) April 30(2) 1993 - May 7 (4) April 3 (44) May 6 (8) 1994 - May 23 (Il} April 14 (2) May 2 (2) 1995 May 20(3) April 4 (5) April 21 (4) 1996 April 28 (3)* April 7 (4) April 30 (2) May 24 (9)

• Due to low abundance during the 1993 sampling, the Gompertz function could not be fitted to the data.

" Corrected from NUSCO(1996).

review of previous annual length-frequency Stage i larvae due to net extrusion was corrected in distributions for the bay indicated a similar bimodal production loss estimates that are discussed below, distribution in 1985, with peaks occurring at the 2.5 An increase in frequency of larger size-classes in to 3.0-mm (23%) and thc 6.0-mm (ll%) size-classes. the river was not as apparent during 1996 as was On collection dates during March and April of 1985 reported in some other years (NUSCO 1987,1988b, at station EN, the flow at Unit 2 was 48% of nominal; 1989, 1991 b, 1992a, 1993, 1994a, 1995a). The about one-third of the samples were collected at this previous f'mdings suggested that some older larvae unit. As was found for 1996, Stage 1 abundance in were imported into the river, import of larger size- ' 1985 at stations EN and NB was similar (Fig.16). classes was also apparent in the length-frequency This suggested that some smaller larvae, primarily Stage 1, may be extruded through the entrainment distribution at a station located in the river mouth which was sampled in 1991-93 during maximum sampling net under normal flow and discharge velocity conditions. flood current (NUSCO 1994a). However, for entrainment impact assessment, the possible undersampling of WinterFlounder 99

. . _ _ _ . . _ . _ _ _ . _ ~ . _ . . _ . .                                         m.._.         _ . . _ . _ _ _ _ . . _ . . . ~ . _ _ . . . . _ . _                                      - _ .        . . _ _ . _ _ _ _ .

i 1 1 1 40 - - STAGE 1 30 - i - RNER i 8 i

                        $                                                                                                        8 20 -                     -

Y ., $ 20 - ,, 6 a g 10- . g 10- , i I i 0 - 0 U """""- l 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 'l j $0 - STAGE 2 Ib -

                                                                                                                                                      ~

I BAY

                          -40 8

• 30 - - 810- .

                                          ~
                     . @ 20 -

6 "

                                                                                                                                                              ~                  '
                                                                                                                                                                                       ~

g @

                                                                                                                              . g 5-10-1
                                                         ,O,n,D,,,,,,,,,

0, , , , _ 2.0 3.0 4.0 5.0 6.0 7.0 - 8.0 0- , , , , , , , , , , , ,h,n, , , 9.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 20 - LENGTH (mm) STAGE 3-Fig. 19. Length-frequency distribution of larval winter 15- . flounder in the Niantic River and Bay during 1996. (Note that the vertical scales differ between the graphs). b< 10- " Length frequency data from entrainment collec-M E 5- tions taken from 1976 through 1996 (station EN) were used to estimate larval winter flounder growth rates in Niantic Bay. Weekly mean lengths during a

n. n 0n ' ' season fonned a sigmoid-shaped curve (NUSCO 2.d 3.$ '4.d 's.$ 's.0 '7 .$ s.d '9.0 1988b). The linear portion of the sigmoid curve usually occurred in the middle of the larval season g, and growth rates were estimated by fitting a linear STAGE 4 model to individual larval length measurements during this time period. This model adequately g 20 -

described growth and all slopes (growth rate as 5 ~ 5

                                                                          ~

mm day") were significantly (p 5 0.001) different E from zero (Table 17). In addition, most intercepts of E go. - the linear regression were about 3, the approximate ' size of winter flounder larvae at hatching. Annual I growth rates for station EN were variable and ranged 0 - n from 0.048 to 0.100 mm day", with 1996 less than 2.$ '3E 'N 5$.6 4 7$ 8.0 Sy ' average. To validate this technique, growth rates LENGTH (mm) were estimated from length data collected at station Fig.18. Combined length-frequency distribution of larval ] NB from 1979 through 1989 (NUSCO 1990); annual winner flounder by developmental stage at all stations sampled in the Niantic River and Bay during 1996. (Note growth rates were highly correlated (r = 0.89; p s

                 - that the vertical scales differ among the graphs).                                                      0.001) with those from station EN.

100 Monitoring Studies,1996

TABLE 17. Annual larval winter flounder growth rows in Niantic Bay as estimated from a linear regression fitted to length data collected at station EN from 1976 through 1996. The 95% confidence intervals and mean water temperatures during the first 40 days of the time period are g!so given. Time period Growth rate 95% confidence Mean water Year included' (mm day") interval temperature (*C)*  ; l 1976 March 21.May 2 0.100 0.098 0.102 7.0 j 1977 April 3. June 5 0.076 0.073 0.079 6.7 1978 March 26. June 11 0.0$$ 0.052 0.056 4.8  ! 1979 March 25. June 10 0.058 0.056 0.060 5.9 J 1980 March 23 June 8 0.060 0.058 0.062 5.9 '! 1981 April 5.May 31 0.064 0.061 0.067 7.3  ; 1982 March 28.May 30 0.063 0.060 0.066 5.8 i 1983 March 6.May 22 0.056 0.054 0.058 5.2 1 1984 March 25 May 13 0.069 0.066 0.072 64 1985 March 17. June 2 0.059 0.057 0.061 6.0 1986 March 30. May 11 0.094 0.087 0.101 7.6 1987 March 22 May 17 - 0.079 0.075 0.083 7.0 1988 March 27.May 8 0.088 0.083 0.093 7.1 1989 March 26 Msy 7 0.069 0.060 0.078 7.0 1990 March 4.May 13 0.071 0 066 0.076 5.3 1991 March 10. April 21 0.059 0.048 0.070 4.7 1992 March 15.May 3 0.064 0.059 0.069 5.5 1993 February 28.May 16 0.048 0.040 0.056 3.3 1994 March 27. June 12 0.076 0.070 0.082 6.5 1995 March 5 - April 30 0.088 0.08i . 0.094 5.8 1996 March 24. June 16 0060 0.056 0.063 5.2

• Time period of the weekly mean lengths used to estimate growth rate.

Mean during a 40 day period starting at the beginning of the week that the first weekly mean length was used in estimating growth rate. In laboratory studies, water temperature affected 0.10- .7s the growth rate of winter flounder larvae (Laurence y r2 , ,,, ,,, j 1975; NUSCO 1988b). To examine the effect of temperature on estimated annual growth rates, mean j

                                                                                      - 0.06 -

8 ' "0 ' g, * ! water temperatures in Niantic Bay determined using W so 8'*77'l v J data collected from continuous recorders in the I l s2 . f * *d *" I A intakes of Units I and 2 were calculated for a 40. day {0.06d s,97*s2*73 's, et period starting at the beginning of the week when the 6 * * " '

                                                                                                                 / 7e ss I

i first larval length measurements were used to estimate the annual growth rate (Table 17). The 5 #3  ! 0.04 ! mean temperatures used may be not be indicative of 3 4 5 6 7 8 j the actual annual seasonal water temperatures MEAN WATER TEMPERATURE (C) i i because annual starting points varied from February i 28 (1993) to April 3 (1977). A positive exponential H E. 2R Be exponential relmionship between mean water l relationship was found between growth rate and tempenture U Q and thusianated gmwth mte G (nun per day) of winter flounder larvae at station EN from 1976 i wIter temperature with the point for 1996 falling through 1996 (G = 0.030 e""). nearly on the line described by the relationship (Fig. 20). A similar exponential relationship of tempera. j Therefore, the mean length of larvac collected at ture to growth was reported for larval plaice by Hovenkamp and Witte (1991). If temperature affects station EN during the period of April 1 15 for each - i year was compared to the mean March water growth rate, then the length of a larva at a specific time during the season should be related to water temperatures (Fig. 21). Again, there was a positive relationship with larger mean lengths associated with temperatures to which it has been exposed. warmer March temperatures. i Winter Flounder 101 4 i m_ % . __ _- - - . . _ - . _ . - _

  - . . - - -         ..----..--_....~.....-........._--..,..-.-.-o                                                                                            . . . -

l l ~ 5.5 : es

• 2

r =0.sa6

• 4 .7s .

                                      """                                               s1                                       77 g 51                                                ,, . 7.

l MAY20 ps.N ,, g 4.5i a4 90 g MAY W 1 3 .s1 \ i Li h 3g i

                                            .,a v.c,..::;"

so g77 w i l APR20"C'"% N. .s w-

                                                                                                                                              . as M     ,g,                            g               j , ,2.o.eu,
                                                                                                                       . o.oo                as       7s ,7.s 2.5 :         .

i i . I 'I C Am W , , , 2 3 4 5 6 7 1 3 4 5 6 MARCH TEMPERATURE (C) MARCH APRIL WATER TEMPERATURE (C) Fig. 21. 'Ihe relauonship between annual mean March water Fig. 22. 'Ihe relationship between March-April mean water

,                    temperature ('C) and the mean length of winter flounder                      temperature (*C) and the annual date of peak abundance i

larvae during April 115 at station EN for 1976 thmugh (estimated from the Gompertz function) of winter flounder , 1996. larvae at station EN from 1976 through 1996. 9 As concluded previously from comparisons of as Chambers et al. (1988) found that, at l annual length-frequency distribution and develop- metamorphosis, age was more variable than length a mental stages, growth and larval development are and larval age and length were indapandaat of one another. ) closely related. If water temperature affects growth

Growth rates were also estimated for Niantic River rates, then it should also affect larval developmental time. The timing of peak larval abundance should larvae using length data from station C with the methods given above. Station C was selected for this therefore be related to the rates of recruitment and loss (including mortality and juvenile meta- analysis because all developmental stages were col-morphosis), which, in tum, would be affected by lected there in abundance (Fig.15). Estimated larval development. Annual dates of peak abundance growth rates for larvae in the river were generally greater than for larvae from the bay, with the 1996 of larval winter flounder collected at EN were negatively correlated with the mean water rate for river larvae close to average for the 14-year i

temperature in March and April; carlier dates of peak series (Table 18). A linear model again provided a good fit and slopes (growth rates as mm dayd ) were 4 abundance were associated with warmer mean water temperatures (Fig. 22). This agreed with the results significantly (p 5 0.001) different from zero. Growth of Laurence (1975), who found that winter flounder oflarvae in the river was similar to laboratory growth

i. larvae metamorphosed 31 days earlier at 8*C than at rates of 0.104 and 0.101 mm. day" at mean water i

5*C, Annual dates of peak abundance varied by 41 temperatures of 6.9 and 7.5*C, respectively (NUSCO days 6uring the 20-year period, possibly because of a 1988b). i 3.6*C difference in the March April water A laboratory study with larval winter flounder held temperature between the earliest (April 13,1991) and at 8'C showed a decrease in growth as prey densities [ the latest (May 23,1978) dates of peak abundance. decreased, suggesting density-dependent growth due The average March and April water temperatures in to food availability (Laurence 1977). To examine

                 - 1996 suggested a later date of peak abundance than                         density-dependent growth in the Niantic River.

annual growth rate was compared to the abundance calculated, but the earlier estimated date of peak abundance may be related to the greater abundance index for Stage 2 larvae. This method also assumed of early larvae (i.e., in Stagel of development), that prey availability was similar from year-to-year. possibly due to the reduction in net exuvsion that The annual index of Stage 2 larval abundance was was discussed above. Despite the wide range in the at parameter (Eq. 2) for all three' river stations (Table 18). The abundance of Stage 2 was examined ' annual growth rates, a consistent relationship was found : between length-frequency distribution and because during this developmental stage larvae begin stage of development (Fig.18). This was consistent to feed. A density-dependent relationship was with laboratory observations oflarval winter flounder apparent during previous years (NUSCO 1990,

                                                                                              }991b,1992a,1993), but since 1993 the relationship 102 Monitormg Studies,1996 t

f

  .. . . _ . . _ _ . _ - _ _ _ . _ _ _ .                                      .._.-________..--___.m                                                        . _ _ _ _ _

i ' 1 i i ' ~ TABLE 18. Annuallarval winter flounder growth rates in the Niantic River as estimated from a linear regression fit to length data collected at i station C from 1983 through 1996. The 95% confidence intervals for the growth rate, mean water temperatures during the first 6 weeks of the i time period, and the annual abundance Indices of Stage 2 larvae in the river are also given. } j Time period Growth rate 95% confidence Mean water Stage 2 i Year included' (mm dayd ) 1 interval temperature (*C)" abundance index' 4 1983 March 20 May 1 0.100 0.096 0.104 6.1 749 l 1984 March 25 May 6 0.100 0.094 - 0.105 6.4 1,501

1985 March 31 May 26 0.084 0.080 0.088 7.7 4,676

1986 March 23 May 4 i 0.109 0.103 0.115 8.0 176 j 198? March 22 May 10 0.099 0.095 0.103 7.2 829 *

1988 March 20 May 21 0.099 0.094 0.104 6.8 4.469 '

i 1989 March 26 May 21 0.087 0.082 0.092 7.4 3,976 1990 March 25 - May 13 0.106 0.099 0.113 7.5 365 I 1991 March 10- April 28 0.123 0.114 0.132 6.9 252 } 1992 March 15 May 17 0.088 0 083 0.093 5.7 1,367 ' i 1993 March 7 -May 16 0 070 0.065 0.075 4.1 133 { 1994 March 20 May 29 0.072 0.068 0.076 4.7 1,248 j 1995 March 12 April 30 n.130 0.121 0.140 ' 6.8 2.023 1996 March 24 -May 19 0.096 , 0.092 0.099 6.7 3,677 4 4 1 I

• Time period of the weekly mean lengths used to estimate growth rate.

7

• Mean during a 6-week period starting the week of the first weekly mean length used in estimating growth rate.

! ' a parameter from the Gomportz function for Stage 2 larvae in the Niantic River (three stations combined). t was no longer significant (p = 0.562) when tested factors controlling larval growth (Buckley 1982). j with functional regression. Because there was a Although Laurence (1975) demonstrated that the i strong relationship between growth and water i metabolic demands of larval winter flounder in- ) t:mperature in the bay, both Stage 2 abundance and creased at higher temperatures, the growth rate also j water . temperature were used as independent increased if sufficient food resources were available, } variables in a multiple regression model to examine and other laboratory studies (Laurence 1977; i growth rates. Prior to conducting the regression Buckley 1980) showed that larval winter flounder i analysis, it was determined that the two independent i growth rates depend upon prey availability. In variables were not correlated (Spearman's rank-order summary, growth and development of larvae in i correlation coefficient r = 0.141; p = 0.631). The Niantic Bay correlated with water temperature, but in ! multiple regression including 1996 data was the Niantic River growth may also be affected by 2 significant (p = 0.035; r = 0.456) with the slopes larval density as well as by water temperature. i being positive for temperature and negative for Stage

2 abundance, although the Stage 2 slope was not 1 Mortality significant (p = 0.141). This relationship suggested j

that winter flounder growth in the Niantic River may From 1984 through 1996, total instantaneous

be a function of both water temperature and larval density.

mortality (Z) for larvae in the Niantic River from i The varying results from year-to-year hatching to just prior to metamorphosis ranged suggested that factors affecting growth, such as prey i abundance, for which no information was collected, between 82.4 and 97.9% (Table 19). Estimated larval mortality in 1996 of 2.96 was greater than the may be more complex than just water temperature ! 12 year mean value of 2.67. Based on larval length-and larval abundance. Slight declines in growth rate frequency distributions in the river during 1996 (Fig. ] c used by less than optimal food, unfavorable

19) and previous years, most winter flounder larval temperatures, disease, or pollution leads to longer mortality occurred between the 3.0- to 4.0-mm size-j developmental times, during which high rates of classes. A 74% decline in occurrence frequency was  :

. mortality have a profound effect on recruitment i found in 1996 between these two size-classes, which (Houde 1987). Food availability and water included yolk-sac (Stage 1) and first feeding Stage 2 temperature appeared to be the two most important larvae. This initially large decline was followed by i WinterFlounder 103 i i

1 I i TABLE 19. Estimated larval winter flounder total instantaneous thought to be determined by the availability of suf- I monahty rate from hatching to the 7. nun size-class from 1984 ficient food after completion ofyolk absorption. through 1996-Predation may be an imponant cause of larval Abundance index winter fl under monality. The escape response of Newly 7-mm Monality Instantaneous larval winter flounder to predators was studied by Year hatched size. class (%) mortality rate Williams and Brown (1992). They found that escape response increased with increasing larval size, but 1984 6.500 654 39.9 2.30 remained slower than that of other larval fishes 19 483 3 4 7 examined. Larval winter flounder may be vulnerable 1987- 6,480 474 92.7 2.62 to both fish and invenebrate predators. Although 1988 24.561 678 97.2 3.59 susceptible to attacks by planktivorous fishes, the , 1989 19,192 394 97.9 3.88 occurrence and abundance of fishes that could 1990 7,915 653 91.7 2.49 potentially prey on larval winter flounder is low, 8 0 Particularly during the early portion of the larval 1993 1,874 88 95.3 3.06 winter flounder season. Most predation is likely by 1994 7,270 761 89.5 2.26 invertebrate contact predators, such as cnidarians. 1995 13.088 1,536 88.3 2.14 Jellyfish predation can affect the abundance of 1996 11,151 576 94.8 2.% flatfish larvae. Evidence of a causal predator-prey mean. u9 relationship on larvae of plaice and European flounder (Platichthys flesus) by the scyphomedusan Aurelia aurita and the ctenophore Pleurobrachia smaller decreases to the 5.5 mm size-class, indicating pileus was reponed by van der Veer (1985). I a reduction in the mortality rate. Pearcy (1%2) However, predation by these species was believed to reported a greater mortality for young winter only terminate the plaice larval season and did not flounder larvae (20.7% day") compared to older ultimately affect year-class strength (van der Veer individuals (9.1% day") in the Mystic River, CT. In 1985; van der Veer et al.1990). Laboratory studies a laboratory study on winter flounder larvae, showed that successful capture of plaice larvae Chambers et al. (1988) reponed that larval mortality increased as medusal size of A. aurita increased was concentrated during the first 2 weeks after (Bailey and Batty 1984). Pearcy (1%2) stated that hatching. Based on the estimated growth rate in the Sorsia subulosa medusae were important predators of river for 1996 of 0.096 mm day" (Table 18), a larva larval winter flounder in the Mystic River and had would require about 10 days to grow from 3 to 4 mm. greatest impact on younger, less mobile larvae. The decline in abundance between these two size- Crawford and Carey (1985) reported large numbers l classes would be equivalent to a mortality of about of the moon jelly (A. aurata) in Point Judith Pond, RI I 4 12.6% day , which is less than the rate reponed by and believed that they were a significant predator of Pearcy (1%2) and for the previous 3 years in the larval winter flounder. A possible predator of winter Niantic River, which ranged from 14.6 to 19.9% flounder larvae in the Niantic River was medusae of (NUSCO 1994a,1995a,1996). Laurence (1977) the lion's mane jellyfish (Cyanea sp.), which was found that winter flounder larvae had a low energy prevalent in the upper portion of the river at station conversion efficiency at first feeding (i.e., Stage 2) A. Marshall and Hicks (1%2) also reported that compared to later developmental stages, and that it jellyfish were abundant in the ' upper river. A was probably a " critical period" in larval winter laboratory study showed that winter flounder larvae flounder development. Hjorleifsson (1992) showed contacting the tentacles of the lion's mane jellyfish that the ratio between RNA and DNA, an index of were stunned and ultimately died, even if not condition and growth rate, was lowest at the time of consumed by the medusa (NUSCO 1988b). first feeding of winter flounder (about 4 mm) and During 6 of the 14 years (1983,1984,1986,1989, that these ratios were affected by food availability. 1990, and 1994), the weekly mean larval abundance The " critical period" concept, hypothesized by Hjort oflarvae at station A was negatively correlated (p s (1926), was discussed by May (1974) for marine 0.05; Spearman's rank-order correlation coefficients fishes. In many cases, the strength of a year-class is having a range of -0.736 to --0.927) with weekly mean jellyfish volume during the period when both 104 Monitormg Studies,1996 l I l

                                                                                                                                                                               *i ]             '; :; j ij                                        4          i!                             !l.           } :i lomsoa                                          aGot he rs e(

ogr aae c : r o fs e cr iq Ew s bbr s e t l 1( aF po wTs wlt not a q g e t i 9wt 9isg i >o3h g%$ Eg g q a hae r r uc b wi oh (F u eljm ant n er np ui oige mp Num.t enTh l cyil ed r t t r ri 6 ha2 t . 0 2 1 1 1 1 t s e t r i e a vmcunwgi unt e e . c rot n rp a i ia 4r e t i 3 6 8 0 2 4 6 iol e o e br kfiu him t ne ic s ee geEr m e t r wan mte) f lp 9o. 5 *9  ;*  ; ni ko fnf i er r e a dr v e 2y l has s mdt e a e z ann lat nt e s p uo o %n AoC F 7 , 1

                                                                                                                                                    .                                                             oel       r       Ast e  r     ws      d      n     f c o  l  ,r       3                 e oxmol cu cy,t Nv                                af ugziin E                                                                             f  y    ,e                                                        ) ian l                                nc     t      or                  s c      im           B                                                                             t                 o            r      dfe          ut                   cp        a e

f n R . fo ns vdib onp n 2, 4 m ~ h e d hc a t c t e e r c nha rp n o o e de daa ne e hwp r sala o ev wa c c t v t aie i f t ar i hi ~ i eu r e uk r e cl a r mf e d e sahs oe ioof ne s oioe t t l oc rhne s eiuday Tndrl i l de e s o 3,3 ~. v t dre ,n o s mwl mp r e t s r e e r l e nk( r t nNn e o e l e i ia e ddal a t i a c es s nt s h e t s he c t i f s Tnl ba a a v of c eai nf o M 1 0,.T. 1 r ag c d libots c d d - e rp n t a e t l a v r olc iwbr ol ne el sg t u c dt ez lbup ee t iansy e it A 9 nr ne e r vokot a r e a o ut oe r o p he f o. e a a wend niw t e c e R 1 7,l-

                                                                                                                                                    .-                                 8 3-                         dad e oh            ft da          p afMr               e al     dt rh e e fal he u u )w a t     t    j                             dr t ohT1 l

nc is s e g ike i wRa vRek r a ivyl C H 1 9 n uirnee o a iir l o ndc o ve dl e n iw jd sti a t a s e l wm ame h )3 2, t y hu e dve s e . na s a e g ner axy wpr g s( t l e s

                                                                                                               ) r de m

4 ^i..

                                                                                                                                                             .                        9 5

ju t ol r l lall imdhr f vyp e s a i t nr h dl a i brr re m t e v a e r p abndLinny y il e brg a v o a9ua 1 7n b a s ot md ip e iwua t ir n 3, 1 " 8r'

                                                                                                                                                          -                                                      e s afp l                 i              - e eorp pu iapl e i ng e e o t

c . yl e u lado f n hl ahapsage Ar 1 a t s e a f m 6 aa dp u doc 9e n en hnv g o . i

                                                                                                                                                                                                                                                                                                ,s e   ll r     a      . c                                                                                                                 7, a e                                                                                                                                                                                                                                       v v nM               o ac uk 7d                                  w                   A                                                                              ig b                    r r f hr miAuiat                                        hc ea es ct nr 1 ,)f.

l ak h e t 5 d e 1 u s bait n s n ke a r s mdis) ue e 89me P R h fnu ce nl c e .r e a ea yre f.oy p o a fl t t e eg t e r t r t e( hvh l a e th s e s uo r uot an z e hn l eg fe s r e oi e t ag m s n it k yl 3s vhr fo t I L 4 2, /-ig I i* pr o a e r mnm dae w c e o eh Oy r l a lj t r e u vg h

                                                                                                                                                                                                                                                                                                       .d fe o           ae ly        7s-     e e    r v

a o h7 - f a st at a f1 mr o emimdr n a o r ft ooc ul uy 1 2, ./

                                                                                                                                                                   ;.A' d

a i t h t e a fn l nAs ilyl l t t t ef iuhoTiv hn l I a lq nl ua r r m oe ibajyp a .e nc e e mghn a 8 nur e i s s s o a 99mma r 7a

         -                wmMe                   l au                t            e           e M

A n. pu nvll a il it o h y y u l v 5 at u s e v r ir tciyi t s day 1 mu r u t r ulon aamr ss y a laa nhr ia e o o f o9s 1 9 a Y 5, Ji

                                                                                                                                                        .          ji p

e 1 9a e fs fe a n nP e 9n i anae a e a m9 9 mlati - dd t de e nl s nf r 5 . p r 6t . a h w d A k d ,r t s r e ,6 s Tc m Np T.a cp c p aaaF oo r it s iox Nc eer io hc er er u f dc 0r e Tas hn dmroie t n eshn ooal r ss c B nhu c c nta o. 0g ei id r nnni ondng b"ga S %w$ gay a ree dr w eg anel si alg n l t dut u. ep at t a l in e a ibl l iu e a c u c aha2l e l 4 2 4 0 8 i nieNg i c e v t t i hsiu rut r c a p m0r e ms t 1 l e r i c ec f rodti e ot l t a t i al w o+j: t as c ed t t de a l s iiiop t t t e) s a s a r ey o e ne onno s ie ag vr e t .

                                                                                                                                                                 ~         -                        t                                                                                    t i

n i u s n bait J i, yl r e mmn is md it r n oat eivd f R o ,e p ny g e e nb

                                                                            )

adl u n vt r n aeer r h . g e r f l c t id t Ra a h o l a vn e e a na e n l t nnl t t s o g t a e l u r t

• oRhi t wt werc hi inirnesy h r a (F n t

e t a r 5 u t e e r e s s s n ade e xh he ne t s t ac ne f l l e 8 3 9 n iv ta v s e,r t i id l a ie r c oa gr ome . hat awrv i 5 am r t f E 9 t vg ti o n eewies ealge t t s 8.y ti d m (Z o t t t t o uinoui Gt 0 4 der an e i a e t ut 9 1 e na) cy e h ndrg not dn fme s Go e e* e .hs*;g pr r e wn o ng ih 81 e nyg t wor 2e s e r e 4 e k -dr ( v E r h 5 *o .= g e ut t si e rf yt s t o awp e a c r t e h e o lapro N7e n t ep g i P R1 5 7 9 *.x B 4 o.0 g er a h e Eop e r a3ot msi 29s 9i s dala ). l t sp l ol o t o c 6 l ne n d f e oms ep r np dr it dn u ifr de u r m i f mo me dp n g (f u n O D e n dN,7 o t 7 p c o r nog e c Nfrot t d l d a n hy i io Ap t t e . n t s e teh d e hids 1 i t t U2 eg hus e ome ir tc c ax otr c oie cdhu ni p e t dsg o 1 h l 1 C0 e dns 9al a T ig e e _ ng e e na ep nui t io 9 n( 7 dl u n doi T s c sp md u i t ie 9n ir ng hp ne e ag t dp r ,d tl c hn e r t n o c a t nom rp nwo x h.5 I O2 t e i rgg e r iv a t 6lg t i e on oe noi. n e ,e wfima s 19a ainr N5 l W no dd e e t a ai nt dnt wpi x c eun hrine r t or ip t - s u s n i t t muc( dr e ha e wrv l r f oincint t i n i m t rg oe h u r mnes ir g I N 3 B0 x e 3* . wc t t i i oi r t ao t e r Bs a i Tp nhe e c d8 s e ,2 f t det o d s n n s e r F otiTr r t oa em odxg d(g e e nt a de g hal r ohi t s I L L N horg r ny i f l ce ft noau a 9d wt oh1 e oi 1 ua r t h v o( wge c u t l a nbar e)g e 1 vfe o t I 3 ' e, e g e a uu nt 9 i 9n = ig io axw 9a 9e he Nib)n o9,3 ,d l o l nt a O5 , i n n t u n i ( e t y T nl uiciads et vo p r a 6 t r a e N is r e ns oiet t de da6 ag s p n c a1 ay t t as S nml afer r e ofrp n 0 fr e n riupb o oant d e r o b )9 l u e l fg u s t hoci t i t w e 4 0 a l atio o ivl e odi l n fau r ng o dg 7n di 6u os t p m t o a c ru v t e l a c e s nmladot l e l d1 wa r r bt i o o ,e e oRn e r t bo e ;1 i cu m t . fht cvrnl e wi v i d 1 s e ft v i t oe r 1 4 s ot a 0 b )3 . er a e vthhel t nus an g heh 5 n nie wt t t 9tip nit a oe i 5 e l e e s. ) g e r e doiha s rp e e egy s r r e 5g = ae e h9n l n. s ll

i 0 ) ! l 5 l j identified in the larval stage, the values of Z were Juvenile Winter Flounder i compared to egg production estimates. I. addition, 4 the effect of water temperature on larval mortality [ was examined. A multiple regre.: ion model used Age-OJuveniles during Summer j egg production with various combinations of a seasonal water temperature recorded at MNPS. The Abundance. Abundance of demersal young of-I combinations included monthly means for March the-year winter flounder was determined using a 1-m though May, combinations of March and April ar.d beam trawl. Although beam trawls are much more ' April and May, and seasonal means for winter efficient than small otter trawls for collecting (January-March) and spring (April-June). The best juvenile flatfish (Kuipers et al.1992), the densities of i model indicated that larval mortality increased as egg young winter flcunder reported herein should be i production (p = 0.014) increased and spring regarded as minimum estimates because ofcollection j temperature (p = 0.007) decreased (Fig. 25). This inefficiencies. For example, using a beam trawl j suggested that density-dependent larval monality Berghahn (1986) caught more young plaice at night i may occur in the Niantic River that is further in comparison to samples taken, during the day and moderated by spring water temperatures. The effect Rogers and Lockwood (1989) showed that replacing of temperature on mortality may be due to its positive tickler chains normally used with even heavier, j relationship to rates of larval growth and 2 spiked chains nearly doubled catches. Efficiency of development. For example, the low mortality rate in the NUSCO l-m beam trawl was discussed in I 1995 was likely related to the lowest annual egg NUSCO (1987,1990). Large mats of the alga

production estimate and relatively warm spring Enteromorpha clathrata, which hampered sampling i temperatures and the relatively high rate in 1988 to efficiency in the Niantic River during much of 1993 4-high egg production and cool temperatures (Tables 7, and occasionally in other years, were not found this 13, and 19). year.

In contrast to the high densities of Stages 3 and 4 { j larvae and metamorphosing young found during 1994 and 1995, larvac were relatively less common A'J1 D in 1996 (Figs.15 and 16). Following metamorpho- l

<sp_ g y y x ~

5 's sis, numbers of young settling in the Niantic River I j were also relativel low in 1996. Densities of about

                                                              /    j'                           -4                    70 fish per 100 m{ were found in mid-June at Wl p a. F "'( g "

g ' l / / -

                                                                                                   ,,9                but at all other times densities were less than 25 fish           I

$ g, N s j Per 100 m' (Fig. 26). Young winter flounder were 1 - g g ,",8 particularly scarce during late summer (August and 5 '5 %- lad j j, f{ 4 September). Unlike most other years, abundance of

                                                                           ,  g      j           Ja g                young in 1996 was usually greater at WA than at LR.

, g ,  ; '

                                                                                                 .g                  The CPUE abundance index for early summer at LR

! N '

                                    '                                         4               ,

g (8.8100m) was the lowest of the time-series dating j

                                                                                      ,  j                           back to 1983 (Table 20). The index for WA (21.7) p jlj,                                 was the fourth lowest. The CPUE of 3.0 for the 8
                                                     ,            g        p{              9 U                        second half of summer at LR was again the lowest saa      ta y , . ] j)"'                                         value for all years sampled there, whereas at WA the d, ao y                                          CPUE of 6.2 was only greater than during 1989 and 1993.

F'ig. 25. Relationship between the instantaneous larval Overall, catches at both NR stations were consid-mortality rate (Z) and annual winter flounder egg production in the Niantic River and spring (April-June) mean water erably less than during 1995 and the year-class pro-duced during 1996 was among the least abundant WmPerature N at the MNPS inh fmm 1984 thmugh 1996 (Z = 10.10.295 + 0.074 egg production - 0.734 spring since 1984 (Fig. 27). A comparison of erely and late

                              **                                                                                 season median CPUE values illustrated that initially strong sets of young during early summer may not necessarily result in high densities of fish at the end 106 Monitoring Studies,1996

d f I i 1 251 LR-996 Mean length of young at LR during late summer ' g

             - 20g:                                                             (July through September) of 1996 was 60 mm, but at WA was only 47 mm (Table 21). His was the

{. h15 , T largest difference in mean length found between the r 8 tog j 1 NT two stations since 1988, when it was 15 mm. Growth i D ^ , was probably affected by water temperature, which i '

• 7 increases respiratory and other metabolic demands. (

5-]! * *  ! 0'. .

                                      '. ;l   , ,     ,       ,

2 Even though faster growth occurs in warm waters, i optimal growth temperatures for young winter MAYJUNE JULY AUGUSTSEPTEMBER flounder can be exceeded (Sogard and Able 1992). Bergman et al. (1988) and van der Veer et al. (1990) i 11 ^ 996 noted that growth of young plaice in northwestem l 90- Europe was not food limited, but was related to [. $ 80- l ! Prevailing water temperatures and the length of the {70- growing season in different nursery areas. b Furthermore, fish grew more rapidly on the warmer l 5 4[o. nursery grounds in embayments than did fish settling b30-  ? on cooler North Sea beaches. As noted above, water 20-

                                          ,   I                     T temperatures in Niantic Bay during the summer of
               'I                         "                                                                                                1 o                                                            1996 were the coolest in 21 years. Some of the                !

MAYJUNE JULY AUGUSTSEPTEMBER mnces b mean le@ noted each year Wm  ; LR and WA (Table 21) were also likely due to water i temperature, as WA was generally warmer (ca. 0.5 - Fig. 26. Weekly mean CPUE (12 standard errors) of age-O < 1.0*C) than LR. However, no consistent differences winter flounder taken at Niantic River stations LR and WA  ! by 1-m beam trawl during 1996. were found between these stations, with differences ' in mean length during some years relatively of summer (Fig. 28) Dese differences were related substantial (8 - 16 mm), but small (-2 4 mm) in  ! to variation in mortality rates (discussed below). some other years. Most fish were produced in the Niantic River during Relatively large mean lengths associated with low i 1988,1992, and 1995, with year-classes from the abundance in some years and small mean lengths in i mid 1980s,1989,1993, and 1996 particularly weak, years when fish were abundant suggested a Growth. Increases in mean length over time were component of growth that was density-dependent. used to express growth of age-O winter flounder. Numerous accaunts on the growth of age-O plaice During 1996, growth rate was greatest m early were inconchesive on the effeo # density and other summer with smaller increases observed in biweekly factors influencmg gcowth. I,ergman et al. (1988)

                                                                                                  ~

means during late summer (Fig. 29). Mean lengths and Zijlstra et al. (1982) re-examined reports of were virtually identical at LR and WA until early density-dependent growth in British waters by Steele July, when fish at LR began to show larger increases, and Edwards (1970), Lockwood (1972), and Rauck De difference between mean length of fish at the - and Zijlstra (1978). De reviewers concluded that two stations was about 20 mm m September. Fast increases in length corresponded to maximum growth growth after settlement followed by a rapid decime m expected from prevailing water temperatures and that growth rate was reported for young winter firwider grawth was not density-dependent. Similarly, Pihl in New Jersey bays by Sogard and Able (1993), wh and van der Veer (1992) determined that growth of found nearly imperceptible growth by the tim; young young plaice n Swedish bays appeared to be affected reached 50 mm m length, it is likely that growth by ambient water temperatures and was not food-l compensation occurs in winter flounder where size- limited. However, Berghshn (1987) and Karakiri et  ! ct-age, which may diverge in larval stages, converges al. (1989) suggested that food limitation and not l during the early juvenile phase and progressive water temperature could have been responsible for declines are seen in size-et age differences (Bertram growth differences of plaice observed among et cl.1993). different years within the German Wadden Sea. I I WinterFlounder 107  ! i

l I 4 I TABLE 20. Seasonal 1-m beam trawl median CPUE (number 100m ) of age 4' winter flounder at two stations in the lower Niantic River (LR and WA) from 1983 through 1996 ' ) Median 95% confidence Coefficient Survey Tows used CPUE interval for of year' Station Season

• for CPUE estimate median CPUE skewness'

         ]983            LR         Early            30              32.7              20.0 50.7                  2.29 LR          Late            27              10.0                8.0 133                  0.49 1984             LR         Early            40              18.8               16.7 25.0                 0.63 LR         Late             36               63                 3.87.5                  0.58 WA          Late             32              llJ                 8.0 I7.5                0.94 1985            LR         Early             40              13J               10.0 163                  0.91 LR          Late             32               7.0                6.08.0                  0.97 WA         Early             40              15.0              10.0 20.0                 0.8 I WA          Late             32               9.0                8.0 10.0                0.70 1986            LR         Early            39              33.8              233 -40.0                  033 LR          Late            36              13.8               12.5 17.5                 0.80 WA          Early            40              21.7               12.5 26.7                 1.49 WA          1 ate            36              18.1               15.0 20.0                2.03 1987             LR        Early             40              59.2              533 733                  -0.12 LR         Late             36              17.9              12.5 26.7                 0.70 WA         Early             40             283                21.7 383                  0.27 WA          Late             36              10.6                6.0 13.8                0.83                    I 1988            LR         Early             40             61 3               52.5 - 72.5               037 LR          Late             36             60.0               50.0 - 70.0               1.17 WA         Early             40             40.0               32.5 51.7                 0.13 WA          Late             36             383                333 51.7                  0.22 1989            LR         Early            40              17.5               11.7 21.7                0.09 LR         Late             36                8.8               7.0 11.3                0.84 WA         Early             40              10.0                83 13.8                  1.16 WA          late             34                5.5               4.0 10.0                0.66 1990            LR        Early             40            I$6.3              137.5 - l87.5               1.05 LR         late             36             20.0               15.0 52.5                 1.10 WA         Early             40             68.8               50.0 - 95.0               0.62 WA          Late             36              13.5              10.0 19.0                  1.20 1991             LR        Early             44             77.5               51.7 90.0                 0.%                      l LR          Late             36             21.7               183 283                   0.75 WA         Early             44             37.9               30.0 433                  134 WA          Late             36             25.8               213 31.7 i

1.27 1992 LR Early 40 90.0 57.5 122.5 1.16 LR Late 36 28.1 238 333 0.51 WA Early 40 74.6 56.7 - 82.5 135 WA Late 36 30.0 27.5-32.5 013 1993 LR Early 20 10.6 7.0 - 15.0 0.68 LR Late 20 5.0 3.07.0 1.15 WA Early 20 5.0 3.87.5 2.57 WA Late 20 5.5 4.0 10.0 0.77 108 MonitoringStudies,1996

TABLE 20. (continued). Median 95% confidence Coefficient Survey Tows used CPUE year

• interval for of Station Season' for CPUE estimate median CPUE skewness' 1994 LR Eariy 20 128.8 125.5 t72.5 0.38 '

LR Late 20 62.9 38.3 75.0 0.26 WA Early 20 126.3 92.5 - 192.5 0.31 WA Late 20 49.2 35.0 - 55.0 -0.79 1995 LR Early 20 87.5 52.5 - 140.0 1.82 LR Late 20 15.8 12.0 26.7 1.% WA Early 20 116.3 85.0 137.5 2.31 WA Late 20 55.0 28.3 - 70.0 0.59 1996 LR Early 20 8.8 5.0 - 15.0 0.27 LR Late 20 3.0 3.0 - 6.0 1.42 WA Early 20 21.7 11.7 - 27.5 1.30 WA Late 20 6.2 i 3.0 - 7.0 1.76

• For age 0 fish, the year class is the same as the survey year.
• Early season corresponds to late May through July and late to August through September.
• Zero for symmetncally distributed data.

       ,                    Early summer (late M ay4uly)                                         Early dashedhne; Late solidkne E 14@,                                                                g1403
                                                        ~
                                                                                " 12%                                 .                 !.

8 12d  ; N [303j f l ; h1 /- g80 23 I 80g /. - 60 i i 1 U sof , l \

• l '

40 1 sl 4 s 20

                                                                                             .._ [             . , -

0 ,,,,,,,,,,,, y O'.

                      , , , , , , ,,,,,,                                                  84     86        88      90         92      94         96 84     86       88     90    92      94    96 YEAR YEAR Fig. 28. Comparison between the early and late summer
      ,                 Late summer (August September) seasonal 1-m beam trawl median CPUEs at Niantic River E 603                      ,

stations LR and WA combined from 1984 through 1996. 8 '

       * $0 :

0: ,, The effects of density (median CPUE during July h 40]} /j and August) and water temperature (cumulative de-j \T [o 30j t gree-days during the period of May 15 through Sep-6 20 ! . /

• tember 30) on growth (mean lengths achieved during i [^ \ i late summer) at each station were examined using a g I 1 multiple linear regression model. Water temperature 2 0 , , , , , , ,

g did not significantly affect growth. Using functional YEAR regression, growth of young at LR, however, was found to be significantly (r -0.587; p = 0.027) Fig 27. Early and late summer seasonal 1-m beam trawl negatively correlated with densities of fish during median CPUE and 95% confidence interval at Niantic late summer (Fig. 30). The relationship between River stations LR and WA combined from 1984 through abundance and mean length was only significant (r = 1996.

                                                                          --0.779; p = 0.005) for station WA if data outliers from 1988 and 1991 were excluded. The pattern of WinterFlounder 109

 . __. _ _ _ _ _._. _                              __.____.___._____,.m.                                                         _ . _ ._ .-.___

i 2 { j LR (sohchine) did not make any difference in the relationship . E [j

                      ~                                                             .

between density and growth at LR. Other factors found to affect growth of young fgsoj

                                                    ,4                     ; .                   winter flounder include physical location and specific y  j                        ,

4 ~ ,,,..V,j,,** y * *...... habitat (Sogard 1990; Sogard and Able 1992). y ^ Benthic food production and its availability also may j 203

• WA (dashed kne) differ among areas within the Niantic River and '

h 10) . likely changes from year to year. Karakiri et al. .! {0 , , , , , MAYJUNE JULY AUGUSTSEPTEMBER

                                                            ,        ,     ,   ,    ,            (1989) reported differences in the size of young Pl aice of similar age between Wadden Sea estuarirw 1

l nursery grounds (larger fish) and coastal waters off Fa.g. 29. Weekly mean length ( 2 tandard errors) of age-O Germany (smaller fish). They suggested that the  ! winter flounder taken at Niantic River stations LR and WA diff*"*'8 were due to lower water temperature,

                                                                                                               .                                          j by I-m beam trawl during 1996.                                                   food limitation, or wave action in the waters outside j

of the Wadden Sea. Al-Hossaini et al. (1989)  ; abundance for age-O winter flounder during 1988 reported greater growth for cohorts of plaice that ) was unlike all other years ssunpled since 1983, settled relatively early in Wales, but these fish also Abundance rose to moderately high icvels by mid- had higher mortality. Conversely, growth was slower June and remained fairly level, without the normal for late-settling cohorts, but survival was higher. decrease seen in all other years (NUSCO 1989). In Mortality. Catch curves constructed from weekly  ; 1991, water temperature, as indicated by degree-days abundance data by year and station were used to from May 15 through September 30, was the obtain estimates of monthly instantaneous mortality  ; warmest of the 14 year series, indicating a possible rate (Z,,,). This method assumed. that young I threshold temperature effect for that year at WA. comprised a single-age cohort throughout the season. l The inclusion of data from these 2 years, however, With some exceptions, the catch curves generally fit the data well with relatively high r* values found, TABLE 21. Comparison of the mean lengths of age-O winter flounder taken at stations LR and WA in the Niantic River dunng late July through September of 1983 through 1996. Mean length *in mm by station and year: 66 61 60 59 5s 57 56 55 51 51 51 50

          . LR          LR       LR      LR            LR             LR         LR    WA   WA          LR       WA     WA 83         93       %       84            89              85        91     91    93        88       88      89 Mean length by station and year (continued):

48 47 47 46 46 45 44 43 43 42 42 42 40 39 1s LR WA WA LR LR LR WA LR WA WA WA WA LR WA WA

           .92          %        87      95            86              87        92     90   84         85       90      86     94         94      95 Difference in mm between the late seasonal means at LR and WA:

Year 84 85 86 87 88 89 90 91 92 93 94 95  % Difference (LR.WA) 16 15 4 -2 0 8 I i 4 10 1 8 13

• Means joined by underlining are not significantly (p 5 0.05) different from each other, as determined by analysis of variance and Duncan's multiple-range test.

110 Monitoring Studies,1996 I

       . .. . - - -                            - - - - - . - -- - - - . - - ---~-

Station LR (Lockwood 1980; Poxton et al.1982; Poxton and I. 70 i r = 027 Nasir 1985; Al-Hossaini et al.1989). Annual l 65j p = 0.027 variation in mortality rate resulted in the differences 1 3 soj .*,*

• observed in year class abundance for Niantic River j

( 55 ! winter flounder. Notably, little observed mortality in 5 '503 . {

      "'                                                                              1988 meant that a modest set of young resulted in a     '

45j ... relatively strong year-class, whereas high mortality

             ,oj                                                     ,               during early summer in 1990 and 1994 reduced 3

i initially high numbers considerably by late summer 0 1O $0 $0 4O 50 60 7o in both years. MTE SUMMER MEDIAN DENstry During 1988-92, when both areas were sampled,  ! mortality of young was much greater at two stations stanon s sampled in Niantic Bay than in the Niantic River  ! g 55- (NUSCO 1994a). Except for a station just outside l es a r =.0.779 the mouth of the Niantic River. in 1988, no young h3 w 50

• es x p = 0.005 were found in Niantic Bay following mid-summer,

                 ]; %                                                               Even in 1988, however, densities at the Niantic Bay f                                                                               station in late summer were only 10 to 15% of those
                                                                                                                                              )

I #9

                             .       ~

s

                                             =

in the river. High natural mortality of young winter 1

                                           'N                                       flounder in Niantic Bay was the probable reason for       {

N  ! N. . declines in density following larval metamorphosis l h 35 i , , . , and settlement to the bottom, rather than from off-5 0 10 20 30 40 50 60 station emigration. Because of the apparent lack of ] MTE SUMMER MEDIAN DENSrTY Production of young in Niantic Bay in comparison to the Niantic River, no further sampling was conducted Fig. 30. The relationship (functional regression) between in the bay. 2 the density (median catch per 100 m ) and mean length of age-0 winter flounder during late summer (August- he cause of high mortality shortly before or after l September) at stations LR and WA in the Niantic River. settlement of young in the Niantic River has not been For WA, data points designated by an 'x, for 1988 and investigated. Of all life stages of marine fishes, least 1991 were not included in the regression. Is . .

                                                                                   . known about larval and early ,Juvemle stages, yet this is likely where relative year-class strengths are particularly for 1994 and 1995 (Table 22). Der2                                    determined (Sissenwine 1984; Bailey and Houde value of 0.47 for station LR in 1996 was the lowest                                1989). Predation by caridean shrimp (Crangon spp.)

determined since 1984 and was indicative of has been suggested as the cause of high mortality rzlatively high variance in abundance this year. No after metamorphosis for both winter flou*Jer mortality estimates were made for LR and WA (Witting and Able 1993, 1995) and plaice during the high abundance year of 1988 as slopes of (Lockwood 1980; van der Veer and Bergman 1987; th se catch curves were not significantly different Pihl 1990; van der Veer et al.1990; Pihl and van der from zero as they were for WA in 1986,1993, and Veer 1992). Van der Veer et al. (1990) speculated 1996 because of considerable variation in weekly that, in general, predation by crustaceans on young rbundance. De 2,,,, estimate for station LR in 1996 may be a common regulatory process for flatfishes. was 0.504 (equivalent to a survival rate S,,,, of 60.4. Witting and Able (1993,1995) found that the size of Long-term averages of Z,,, at LR and WA were age-0 winter flounder significantly affected their 0.631 and 0.557, respectively, equivalent to survival Probability of predation by sevenspine bay shrimp rat:s of 53.2% and 57.3%. Mortality estimates for (Cr8"E0" 8'Ptemspinosa), with predation greatest at Ni ntic River winter flounder were usually greater settlement for the smallest fish. Mortality decreased than the equivalent 2,,, value of 0.371 reported by with size and young apparently outgrew predation by P arcy (1962) for the Mystic River, CT estuary, but shrimp when they reached 17 to 20 mm in length, were similar to various estimates (0.563 . 0.693) which meant that fish would have to double in length made for young plaice in British coastal embayments after settlement before attaining a size refuge from WinterFlounder 111

l TABLE 22. Monthly instantaneous total mortality rate (Z) estimates as determined from catch curves for age-O winter flounder taken at two stations (LR and WA)in the Niantic River from 1984 through 1996. Standard Standard Year Station n' slope' error r' Station n' slope' crror r 2

                                                                                                                                                 )

1984 LR 16 0.129 " 0.017 0.80 WA - - - - l 1985 15 -0.118 " 0.015 0.82 16 -0.084 " 0.023 0.51 1986 15 0.127 " 0.012 0.89 - 1987 15 -0.108 " 0.021 0.67 16 -0.139 " 0.016 0.84 1988 19 NS - - 19 NS - - 1989 12 -0.154 " 0.022 0.84 13 -0.14 5 *

• 0.028 0.71 1990 13 -0.322 " 0.028 0.92 15 -0.235 " 0.028 0.84 1991 18 0.140 " 0.016 0.82 18 -0.049 " 0.011 0.54 1992 18 0.129 " 0.019 0.74 16 -0.112 " 0.009 0.91 1993 9 -0.087
• 0.028 0.57 10 NS - -

1994 9 -0.110 " 0.008 0 96 9 -0.124 ** 0.020 0.84 1995 9 -0.203 "  ! 0.010 0.98 9 -0.138 " 0.018 0.89 1996 9 -0.116

• 0.046 0.47 l 8 NS - -

l Mortality (Z,,,) I Survival (S ,,) Mortality (Z,,,) Survival (S,,,) 1984 LR 0.560 57.1% WA - - 1985 0.512 59.9 % 0.363 69.9 % 1986 0.552 57.6% -' - 1987 0.469 62.6 % 0.604 54.7 % 1988 - - I989 0669 51.2 % 0.630 53.3 % 1990 1.398 24.7% l.021 36.0 % 1991 0.608 54.4 % 0.213 80.8 % 1992 0.560 57.1% 0.486 61.5% 1993 0.377 68.6 % - - 1994 0.476 62.1% 0.538 58.4 % 1995 0.883 41.4 % 0.600 54.9 % 1996 0.504 60.4 % - - Mean 0.631 53.2 % Mean 0.557 57.3 % SD 0272 SD 0235 CV 43 % CV 42 %

• Weekly sampling during 1984-92 and biweekly sampling during 1993-96. WA was not sampled in 1984.
• Probability level that the slope of the catch curve differs from zero is shown:

NS not significant (p > 0.05),

• significant at p s 0.05, " significant at p s 0.01.
• Although having a significant slope, the catch curve for 1986 at station WA did not provide a reliable estimate ofZ because of considerable variation in weekly abundance.

shrimp. Predation was also related to shrimp density significant implications for young winter flounder and steadily increased until reaching an asymptote at survival after metamorphosis. shrimp densities 210.6.m'* (Witting and Able 1995). Recruitment of many fishes is greatly affected by Therefore, the duration of time spent in a vulnerable density-dependent processes occurring during their size range, which is related to growth rate, affects the first year oflife following the larval stage (Bannister vulnerability of young winter flounder to predation et al. 1974; Cushing 1974; Sissenwine 1984; by shrimp and other organisms. Variation in growth, Anderson 1988; Houde 1989; Myers and Cadigan which can depend upon specific location of settling, 1993a,1993b). Bannister et al. (1974), Lockwood specific habitar within a location, or temperature (1980), and van der Veer (1986) all reported density. (Sogard 1990; Sogard and Able 1992) may have dependent mortality for young plaice, although 112 Monitoring Studies,1996

examination of their findings indicated that greatest sampling of young in summer and precedes the catch rates of mortality occurred only when extremely of this cohort of fish as age-I juveniles during the large year-classes of plaice were produced (three to intensive late February early April adult winter more than five times larger than average). High (> flounder survey in the Niantic River. Based on the 2.md ) densities of young winter flounder at LR availability of data for this report, the most recent during 1990,1994, and 1995 produced the steepest A-mean CPUE is for the 1995 year-class (Table 23). declines in abundance at that station. However, The A-mean CPUE of 4.E for 1995-96 was the lowest mortality rate at WA was also high in 1990, despite of the time-series, which was unexpected, given the having only moderate densities of young there. relatively high abundance of the 1995 year-class Furthermore, mortality rates during 1994 and 1995 at (Table 20; Fig. 27). In 1994-95, the A-mean CPUE WA were about average, despite having the highest of 31.7 did reflect the strong 1994 year-class, observed abundances at that station. Estimated however. Also in recent years, strong production of production of young was high in 1988 because of young in 1988 and 1992 as well as the weak 1993 rpparently very low mortality, but no sharp peaks in year-class were also evident in the corresponding abundance were observed that year and densities A-means of 29.6,31.1, and 7.4, respectively. generally remained below 1m 4 Thus, the Since 1983, when data were first available from relationship between density and mortality rate for Niantic River beam trawl sampling, the fall-carly young winter flounder may be subject to considerable winter A-means were compared to a 1 m beam trawl variability (i.e., regulatory mechanism not well- median CPUE from late summer using data established). Mortality rates for demersal age 0 combined from both stations LR and WA (Fig. 31). winter flounder only occasionally (e.g., 1988, 1990) These abundance indices track each other closely and showed large deviations from average. However, are significantly correlated (Spearman's rank order these occurrences, in addition to events occurring correlation coefficient r = 0.6667; p = 0.0171). The during the larval phase of winter flounder life history, relationship between the TMP A-mean CPUE and the determine the potential for a year-class to become A-mean CPUE of winter flounder smaller than 15 cm exceptionally strong or weak, taken in the Niantic River during the subsequent (late February-carly April) adult winter flounder survey Age-OJuveniles during Late Fall (see below) was negative (r = -0.3524; p - 0.1215; andEarly Winter Fig. 32), although no longer significant with the inclusion of data from this year. More juvenile As water temperatures decrease m, fall, young winter flounder of the 1985 and earlier year-classes winter flounder disperse from shallow waters near were taken in the river than at the six TMP stations  ; the shoreline to deeper waters and become @ve of which are outside of the Niantic River) susceptible to the otter trawl used in the year-round during the preceding months. Since the 1988 year- ) trawl monitoring program (TMP). Young are first class was produced, abundance of young has been regularly captured by trawl at the two shallower higher in the TMP, with the exception of the 1993 I mshore stations (NR and JC) adjacent to inshore and 1995 year-classes, when smaller differences were l nursery grounds m, November, the near-shore Niantic observed. The reason for the relative shift in l Bay stations (IN and NB) m December, and at the magnitude for these abundance indices is unknown. I deeper-water stations in LlS (TT and BR) in January. The numbers of young taken during late fall and I A A mean (NUSCO 1988c) mdex of relative carly winter by the TMP should be a predictor of abundance was developed for these age-0 fish using age-1 fish abundance in the Niantic River during late TMP catch data, begmnmg with the months given w nter and early spring. However, this presumes that above and continuing through the end of February. the relative distribution of fish both inside and in January 1996, stations BR, TT, and NB were outside the river remains constant each year, which likely does not occur. deleted from the TMP (see Fish Ecology section). Therefore, sample size was reduced from 42 for Relationships among abundance indices ofjuvenile 1976-77 through 1994-95 to 28 for 1995-96. winter flounder may have been obscured by The months of November through February form a differences in sampling gear used and variations in transitional period that follows the 1 m beam trawl fish behavior in response to environmental conditions. Major biases in abundance estimation Winter Flounder i13

_ _ _ _ _ . _ - . . - _ _ . _ _ - . - - . - - . - . _ - - . - . . . _ _ ~ _ - . _ . . . - . . . - . - _ - . _ - . . . . - . _ - . _ . . i I I TABLE 23. . The late fall early winter seasonal

• a-mean CPUE' of age O' winter flounder taken at the six trawl monitoring stations in ti vicinity of MNPS from 1976-77 through 1995 96.

( Surysy year' Number of samples Non-zero observations .1-mean* 95% confidence interval 5 1976 77 42 36 6.1 2.0 10.3 1977 78 42 38 - 5.1 2.3 - 7.9 1978 79 42 36 4.2 2.0 - 6.4 19794 0 42 38 198041 42 4.2 2.26.2 39 10.1 198142 4.3 15.9 - 42 39 7.7 1982-83 2 3 - 12.5 42 37 l9.6 1983-84 9.0 30.3 42 39 6.6 1984-85 3.2 10.0 42 35 7.4 198546 1.7 13.1 42 39 198647 8.1 4.4. I1.7 ' 42 39 t 11.7 3.4 19.9 ' 198748 42 41 4.8 ' 2.17.5 19884 9 42 41 l 29.6 I1.8 47.3 1989-90 42 42 Il.3 6.7 -15.9 ' I990-91 42 40 21.7 6.7-36.8 1991 92 42 4I 4 i90 7.6 30.3 1992-93 42 39 31.1 7.4 54.8 1993-94 42 38 i 7.4 3.4 . I 1.4 1994-95 42 41 { 31.7 7.3 56.1 1 1995-96 28 25 4.8 1.08.6 1

• Data restricted during 1976-77 through 1994-95 to November-February for NR and JC, December-February for IN and NB. :

February for 7T and BR and during 1995-96 to November-February for NR and JC and December-Fehruary for [N.

• Catch per standardized low of 0.69 km (see Materials and Methods of Fish Ecology section). i
• For age-O fish, the year-class is the same as the frst year given. '

j can arise from size selectivity of the gear, spatial 1963). Mean lengths of age-O winter flounder taken I distribution ofindividuals in relation to the gear, and by otter trawl in fall were usually about 15 to 25 mm i behavior of fish in the vicinity of the gear (Parrish larger than those taken during the immediately w w 100-  ? O 60 i  ? - 5 mi i @MJ a a: i

                     %*3i                                                 A               ,E l'

I 700 7i A. g 30i  ; m7 I in 8 80 i N.jSi1..I 50 - -ja . j }.

                                                                                  . Q, hit ii I
                                                                        !u .                    a                               40-
                     *                            ;         . .I .                ,

• f. l- .y i d

f 0[ g dj~ { - M 105 9.

                                                                                                                          $ 01                              ' 4..[k.

M k., , , , , , , , ?, , , , , r I g.y..g,{ ( 76 78 80 82 84 86 88 90 92 94 2 76 78 80 82 84 86 88 90 92 94 j YEAR-Cl. ASS YEAR-CLASS

                                                                                                                                                                                                     )

Fig. 31. Comparison between the late fall-carly winter Fig. 32. Comparison between the late fall-carly winter seasonal A-mean CPUE (solid line) of age-O winter floun-seasonal 4-mean CPUE (solid line) of age-0 winter floun-der (all trawl monitoring program stations) and the 1983-der (all trawl monitoring program stations) and the Niantic 95 late summer Niantic River (stations LR and WA com- River (stations I and 2) spawning survey 6-mean CPUE bined) age-O 1.m beam trawl median CPUE (dashed line) for the 1976-95 year. classes. (dashed line) of winter flounder smaller than 15 cm for the 1976-95 year-classes. 114 Monitoring Studies,1996 i i

   ~.- - . - - - -. -                                   - - . - - . -             . _ _ - - - - - _ . - - - -                                             - _         _ . _

l preceding months by 1-m beam trawl. His size - g 607 ,,,,,,,,,,,,,,,,, ,

,                        difference was greater than would have been                  c soy                                                            !

expected from growth alone and suggests that CPUE 1 indices were biased because smaller individuals were @ jc g;.

                                                                                                                  %nne                                /f I                                                                                                                                                  /'

excluded from the catch. Differences in sizes g 30; - - M stations , ../ /

achieved by age-O winter flounder each year (Table i 21) may have ' differentially biased . the trawl {20-' g.'. kj/ ' '

I

                                                                                                                                                /

y ! sampling. De fixed locations of the otter trawl 8 / V 0 { sampling stations in relation to the habitat available ' ' i to juveniles also may have affected catches and the j, ' g, ' fg ' f, ' ,, ' ,, ' f, ' fg ' 92 d4 96 reduction in the offshore TMP stations may also have } Effected du canparisons. Fig. 33. Percentage of tows with no fish smaller than 15 Movements of small juveniles were probably

;                                                                                cm collected in the Niantic River at all stations or in the influenced by factors such as water temperature and       navigational channel (stations I and 2) by year from 1976 i-                   . tide. Moreover, their availability to sampling gear in     through 1996.

fill and winter appeared to have varied from week to week and year to year. Relatively large Cis around channel (Fig. 2) because the distribution of small i the A-mean CPUE values were probably a winter flounder generally varied more so than for j consequence of this variation. In contrast, variation adult fish and, moreover, no tows were made in the was less in data collected during summer by the upper river from 1977 through 1980. De A-mean relatively efficient 1-m beam trawl. Furthermore, CPUE was highly correlated (Spearman's rank-order i sampling in summer occurred weekly or biweekly correlation coefficient r = 0.9786; p = 0.0001) with j during the same tidal stage and in areas known to be median CPUE values (Fig. 34). The A-mean index

+

preferred habitat of young winter flounder. Finally, a was slightly greater in magnitude than the median for mixture ofjuveniles from a number of sources most all years, with largest differences occurring during i likely was present throughout LIS during the winter, 1976-83. j which would have influenced measures of abundance The A-mean CPUE for age-1 juveniles taken in the

because of potential variable contributions from navigational channel of the lower Niantic River i

different stocks. nese factors all contributed to during 1996 was 1.6, the lowest value of any year 4 weakening the strength of correlations among since 1976 (Table 24). When tows from throughout I juvenile winter flounder abundance indices. the river were considered in the calculation, the  ! ] median CPUE was 0.8, which again was a series low Age-1 Juveniles duringLate Winter l l and continued the trend in relatively small CPUE for  ; j this time-series (Table 25). Small (<l5 cm) winter flounder, which include { { i Wnternoundes s em mostly age-1 fish spawned during the previous year, j j are incidentally captured each year during the February-April adult winter flounder surveys in the h 7o; f.

                                                                                                              /\

Niantic River. As for the adult CPUE, because of Soj j increasing percentages of tows with no fish smaller p 50; ;p/y,i .

                                                                                                                                       ,,,,,n ,,,,

l than 15 cm (Fig. 33), the annual age-1 juvenile u 40;  ! i, dena-mean j winter flounder abundance index was changed from a 30 L <*' i' t. median to a A-mean estimator this year. During $ A j ' . j. 1996,58% of the tows made throughout the Niantic River and 45% of the tows made in the river channel o ,,,,,,,,,h,(,;. *;*, 76 78 80 82 84 86 88 90 92 94 96 1

                  - had no small winter flounder, by far the largest                                                      YEAR numbers of zero catches. Also, before calculating CPUE, adjustments were made to the catch data that        Fig. 34 Comparison of annual median and A mean CPUEs j                    were similar to those previously discussed for aduk        for winter flounder smaller than 15 cm taken in the Niantic

( fish. In some annual comparisons, data were River at stations I and 2 from 1976 through 1996. l restricted to stations I and 2 in the navigational WinterFlounder 115

                                                                                                                                                              . . ~ -

l TABl.E 24. Annual 9.I-m otter trawl adjusted a-mean CPUE' of winter flounder smaller than 15 cm* taken in the navigational channel of the l lower Niantic River during the 1976 through 1996 adult population abundance surveys. l Weeks used Tows Adjusted a-mean 95% confidence Survey for CPUE acceptable number of Non-zero d CPUE Standard interval for ' year computation

• for CPUE tows used* observations estimate error A-mean CPUE 1976 7 98 154 152 25.2 1.9 21.5 -28.9 -

1977 6 166 228 222 26.3 2.8 20.9 31.7 ( 1978 6 129 156 152 32.5 43 24.1 -40.9 1979 5 107 135 134 66.7 l 8.6 49.9 83.4 1980 5 110 145 144 57.2 4.9 47.7- 66.7 1981 7 93 140 140 86.2 7.4 71.7 - 100.7 1982 5 50 70 70 57.4 10.4 37.0 77.8 1983 7 77 77 76 52.5 6.4 39.9 - 65.0 1984 7 72 77 76 25.0 2.8 19.6 -30.4 1985 7 82 84 84 34.0 3.7 26.7 4IJ 1986 7 72 119 109 6.0 0.6 4.8 - 7.1 1987 5 41 50 44 6.6 0.9 4.9 83 1988 6 49 54 52 17.0 3.1 11.0 23.1 1989 6 50 56 50 10.6 1.9 6.9 - 14 3 1990 7 65 91 88 14.7 1.9 10.9 18.5 1991 6 45 60 56 7.4 i .2 5.09.8 1992 7 35 49 44 11.9 2.1 7.8 -16.1 1993 7 36 49 45 6.6 1.0 4.6 - 8.5 1994 4 22 24 24 5.6 13 3.1 - 8.1 1995 6 39 54 50 6.4 1.1 43-8.4 1996 6 49 60 38 1.6 03 1.02.2

• Catch per standardized tow (see Materials and Methods); differs from NUSCO (1996) because median CPUE was replaced by a the index of abundance.
• Mostly age I fish; predominant age-class was produced I year before the survey year.

    *8 Effort equalized among weeks; during several years weeks with very low effort were not used for computing CPUE.

Only tows of standard time or distance were considered. Distribution of juvenile winter flounder largely recent years, relative abundance of these fish in influences their availability to sampling and Niantic Bay has increased. Except for 1988, the apparently differs from year to year, probably as a CPUE of fish found in the bay from 1986 through result of variable environmental conditions, which 1996 was greater than that of fish taken in the river. includes water temperature and winter storm events. The catch outside the river in 1995 was the highest of The relative abundance of small winter flounder has the time-series, indicating that most fish from the not been consistent between Niantic Bay and River strong 1994 year-class did not remain within or re-from year to year (NUSCO 1993). A A mean CPUE enter the Niantic River during the spawning season. was computed for winter flounder smaller than 15 cm However, the low CPUEs both inside and outside the taken by the TMP from January through April at river during the fall and winter of 1995 96 did not stations outside of the Niantic River (five for 1976- appear to reflect the relative abundance of fish 95; two, IN and JC, for 1996). This tirr,e span produced in the Niantic River during 1995, overlapped the spawning period and also served to A small CPUE value determined for the Niantic increase sample size. The TMP A-mean was then River may not necessarily represent low abundance compared to the A-mean CPUE for fish found within of a yesr-class. Even a relatively small increase in the river during the spawning season (Fig. 35). catch for the much larger geographical area of Generally, the catch of age-1 winter flounder in the Niantic Bay and nearby LIS could account for a low winter and early spring fluctuated less outside than abundance index in the river as fish dispersed from a inside the Niantic River. As the number of age-1 relatively confined area to more open waters. As a winter flounder in the river declined to low levels in result of the differential distribution and abundance 116 Monitoring Studies,1996

TABLE 25. Comparison of annual 9.1-m otter trawl adjusted a-mean CPUE'of winter flounder smaller than 15 cm' taken in the navigational channel of the lower Niantic River with those caught 'throughout the entire sampling area of the nver during the 1976 through 1996 adult population abundance surveys. Nes r- al" t ontv: Fatire area of river ea =alad Adjusted A mean Adjusted A-mean Survey number of Non zero CPUE number of Non-zero CPUE year

• tows used' observations estimate tows used* observations estimate 1976 154 152 25.2 224 219 20.5 1977 228 222 26.3 Insufficient tows made in upper river 1978 156 152 32.5 Insufficient tows made in upper river 1979 135 134 66.7 Insufficient tows made in upper river 1980 145 144 57.2 Insufficient tows made in upper river .

1981 140 140 86.2 231 225 45.5 1982 70 70 57.4 150 149 27.8 1983 77 76 52.5 238 230 24.9 1984 77 76 25.0 287 272 10.4 1985 84 84 34.0 280 272 18.7 1986 1l9 109 6.0 336 301 6.4 1987 50 44 6.6 239 198 1988 -4.2 34 52 17.0 312 281 6.5 1989 56 50 10.6 271 246 8.7 1990 91 88 14.7 315 255 3.9 1991 60 56 7.4 330 263 2.4 1992 49 44 11.9 406 327 4.0 1993' 49 45 6.6 392 312 3.1 1994 24 24 5.6 212 163 2.9 1995 54 50 6.4 342 e 232 1.8 1996 60 38 1.6 342 139 0.8 Catch per standardized tow (see Materials and Methods) differs from NUSCO (1996) because median CPUE was replaced by a A the index of abundance

• Mostly age I fish; predominant age class wu produced I year before the survey year.
• Effort equalized among weeks; during several years weeks with very low effort were not used for computing CPUE. Only tow time or distance were considered.

     'd                                                                   of age-1 juveniles, perhaps as a result of variable environmental conditions influencing their behavior k100                 T                                               and availability to sampling, the abundance indices lh              .

i determined from data taken during the TMP and the I 70.f g 60i p?[1 h~ j ,I adult spawning surveys remain generally unreliable predictors of future population size. For example, 501 l }. } the low abundance ofjuvenile winter flounder found h 203 E

          'k       ,*{l^ ^^ 1,.}

T  ; ! 5.- 7

                                                       ,                 during the fall-winter season of 1995-96 may have been the result of their movement into deeper water i 10h,:,,               ., h /f, N- i8.i.y..x,\
                     ,,,,,,,,'d.                                         beyond the sampling stations because of colder water
    @     0      ,                      ,  ,,,,,.,,T g         76 78 80 82 84 86 88 90 92 94 96                           temperatures that occurred during these months (Table 6).

YEAR (.lANUARY APRIL) Fig. 35. Comparison between the annual January April Comparisons among Life-Stages A-mean CPUE (solid line) at all trawl monitoring program Of Winter Flounder Year-Classes stations except NR and the Niantic River (stations 1 and 2) spawning survey A-mean CPUE (dashed line) for winter flounder smaller than 15 cm from 1976 through 1996. A.bundance indices for various life-stages of the 1976 6 @ 1996 m h d h & h WinterFlounder 117 l j i

                                                                                                                                            ]

                                                                                                                              -                    ~
                                                                                                                                                     )

l l e i- winter flounder given previously in this report are of 78% determitied in 1994. Variability in winter  ; summarized in Table 26. Coefficients of variation flounder abundance decreased to 82% following i j^ (CV) were used to examine annual variability in larval metamorphosis and settlement of demersal l these abundance indices (Table 27). Changes young, but became relatively higher (95%) for age-0 (mostly inersases) from CV values given in NUSCO young in late summer, illustrating effects of variable i (1906) differed by 7% or less. Variability in mortality rates during this life stage. The CV 1 abundance was lowest (CV _70%) for the number of decreased to 74% after young left shallow inshore females spawning in the Niantic River and for waters during fall and early winter. Another increase associcted egg production (64%). In the first three was seen in CV for age-1 juveniles in the Niantic  :' 3 adult female age; classes, variability decreased from River during the adult surveys (CV = 93%) that was i age-3 (102%) to age-4 (84%) and age 5 (81%). This probably related to the previously discussed annual l likely reflected variation in recruitment of year- differences in distribution related to behavior as 1 4 classes as well as variable numbers of immature much as from actual variation in year-class strength. I j ages-3 and 4 fish present in the river each year. Greatest variation in abundance during early life Miller et al. (1991) noted that interannual variability history was expected for Stage 2 larvae, because 1 of many flatfishes appeared to decrease with age. much of the compensatory mortality was believed to l CVs for larval abundance indices were 75% for occur during this stage of development. However, Stage 1,92% for Stage 2,98% for Stage 3, and 155% considerable variation was apparent in other larval i for Stage 4. The a parameter for Stage 4 larvae in { and juvenile life history stages, which indicated that  ; { 1995 was about 3.75 times larger than the next compensation likely takes place throughout the early i j highest value and, as a consequence, the CV for this life history of winter flounder. l l measure of abundance nearly doubled from a value ' { TABLE 26. Comparison ofindices of abundance of various life-stages of winter flounder for the 1976 through the 1996 year-classes *. l AMr inAm Larval iaam Juvenile indim Female Annual Niantic River stations (Feb-Jun) MNPS Lower Lower River / bey Age-1

Year- spawners egg Stage i Stage 2 Stage 3 Sta8e 4 (EN) river river a-mean CPUE

} class (Feb Apr) prod. (3 mm) (3.5 mm) (6 mm) (7.5 mm) (27 mm) (May Jul) (Aug Sep)(Nov Feb)(Feb-Apt) 76 - - - - - - 854 - - 6.1 263 j 77 889 394.2 - - - - 567 - - 5.1 32.5 ! 78 1,415 716.2 - - - - 754 - - 4.2 66.7 1 79 1,129 535.7 - - - - 641 - - 4.2 57.2 l 80 916 425.0 - - - i

                                                                                       -           845       -         -         10.1   86.2 81        2,683     1.3823       -            -           -         -

561 - - 7.7 57.4 , 82 2.756 1,594.4 - - - - 610 - - 19.6 52.5 83 1,873 1,081.2 - 749 408 56 1,215 32.7 10.0 6.6 25.0 i 84 872 500.7 2,601 1,501 573 67 917 18.8 63 7.4 34.0 4 85 931 564.9 6,260 4,676 584 35 312 133 7.0

'                                                                                                                                 B.1     6.0 86          654      436.1     1,279         176        301        24            510    33 8      13.8       11.7      6.6         !

! 87 852 530.9 3,218 829 1,036 48 315 59.2 17.9 4.8 17.0 , 88 1,278 865.7 14,491 4,469 1,531 210 419 1 61 3 60.0 29.6 10.6 89 983 715.4 12,463 3,976 589 73 327 17.5 8.8 113 14.7 90 580 370.1 4,728 355 258 57 508 1563 20.0 21.7 7.4  ; 91 1,060 638.7 3.248 252 343 112 439 77.5 21.7 19.0 11.9  ; 92 533 390.7 5,476 1,367 2,339 195 1,003 90.0 28.1 31.1 6.6 1 93 273 223.4 1,187 133 til 6* 130 10.6 5.0 7.4 5.6 94 507 329.4 3,692 1.248 429 90 834 128.8 62.9 31.7 64 95 218 1693 5,580 2,023 2,615 787 1,804 87.5 15.8 4.8 1.6

                   %            97       82.0     4373       3,677         265        31           462      8.8       3.0          -        .

• Differs from NUSCO (1996) for female spawners and annual egg production because of changes in the length-age key used and in age CPUE because median CPUE was replaced by a 6 mean as the index of abundance ,

l

• An approximation based on cumulative geometric weekly means. Gompertz function could not be fit to the data as larvac were '

collected during 2 weeks of sampling. 118 Monitoring Studies,1996

TABLE 27. Coefficients of variation (CV) of annual abundance indices' for various life stages of Niantic River winter flounder. Number of Life stage Abundance index used observations CV i Female spawners Annual standardized catch 29 70% Age.3 females Annual standardized catch 18 102 % Age-4 females Annual standardized catch 17 84 % Age 5 females Annual standardized catch 16 81% Eggs Egg production index 20 64 % Stage I larvae a parameter of Gompertz function 13 75 % Stage 2 larvae a parameter of Gompertz function 14 92 % Stage 3 larvae a parameter of Gompertz function 14 98% Stage 4 larvac a parameter of Gompertz function 14 155 % Age-O young Median CPUE at station LR (May4uly) 14 82 % Age-O young Median CPUE at station LR (August-Sept) 14 95 % Age-0 young Fall-winter A-mean at trawl stations 20 74 % Age Ijuveailes A-mean CPUE of fish < 15 cm in Niantic River 20 93 %

             ' Indices used correspond to those given on Table 26. except for age-3 through age 5 females.

Rothschild and DiNardo (1987) reported a median crustacean predation on newly metamorphosed CV for recruitment indices of various marine fishes young in more northerly waters than The Netherlands of 70%, although various flatfishes had CV values (Pihl 1990; Pihl and van der Veer 1992). Thus, mostly less than 75%, which is consistent with values population regulation in flatfishes may be coarsely of Niantic River winter flounder. De CV of determined in earliest life history by variable survival European flounder abundance decreased from 172% of eggs and larvae and then fine-tuned by density-(n = 9) in the larval stage to 99% (n - 8) in newly dependent mortality of newly metamorphosed settled young to 80% (n = 8,12) in young during juveniles (van der Veer and Bergman 1987). September and again at age-1 (van der Veer et al. Relationships among abundance indices of winter 1991). As summarized by van der Veer (1986), the flounder for the same year-class are of interest for highest CV for yearly abundance estimates of impact assessment. Knowledge of the earliest different life stages of plaice in The Netherlands possible measure of relative year-class strength is occurred during larval development in late winter (n desirable because it enables predictions of future

             = 4, CV = 95%) and at first settlement of pelagic                                                   recruitment to the adult stock, thus providing an early juveniles in spring following larval metamorphosis                                                   warning of decreases in stock abundance. If indices and settling (9,62%). Smaller variation was found in                                                 for all life-stages are assumed to be accurately and post-larval young during mid-summer (9, 30%) and                                                     precisely measured each year, they should be age-2 recruits (9,35%), which is less than was found                                                correlated after applying appropriate time lags, for winter flounder. He attributed the decline in                                                   except when processes such as density-dependent variation of abundance for older juveniles to a                                                     mortality or size-selective fishing result in a lack of density-dependent regulatory mechanism that                                                         colinearity between two consecutive life-stages, operated during and shortly after larval settlement.                                                Indices of female spawners and egg production were Van der Veer (1986), van der Veer and Bergman                                                       highly correlated (Spearman's rank-order correlation (1987), and Bergman et al. (1988) noted that                                                        coefTicient r = 0.9264; p = 0.0001), which was recruitment variability in plaice found in The                                                      expected because calculation of the latter included Netherlands was stabilized between years as a result                                                female spawner abundance as part of the of density-dependent regulatory processes (i.e.,                                                    methodology of estimation (Table 28). Significant or shrimp predation) on newly metamorphosed fish. In                                                   near-significant correlations were also found among        .

contrast, year-class strength of plaice in Swedish most larval stage abundance indices. Niantic River l bays veried to a greater degree (CVs = 67-118%), Stage 4 larval abundance was also significantly which was thought to be related to temperature correlated with age-0 juveniles collected during early efTects during the larval stage and more variable (r = 0.6528; p = 0.0114) and late (r = 0.6747; p = l l Winter Flounder 119

     ~   . . . . ~         . . . . - - . .         - ._- - , .                      . _

i ' 4 i TABLE 28. Matrix of Spearman's rank order comiations among various winter flounder spawning stock and l

• 4 indices refer to adults or larvac collected in the Niantic River, except for larvac 7 mm and larger taken at th Adult egg Stage 1 Stage 2 Stage 3 Stage 4 l Index' production larvac Larvae (27 mm)at

;                                                                                                             larvac -           larvae                      larvae            MNPS discharge    ;

E

• Femaic 0.9624' O.2%7 0.1253 0.1736 0.1692 spawners 0.0001 " 0.0211 0.3249 NS 0.6696 NS 0.5528 NS 20 0.5630 NS 0.9298 NS i 13 14 14 14

, 20 Adult egg 0.3462 i production 0.1692 0.2571 0.1692 -0.0767 ] 0.2466 NS 0.5630 NS 0.3748 NS 0.5630 NS 0.7479 NS 13 14 14 14 20 ! i

Stage 1 0.8187 0.5934 0.5879 tarvac 0.0165 0.0006 " 0.0325

• 0.0346
• 0.9574 NS 13 13 i 13 13 1

Stage 2 i l larvae 0.6176 0.3626 0.0550

j. 0.0186

• 0.2026 NS 0.8520 NS {

14 14 14 4 Stage 3 , i larvae 0.6967 0.2440 i 0.0056 " 0.4006 NS I 14 14 i } Stage 4 larvae 0.4593 ' 0.0985 NS 4 14 4

• Indices used correspond to those given on Tables 26 and 27.
• The three statistics shown in each correlation matrix element are:

correlation coefficient (r), i probability of a larger r (NS not significant (p > 0.05),

• significant at p 5 0.05, " significant at p s 0.01), and number of annual observations (sample size).

f 0.0081) summer (Table 29). Age-0 juvenile abun-j because only a fraction of these fish likely are present dance during late summer and late fall-early winter l was also correlated (Table 29; Fig. 31). The densities on the spawning grounds each year. Furthermore. of larger (2 7 mm) larvae collected in entrainment the presence ofimmature fish may vary from year to year because of environmental conditions. Several samples at MNPS were not significantly correlated

significant correlations were found between the with most adult, larval, or juvenile abundance i indices, ahhough abundances of ages-3 through 5 female spawners and some correlations have 1 those of juvenile winter flounder and larvae 7-mm 1 strengthened over time (Tables 28 and 29). As i and larger taken at MNPS (Table 30). He CPUE of discussed previously, although not significantly age-I fish taken in the river during the adult correlated, the abundance the Niantic River spawning  ;

spawning surveys was significantly correlated with

survey age-1 A-mean CPUE was inverse with that of  !

both age-3 and 4 female spawners, but not with age-5 4

                 . young during fall and early winter in the TMP (Fig.                                               females.       Significant negative correlations were                       !'

32). His was probably mostly related to changes in

found between ages-3 through 5 females and the distribution rather than an indication of any age-O fall winter A-mean CPUE. However, the form

] compensatory mortality, ] of these relationships is unclear and may be a Niantic River winter flounder are not fully i recruited until age 5, if catch indices are assumed to statistical artifact (Fig. 36). Persistence of negative i correlations in future years perhaps result from be representative of annual relative abundances 1 unknown processes operating after winter flounder (NUSCO 1990). Thus, age-3 or age-4 fish probably 2 become age-1 that produce fewer age-5 adult recruits should not be used as an index of year-class strength 4 from more abundant year-classes of juveniles. 120 Monitoring Studies,1996 . 1 a

l TABLE 29. Matrix of Spearman's rank order correlations among various larval and juvenile winter flounder abundance indices. Niantic River Lower river Lower river Fall-carly winter Niantic River Stage 4 carly age-O late age-O river-bay winter spring index* larvac juveniles juveniles juveniles age-Ijuveniles Larvae 0.4593' O 4769 0.2528 -0.1536 0.2377 (27 mm) at 0.0985 NS 0.0846 NS 0.3833 NS 0.5180 NS 0.3129 NS Millstone discharge 14 14 14 20 20 Niantic River 0.6528 0.6747 0.3333 -0.0138 Stage 4 0.0114

• 0.0081 " 0.2657 NS 0.9644 NS I larvac 14 14 13 13 i Lower river 0.8901 0.5179 -0.1238 carty age-O 0.0001 " 0.0698 NS 0.6870 NS juveniles 14 13 13 Lower river 0.6667 4 0578 late age-O 0.0128
• 0 8513 NS

juveniles 13 13 Fall-early winter

                                                                                                                            -0.3524 river-bay 0.1275 NS age-0 juveniles 20

• Indices used correspond to those given on Tables 26 and 27.
• The three statistics shown in each correlation matnx clement are:

correlation coefficient (r), probability of a larger r (NS not significant [p > 0.05),

• significant at p 5 0.05, " significant at p s 0.01), and number of annual observations (sample size).

3 Possibilities include variable discard mortality of mates provided by CT DEP, and a change in the age-juveniles in the commercial fishery; high rates of length key used resulted in some differences between fishing; and non-random fishing effort, which may current estimates of spawners and recruits and those occur in overfished stocks. Meanwhile, none of reported in NUSCO (1996). Ages were formerly these life-stage indices can presently be used as a assigned to female winter flounder using an age-reliable measure of year-class strength. length key described in NUSCO (1989). Briefly, this key made use of derived probability density functions Stock-Recruitment Relationship (SRR) approximated by fitting a two-parameter compertz cumulative function (Draper and Smith 1981) to Sampling based estimates- Egg production Niantic River female winter flounder age and length estimates from annual spawning surveys were used to data. However, it appeared that this method did not determine recruitment because the abandance of adequately reflect the proportions of females in ages cther early life-stages have not been reliably 3 through 5. A re-examination of the data resulted in correlated with adult spawners. Both recruitment and the creation of a revised age-length key that more the parental spawning stock indices were scaled to closely matched empirical age-length data (Fig. 37). tbsolute population size as described previously (see A two-parameter SRR model (Eq. 6) was initially Absolute Abundance Estimates section, above). The fitted to the spawner and recruit data. The stock resulting annual values were used with the Ricker growth potential parameter a (scaled as numbers of SRR model as estimates of adult female spawning fish) for this model was estimated as 1.4L with a stock and potential female recruitment (Table 31). standard error of 0.525 (37% of the parameter value). The addition of new catch data from the 1996 adult The two-parameter model estimates were used as winter flounder survey along with a change in the initial values for fitting the three-parameter SRR scaling factor used, updated fishing mortality esti-WinterFlounder 121

_ . _ . __ __ _ _ - - _ _ _ _ _ _ . , _ . . _ _ _ _ . _ . ~ . - . _ _ ~ ~m .._ _ _ _ _ ___.-___ _ __ _ _ _ i l j- TABl.E 30. Matrix of Spearman's rank order correlations among various winter flounder larval and female spawner abundance indices.

                                                                                                                                                                                         \

Larvae IAwer river lower river Fall <arly winter Niantic River (27 mm) at early age-O late age-0 river-bay winter-spring

index* MNPS discharge juveniles juveniles juveniles age I juveniles

, Age-3 0.3375' -0.1273 0.0818 -0.5227 0.7734 female 0.1708 NS 0.7092 NS 0.8l10 NS 0.0260' O.0002 " spawners" 18 11 11 18 18 Age-4 0.2500 -03818 -0.5273 -0.8866 i 0.6806 female - 0.3332 NS 0.2763 NS 0.1173 NS 0.0001 " 0.0026 "

. spawners" 17 10 10 17 17 Age-5 0.7088 -0.5333 0.5833 -0.6534 0.4588 femaic 0.0021 " 0.1392 NS ' O.0992 NS 0.0061 " 0.0738 NS spawners" 16 9 9 16 16

• Early life history indices used correspond to those given on Tables 26 and 27.
• Desermined by applying an age-length key (NUSCO 1989) to the length distribution of annual standardized female abundances.

              ' The three statistics shown in each correlation matriz element are:

correlation coefficient (r), probability of a larger r (NS not significant [p > 0.05),

• significant at p s 0.05, " significant at p 5 0.01). and l number of annual observations (sampic size).

model with temperature effects (Eq. 7). He three- spawning population. However, these recruitment parameter SRR explained 71% of the variability indices were much below expected values, likely the ] associated with the recruitment index. Relationships result of high fishing mortality rates in recent years resulting from fits of both Ricker models are shown (Fig. 5).

           ' as the curved lines in the central pottion of Figure 38                                        The estimate of Ricker's p parameter, which as follows: the unadjusted SRR (two-parameter                                          describes the annual rate of compensatory mortality model; Eq. 7) is shown as the thinner solid line and as a function of the stock size is important in SPDM the three-parameter model (SRR adjusted for T,.) is                                    simulations. De value of 2.450 x 10 was indicative represented by the thicker solid line. He outermost                                    of the consistency found for S parameter estimates                                   !

two dashed lines illustrate low recruitment in the (range of 2.140 2.583 x 10; NUSCO l990,1991b, ' warmest year (1991; T,. = +1.95) and high 1992a,1993,1994a,1995a,1996). The parameter $, j recruitment in the coldest year (1977; T,.= -2.45). which is an estimate of the effect of February Using the three-parameter model, a was estimated temperature deviations (Tr.) from the 1977-92 mean at 1.473, with a standard etror of 0.306, which is of 2.81*C, was -0.418, which was quite similar to the i 21% of the parameter value (Table 32). This was the estimate of-0.415 given in NUSCO (1996). Other  ; lowest estimate of a made to date, reflecting low estimates of $ ranged from --0.412 through -0.259 recruitment in the early 1990s; previous values (NUSCO 1990,1991b,1992a,1993,1994a,1995a). ranged from 1.710 through 2.646 (NUSCO 1990, ne values for $ are negative and although the i 1991b,1992a,1993,1994a,1995a,1996). Variation reasons for the apparent relationship between winter in a estimates could be caused by increasing fishing flounder recruitment and February temperatures mortality rates on winter flounder in addition to the remain unknown. February coincides with most inherent instability of parameter estimates fitted to spawning, egg incubation, and hatching, which along i small data sets. in particular, the influence of the with larval growth are all temperature-dependent. 1988-92 data points on the estimate of a were Buckley et al. (1990) noted that the winter flounder illustrative of higher recent exploitation and poor reproductive process appears optimized for cold  ; recruitment (Fig. 38). Because of the relatively high winter temperatures that are followed by a gradual abundance ofjuvenile winter flounder from the 1988 - spring warming. Adult acclimation temperatures and year-class, numbers of females were expected to egg and larval incubation temperature affected larval increase during 1992-94 and form the bulk of the size and biochemical composition. Cold winters and 122 MonitoringStudies,I996

                                                                    ,----,v -yy,a-      ,                         , - _ , -              ,.

g + - - --i--.-- T"'-*-

w - . bloom more in response to the amount of solar E1* radiation received, which is generally consistent over 1 h1200]j , time each year. Therefore, a bloom in a cold year has a toooi , the post.sility of lasting longer before being grazed [ 800 j . . down by zooplankten. This allows for a greater

     ' eoo :                                                                            contribution of organic matter to the benthos than in 400 3                   ,
                                                       .                                other years, benefiting juvenile demersal fishes that 5          * **
                             .             .                                            metamorphose just after the spring bloom of i

f'200of , .,? ., ,.- , .,n, Ph ytoplankton and have to outgrow various i o 5 to 15 20 25 30 35 Predators. As noted previously, the effect of AGE-o TM P M EAN CPUE temperature on potential prey or predators of larvae 4 and newly metamorphosed juveniles, such as the sevenspine bay shrimp, may be an additional means y *0] for control of population abundance. Strong year-

g classes of plaice were also associated with cold K ,ooj j .

winters, likely because the predatory brown shrimp

8 sooi , (Crangon crangon) sufTers high mortality or migrates

uf j out of plaice nursery areas (Zijlstra and Witte 1985; I *1 van der Veer 1986; Pihl 1990; Pihl and van der Veer d

200 3 **.. I992). 3 j * ** . In addition to parameters directly estimated from 8 o1- , ,

                                                         ,.         , .,e  .,          stock-recruitment data (Fig. 38), Table 32 includes o          5             to        15    20           25   30   35         four derived biological reference points. Ricker's AGE-o TM P M EAN CPUE stock-at-replacement, or P,,, (Eq.11), was estimated at 72,239 female spawners and is the unfished W 700 -                                                                            equilibrium spawning stock size, also known as the E                  .

maximum spawning potential (MSP). This reference h sooj  !

• point, expressed in units of biomass, is 113,415 lbs.

a *l Stocks with biomass less than the critical size of 25% [*j

• of MSP (here, 28,354 lbs) are considered to be l 300 j .

overfished (Howell et al. 1992). The present M200: . . , . equilibrium size Par; (Eq. 9) of 15,808 spawners 33o0) . . . refers to the sustainable or equilibrium size to which goj ' ' the stock could converge if present (through 1992) b 5 1o 1s 2o 2s b 3s eXP l oitation and other conditions remained AGE-o TM P M EAN CPUE unchanged. The calculated (Eq.10) value of F that would achieve equilibrium stock size was 1.38, which is consistent with the high estimates of F made Fig. 36. Comparison between the late fall-early winter seasonal A-mean CPUE of age-O winter flounder W n recent years. As mentioned previously in the data) and the relative annual abundance of ages-3,4, and 5 Materials and Methods section, these reference points Niantic River female winter flounder. (Note that the verti- derived from fishery data are only determm. is . tic col scales differ among the graphs). approximations useful for comparative purposes across stocks and were used in this study as a warm springs produced large larvae that were in the comparison to more realistic values derived through best condition at first feeding, which favored high simulation using SPDM. survival and partly explained the observed Estimation of a for SPDM simulations. The correlation between cold years and strong year. above stock-recruitment-based estimates of a for the classes of winter flounder. Townsend and Cammen Niantic River winter flounder underestimated the true (1988) noted that the metabolic rates of pelagic slope at the origin for this stock. The method of consumers are more sensitive to lower temperature calculating annual recruitment included the effects of than rates of photosynthesis by phytoplankton, which fishing on winter flounder age-2 and older as well as WinterFlounder 123 1

    .   .--.- ... _ . . . -                          . - - _ -                   - - . ~ . . - - - . - . . _ - .                 . - - . . ~ ~             . - - - .      -        . - ~

l TABLE 31. Annual Niantic River winter Sounder stock-recruienent data based on indices of egg production for the 1977 through the 1992 year classes with rnsen February water temperature and deviations (T,.) from the mean. Mean February Deviation from

                                                            -Index of female                Index of female           R/P           water .      mean February water                     !
                                       . Year class                spawners (P)*               recruits (R)'         ratio    temperasure ('C)    temperature (Tr.)

( 1977 17,565 51,250 2.92 036 2.45 1978 31,915 39,781 1.25 1.09 ' 1.72 1979 23,874 32,827 138 1.48 -133 1980 18,939 26,153 1.38 238 0.43 1981 61,599 24,345 0.40 2.63 -0.18 o 1982 71,052 . 30.997 0.44 1.56 1.25 1983 48,182 30,205 0.63 3.74 0.93  ; 1984 22,315 23,537 1.05 i 4.02 1.21 i 1985 25,175 22,998 0.91 2.36 -0.45 I I 1986 19,433 20,002 ' 1.03 338 0.57 1987 23,659 16,737 0.71 3.27 0.46 1988 38,577 14,137 i 037 2.67 0.14 ! -- 1989 31,881 9,066 0.28 3.24 0.43  ; 1990 16,494 6,293 038 4.28 - 1.47 i j 1991 28,463 3,600 0.13 ' 4.76 1.95 1992 17,409 1,625 0.09 3.68 0.87 Mean 31,033 22,097 0.60 2.81 CV $2% - 61 % 43 % r

• Scaled number of female spawners and recruits from expected egg production: scaling factors used were 561,000 eggs per fema, multiplier of 25 to convert relative abundance to an absolute population size. Indices of female spawners and recruits differ from those reported in NUSCO (1996) because of data added from the 1996 adult winter Sounder population survey and changes in the mortality rates, age-length key, and in the scaling factor used in the calculauons the entrainment oflarvae at MNPS nerefore, these fishing rates are avoided and independent means of direct estimates of a correspond to a compensatory validating SRR based estimates are provided. De reserve diminished by existing larval entrainment and present study used a Ricker SRR a parameter exploitation rates. De concept of compensatory estimate derived from the value of 3,74 (in biomass reserve in fishing stocks and the effect of exploitation units) reported by Crecco and Howel! (1990: Table on the shape of the reproduction curve when the 2). The value of 3,74 was re-scaled for numbers of recruitment index is based on the exploited stock was fish on the basis of the following relationship:

discussed by Goodyear (1977: Fig.1). Dus, iflarval entramment and fishing rates increase, the field am= a /(mean weight per mature female estimates of recruitment will be smaller and so will fish) (19) , the estimates of a (i.e., the " remaining" compensatory reserve). To assess impacts - where the mean weight was calculated for a popula-appropriately, the mherent potential of a stock to tion at equilibrium and one for which only natural increase in the absence of fishing and plant effects mortality was assumed to have occurred (i.e., the ' must be determined. Crecco and Howell (1990) unfished population). Dese calculations used popu-  ! investigated the possibility of using indirect methods lation data previously reported (NUSCO 1990) and . to estimate the true a parameter (i.e., for the unfished an M of 0.25 (NUSCO 1996). Small changes in the stock when F = 0). Dey used four indirect methods fraction mature from a review of the age-length data (Cushing 1971; Cushing and Harris 1973; Longhurst previously discussed resulted in a mean weight of 1983; Hoenig et al.1987; Boudreau and Dickie 1.57 lbs per female spawner for the Niantic River 1989) based on different life history parameters. unfished winter flounder stock and a mean fecundity Because these methods do not depend upon direct of 972,205 eggs per female spawner (Table 33).

                        ' estimates of recruitment, biases caused by changing 124 MonitonngStudies,1906 I

_ _. __ _.. . __ _ = . _ - _ . _ - . - - . _ _ _ . - _ - _ _ . - . .. - New matted of assa0'ung ageto fer,uielen0th* trem aner csa dais towves sanothed by ere) Cxploited stocks of winter flounder. Using an I unfished stock as a starting point for a population

   >.     ,                                                                                                                                                         {

oel , ... dynamics simulation has other advantages, depending upon the particular scenario selected. For

                                                                                                                                                                    )

ca- 1 example, the simulation in this report includes [;7j 4- initially moderate fishing rates that r.re much lower osl 5 than those affecting the data on which the regression g 04- , ' . . . . ,? ., estimate of a was based. The data-based estimates of g,8[ o.i .

                                 /j .           ,

N. , the other two SRR parameters (p and (), which do not depend upon fishing and entrainment rates, were 2o ,,,,,,, ,,,,',,, used in the population simulations as given in Table as as 27 as ne so at 32 ss 34 as se s7 se se 4o 37 LENGTH (cm) ro, m.ndof.ss.eng ..v i.nem. MNPS Impact Assessment from calculated probabdaty cwves (NUSCO 1009) , l

             '?    s o.9 H
                                                                                     ,s+                             LarvalEntrainment 1

i o8i A

                                                                                 ,J 5

w o.7 / g oe- Estimates oflarval entrainment at MNPS. De

                                                /N                 '

number of winter flounder larvae entrained in the

                                              ."               y.[

S o,",3$ ': J / , ', condenser cooling water of MNPS is the most direct measure of potential impact on the Niantic River "3 022 / Y '. . winter flounder stock. Annual totals of entrained 6"

                                                                                    \               larvae were related to both larval densities in Niantic EsAi7isWEo'5tE2EE4 AsAlrAEe'lo                                                       Bay and plant operations (i.e, cooling water volume).

1.ENGTH (cm) Nearly all winter flounder larvae collected at station EN were taken from February through June, mostly Fig. 37. Comparison of female winter flounder length-age (> 90%) during April and May. He entrainment distribution as presently determined (from empirical data - with curves smoothed by eye) and as formerly determined estimate for 1996 of 53.9 million was the second (calculated from probability curves; NUSCO 1989). lowest since three-unit operation began in 1986 (Table 34). His largely can be attributed to plant Using the derived mean. weight, the re-scaled a padon buse the coohg wata volume during parameter for this study was obtained as: was k lownt of h h, unit operatsnal period. The a parameter, an index of larval a=ao (mean weight) = 3.74.(1,57 lbs) = abundance (Eq. 2), was 1,388 in 1996, which was

                         '                                                                         lower than the 21 year average of 1,685.

Due to refueling outages, Units I and 2 did not - his parameter describes the inherent potential of a F*** 8 * '"' I*"*' <

                                                                                                                                        *I"'". season and stock for increase because the natural logarithm of a                                              Unit 3 was shut down after March 30 (Fig. 39). The is the slope of the SRR at the origin for the unfished                                            decrease in cooling water usage resulted in a stock (Ricker 1954) and that slope, m turn,                                                       calculated reduction in larval entrainment of about               >

corresponds to the mtrinsic rate of natural increase of 72% (equal to 138 million larvae) from that expected l the population (Roughgarden 1979). Using field if all three units had operated fully during this period. dat:, the slope of the SRR at the origin decreases As in previous years, Stage 3 larvae predominated with increasing exploitation rates and a can be in entrainment collections. In 1996, the percentages considered as the _ remaining growth potential" or of each developmental stage entrained were 25% for

" growth reserve" of the stock. Consequently, the                                                 Stage 1,30% for Stage 2,34% for Stage 3, and 11%

largs difference between the derived value of a for Stage 4. Percentages for 1983 95 combined were (5.87) and regression estimates of a based on ficki 3% for Stage 1,20% for Stage 2,65% for Stage 3, data (e.g.,1.47; Table 32) reflects the difference in and 12% for Stage 4 of development. Compared to growth reserves between unfished and highly WinterFlounder 125  !

_. .. _ _ _ . _. _ _ . _ - _ _ ~ _. _ . _ _ - _ _ . . - _ ..m ___ ._- 709 G i s 80- ~ g 50

• 4 4

77 * * / p ... ,,.....- ~ ~ ......,,, ~ s...,.....

                                                                                                                                    %.~~

A ' f

                                                     '                             e 78 u.
                                                   /

i '"...,n.. v>

                                                 !                            79 o 30 9
                                                /                                                    e83                         e82 h                        ,/        80 e 84 85p'

'

• BI - ~ . _ _ ~ ~

                     )20i.i !             .
                                           ,f 88 6                                                                                      ~' '-

, 0 w . , 89

                                                                                           .=

, @ 1& : a: o 2 .* at .......'e.....~"*,**.................................................................., 90 Ec - i - e 91 g ' y ' 60 70 80 90 i 10 t FEMALE SPAWNERS IN THOUSANDS Fig. 38. The Ricker SRRs of Niantic River winter flounder (seu text for explanation of the four curves plotted). Calculated recruitment indices of the 1977 through the 1992 year-classes are shown. previous years there was a noticeably large increase due to collections made under reduced cooling-water for developmental Stage 1 in 1996. As noted flow, which implies that many smaller larvae likely previously, the greater frequency of small larvae are extruded through the entrainment sampling net collected in entrainment samples in 1996 was likely under normtl operation with higher flows. TABLE 32. Parameters of the Ricker stock-recruitment model fitted to data for Niantic River female winter flounder spawners from 1977 through 1992 and some derived points ofreference. Model parameters and reference points Model ._. . t - :- :-- 4 from - '

                                                                    - of nihi-                        Estimased value               Standard error                     /

4 n o(compensatory reserve for unfished stock) 5.87 . - a (current compensatory reserve) 1.473" 4 0.306 4.82 " p (stock. dependent mpensato y rate) 2.450 X 10 4 88 X 10* 5.02 " 6 (environmental ltemperature] effect) -0.418 0.075 5.59 " Scaled as Scaled as Derived noints of reference-numbers of fish biomass (lbs) Unfished stock equilibrium size (P4 called maximum spawning potential by Howell et al.1992) 72.239' t 13,415 Present (through 1992) equilibrium size (Pan) 15.808' 15,808 F for Pan = 15,808 female spawners 1.38 - Estimate of critical stock size (25% of maximum spawning potential) - 28,354 8

     * (t-test for Ho* parameter estimate = 0); t =13 with df = n-3 = 13 rejects He                   a t p s 0.01. R for the fit to the model was 0.71.
     ' includes the effects of recent exploitation rates.

• Mean weight of a female spawner for the unfished stock is 1.57 lbs and for the current exploited stock is I lb.

f 126 Monitoring Studies,1996

.  - . . .               . .           - . .     . ~ . -          .         -- . - . . .                          . - _ _ - - ~ . .                             . _ _ _ . . - - _ _ _ -

i I 4 l I i i TABLE 33. Biomass calculations for the Niantic River winter flounder female spawning stock at equilibrium based on an instantaneous l natural mortality rate of M = 0.25 and an instantaneous fishing mortality rate of F = 0 (i.e., an unfished stock). I l 4 Female Number of Weight of Eggs per Spawning stock Egg < population Fraction mature mature females mature biomass production j Age size mature females (lbs per fish) female (Ibs) (millions) l 2 1,000.00 0.00 0.00 - - - 0.000 1 3 778.80 0.10 77.88 0.554 223,7J5 43.15 17.424 1 i 4 606.53 038 230.48 0.811 378.584 186.92 87.257 j $ 47237 0.98 462.92 1.088 568,243 i 503.66 263.053 6 367.88 1.00 367.88 1377 785,897 506.57 289.!!6 7 286.50 1.00 286.50 1.645 1,004,776 I 471.29 - 287.868 8 223.13 1.00 223.13 1,873 1,201.125 417.92 268.007 l 9 173.77 1.00 173.77 2.057 1,366,951 357.44 237.535 10 13534 1.00 13534 2.203 1,502,557 298.15 203356 11 105.40 1.00 105.40 2304 1,598,597 242.84 168.492

,                    12             82.08             1.00               82.08                2390           1,682.208                  1 %.17            138.076 l                     13             63.93             1.00               63.93                2.461          1,754,800 i                                                                                                                                        157.33            112.184 14             49.79             1.00               49.79                2.516          1,809,000                  125.27             90.070 3                     15             38.77             1.00               38.77                2.552          I 845,800                    98.94            71.562

Total 4,384.29 2.297.87 3,605.65 2,234.000 4

Mean weight per mature female fish = (3,606 lbs + 2,298 mature females) = 1.57 lbs (-38.6 cm fish) Mean fecundity (unfished stock) = 972,205 eggs per female spawner J-i 5 4 TABLE 34. Annual abundance index (a parameter of the Gompertz function) wish 95% confidence interval of winter flounder larvae in l entrainment samples and total annual entrainment estimates during the larval season of occurrence, and th: volume of seawater entrained at j MNPS cach year from 1976 through 1996 during an 136-day period from February 15 through June 30. 4 a Standard 95% confidence Number entrained Seawster volume Year parameter error interval (X 10') entrained (m' X 10') l 1976 1.656 32 1,588 1,724 107.6 662.8 1977 751 47 650 852 312 585.6 ! 1978 1,947 352 1,186 2,706 87.4 490.9 6 1979 1,296 81 1,121 1,470 47.7 474.1 j 1980 2,553 37 2,475 2,632 175.7 633 3 j 1981 1,163 23 1,113 - 1,213 47.7 455.2 j 1982 2,259 36 2,184 2,334 170.4 674.1 1983 2.966 21 2,921 3,012 2193 648.0 1 1984 1,840 47 1,741 - 1,939 88.1 573.8 1985 1,585 . 48 1,483 - 1,686 83 3 528.1 1986 903 31 837 - % 8 130.6 1,353.4 1987 1,194 23 1,145 1,242 172.0 1,323.6 i 1988 1,404 42 1,315 -1,493 ' 1933 1,381.7 1989 1,677 13 1,650 1,704 175.0 1,045.9 l 1990 1,073 25 1,021 - 1,125 1 138 8 1,302.7 1991 1,149 18 1,110 1,189 1213 934.4 l 1992 3,974 76 3,812 4,136 513.9 i 1,199 3 1993 328 23 i 280-377 45.1 1,4123 1 1994 1,709 38 1,626 - 1,790 182.1 I,174.6 ' i 1995 2,571 47 2,470 - 2,671 222.9 1,133.8 j 1996 1,388 78 1,222 - 1,554 53.9 544.7 4 4 i 4 WinterFlounder 127 d

a i i c 20 i

J ' ' )

Unit 3 Outage _

                                       .                                           I j                                       -

1 t As C 15- 1 E I i g - g - 10-Y t i t 3 8

,                       z                                                        ,

} $ 5- I 1 i s t ! t 1 0

                                          #                                     1

i  ; 4 i ,  ; , , ,  ;

i 1 2 31 5 30 5 30- 14 29 ! l MAR MAR MAR APR APR MAY MAY JUN JUN Fig. 39. Abundance curve of entrained larval winter flounder in telation to the period of MNPS unit outages during .anit 199'3 t I (shutdown on November 4,1995) and Unit 2 (February 20,1996) did not operate during the entire larval winter flounder sen-son. f Effect of entrainatest on a year-class. To abundance indices ofjuvenile winter flounder, which determine the effect of winter flounder entrainment were the median CPUE of age-0 fish taken in both on a year class, the relationship between entratament early and late summer at station LR (Table 35). estimates and various indice ofjuvenile abundance Entrainment estimates were also significantly were examined. Annud entrainment estimates were negatively correlated with a calculated apparent significantly positively correlated with two larval survival rate (defimed as the A-mean CPUE of TABLE 35. Speannan's rank order conelations between the annual estunates of larval winter Gounder entramment at MNPS and the abundance indices of several post-entrainment early life history stages. I i Lower river Lower river Fall 4arly winter Niantic River Apparent larval early age-O late age-0 river-bay winter 4pring survival index* juveniles juveniles

                                                                       . Juveniles                                            age Ijuveniles           rate Annual                        0.6484*               0.6659                           0.4349                  0.2912           -0.5386 estunate of                    0.0121

• 0.0093 " 0.0553 NS - 0.2129 NS 0.0143
• entramment 14 14 20 20 20 Indices used conespond to those given on Tables 26. 27, and 34, except for the apparent survival rate, which is the A-mean CPUE of winter nounder taken in the Niantic River during adult population surveys divided by the a abundance index of 7 mm and larger larvac at the MNPS discharge (etsainment station EN).

1he three naamia shown in each correlation matrix element are: conelation coefficient (r), probability of a larger r (NS not significant [p > 0.05), * - significant at p 5 0.05, " significant at p s 0.01), and 1 number of annual observations (sample size). I 128 Monitonng Studies,1996 ' i

NSdwel i oe chc J a a wthwr(e c bco( wfsrp Nt i fbis ee au nr l ehvdDe r a a ig r s n is u8 9nyoaofNio 1 d u l wUt n mwn aai hrdHef Mmtawpg e oy nlom g dr r t up a na e e mtatnvmni g i es 4nal o b ar :n ga er ey a anl aia si n a oOi- eni-n1 r ep r nl t t u vt e aagt rur e l l ada r bug t ci ce/ e t in n 5eyueet r cr s - el 6u t c a diSe e s nCr db t nti c u o e de e oindcal e c s ae i y if r de h - 9tOf i c n t ss e et n i nivg a b ocr dg a e adicniaf udm ng nma t Bio a S n n hg vlecl f6ec at oo ue dhp r l olaoBeun dsy t dr o fbs-f n i a ,h hl n oe en c a nn7vanlat nthw eic ibs s a bp igNnh nnti lu u rc)or oyf ol 9nt usha nda y kna y ande p f 1 t y s s t t n ge nnt oc emu os nuhs f i p ). i kce e e n i e el oe t a ot c r ar iml o n a ws s n t dt ae r o9 dutofr ees ntam( ds 'niandr e u eAue n( 9Nor r r imp riad n nr it d o ic i. 9Us se hr e 1 n iea c ume c r t at n u2ed

                                                                                             ,r h d ,a rl it eh s   e r

omo iact r c vkr c h c (N m ae dnieichaon n a ep c es oaiaao t nvp t dep de n r r t t hyl h e a ic Fnb e e a Aadul l t ext f g n) . 6 S( a5d 1 ee1r a l e ct r ng t wht t p ou l pema il n B ,e 9 9r a ae oUo r s e r a d. t c l 8 n 9vs l aLiSe l d el r yal i c ebcng a eat aoh 4ae o ho o hr aCTsOaodd4mul hd 2e s at s Sg nCl ir oent nu c e t ig r a a0l neu efp n k 0 o hiy a2aot ivs uot e - no -e enr) r wud I e e v bnaut ar -hs nrvbe dng f i r e f ft ea t d is a( eay l 1 93epg l r obe i n e r -s b a ). ed n n ia_. w Onidgg i s o a l af c e l l i e s nr adeneO r r c 0aoe l t nr g

                                                                                                                                                                                                                                                   .         t t hh          hMiarNntl e t i s

t M ar (T a d t f e96 l e t g ua r l Tt ni c d Oa 9 (N na u nikoihoyi 1 t c al nar a t oi e e De es t e aegS iPeoa r t c it ai be f nd t l p a b ao v rf eher fs a ,3 . r )r p ihhal oeor et eawt e 9 nk a r nlee mec b n ,o 1n9Rivde1 e oor nfa cl vd e s ehr u 2Uep t do h.sge t n nir s t er eonuf dwf i oa np nni dou n ksp t f l s e s )b S ei r a mo nmnic c i e ntiaai iur r f i t e ieNt t r n ( ol s edieths ot fl 3npg e t i h 9edeou9 4s e 9ea vg rnt r e a ihos nt af t odhOfhi t es C N H aTy ol t ip f n caes nn i as e enac nvide n t r ic s s n i e ,a lu d f n,it;6 a oae bs ny ebee s (T u taEfs e amt ht o s )6 oi c a miahct o i v t t b ot n m ,n a t h ey r r s h wtimel t 1 l e hic t n t e kbw19m t e s . hinn t s a t as a e f e t i nt s s ho1 coe n sy r a ne nt g r f - beg hgi 1 l de l i e 7 c a 3g g 0es c al t f ut he a ny we c wol dd N w: ve e ai a o v eBrab fb l 9fa ep ub d )6 r r v a e n9o rn e ol s fa ei Niag f a aMt r m ). e t e ue a ni( nis i f vin vlaAy 5 t ir at y ey h a h a R i in a n e l e. r n t foa di a l r t a ond u n fo ,a oe 9a hovnh t x9i 1 r t t e ne t e f r u nkPt na eNa nin d sy edr v t u n nfin e d2c ep mat ne r vlamr t ior e e a 7 dn s v nbn e a 1 ha 6o d bt e r e H 9 a amt t e e e mth 9 rem nes s dicn omdnSm i a e ge g g nbe e 9 r o c o e a op b ). oc nfr l a t es o e hmy bi c ofi a a s e )6 y pl a . yl w i nrm umbe es we nc u us r f wo mfninihawsoe aoms t f )t eue S dot wxndwi w t a mthR ivt iadr t t

  ,n v         t    nt                                                   r aet ivb otl                  Lr u nte hreche                     ot h nt        if                                      e iw                                                    fs ffm . e e vHe t

u u eot nu e t t r a vieHg a e ae ve e oe v oo i i r oa es eer r r rheeI S fs os t r onk t r n nr v e e s fl e dt y a ,r nee e fr oeat r s he e hmn ii ft . e r e l , dc x r y awi5b fl 1 s a o9uCs mF n, b ur d e t u e u7r Pr uvagi t

                                                                                                                                                                                                                                     $<gy gg efW It                                         "

n6e vUie. t r ne khb - 4

                                                                                                          *gyl 4 s'                      5ips                                    {yk "gh*g                                                o c

l de - o o ;0 o. o. o d a r ee ge a s r 9y Ev 5dni a o l f .0 2 04 oo t s so o0 20 4o s 1 1 i 0o.o 0 o j ,0 2ss 0 o I 0 o 0 o. ne d Som. cq s t i vt r'w o 0 o 1 0 0 2 * , s s 7 o o o 0 0s o 5o s 5' s o e u t u n. e ay ihaiC s oiijj4iji - o jjiij i43 o ;j4 jI 13 aihr ). eae kedet noe t ner a mceg dd t nr dN - i o em rp ' b l a ow dcl u ay nbafloi a r f r E N i f E E , i ss t t s Tm, N i . N i

                                                                                                                                                                                                                                                          ,e                .

lat r r i. t n iau o Tm a ovaS it h g sehcg e n n R , . R * .* n cs e nh b ). aRed i I- e rb e AI A Mn

• e pu r f ke o ap e v N . . N I

                                                                                                                                                                                   . ,                                            N                     *       ,

t l t Ma, .

• M2
• M2 ,'* ,

hp ot e5 - t hn edadiv ar r wa en r e w E m , Em . Em eomr dma a N * +, N -

• e t r

y r t yl heiee mu r t an T N T T q o et rg t E t a enr ah S a E E ah hebp ut f t e Tm, S a Tm Sa i, e re cr e nCf t l PohoMa I t d M I W o ai o laUl a uNn r e n n A M A

     . ho e md nc        s Nm i

n t dPu T T e t a u s seEad v u eSal r E 4 Nm, I E 4 4 r T, g n cpr a t uoatl r ae Nm I I F l i m, e e or t mfc 2ig ma w (adtnnr M M o s s i Nm I e t i a L I u m in t eg 7n L s L r e me a m t ( i g hn L n g i, i. v e I Oo, I s  : s o n ,.e 1 Omf ap I p en r d e .e o n ( - No

• Ou N *
• r 9 8wao l

S S Me fhr la L HpA t t u t 1 2 i s vI la )S si, lri 3i

                                           -    nEnma t              de esr e                             e m,                                                                 e                                                             e 9                                        % r- ;Ne a ni          r nt t
                                                                                     -                                                                                m                                                              m
                                        !,                'j l                               ,          )            'j                                               I!!                   iijj                                               'i                   '.                        ll!I;l >

 . _.__.._.__ _ _._._ _ _ _ _ . _ . _ _ _ . _ . . - . . __ _ .. _ .-._ m _ _ -_                                                      - ...__..._.m_.___.                - - _

1 l l TABLE 36. Results of mans-balance calculations for each 5-day period in 1996. # Number less due Number from the . Number to the Stan of J. day entrained to mortality . Niantic River Niantic River i 5 day change (Ent) (Mort) (FromNR) (ToNR) Source orSmk period - (X 10') (X 10') (X 10') (X 10') (X 10') (X 10') i j 2-15 0.0" 0.1 a0 0.6 0.0 -0.5 2-20 0.0 0.2 0.0 1.5 6.7 5.4 2 25 0.1 0.4 0.0 5.5 6.8 1.8 3 01 0.4 0.7 0.1 14.5 . 7.0 3 06 0.8 13

                                                                                                                                                                -63 0.3                   26.6                    7.6 3 11                          1.4                                                                                            16.7

2. t 0.7 37.1 8.7 -24.1 3 16' 2.1 3.1 13 42.2 10.6 25.2 3-21 2.6 43 2.7 41.8 13.1 3-26 2.8 -19.0 5.8 3.1 373 16.0 -9.7 3 31 2.6 3.9 4.6 31.1 19.0 4 05 2.1 3.8

                                                                                                                                                               -1.0 5.1                    24.7                  21.6 4-10                          1.4                                                                                              7.9 3.7              5.6                    18.9 4-15                                                                                                           23.6          153 0.6                33               5.1 l

14.2 24.8 4 20 -0.1 19.6 1 3.6 3.9 10.5 4 25 25.1 22.0

                                                             -0.6                 43               3.6                     7.7                  24.8         24 3 4 30                           1.0               3.9              3.2                      5.7                  24.0 5-05                           13                                                                                            24.4 13              2.9                      4.2 5-10                                                                                                            22.7         21.4                   I 1.4               0.9              2.6                      3.1                 21.2                                  '

5-15 1.5 20.2 0.9 23 23 5 20 19.7 19.1

                                                             -1.4                0.5              2.1                                                                                l 1.8                  18.1 5-25                         -1.3                                                                                             17.5                   i 0.5               1.7                     l .4 5 30                                                                                                            16.6         16.0 1.2                0.9              1.5                                                                              {

1.1 15.2 6-04 15.3 ' 1.1 0.7 13 1.0 6-09 14.0 13.9 I

                                                            -0.9                 1.0              1.1                      0.8                 12.9          13.1 6 14                        -0.8                 1.1             0.9                      0.7                 11.9 6-19                        -0.7                 1.0
                                                                                                                                                          ' 12 3 0.7                      0.7                 11.1 6-24                          0.6                                                                                            11.4 0.7             0.6                      0.6                 103            10.5

• Due to roundin8, any zero value represents less than 50,000 larvae.

change in the Source or Sink term was similar to period when there was a net loss (negative Source or previous yean (NUSCO 1993,1995a,1996), except Sink term) or when the proportion from the river was for an earlier date of February 25 in 1993 (NUSCO greater than one, all larvae entrained during that time 1994a). Considerably fewer larvae were entrained were assumed to have originated from the Niantic (Eni) than were imported from LlS (i.e., positive River. His estimate was conservative because the Source or Sink), starting in early April. Also in 1996, results of a dye study and larval dispersal modeling the weekly estimates of Ent were comparable or (Dimou and Adams 1989) showed that only about smaller than losses due to natural mortality (Mort) 20% of the water discharged from the Niantic River and until early May, were considerably smaller than passed through MNPS during full three-unit the number flushed from the river into the bay operation. Estimates of annual total entrainment and (FromNR). the annual number entrained from the Niantic River During each 5-day period of the season, the were then determined by summing over all 5-day proportion of entrained larvae from the Niantic River l~ periods. In 1996, the estimated number of 31.6 was estimated from the ratio of larvae entering the million larvae entrained that were from the river was l bay from the river (FromNR) to the total input from the second smallest of the 13 year time-series (Table both sources (fromNR + Source or Sink). His -37). However, the percent entrainment attributed to proportion was applied to the total number entrained the Niantic River in 1996 was 58.6%, by far the in the same 5 day period to estimate the number largest estimate, t.s in 1984-95 approximately 14 to entrained from the Niantic River. During any 5-day 38% of winter flounder larvae originated from the i 130 Monitoring Studies,1996

j I TABLE 37. Estimates of the total number of larval winter flounder entrained, number oflarvae entrained from the Niantic River, and the percentage of total entrainment enributed to the Niantic River from 1984 through 1996. i Niantic River  % entrainment Total entrainment larval entrainment attributed to I Year (X 10') (X 10') the Niantic River 1984 88.1 33.1 37.6  ! 1985 83.3 28.8 34.6 1986 130.6 28.9 22.1 1987 172.0 42.8 24.9 , 1988 193.3 40.8 21.1 1989- 175.0 34.5 19.7 1990 138.8 39.7 28.6 1991 121.3 36.3 29.9 1 1992 513.9 82.5 16.1 l 1993 45.1 6.2 13.7 i 1994 182.1 52.0 28.6 1995 222.9 80.4 36.1 1996 $3.9 { 31.6 58 6 i } Niantic River, with the remainder coming from other Niantic River. Consistent with previous years, older i sources. The large value for 1996 was a result of the Stage 3 and 4 larvae entrained were determined to

conservative nature of the mass-balance calculation originate mostly from other sources.

i in that all larvae were assigned to the Niantic River De interpretation of mass-balance calculation ! unless other factors in the model could have results has been substantiated by results from i eccounted for them. De model was also concep- ongoing sampling and during several special studies. j . tualized for normal MNPS operating conditions and Some of the larger larvae from other areas may enter !. under the unusually low flow at MNPS in 1996, the the Niantic River during a flood tide and caused the

model appeared to overestimate the contribution of increased frequency noted in larger size-classes

larvae from the river. This led to an unrealistically ~ during some years. In special bay-wide sampling in i high estimate of Niantic River larval entrainment that April and May of 1991 (NUSCO 1992a), when about

] occurred concurrently with the second lowest 75% of Stage 3 larvae were entrained, more larvae j estimate oflarval entrainment. entered Niantic Bay from LIS cast of Millstone Point The potential impact of larval entrainment on the and passed by the MNPS intakes during a flood tide i population depends upon the age of each larva at the than were flushed out of the bay to LIS during an ebb i time it is entrained. Older individuals have a greater tide. Therefore, greater densities of Stage 3 !arvae j probability to contribute to year-class strength than were expected at station EN during a flood tide than j' younger ones. Therefore, the estimated number of during an ebb tide. This was confirmed in NUSCO 4 each developmental stage entrained during each (1993), where significantly (p s 0.05) greater Stage 3 , 5-day period was based on the proportion of each densities found in April and May from 1983 through ' stage collected at station EN. By applying the 1992 at station EN were from collections made proportion of entrainment attributed to the Niantic - during flood tides as compared to ebb tides. j River (FromNR / [FromNR + Source or Sink]), the Estimated prodnetion loss from the Niantie number of larvae in each stage was allocated to each River stock. Estimates of larvae entrained by  ; L of the two sources (Niantic River or other) for every different larval stages from the river were compared I i 5-day period. The annual total of each larval stage to annual abundance estimates for each stage in the I

entrained from either source was esti:nated by Niantic River. The latter were computed by applying  !

[ summing over all 5-day periods (Table 38). In most mortality coefficients to each stage of early life

y

ars, Stage ", was the predominant stage entrained history, beginning with total annual egg production j whose source was attributed to the Niantic River, (Table 13), which differed somewhat from estimates 4 However, in 1996 developmental stages I and 2 reported in previous years (see Spawning Stock Size

predominated entrained larvae attributed. to the and Egg Production). This process allowed the i

1 i WinterFlounder 131 i

                                                                 -,.,v                                                           -- .,,-,    , ,,-.-r-.

TABLE 38. Estimated number of winter flounder larvae entrained at MNPS by developmental stage from the Niantic River and other sources, based on mass-balance calculations for 1984 through 1996. Stage i Stage 2 Stage 3 Stage 4 Year Source (X 10') (X 10') (X 10') (X 10') 1984 Niantic River 0.2 15.4 14.4 3.2 Other 0.1 25.2 25.9 3.7 1985 Niantic River 3.5 17.9 7.1 0.4 Other 0.8 11.1 35.9 6.7 1986 Niantic River 0.7 7.7 15.9 4.5 Other 1.5 25.6 63.1 11.4 1987 Niantic River 0.8 15.6 24.5 1.9 Other 0.6 31.5 89.1 7.9 1988 Niantic River 4.1 9.8 253 1.6 Other 1.2 8.1 119.4 23.9 1989 Niantic River 2.9 l 1.5 19.7 0.5 Other 43 42.4 85.0 8.8 1990 Niantic River 1.0 6.4 28.5 Other 3.8 0.9 12.8 76.1 9.4 1991 Niantic River 0.3 3.7 27.5 4.9 Other 0.7 9.2 68.5 6.7 1992 Niantic River 5.8 10.4 $7.2 9.0 Other 31.4 56.5 308.8 34.6 j 1993 Niantic River 03 1.2 4.1 0.5 Other 13 5.4 24.2 8.1 { 1994 Niantic River 2.9 12.8 29.9 64 Other 2.7 25.5 84.5 17.4 1995 Niantic River 0.6 7.1 57.5 153 Other 1.1 14.1 109.0 18.2 1996 Niantic River 13.7 13.2 4.3 03 Other 1.7 43 12.0 4.1 determination of percent production loss of larvae three-unit operation, 20% of the daily density from the Niantic River stock (Table 39). Estimates estimates of Stage I larvae at station C were used to of Niantic River Stage 1 larvae entrained were determine Stage I larval entrainment from the calculated from daily abundance estimates (Eq. 3) at Niantic River. During periods of reduced plant station C following an evaluation presented in operation, estimates were proportionally reduced NUSCO (1993). This study indicated that based on daily water volume use. Entrainment entrainment sampling may underestimate Stage 1 estimates for Niantic River Stages 2,3, and 4 larvae larval abundance because of net extrusion, which was were from the results of mass-balance calculations, verified by sampling this year. As noted above,20% which used entrainment sampling densities. The of the Niantic River discharge passes through MNPS estimated percentage of the Niantic River winter during full three. unit operation. Therefore, for full flounder production entrained in 1996 was 25.7%, 132 Monitoring Studies,1996

TABLE 39. Estimated abundance of winter flounder larvae in the Niantic River and the number and percentage of the production entrained from the Niantic River by developmental stage from 1984 through 1996. Numbers oflarvae from the Niantic River were based cn the most recent mass-balance calculations. Projected full

• MNPS Aetn=1 MNPS operatino conditiano three-unit operatina condisiaar Niantic River Entrainment from Entrainment from Stage of abundanec* the Niantic River"  % of the the Niantic River '  % ofthe Year development (X 10') (X 10') production * (X 10') production
• 1 1984 Stage 1 2502 103 0.4 22.6 0.9 Stage 2 599 15.4 I

2.6 32.7 5.5 S'.ge 3 294 14.4 4.9 34 9 11.9 Stree 4 205 3.2 1.6 9.0 4.4

                 ~ otal                                          433                          9.4           99.2 i

22.6 l l 1985 Stage 1 2823 15.6 0.6 44.2 1.6 Stage 2 676 17.9 2.6 44.7 l 6.6  ; Stage 3 332 7.1 2.1 15.2 4.6 l Stage 4 232 0.4 0.2 1.0 04 Total 41.0 5.5 105.1 13.2 1986 Stage 1 2179 11.6 0.5 14.4 0.7 Stage 2 612 7.7 i IJ 8.5 1.4 Stage 3  ! 319 15.9 5.0 15.7 4.9 I Stage 4 223 4.5 2.0 5.1 23 Total 39.7 gJ 43.7 93 1987  ! Stage 1 2653 34.4 1.3 39.8 1.5 Stage 2 745 15.6 2.1 18.1 2.4 Stage 3 389 24.5 63 25.4 6.5 Stage 4 i 271 1.9 0.7 2.0 0.7 I Total 76.4 10.4 853 11.2 1988 Stage 1 4326 83.7 1.9 92.l 2.1 Stage 2 647 9.8 1.5 103 1.6 Stage 3 233 25.3 10.8 27.1 11.6 Stage 4 168 1.6 1.0 1.7 1.0 Total 120.4 15.2 131.2 16J 1989 Stage 1 3575 66.5 1.9 84 3 2.4 Stage 2 499 11.5 23 14 3 2.9 Stage 3 164 19.7 12.0 24.1 14.7 Stage 4 110 0.5 0.5 0.7 0.6 Total 98.2 16.6 123.4 20.5 1990 Stage 1 8850 33.2 1.8 36.7 2.0 Stage 2 760 6.4 0.8 7.6 1.0 Stage 3 209 28.5 13.6 32.4 15.5 Stage 4 180 3.8 2.1 43 2.4 Total 71.9 I8.4 81.0 20.9 1991 Stage 1 3192 8.0 03 13.0 0.4 Stage 2 2227 3.7 0.2 5.2 0.2 Stage 3 677 27.5 4.1 36.2 53 Stage 4 549 4.9 0.9 6.4 1.2 Total 44.1 5.4 60.8 7.1 Winter Flounder 133

   -_ _ _ _ _ _ _ _ . . . - - _ . . _                        __ _ . _ . _ _                 m - - . - .             .-.____.~..____._____m__.~._--_                     ._~

1 1 u a TABLE 39. (cont.). 4

                                                                                                                                                                            .I i

< Projectedfulld MNPS p.ia MNps namei.. ,o.x,w. eh .unir onmei== caadialaan-I Niantic River Entramment from i Entrainment from Stage of abundance

• the Niantic River"  % of the the Niantic River  % of the Year development (X 10') production' (X 10*) (X 10') production g

                      *)92               Stage i        1952                       23.0                         1.2 4

28.6 1.5 Stage 2 818 10.4 13 11.8 1.4 Stage 3 301 57.2 l 19.0 64.4 21.4 Stage 4 24I 9.0 l 3.7 ' 10.5 43 1- Total l

+                                                                                 99.6                       25.2                    l153                       23.7 f                     1993                Stage 1       1816                        11.7                        1.0                     133 Stage 2                                                                                                                  1.2

] 577 1.2 0.2 13 0.2 g Stage 3 104 4.1

 ^

3.9 4.2 4.0 Stage 4 73 0.5 4 0.7 0.5 0.7 Total 17.5 5.9 19 3 6.1 i 1994 Stage i 1646 27.4 1.7 36.1 { Stage 2 2.2 903 12.8 1.4 16.2 1.8 ,' Stage 3 453 29.9 6.6 38.6 8.5 Stage 4 394 6.4 1.6 7.5 1.9

                                      - Total                                     76.5                       11J                     98.4                      14.4 1995                Stage 1         846                      39.6                         4.7                    44.2                       5.2 Stage 2         534                        7.1                        13                      10.1                       1.9 Stage 3         222                      57.5                       26.0                     84 3                      38.1 Stage 4         148                       15.3                      103                      23.6                      15.9 Total                                   119.5                       42J                    162.2                      61.1 1996                Stage i          410                      24 3                        5.9                    39.0 Stage 2                                                                                                                  9.5 145                      13.2                        9.1                    25.5                       17.6 Stage 3           45                        43                        9.7                     12.2                     27.4 Stage 4           30                        03                         1.0                      1.9                      6.2 Total                                     42.1                      25.7                     78.6                      60.7 Geometnc mean 12.7                                              17.5

• Abundance estimates for 1984-89 were from Crecco and Howell (1990), for 1990 from V. Crecco (CT DEP, Old 1.

and those for 1991-% were calculated by NUSCO staff.

• Entrainment estimates attributed to the Niantic River are higher than those in Table 37 due to adjustments mad
• Valacs changed from those reponed in NUSCO (1996), based on revised estimates of total egg production (see Table 1 d .

Although only MNPS Units 1 and 2 operated in 1984 and 1985, the projected values assume full three-unit operation for all y which was the second largest conditional mortality ne entrainment estimate for 1996 was the second rate (ENT) of the ~13-year period. His large lowest since three-unit operation began in 1986, production loss estimate was mostly due to the which was related to moderate larval abundance and entrainment of Stage 2 and 3 larvae attributed to the Niantic River stock, which, as discussed above, is lowest entrained seawater volume since 1984 (Table 34). Because annual egg production in the river likely an overestimate because of the conservative during both 1995 and 1996 were the lowest of all the nature of the mass-balance calculation. Values of estimates (Table 13), high larval abundance ENT were about 5% higher than reported in NUSCO suggested substantial survival that resulted from (1996) due to a decrease in estimation of annual egg several factors. Egg hatchability for these years was production, he geometric mean of the time-series was 12.7%. apparently higher than usual with more Stage I larvae than would have been expected (Fig.17). De 134 MonitoringStudies,1996

larval mortality rate in the river during 1995 was the. Materials and Methods section (Tables 1-5; Figs. among the lowest found, but in 1996 was greater than 4-6). Simulations were made from 1960, a decade the long-term average (Table 19). The larval before Unit I went on-line, until 2060,35 years after recruitment indices at station EN (Fig. 24) for both Unit 3 is scheduled to be retired. The model years were the highest calculated. Larval accessed a secondary input file, which included development is directly related to growth rate and fishing (plus impingement mortality) rates and the shorter larval periods likely result in better survival larval entrainment losses (i.e., ENT, the percent (Houde 1987). For example, during 1995, larval Niantic River annual larval production loss) assumed growth and development were among the fastest in for each year of the simulation. Values of ENT comparison to previous years, as indicated by greater during 1984-96 were based on known rates of MNPS annual growth rates in the bay and river (Tables 17 cooling-water flow (Table 5) and calculated and 18). This resulted in a large mean larval length entrainment of Niantic River winter flounder larvae tt station EN during the first part of April and a as derived from the mass-balance calculations i relatively early date of peak abundance (Figs. 21 and discussed above (Table 39). Larval losses for 1971-22). Both growth and development appeared to be 83 were simulated by modifying a randomly chosen related to the warmer than average water temperature i value of ENT by known condenser cooling water in 1995 during the larval sease (Table 6), with high flows at MNPS for each of those years. Similarly, larval survival apparently rr hg in large rumbers entrainment rates for 1997 through 2025, which also of Stage 3 and 4 larvae available for entrainment. depended upon a unit retirement schedule (Table 1), The relatively large entrainment estimates in were estimated by randomly selecting both records of comparison to low egg production resulted in a high cooling-water flows for each unit during 1974 96 11rval production loss in 1995. Conversely, the mean temperature in spring of 1996 was the second coldest (Table 5) and the historic time-series values of ENT for full three-unit operation (Table 39). The flows since 1976 (Table 7) with a late date for peak were used to adjust the values of ENT to simulate abundance (Table 16) and a longer exposure of year-to-year variation in cooling-water use during the larvae to predators, which probably reduced some of larval winter flounder season. In this simulation, the advantages of relatively low larval densities. MNPS units were assumed to operate during a larval ne mass-balance calculations given above were winter flounder season in the future as they had in the based on actual daily condenser cooling-water past. However, neither the estimate of ENT nor unit-volumes. To determine annual percentages of the specific cooling-water flows for 1996 were used in Niantic River winter flounder production that would the random selection process because the hive been entrained since 1984 under simulated full simultaneous shutdown of all three units for nearly (100% capacity) three-unit operation, the calculations all of the larval winter flounder season resulted in were recomputed based on a maximum daily atypically low flows (Table 5) that in all likelihood condenser cooling water volume of 11.1 million d (Table 39). To have a longer time-series, will not reoccur after 1997 as well as an overly ml ay" conservative (i.e., high) estimate of ENT. All values three-unit operation was simulated to include 1984 of ENT calculated or selected for use in the SPDM and 1985, prior to Unit 3 start-up Estimated simulation presented in this particular report are reductions in year-class strength under three-unit given in Table 40. operation ranged from 7.l to 61.1% (geometrie mean A combined mortality of fishing (F) and

   = 17.5%), with the highest values found in 1995 and impingement (IMP) was used in the simulations only 1996. He annual estimates of ENT were used in              during 1971-2025, years corresponding to actual or impact assessment simulations with the SPDM as expected MNPS operation. Expected changes in the described below.

values of F over time were determined after consultation with DEP Marine Fisheries (P. Howell, Stochastic Simulation ofthe CT DEP, Old Lyme, CT, pers. comm.) and reflected Niantic River Winter Flounder Stock recent estimates of mortality and changes in regulations designed to considerably reduce F in the Model simulation of MNPS impact. He initial future (Table 2). Nominal fishing mortality rates input data used to run the SPDM were described in were initially set at F = 0.40, remained unchanged through the 1960s, mcreased abruptly to 0.50 m, Winter Flounder 135

l i TABLE 40. Schedule of conditional entrainment (ENT values), fishing (F) mortalities with adjustments for impingement (IMP), and fishing discard mortaines as implemented in the 1996 SPDM simulations. ,

                                % of year-class reduction Time simulation              based on calculated or      Nominal F                        Fractional fishing discard F for :

nen wear du-aad leveh of ENTS inlm IMP @ Ame-! Ame-2 Ane 3 Ame 4 0 1960 0.0 0.40 0.036 0.240 0.400 0.400 1 1961 0.0 0.40 0.03 6 0.240 0.400 0.400 2 1%2 0.0 0.40 0 036 0.240 0.400 0.400 3 1%) 0.0 0 40 0.036 0.240 0.400 0 400 4 1964 0.0 0.40 0.036 0.240 0.400 0.400 5 1%5 0.0 0.40 0.036 0.240 0.400 0.400 6 1966 0.0 0.40 0.036 0.240 0 400 0.400 7 1%7 0.0 0.40 0.036 0.240 0.400 0.400 i 8 1%8 0.0 0.40 0.036 0,240 0.400 0.400 9 1%9 0.0 0.40 0.036 0.240 0.400 0.400 10 1970 j 0.0 0 40 0.036 0.240 0,400 Il 0 400 ) 1971 0.1530 X ENT= 2.122 0.51 0.045 0.300 0.500 0.500 12 1972 0.2262 X ENT = 1.606 0.51 0.045 0300 0.500 0.500 13 1973 l 0 0767 X ENT= 1.250 0.51 0.045 0.300 0.500 ' 14 1974 0.500 0.1895 X ENT = 1.423 0.51 0.045 0300 0.500 0.500 15 1975 0.2262 X ENT= 2.104 0.61 0.054 0360 0.600

                                                                                                                                         ]

16 1976 0.600 i 0 4421 X ENT = 3.139 0 61 0.054 0360 0.600 17 0.600  ; 1977 0.4232 X ENT = 2.582 0.61 0.054 0360 0.600 0.600 18 1978 i 03018 X ENT = 4346 0.61 0.054 0.360 0.600 0.600 i 19 1979 03133 X ENT = 4.136 . 0.61 0.054 0360 0.600 20 0.600 1980 0.4810 X ENT = 9.861 0.61 0.054 0360 0 600 0.600 21 1981 0.2873 X ENT- 8.246 0.71 0.063 0 420 0.700 0.700 22 1982 0 4857 X ENT = 2.963 0.71 0.042 0343 0.700 0.700 23 1983 0.4675 X ENT= 5.236 0.71 0.042 0301 0.700 0.700 24 1984 9.4 0.90 0.053 0383 0.890 0.890 25 1985 5.5 0.72 0.043 0.256 0.710 0.710 26 1986 8.8 0.72 0.043 0.256 0.710 0.710 27 1987 10.4 0.72 0.043 0.256 0.710 0.710 28 1988 15.2 0 95 0.056 0.254 0.912 0.940 29 1989 16.6 1.07 0.064 0.286 1.028 1.060 30 1990 18.4 134 0.080 0.239 1.290 1330 31 1991 5.4 130 0.071 0.214 1.154 1.190 32 1992 25.2 1.27 0.076 0.227 1.222 1.260 33 1993 5.9 1.13 0.067 0.202 1.086 1.120 34 1994 11 3 1.31 0.078 0.234 1.131 1.248 35 1995 423 3 1.11 0.066 0.077 0605 0.935 36 1996 25.7  ! 0.78 0.046 0.054 0.246 0.608 37 1997 UI, U2, U3 flow X ENT = 8323 0.81 0,056 ] 0.048 0.256 0.632 38 1998 UI, U2, U3 flow X ENT= 10.905 0.81 0.048 0.056 0.256 0.632 39 1999 Ul, U2, U3 flow X ENT = 5.623 0.81 0.048 0.056 0256 0.632 40 2000 Ul, U2, U3 flow X ENT = 7.532 0.71 0.042 0.049 0324 41 2001 UI, U2, U3 flow X ENT = 14.938 0.553 j 0.71 0.042 0.049 0.224 0.553 42 2002 Ul, U2, U3 flow X ENT = 18.361 0.71 0.042 0.049 0.224 0.553 43 2003 UI, U2, U3 flow X ENT = 16354 0.71 0.042 0.049 0.224 0.553 44 2004 UI, U2, U3 flow X ENT = 14.228 0.71 0.042 0.049 0.224 0.553 45 2005 Ul, U2, U3 flow X ENT = 9355 0.71 0.042 0.049 0.224 0.553 46 2006 Ul, U2, U3 flow X ENT = 5.623 0.61 0.036 0.042 0.192 0.474 i 47 2007 Ul, U2, U3 flow X ENT = 17.541 0.61 0.036 0.042 0.192 0.474 48 2008 UI, U2, U3 flow X ENT = 23.228 0.61 0.036 0.042 0.192 0 474 I 49 2009 UI, U2, U3 flow X ENT = 15.072 0.61 0.036 0.042 0.192 0 474 50 2010 UI, U2, U3 flow X ENT = 10.799 0.61 0.036 0.042 0.192 0.474 1 51 2011 U2, U3 flow X ENT = 7.687 0.61 0.036 0.042 0.192 0.474 . 52 2012 U2, U3 flow X ENT = 14.083 0.61 0.036 0.042 0.192 0.474  !

   $3         20f3          tD in flow X ENY= 12 774          0 61      0DM             0042              0 192            0 474 136 Monitoring Studies,1996                                                                                                                1 1

l i

TABLE 40. (continued).

                                    % of year-class reduction Tirne Simulation              based on calculated or        Nominal F sten         vear                                                                                  Fractional fishing discard F for :

simuisted ieve!s of EW (nlut IMPN A ce-l Aced Ape-3 54 2014 U2, U3 flow X ENT = 4.607 Ace-4 0.61 0 036 0.042 55 2015 U2, U3 flow X ENT = 13 625 0.192 0.474 0 61 0.036 0 042 56 2016 0.192 0 474 U3 flow X ENT = 2.819 0.61 0.036 0 042 57 2017 0.192 0 474 U3 flow X ENT = 6.500 0.61 0.036 0 042 58 2018 0.192 0474 U3 flow X ENT = 13.115 0.61 0.036 0 042 59 2019 0 192 0 474 U3 flow X ENT = 6.277 0.61 0.036 0.042 60 2020 0.192 0 474 U3 flow X ENT = 24 629 0.61 0.036 0 042 61 2021 0.192 0474 U3 flow X ENT = 4.179 0.61 0.036 62 0 042 0.192 0474 2022 U3 flow X ENT = 2.819 0.61 0.036 0.042 0.192 63 2023 0.474 U3 flow X ENT = 11.396 0 61 0.036 64 2024 0.042 0.192 0.474 U3 flow X ENT = 6.277 0.61 0.036 0.042 65 2025 0.192 0.474 U3 flow X ENT= 2.787 0.61 0.036 66 2026 0 042 0.192 0 474 00 0.60 0.036 67 2027 0.042 0:192 0.474 0.0 0.60 0 036 68 2028 0.042 0.192 0 474 00 0.60 0.036 69 2029 0 042 0.192 0 474 0.0 0.60 0.036 70 2030 0.042 0.192 0 474 0.0 0 60 0 036 71 2031 0.042 0.I92 0 474 0.0 0.60 0.036 72 2032 0.042 0.192 0 474 0.0 0.60 0.036 73 0.042 0.192 0 474 2033 00 0.60 0.036 0.042 0.192 74 2034 0.0 0.474 0.60 0.036 0.042 75 2035 0.192 0.474 0.0 0.60 0.036 76 2036 0 042 0.192 0.474 00 0.60 0.036 77 2037 0.042 0.192 0 474 00 0 60 0.036 78 2038 0 042 0.192 0.474 0.0 0.60 0 036 79 2039 0.042 0.192 0.4 74 0.0 0 60 0.036 0.042 0.192 80 2040 0.0 0 474 0.60 0.036 0 042 81 2041 0.192 0.4 74 A.0 0 60 0.036 0.042 0.192 82 2042 0.0 0474 0 60 0.036 0.042 83 2043 0.192 0.4 74 00 0.60 0.036 84 0.042 0192 0.4 74 2044 0.0 0.60 0.036 0.042 0.192 85 2045 0.4 74 00 0 60 0.036 86 2045 0 042 0.192 0 474 0.0 0.60 0.036 87 0.042 0.192 0.474 2047 0.0 0.60 0.036 0.042 0.192 0 474 88 2048 0.0 0.60 0 036 0.042 89 2049 0.192 0 474 00 0.60 0.036 90 2050 0 042 0.192 0 474 0.0 0.60 0.036 91 0.042 0.192 0.474 2051 0.0 0.60 0.036 0.042 92 2052 0.192 0 474 00 0.60 0.036 03 0 042 0.I92 0 474 2053 0.0 0.60 0.036 0 042 94 2054 0.192 0 474 0.0 0.60 0.036 95 0.042 0.192 0.474 2055 00 0.60 0.036 0.042 0.192 0.474 96 2056 0.0 0.60 0.036 0.042 97 2057 0.192 0 474 0.0 0.60 0 036 98 0.042 0.192 0 474 2058 0.0 0 60 0.036 0 042 0 192 0 474 99 2059 0.0 0 60 0.036 0 042 100 2060 0 192 0 474 0.0 0.60 0.036 0.042 0.192 0474

• ENT values for 1984-96 were estimates made under actual MNPS operating conditions as shown on Table 39 For 19 . - -

ENT values were randomly selected from projected rates determined from mass-balance calculations for full three-un 1984-96 (Table 39). To adjust the chosen values of ENT, actual MNPS flow values were used for 1971-83 and randomly selec from Table 5 (except for 1996) were used for 1997-2025. for this particular report only. The values of ENT given in the table above were used in the SPDM simulations D F values were obtained from the DEP(P. HowcII CF DEP, Old Lyme, CT, pers comm.). Impingement mortality was implem equivalent instantaneous mortality rate (0.01) held constant throughout the MNPS operational penod (1971 2025). WinterFlounder 137

1971, to 0.60 in 1975, and to 0.70 in 1981 (Table 40; MSP) of 28,340 lbs, shown as the dashed line in Fig. 5). Fishing mortality increased more rapidly in Figures 41A and B; this critical stock size will be subsequent years to a maximum of 1.33 in 1990. The discussed in greater detail below. Allowing for values of F also included an additional mortality of natural variation in the simulation, even the largest 0.01 to account for average impingement losses values of random stock sizes for 1991-93 were below (lMP) during the years of MNPS operation (NUSCO 25% of MSP and the minimum value found in 1995 1992a). As a result of already implemented or was only 6.5% of MSP. The simulation illustrated proposed regulatory changes to the fisheries in 1996 that the baseline population could fall (with varying and thereafter, F was projected to decrease to 0.80 in probabilities) below the critical stock size at any time the late 1990s, to 0.70 during 2000-05, stabilize at from 1975 through 1996. However, if reductions in 0.60 in 2006, and remain unchanged throughout the F are realized as planned, the stock should recover rest of the simulation period. rapidly following its lowest point in 1993. Simulation results. Three stochastic time-series of To determine the effect of MNPS on the Niantic female spawning stock sizes were generated during River female spawning stock, the baseline time-series the three SPDM simulation runs: a theoretical was compared to the impacted time-series (Figure unfished stock, whose size was dependent only upon 41B), which is also shown as the dashed line in the dynamics of winter flounder reproduction and Figure 42. The impacted series corresponds to environmental variability; a baseline stock, whose projections of the baseline stock, but with additional size was affected by rates of fishing in addition to the annual losses due to MNPS operation (i.e., ENT + above; and an impacted stock, which further added IMP). In this impacted population projection, the the effects of MNPS entrainment and impingement to stock did not respond to larval losses due to those of fishing and natural variation. entrainment until 1974 (the fourth year of Unit 1 Baseline stock projections include fishing, but no operation), when biomass began to decline below power plant effects (Fig. 41 A). Rus, this time-series baseline levels (Fig. 42). ne lowest projected stock was used as the reference against which the impacted biomass (10,604 lbs) was again reached in 1993, stock projections were compared so that past and whereas the greatest absolute decline relative to the projected trends of Niantic River winter flounder baseline occurred in 2000 (a difference of 18,682 abundance would be taken into account. Based on Ibs), when the effects of reductions in F beginning in the age and size structure of an unfished female 1996 propagated through the spawning population, winter flounder stock at equilibrium, the unfished From this point, biomass of the impacted stock stock uze used initially in the simulations was generally paralleled that of the baseline, except for 113,415 lbs (value of P,,), which was equivalent to several periods. Relatively large (ENT > 20) 72,239 female spawners (Tables 4,32, and 33). His production loss estimates were selected for 2008 initial stock size represented the maximum spawning (23.228) and 2020 (24.629) in the SPDM simulation potential (MSP) for the unfished Niantic River time-series (Table 40). nese production losses were female spawning stock. The geometric mean reflected by decreases in biomass found mostly in estimate of MSP from the SPDM simulations was 2012-14 and 2023-25. Differences between the 113,360 lbs, which was remarkably similar to the baseline and impacted projections were 6,756 to deterministic estimate of P,, (Table 32) used to 7,843 and 4,484 to 5,787 lbs, respectively, for those initiate the model runs. By 1970, the stochastic mean years before beginning to recover under more size of the exploited stock under the starting nominal moderate estimates of ENT. De large value of ENT fishing rate of F = 0.40 was quickly reduced to chosen for 2020, however, had a smaller effect as 56,243 lbs. The simulated baseline (the solid line in only Unit 3 remained on-line by then and flow Figs. 41 A and 42) responded as expected to the high l through MNPS had been reduced by about half.  ! ra'ies of fishing through the mid-1990s as the stock As MNPS units were retired, impacted stock size  ! steadily declined to its lowest point of 12,880 lbs in began to approach that of the baseline. Impacted I 1993. Stock biomass increased only slightly to stock biomass was less than 1,700 lbs of the baseline 14,405 lbs in 1994 and 14,108 lbs in 1995. The  ; in 2030 (5 years after the end of Unit 3 operation in l annual estimated biomass from 1992 through 1995 2025) and became virtually identical to it by 2033, was only about half or less of the critical stock size These projections are only realistic for the fishing l (defined as a stock biomass equal to 25% of the rates simulated, but actual winter flounder abundance ' 138 Monitoring Studies,1996 l

1 l l 120000 - j n ] . +  ! 3 c,100000 - 1 m m J g 80000 , * . **. * * .* I g ,

                                                                                      *.           ,,.. e . . *=,
                                                                                                                                                       .   .,'f." .          :
                                                                                       .+                                                                                    j e                                          e *.    .

W 60000 - g *.,. I mun g j . m g 40000 ., ,,,,,,. .

                ,i............                ........     ........e o ,f*                    ,           -,              j a j. ,.                l
 , 20000 -                                              .                  .

I .

                )                                                *"***

O ', A. BASELINE 1960 1970 1980 1990 2000 2010 2020 2030 2040 2050 2060 YEAR 120000 - k 100000p . . . s .

                                                                                                                                            .=~. .. * ~       . . *= .

e *

• O 80000j = .
• g .

g . ,* * .. . , . * , . . *. W 60000- . g . 40000, ,. , - s i............*.,......... g w

                                                                                                                  .             w a ===e
                                                                                                                                                   /        g.g,

u. 20000 d

                                                       *=""                       .

4  %#e B. IMPACTED 0,1 , , , , , , , , , , , , , , , , ,,,, 1960 1970 1980 1990 2000 2010 2020 2030 2040 2050 2060 YEAR Fig. 41. The stochastic variability associated with projected Niantic River female winter flounder stocks expressed as biomass in Ibs for: A. The baseline stock with simulated sport and commercial fishing rates, but no effects from MNPS operation. and B. The impacted stock with both fishing effects and MNPS impact. The solid lines are the geometric means and 95% confidence interval (100 Monte Carlo replications) of each stock size trajectory and are equal to the baseline and impacted stocks illustrated on Figure 42. The symbols above and below each solid line correspond to the largest and smallest stocks among the 100 repli. cates generated each year. The horizontal dashed line represents the critical stock size (here,28,340 lbs), defm' ed in Howell et al. (1992) as stock biomass equal to 25% of the maximum spawning potential. WinterFlounder 139

60000 , i e i i l e i i

                                                 ,                     ,      No impact                  ,                ,
                ! '""i m                 .

l l s N u... .&..s i <ww i q e i i .. l 1,. l

                                                       .

• s, i ,#
• I a- ' i s

e I

                                                                                        !                                i y 30000]                       i

( i i e

                                                                                                                                                                  )

R 2 i M. ,! i k f i i E l i M 20000- e i i i u;  ; , , , , , I@N 5 e i

                                                                            \./

i i

u. l No 4

                                             ' 1,2-i                      a 3-unit                  '

1,2- ' Recovery period e j ops , unit ops , operation , unit ops i Da , , (no units online)

                                                                                                                                      ,   ,        ,               l 1960      1970           1980 i

1990 2000 2010 2020 2030 2040 2050 2060 l YEAR Fig. 42. Results of the SPDM simulation showing the combined effects of fishing and ca:culated larval entrainment impingement (dashed line labeled "ENT + IMP") on the biomass in Ibs of Niantic River female winter flou Entrainment rates changed annually according to the number of MNPS units in operation and fishing rates were al (see text and Table 40 for details). The solid line (labeled "No impact") is the baseline with fishing effects only. All s are averages of 100 Monte Carlo replicates. could depart considerably from predictions if fishing and, then, very early in the life history of a fish. The rates and other simulated conditions are not matched relative effects of stock reductions due to fishing and by actual conditions. For example, should fishing MNPS impact can be assessed by comparing the rates remain high into the late 1990s, the difference unfished stock projection line to those for the fished between the lowest points in the baseline and stock with and without plant effects (Fig. 43). Most impacted stock series would become wider and biomass reductions were attributed to fishing. recovery would take longer, assuming that fishing However, as fishing mortality was reduced and stock j would eventually decrease to projected levels. biomass increased, reductions in winter flounder ' The different nature of stock reductions caused population size caused by larval entrainment at directly by fishing and impingement and those MNPS became larger relative to the baseline until resuhing from larval losses through entrainment at MNPS units ceased operation. MNPS is related to the age structure of the spawning Stock sizes projected for each simulation scenario stock. Fishing reduces biomass of the stock at a at nine selected points in time are given in Table 41; greater rate than it reduces the number of spawners losses relative to the theoretical unfished stock for because it tends to select for larger fish and, thus, each particular year are shown as percentages in this reduces the average weight of the spawners table. Stock sizes representing the 5th and 95th remaining in the stock. However, the most important percentiles for the 100 Monte Carlo replicates difference between fishing (with an added generated for each year are also given. The component accounting for impingement) and larval theoretical unfished stock in each of the years shown entrainment is that the former process removes varied little and averaged about 121,500 lbs. Prior to individuals from each year-class every year for as MNPS operation in 1970, the baseline and the long as any fish remain, while the latter causes a impacted stocks were identical (56,993 lbs) and reduction only once in the lifetime of each generation represented about 47% of the unfished stock. By 140 Monitoring Studies,1996

i 140000 1 Unfished stock [120000 W 8 100000i - [80000 Fish <d stock 60000 g 4 No impact m 3 _-

                                                                                                *,-      .m_              y y goooo                                                      ,,...,
                                                                                  ~....

f

u. 20000 -

j .- f ENT + IMP

                        ]                               ~..*

o, , , , , , , , , , , , , , , , , , , , , { 1960 1970 1980 1990 2000 2010 2020 2030 2040 2050 2060 YEAR Fig. 43. Comparison of Niantic River winter flourder female stock biomass in Ibs after effects of fishing (solid line labeled "No impact") and MNPS operation under calculated entrr.inment and impingement rates (dashed line labeled "ENT + IMP") with the theoretical (SRR based estimate) unfished stock All stock sizes are averages of 100 Monte Carlo replicates. 4

1990, winter flounder spawning stocks under full These analyses took into account not only the mean

, MNPS three-unit operation declined to about 38% of stock biomass predicted for each year, but also the the 1970 size. However, this was mostly the result of empirical frequency distribution of 100 replicate increased fishing as the impacted stock was only predictions, including stock sizes both smaller and a 2,164 lbs less than the baseline. As noted previously, larger than the mean. To assess effects of MNPS 4 smallest stock sizes were predicted for 1993 as a operation, the probability that the Niantic River result of high rates of exploitation during the early female winter flounder spawning stock would fall 1990s. .In 1996, the baseline and impacted stocks below three selected reference sizes was determined J increased 49% and 41%, respectively, relative to directly from the frequency distribution of 100 1993, but biomass remained at only about 16% and replicates of selected annual stock sizes. The refer-13% of the unfished stock. In following years, the ence sizes were percentages (25,30, and 40%) of the , baseline stock responded more rapidly to decreased biomass of spawning females for the unfished stock i fishing than the impacted stock, increasing to 40 to (i.e., the MSP) as suggested in the Atlantic States 43% of the unfished population after 1998. In Marine Fisheries Commission management plan for contrast, the impacted stock remained at 27 to 40% inshore stocks of winter flounder (Howell et al. of the unfished stock through the remaining years of 1992). A stock that has been reduced to less than three-unit operation. The gap between the two 25% of the MSP is considered overfished and its populations began to close in 2010 and, as noted continued maintenance is questionable; spawner above, became almost identical in 2033, after which abundance may decline to even lower levels. the impacted stock could be considered fully Alternatively, conservative fishing rates that preserve , recovered. 40% of MSP allow for sustainability of stocks and Probabilistic assessment of MNPS effects. The maximize yield to fisheries in the long term. i stochastic variability associated with stock According to the management plan for eastem LIS,

~

projections for the baseline and impacted stocks (Fig. values of F ranging from 0.37 to 0.68 would be

41) formed the basis for probabilistic analyses, necessary to achieve maximum yield, depending WinterFlounder 141

TABLE 41. Expected biomass in pounds of female winter flounder spewners at seven selected points in time during SPDM simulations of th Niantic River population (see 8 Fi ures 41 and 42). Expected mean stock sizes are geometric means of 100 Monte Carlo replicates and t and ninety-nith percentiles of stock siaes of the 100 replicates of each year are Ri ven. Type of population simulated 1970 1980 1990 1996 2000 2010 2020 2030 2040 neoretical unnahrd stock

• Geometric aean 121,067 120,051 123,110 120,929 122,600 121,375 121,927 119,712 122,755 5th percendte ILt,757  %,231 101,968 99,563 95,159 99,710 99,271 98,661 98,845 95th percentile 14!,,788 146,014 152,778 148,386 157,449 156,959 150,488 145,411 147,939 Baseline
• Geometric mean 56,993 37,656 24,042 19,828 52,980 50,746 48,986 49,147 50,544 Sth percentile 43,807 26,050 16.038 13,073 31,248 36,467 33,737 34,911 95th percentile 76,133 49,136 33,909 32.935 28,303 85,365 68,323 69,% 7 69,162 69,290 Baseline mean as a % of the unnshed stock 47.1 % 31.4 % 19.5 % 16.4 % 43.2 % 41.8% 40.2 % 41.1 % 41.2%

impact (ENT + IMP)* Geometric mean 56,993 36,834 21,878 15,449 33,439

                                                                                                                                           )

45,043 44,735 47,468 50,539 5th percentile 43,807 25,417 14,533 { 10,432 20,232 31,725 30,086 95th percentile 76,133 34,056 32,940 47,919 30,825 22,494 54,072 j 61,528 62,857 67,072 69,170 Impacted mean as a % of the baseline stock 100 % 97.8% 91.0% 77.9 % 63.1% 88.8 % 91.3 %  %.6% 100 % 1mpacted mean as a % of the unnshed stock 47.1 % 30.7% 17.8% i 12.8 % 27.3% 37.1% 36.7% 40.0 % 41.2% l

• No Ashing or MNPS effects.
• Fishing effects, but no MNPS impact.

Combmed effects of entrainment and '; ;

(E!G + IMP) at MNPS in addition to Ashin8 upon various combinations of length (10,11, or 12 stock still had a relatively high probability of being i inches) and trawl codend mesh (3.5,4,5, 5.0, or 5,5 '

less than 25% and 30% of MSP (0.35,0.62, respec-inches) restrictions imposed on the commercial fishery, tively). By 2010, spawning biomass of the impacted  ; stock was likely (p = 0.91) greater than 30% of MS 1 in 1970, both the baseline and impacted stocks and had a probability of 0.42 of being greater than were likely (p = 0.92) larger than 40% of MSP (Table 40% of MSP. In 2020 and 2030, both the baseline 42)J However, by 1980 both stocks were probably (p and impacted stocks were most likely greater than i i = 0.85,0.86, respectively) smaller than 40% of MSP 30% of MSP, although the chances of being greater and had increased probabilities (0,32,0.38) of falling ' below 30% of MSP. In 1990, the stocks were almost than 40% of MSP did not improve substantially, SPDM output shows the Niantic River stock stabiliz-cenainly (p = 0.94) less than 30% of MSP and likely ing at a biomass of about 47 to 49 thousand Ibs fol-(p = 0.81,0.87) less than 25% of MSP, At the lowest lowing the shutdown of MNPS in 2025. For a winter points of both stock projections in the mid-1990s, all , flounder stock to reach a more desirable size, which replicates were below 25% of MSP (p = 0.99,1,00). { according to Howell et al. (1992) is greater than 40%  ! Relatively large reductions in fishing rates in the late of MSP, fishing mortality would have to be further 1990s allowed for an increase in spawning biomass reduced as in 2040 there still was a one in three  : of the baseline stock above 25% of MSP to more chance of stock sizes being smaller than this refer- I optimal stock sizes by 2000, However, the impacted ence level. 142 Monitoring Studies,1996 I

1 l l l TABLE 42. Probabilities of Niantic River femaic spawning stock biomass falling below three selected reference sizes at seven selected points la time. Reference sizes are expressed as a percentage of the maximum spawning potential (MSP) of 113,360 lbs for the theoretical unfished stock (F = 0). Probabilities were based on the empirical probability distribution function corresponding to 100 Monte Carlo replications. Type of population Reference simulated stock size" 1970 1980 1990 1996 2000 2010 2020 2030 2040 Baseline" 25% of MSP 0.00 0 08 0.81 0.94 0.01 0.00 0.00 0.01 0.01 Impacted' . 25% of MSP 0.00 0.10 0.87 0.98 035 0.00 0.01 0.01 0.00 Baseline 30% of MSP 0.00 032 0.94 0.97 0.08 0.03 0.05 0.03 0.08 Impacted 30% ofMSP 0 00 038 1.00 0.99 0.62 0.09 0.10 0.04 0.07 Baseline 40% of MSP 0.08 0 85 0.99 0.99 036 036 0.40 0.39 034 Impacted 40% of MSP 0.08 0.86 1.00 1.00 0.88 0.58 0.58 0.47 034

• Corresponds to reference stock sizes given in Howell et al. (1992) of 25%,30%, and 40% of the MSP (28,340 lbs. 34.008 lbs, and 45,344 lbs, respectively).
• Fishing effects, but no MNPS impact.
• Combined effects of entrainment and impingement (ENT + IMP) at MNPS in addition to fishing.

demersal predators (e.g., sevenspine bay shrimp) of l Conclusions newly metamorphosed setthd juveniles, another life I stage critically important to winter flounder year. Abundance of adult winter flounder spawners in the class formation.  ! Niantic River has been depressed since 1992. De MNPS operations were considerably reduced this current size structure of the spawning population is year, resulting in the lowest cooling-water flows in heavily skewed towards larger fish and may be an 15 years. However, because of the conservative indication of potential stock collapse. Nevertheless, assumptions of the mass-balance calculations used to despite projected low egg production,' densities of estimate production loss of larvae from the Niantic Irrvae found in Niantic River and Bay remained River (i.e., the conditional entrainment mortality - , relatively high, as did entrainment of larvae at rate), the value for 19% was unrealistically high. I MNPS. High larval densities were likely the conse. His was considered to be an overestimate because  ; quence of both high egg hatchability and successful the mass-balance model was not developed for the larval survival. Warm water temperatures during abnormally low flow conditions _ encountered at winter promotes relatively fast growth and MNPS during 1996. Nevertheless, relatively weak development of larvae and probably enhances egg larval production and settlement of young in the and larval survival. Conversely, cooler temperatures Niantic River resulted in a poor winter flounder year-retard larval growth and development and likely class produced during 1996. increase mortality by exposing larvae to increased The NUSCO stochastic population dynamics model predation while part of the plankton. This is (SPDM) was used to assess the long-term effects of apparently contradictory to other findings, as stock MNPS operation (predominantly entrainment) on and recruitment analyses suggest that largest year. Niantic River winter flounder concurrently with classes are produced in cold years. However, this relatively high rates of fishing mortality. Annual temperature effect may enly operate during larval production loss estimates of 7.1 to 61.1%, tbnormally cold years (e.g., 1977-78) as spawning, calculated for projected full MNPS three-unit hatching, and larval development periods are Operation, were used in the SPDM, along with the considerably lengthened, thereby reducing density. annual cooling-water flow histories for each MNPS dependent risks of predation and starvation and also unit. Simulations illustrated that fishing alone li:niting predator activity. Further, very cold winters reduced the stock to less than half of the unfished appear to also negatively affect the abundance of equilibrium spawner biomass (i.e., MSP or maximum WinterFlounder 143

{

                                                                                                                                                                   \

spawning potential) from about 113 thousand Ibs m References Cited > 1960 to 56 thousand Ibs by 1970. Increases in 1 fishing during the 1980s and 1990s further eroded the Al-Hossaini, M., Q. Liu, and T.J. Pitcher. 1989. i baseline stock projection (with fishing but no plant effects) to a level of about 13 thousand Ibs during the Otolith microstructure indicating growth and  ; mid-1990s. MNPS impact decreased the winter mortality among plaice, Pleuronectes platessa L., i post-larval sub-cohorts. J. Fish Biol. 35(Suppl. flounder female spawner stock by an additional ' several thousand Ibs, with biomass reduced to as low A):81 90. p as 9% of MSP. However, projected substantial Anderson, J.T.1988. A review of size dependent l reductions in fishing beginning in the late 1990s survival during pre-recruit stages of fishes in allowed the simulated stocks to recover quickly and relation to recruitment. J. Northw. Atl. Fish. Sci. 8:55-66. MNPS impact became proportionately larger m, terms Amason, A.N., and K.H. Mills.1981. Bias and loss of absolute loss of biomass as the population ' of precision due to tag loss in Jolly-Seber l rebounded. Even so, the baseline and impacted estimates for mark recapture experiments. Can. J. biomass time-series became idenocal withm a few Fish. Aquat. Sci 38:1077-1095. l years of the cessation of MNPS operation in 2025 l Bailey, K.M., and E.D. Houde. 1989. Predation on I and stock biomass leveled off at about 40% of the unfished populatson. eggs and larvae of marine fishes and the  ! Recovery of the Niantic River winter flounder recruitment problem. Adv. Mar Bio.25:183, l stock from present low levels depends upon fishing Bailey, K.M., and R.S. Batty. 1984. Laboratory mortality being reduced as proposed. The SPDM study of predation by Aurelia aurita on larvae of I outputs have projected this uptum each year since cod, flounder, plaice and herring: development relauvely optimistic forecasts of future fishing and vulnerability to capture Mar. Biol. (Berl.) mortality. rates were first presented in NUSCO 83:287-291. Bannister, R.C.A., D. Harding, and S.J. Lockwood. (1993). However, these projections have not yet been matched by reality. The lack of a recovery 1974. Larval mortality and subsequetit year-class despite efforts to reduce fishing and the mability to strength in the plaice (Pleuronectes plasessa L.). correlate abundance indices of j,uvenile winter Pages 21-38 in J.H.S. Blaxter, ed. The early life history of fish. Springer Verlag, New York, flounder with adults from specific year-classes may be analogous to the example of Atlantic cod (Gadus Begon, M. 1979. Investigating animal abundance: morhua) m Canada, although certainly to a lesser capture-recapture for biologists. University Park degree. Meyers et al. (1997) suggested that an Press, Baltimore. 97 pp. Berghahn, R. 1986. Determining abundance, unaccounted fract,on i of mortality from increasing discards ofjuvenile cod occurred as fishmg mortality distribution, .and mortality of 0-group plaice of adult stocks meressed. As cod populations (Pleuronectes platessa L.)in the Wadden Sea. J. Appl. ichthyol. 2: 11-22. declined, fishing mortality of adults not only kept Berghahn, R. 1987. Effects of tidal migration on increasing, mortality of younger fish also mcreased. This reduced recruitment exacerbated the situation growth of 0-group plaice (Pleuronectes platessa until the commercial fishery had to be closed. A L.) in the North Frisian Wadden Sea. I similar process, however, remains to be demonstrated Meeresforsch. 31:209-226. (Not seen, cited by I for the winter flounder. Even though fishing rates Karakiri et al.1989).  ! remain high at present, the Niantic River population Bergman, M.J.N., H.W. van der Veer, and JJ. Zijlstra. 1988. Plaice nurseries effects on of wmter flounder has remained resilient and very small adult spawning stocks m some recent years recruitment. J. Fish Biol. 33 (Suppl. A): 210-218. I have produced relatively abundant year classes of Bertram, D.F., R.C. Chambers, and W.C. Leggett. 1993. Negative correlations between larval and young fish. Contmued efforts m reducing fishmg mortality are necessary, however, to ensure a juvenile growth rates in winter flounder; recovery and avoid a stock collapse. implications of compensatory growth for variation in size-at-age. Mar. Ecol. Prog. Ser. 96:209 215. i f 144 Monitoring Studies,1996 i l l

Bertram, D.F., T.J. Miller, and W.C. Leggett. 1996. of variation in other species. Can. J. Fish. Aquat. Individual variation in growth and development Sci. 44:1936-1947, during the early life stage of winter flounder, Chambers, R.C., W.C. Leggett, and J.A. Brown. Pleuronectes americanus. Fish. Bull., U.S. 95:1- 1988. Variation in and among early life history

10. traits of laboratory-reared winter flounder l

Bigelow, H.B., and W.C. Schroeder.1953. Fishes of Pseudopleuronectes americanus. Mar. Ecol. I the Gulf of Maine. U.S. Fish Wildl. Serv. Bull. Prog. Ser. 47:1 15. I 53:1 577. Christensen, S.W., and C.P. Goodyear. 1988. Bishop, J.A., and P.M. Sheppard. 1973. An l Testing the validity of stock-recruitment curve evaluation of two capture-recapture models using fits. Am. Fish. Soc. Monogr. 4:219-231. the technique of computer simulation. Pages l Cormack, R.M. 1968. The statistics of ! 235-253 in M.S. Banlett and R.W. Hiorns, eds. mark-recapture methods. Oceanogr. Mar. Biol. The mathematical theory of the dynamics of l Ann. Rev. 6:455-506. biological populations. Academic Press, London. Crawford, R.E. 1990. Winter flounder in Rhode Boudreau, P.R., and L.M. Dickie. 1989. Biological Island coastal ponds. Rhode Island Sea Grant, model of production based on physiological and Univ. of Rhode Island, Narragansett, RI. RIU-G-ecological scaling of body size. Can. J. Fish. 90-001. 24 pp. Aquat. Sci. 46:614-623. Crawford, R.E., and C.G. Carey. 1985. Retention of Buckley, L.J. 1980. Changes in ribonucleic acid, winter flounder larvae within a Rhode Island salt deoxyribonucleic acid, and protein content during pond. Estuaries 8:217-227. ontogenesis in winter flounder, Pseudo. Crecco, V.A., and P. Howell. 1990. Potential effects pleuronectes americanus, and efTect of starvation. of current larval entrainment mortality from the Fish. Bull., U.S. 77:703-708. Millstone Nuclear Power Station on the winter Buckley, L.J. 1982. Effects of temperature on flounder, Pseudopleuronectes americanus, growth and biochemical composition of larval spawning population in the Niantic River, winter flounder Pseudopleuronectes americanus. Connecticut Dept. Envir. Prot., Bu. Fish., Spec Mar. Ecol. Prog. Ser. 8:181-l86. Pub. 37 pp. Buckley, LJ., A.S. Smigielski, T.A. Halavik, and Crecco, V.A., and T. Savoy. 1987. Fishery G.C. Laurence. 1990. Effects of water management plan for the American shad in the temperature on size and biochemical composition Connecticut River. Connecticut Dept. Envir. of winter flounder Pseudopleuronectes Prot., Bu. Fish., Spec. Pub. 140 pp. americanus at hatching and feeding initiation. Cushing, D.H.1971. The dependence of recruitment Fish. Bull., U.S. 88:419-428. on parent stock in different groups of fish. J. Buckley, L.J., A.S. Smigielski, T.A. Halavik, E.M. Cons. int. Explor. Mer 33:340-362. Caldarone, B.R. Burns, and G.C. Laurence. 1991. Cushing, D.H. 1974. The possible Winter flounder Pseudopleuronectes americanus density-dependence of larval monality and adult reproductive success. 11. Effects of spawning mortality in fishes. Pages 103-111 in J.H.S. tin e and female size on size, composition and Blaxter, ed. The early life history of fish. viability of eggs and larvae. Mar. Ecol. Prog. Ser. Springer-Verlag, New York. 74:125-135. Cushing, D.H., and J.G.K. Harris. 1973. Stock and Burton, M.P., and D.R. Idler. 1984. The recruitment and the problem of density  ; reproductive cycle in winter flounder dependence. Rapp. P.-v. Rdun. Cons. int. Explor. Pseudopleuronectes americanus (Walbaum). Mer 164:l42155. Can. J. Zool. 62:2563-2567. Cushing, D.H., and J. W. Horwood. 1977. Carothers, A.D. 1973. The effects of unequal Development of a model of stock and I catchabihty on Jolly-Seber estimates. Biometrics recruitment. Pages 2135 in J.H. Steele, ed. 29;79-100.  ; Fisheries mathematics. Academic Press, New l Chambers, R.C., and W.C. Leggett. 1987. Size and York. age at metamorphosis in marine fishes: an DeBlois, E.M., and W.C. Leggett. 1991. Functional l analysis of laboratory-reared winter flounder response and potential impact of invertebrate (Pseudopleuronectes americanus) with a review predators on benthic fish eggs: analysis of the Winter Flounder 145

1 l Calliopius laeviusculus-capelin (Mallotus J.R. Stauffer, eds. Biological monitoring of fish. villarus) predator-prey system. Mar. Ecol. Prog. Lexington Books, Lexington, MA. Ser. 69:205-216.' Goodyear, C.P., and S.W. Christensen. 1984. Bias- , Dimou, N.K., and E.E. Adams.1989. Application of elimination in fish population models with a 2-D ' particle tracking model to simulate stochastic variation in survival of the young. entramment of winter flounder larvae at the Trans. Am. Fish. Soc. 113:627-632. Millstone Nuclear Power Station. Energy Hess, K.W., M.P. Sissenwine, and S.B. Saila. 1975. Laboratory Report No. MIT-EL 89-002. Simulating the impact of entrainment of winter Massachusetts Institute of Technology, flounder larvae. Pages 130 in S.B. Saila, ed. Cambridge,MA. 73 pp. Fisheries and energy production: a symposium. Draper, N., and H. Smith. 1981. Applied regression D.C. Heath and Co., Lexington, MA. analysis. John Wiley and Sons, New York. 709 Hightower, J.E., and RJ. Gilbert. 1984. Using the pp, Jolly-Seber model to estimate population size, Dunn, R.S.1970. Further evidence for a three-year mortality, and recruitment for a reservoir fish oocyte maturation time in the winter flounder -l population. Trans. Am. Fish. Soc. ll3:633-641. J (Pseudopleuronectes americanus). J. Fish. Res. Hjorleifsson, E. 1992. Abundance, condition, , Board Can. 27:957 960. growth and- mortality of winter flounder l Dunn, R.S., and A.V. Tyler. 1%9. Aspects of the (Pleuronectes americanus Walbaum) larvae in j anatomy of the winter flounder ovary with Narragansett Bay during spring of 1988. Ph.D. hypotheses on oocyte maturation time. J. Fish. Dissertation. Univ. of Rhode Island, Res. Board Can. 26:1943-1947. _{' Nanagensett, RI. 259 pp. Garrod, DJ., and B.W. Jones. 1974. Stock and Hjort, J.1926. Fluctuations in the year classes of recruitment relationships in the Northeast Arctic l important food fishes. J. Cons. int. Explor. Mer j cod stock and the implications for the 1:5-38. (Not seen, cited by May 1974). i management of the stock. J. Cons. int. Explor. Hoenig, LM., D.M. Heisey, W.D. Lawing, and H.D.  ! Mer 36:35-41. Schupp. 1987. An indirect rapid methods Gendron, L. 1989. Seasonal growth of the kelp approach to assessment. Can. J. Fish. Aquat. Sci. Laminaria longieruris in Baie des Chaleurs, 44 (Suppl 2):324-338. Quebec, in relation to nutrient and light Houde E.D.1987. Fish early life history dynamics availability. Bot. Mar. 32:345-354, and recruitment variability. Am. Fish. Soc. Gibson, M.R. 1987. Preliminary assessment of Symposium 2:17-29, winter flounder (Pseudopleuronectes americanus) Houde, E.D. 1989. Subtleties and episodes in the stocks in Rhode Island waters. Rhode Island Div. early life history of fishes. J. Fish Biol. 35(Suppl. Fish Wildl., Res. Ref. Doc. 87U. 51 pp. A):29-38. Gibson, M.R.1989, Stock-recruitment relationships Hovenkamp, F., and J.IJ. Witte. 1991. Growth, for winter flounder in the S. New England area otolith growth and RNA/DNA ratios of larval and revised fishery reference points. Rhode plaice fleuronectes platessa in the North Sea Island Div. Fish Wildt., Res. Ref. Doc. 89/9. 10 1987 to 1989. Mar. Ecol. Prog. Ser. 70:105-116. pp + 5 fig. Howe, A.B., and P.G. Coates. 1975. Winter Gilbert, R.O. 1973. Approximations of the bias in flounder movements, growth and mortality off the Jolly-Seber capture-recapture model. Massachusetts. Trans. Am. Fish. Soc. 104:13-29. Biometrics 29:501-526. Howell, P., A. Howe, M. Gibson, and S. Ayvazian, Goodyear, C.P. 1977. Assessing the impact of 1992. Fishery management plan for inshore power plant mortality on the compensatory stocks of winter flounder. Fisheries management reserve of fish populations. Pages 186-195 in W. rep. no. 21 of the Atlantic States Marine Fisheries , Van Winkle, ed. Proceedings of the conference Commission.138 pp. on assessing the effects of power-plant-induced Howell, W.H., and R. Langan. 1987. Commercial i mortality on fish populations. Pergamon Press, trawler discards of four flounder species in the New York. 1 Gulf of Maine. N. Am. J. Fish. Man. 7:6-17. _l Goodyear, C.P. 1980. Compensation in fish Howell, W.H., and R. Langan. 1992. Discarding of populations. Pages 253-280 in C.H. Hocutt and commercial groundfish species in the Gulf of 3 146 Monitormg Studies,1996 i i i

____.-.-.-._._.-_______.__._.___.m _.___ 4 h i

Maine shrimp fishery. N. Am. J. Fish, Man. Lorda, E.C., and V.A. Crecco.

' 1987. 12:568-580. Stock-recruitment relationship and compensatory l Jolly, G.M. 1%5. Explicit estimates from mortality of American shad in the Connecticut j capture recapture data with death and River. Am. Fish. Soc. Symposium 1:469-482.  ! ) immigration stochastic model. Biometrika Manly, B.J.F.1971. A simulation of Jolly's method { $2:225-247. for analysing capture-recapture data. Biometrics i Karakiri, M., R. Berghahn, . and H. von 27:415-424. l Westernhagen. 1989. Growth differences in Marine Research, Inc. 19 0. Brayton Point  ! ! 0-group plaice Pleuronectes platessa as revealed investigations semi-annual report. January-June by otolith microstructure analysis. Mar. Ecol. 1992. Submitted to New England Power Co. 1 Prog. Ser. 55:15-22. Mar. thall, N., and S.D. Hicks. i Klein-MacPhee, G. 1%2. Drift of 1978. Synopsis of biological raedusae and their distribution in relation to the l data for the winter flounder, Pseudopleuronectes E.ydrography of the Niantic River, Connecticut. j americanus (Walbaum). NOAA Tech. Rep. Limnol. Oceanogr. 7:268-269. i NMFS Circ.414. 43 pp. May, R.C. 1974. Larval mortality in marine fishes  !

Koll neyer, R.C.1972. A study o' ne Niantic River and the critical period concept. Pages 3-20 in

{ estuary, Niantic, Connectim .nal report phases J.H.S. Blaxter, ed. The early life history of fish. j i and II, physical aspects oi che Niantic River Springer-Verlag, New York. j estuary, Rep. No. RDCGA 18. U.S. Coast Guard McConnaughey, R.A., and L.L. Conquest. -1993. I 3 Academy, New London, CT. 78 pp. Trawl survey estimation using a comparative 3 Kuipers, B., B. MacCurrin, J.M. Miller, H.W. van der approach based on lognormal theory. Fish. Bull., i Veer, and J.13. Witte. 1992. Small trawls in U.S. 91:107 Il8.  ! j juvenile flatfish research: their development and McCracken, F.D.1%3. Seasonal movements of the efficiency, Neth. J. Sea Res. 29:109117. winter flounder, Pseudopleuronectes americanus Laurence, G.C. 1975. Laboratory growth and (Walbaum), on the Atlantic coast. J. Fish. Res. metabolism of the winter flounder Board Can. 20:551-586. Pseudopleuronectes americanus from hatching Miller, J.M., J.S. Burke, and G.R. Fitzhugh. 1991, through metamorphosis at three temperatures. Early life history patterns of Atlantic North Mar. Biol. (Berl.) 32:223-229. American flatfish: likely (and unlikely) factors 1 Lturence, G.C.1977. A bioenergetic model for the l controlling recruitment. Neth. J. Sea Res. 27:261- ' analysis of feeding and survival potential of 275. winter flounder, Pseudopleuronectes americanus, Morrison, J.A., I.R. Napier, and J.C. Gamble. 1991. larvae during the period from hatching through Mass mortality of herring eggs associated with a metamorphosis. Fish Bull., U.S. 75:529-546. sedimenting diatom bloom. ICES J. Mar. Sci. Lobell, M.J. 1939. A biological survey of the salt 48:237-245, waters of Long Island,1938. Report on certain Myers, R.A., J.A. Hutchings, and N.J.Barrowman. fishes. Winter flounder (Pseudopleuronectes 1997. Why do fish stocks collapse? The example americanus). Suppl. 28th Ann. Rep., N.Y. Cons. of cod in Atlantic Canada. Ecol. Appl. 7:91-106. Dep., Pt.1:63-96. Myers, R.A., and N.G. Cadigan. 1993a. Density-Lockwood, S.J. 1972. The settlement, distribution dependent juvenile mortality in marine demersal i and movements 010-group plaice (Pleuronectes fish. Can. J. Fish. Aquat. Sci 50:15761590. platessa L.) in Filey Bay, Yorkshire. J. Fish. Biol. 6:465-477, Myers, R.A., and N.G. Cadigan. 1993b. Is juvenile mortality in marine demersal fish variable 7 Can. Lockwood, SJ.1980. Density-dependent mortality in J. Fish. Aquat. Sci. 50:1591-1598.

      ' 0-group plaice (Pleuronectesplatessa L.) popula.                 Nichols, J.D., B.R. Noon, S.L. Stokes, and J.E.

tions. J. Cons. int. Explor. Mer 39:148-153. Hines. 1981. Remarks on the use of capture-Longhurst, A. 1983.' Benthic-pelagic coupling and recapture methodology in estimating avian export of organic carbon from a tropical Atlantic population size. continental shelf, Sierra Leone. Est. Coast. Shelf Studies in Avian Biol. 6:121 136. (Not seen, cited by Hightower and Sci.17:261-285. Gilbert 1984). WinterFlounder 147

  .- - - . . .     -       --        . - _-.~. - - - - - - - - ~ .                                                 - - - _ .- ~. _

4 t t NUSCO (Northeast Utilities Service Company). NUSCO. 1991a. Evaluation of the larval winter 1976. Environmental assessment of the flounder sampling program in the Niantic River. ! condenser cooling water intake structures (316b 1 Enclosure to letter D04343 dated January. 23, demonstration). Vol. l and 2. 1991 from E.J. Mroczka,- NUSCO, to L. i NUSCO. 1981. Plankton ecology. In Monitoring Carothers, Commissioner, CT DEP. the marine environment of Long Island Sound at NUSCO.1991b. Winter flounder studies. Pages 9-Millstone Nuclear Power Station, Waterford, 86 in Monitoring the m .rine environment of Long . Connecticut. Annual report, !?80. 40 pp. Island Sound at Millstoce Nuclear Power Station, 1 NUSCO. 1985. Winter floutder studies. In ! Waterford, Connecticut. Ae.ual report 1990. Monitoring the ' marine environment of Long NUSCO.1992a. Winter flounder studies. Pages 7-l Island Sound at Millstone Nuclear Power Station, 109 in Monitoring the marine environment of Waterford, Connecticut. Annual report, 1984. 74 Long Island Sound at Millstone Nuclear Power

pp. j NUSCO.. Station, Waterford, Connecticut. Annual report 3

1986a. Winter flounder studies. In 1991. l Monitoring the marine environment of Long i NUSCO.1992b. Niantic Bay current studies. Pages  ! Island Sound at Millstone Nuclear Power Station, 317-331 in Monitoring the marine environment of y Waterford, Connecticut. Annual report, 1985. 69 Long Island Sound at Millstone Nuclear Power

pp. j Station, Waterford, Connecticut. Annual report '

NUSCO.1986b. The effectiveness of the Millstone 1991. l Unit I sluiceway in returning impinged organisms l NUSCO.1993. Winter flounder studies. Pages 191-to Long Island Sound.18 pp.  ! 269 in Monitoring the marine environment of NUSCO. 1987. Winter flounder studies. In j Long Island Sound at Millstone Nuclear Power i Monitoring the marine environment of Long Station, Waterford, Connecticut. Annual report

1sland Sound at Millstone Nuclear Power Station, 1992 Waterford, Connecticut. Summary of studies i NUSCO. 1994a. Winter flounder studies. Pages prior to Unit 3 operation.151 pp. 141 228 in Monitoring the marine environment of NUSCO.1988a. The effectiveness of the Millstone Long Island Sound at Millstone Nuclear Power Unit 3 fish return system. Appendix 1 to Station. Annualreport 1993.

Enclosure 3 to Letter D01830 dated January 29, NUSCO.1994b. Progress report on the MNPS fish 1988 from E.L Mroczka, NUSCO, to L. retum systems. Enclosure I to letter D08071 j Carothers Commissioner, CT DEP. 21 pp, dated October 20,1994 from D. Miller, NNECO, i NUSCO. 1988b. Winter flounder studies. 'Pages j to T. Keeney, Commissioner, CT DEP. 149-224 in Monitoring the marine environment of NUSCO.1995a. Winter flounder studies. Pages 9-Long Island Sound at Millstone Nuclear Power . Station, Waterford, Connecticut. 92 in Monitoring the marine environment of Long 3 Three-unit operational studies, 1986-1987. Island Sound at Millstone Nuclear Power Station. Annual report 1994. NUSCO. 1988c. The usage and estimation of NUSCO. 1995b. The 1994-95 winter-spring DELTA means. Pages 311-320 in Monitoring the refueling outages at Millstone Nuclear Power

marine environment of Long Island Sound at i

Station Units 2 and 3 relative to the larval winter Millstone Nuclear Power Station, Waterford,

Connecticut. flounder season. Enclosure I to letter D08983

' Three-unit operational studies, dated August 3,1995 from D. Miller, NNECO, to 1986-1987. T. Keeney, Commissioner, CT DEP. j NUSCO. 1989. Winter flounder studies. Pages NUSCO.1996. Wimer flounder studies. Pages 109-239-316 in Monitoring the marine environment of 197 in Monitoring the marine environment of , Long Island Sound at Millstone Nuclear Power Long Island Sound at Millstone Nuclear Power Station, Waterford, Connecticut. Annual report Station. Annual report 1995, i 1988. O'Brien, L., J. Burnett, and R.K. Mayo. 1993. ] NUSCO.1990. Winter flounder studies. Pages 9-77

Maturation of nineteen species of finfish off the in Monitoring the marine environment of Long northeast coast of the United States, 1985-1990.

Island Sound at Millstone Nuclear Power Station, NOAA Tech. Rep. NMFS 113. 66 pp. 1 Waterford, Connecticut. Annual report 1989. 1 i

!                148 MonitoringStudies,1996 I

Olla, B.L., R. Wicklund, and S. Wilk. 1969. Ricker, W.E. 1954. Stock and recruitment. J. Fish. Behavior of winter flounder in a natural habitat. Res. Board Can. 11:559-623. j Trans. Am. Fish. Soc. 98:717-720. Ricker, W.E. 1973. Linear regressions in fishery  ; Parrish, B.B.1963. Some remarks on the selection research. J. Fish. Res. Board Can. 30.409-434. processes in fishing operations. Int. Comm. Ricker, W.E.1975. Computation and interpretation l Northwest Atl. Fish. Spec. Pub. 5:166-170. of biological statistics of fish populations. Bull. Pearcy, W.G. 1962. Ecology of an estuarine Fish. Res. Board Can. 191. 382 pp. population of winter flounder Pseudopleuronectes Ricker, W.E.1984. Computation and uses of central americanus (Walbaum). Bull. Bingham trend lines. Can. J. Zool. 62:1897-1905. Oceanogr. Coll. 18(1):1-78. Raff, D.A. 1973. On the accuracy of some Pennington, M. 1983. Efficient estimators of mark-recapture estimators. Oecologica (Berl.) l abundance for fish plankton surveys. Biometrics 12:15-34. 39:281 286. Rogers, C.A. 1976. Effects of temperature and Pennington, M. 1986. Some statistical techniques salinity on the survival of winter flounder for estimating abundance indices from trawl embryos. Fish. Bull., U.S. 74:52-58. , surveys. Fish. Bull., U.S. 84:519-525. Rogers, S.I., and S.J. Lockwood. Perlmutter, A.1947. The blackback flounder and its 1989. ( Observations on the capture efficiency of a fishery in New England and New York. Bull. two-metre beam trawl forjuvenile fiatfish. Neth. Bingham Oceanogr. Coll.11:1-92. J. Sea Res. 23:347-352. Pihl, L. 1990. Year-class strength regulation in Rothschild, B.J., and G.T. DiNardo. 1987. l plaice (Pleuronectes plateJsa L.) on the Swedish Comparison of recruitment variability and life  ! west coast. Hydrobiologia 195:79-88. history data among marine and anadromous I Pihl L., and H.W. van der Veer. 1992. Importance fishes. Pages 531546 in M.J. Dadswell, R.J. of exposure and habitat structure for the  ! Klauda, C.M. Moffitt, R.L. Saunders, R.A. l population density of 0-group plaice, Rulifson, and J E. Cooper, eds. Common Pleuronectes platessa L., in coastal nursery areas. strategies of anadromous and catadromous fishes. Neth. J. Sea Res. 29:145 152. Am. Fish. Soc. Sym.1. Pollock, K.H., J.D. Nichols, C. Brownie, and J.E. Roughgarden, J. 1979. Evolutionary ecology of I Hines. 1990. Statistical inference for capture- single populations. Pages 295-408 in The theory ' recapture experiments. Wildl. Monogr.107. 97 of population genetics and evolutionary ecology: pp. an introduction. MacMillan Publishing Poxton, M.G., A. Eleftheriou, and A.D. McIntyre. Company, Inc., New York. 1982. The population dynamics of 0-group j Rubinstein, R.Y. 1981. Simulation and the Monte ' fiatfish in the Clyde Sea area. Est. Coast. Shelf Carlo method. John Wiley and Sons, New York. l Sci.14:265-282. 278 pp. Poxton, M.G., and N.A. Nasir. 1985. The Saila, S.B. 1961. A study of winter flounder i distribution and population dynamics of 0-group movements. Limnol. Oceanogr. 6:292-298. plaice (Pleuronectes platessa L.) on nursery Saila, S.B. 1962a. The contribution of estuaries to l grounds in the Firth of Forth. Est. Coast. Shelf the offshore winter flounder fishery in Rhode Sci. 21:845-857. Island. Proc. Gulf Caribb. Fish. Inst.14th Annu. Rauck, G., and J.J. Zijlstra. 1978. On the nursery Sess.1961:95109. l 1 aspects of the Wadden Sea for some commercial Saila, S.B. 1962b. Proposed hurricane barriers fish species and possible long-term changes. related to winter flounder movements in Rapp. P.-v. R6un. Cons. int. Explor. Mer Narragansett Bay. Trans. Am. Fish. Soc. 172:266-275. (Not seen, cited by Zijlstra et al. 91:189-195. 1 1982). Saila, S.B., and E. Lorda. 1977. Sensitivity analysis Reed, M., M.L. Spaulding, E. Lorda, H. Walker, and applied to a matrix model of the Hudson River S.B. Saila. 1984. Oil spill fishery impact striped bass population. Pages 311-332 in W. assessment modeling: the fisheries recruitment l Van Winkle, ed. Assessing the effects of power- ' problem. Est. Coast. Shelf Sci. 19:591-610. plant-induced mortality on fish' populations. Pergamon Press, New York. Winter Flounder 149 l l

SAS lastitute lac.1985. SAS user's guide: statistics. Spaulding, M.L., S.B. Saila, E. Lorda, H. Walker, E. Version 5 edition. SAS Institute Inc., Cary, NC. 956 pp. Anderson, and J.C. Swanson. 1983. Oil-spill fishery impact assessment model: application to Scarlett, P.G., and R.L. Allen. 1992. Temporal and selected Georges Bank fish species. Est. Coast, spatial - distribution of winter flounder Shelf Sci.16:Sil 541. (Pleuronectes americanus) spawning in Steele, J., and R.R.C. Edwanis. 1970. The ecology Manasquan River, New Jersey. Bull. N.J. Acad. of 0-group plaice and common dabs in Loch Ewe, Sci. 37:13-17. IV. Dynamics of the plaice and dab populations. Scott, W.B., and M.G. Scott. 1988. Atlantic fishes J. Exp. Mar. Biol. 4:174-187. of Canada. Can. Bull. Fish. Aquat. Sci. 219. 731 Stuart, A., and J.K. Ord. 1987. Kendall's advanced pp. Simpson, D.G. 1989. Codend selection of winter theory of statistics. Vol.1. ' Distribution theory. flounder Pseudopleuronectes Oxford University Press, New York. 604 pp. americanus. Townsend, D.W., and L.M. Cammen. NOAA Tech. Rep. NMFS 75,10 pp. ~

                                                                                                             -1988.

Sissenwine, M.B. 1984. Why do fish populations Potential importance of the timing of spring vary? Pages 59-94 in R.M. May, ed. plankton blooms to benthic-pelagic coupling and Exploitation recruitment of juvenile demersal fishes. Biol. of marine communities. Springer Verlag, New York. Oceanov.5:215 229. Vaughan, D.S. 1981. An age structure model of Smigielski, A.S.1975. Hormonal-induced ovulation yellow perch in westem Lake Erie. Pages 189-of the winter flounder, Pseudopleuronectes americanus. Fish Bull., U.S. 73:431-438. 216 in D.G. Chapman and V.F. Gallucci, eds. Quantitative population dynamics. International Smith, E.M., E.C. Mariani, A.P. Petrillo, L.A. Gunn, Co-operative Publishing House, Fairland, MD. and M.S. Alexander.1989. Principal fisheries of Veer, H.W. van der. 1985. Impact of coelenterate Long Island Sound,1%I 1985. Connecticut predation on larval plaice Pleuronectes platessa Dept. Envir. Prot., Bu. Fish., Mar Fish. Program 47 pp. + app. and flounder Platicht/rys flesus stock in the Smith, T.D. 1988. Stock assessment methods: the westem Wadden Sea. Mar. Ecol. Prog. Ser. 25:229-238, first fifty years. Pages 1-33 in J.A. Gulland, ed. Veer, H.W, van der. 1986. Immigration, settlement, Fish population dynamics (second ed.). John , Wiley and Sons, New York. and density-dependent mortality of a larval and early postlarval 0-group plaice (Pleuronectes Smith, W.G., J.D. Sibunks, and A. Wells, 1975, platessa) population in the westem Wadden Sea. Seasonal distributions of larval flatfishes Mar. Ecol. Prog. Ser 29:223-236. (Pleuronectiformes) on the continental shelf Veer H.W. van der, and MJ.N. Bergman. 1987. between Cape Cod, Massachusetts and Cape Predation by crustaceans on a newly settled Lookout, North Carolina, 1 % 5-1966. NOAA 0-group plaice fleuronectes platessa population Tech. Rep NMFS SSRF 691. 68 pp. Snedecor, G.W., and W.C. Cochran. in the westem Wadden Sea. Mar. Ecol. Prog. Ser. 1%7. 35:203-215. Statistical methods. The Iowa State University j Veer, H.W. van der, M.J.N. Bergman, R. Dapper, and Press, Ames,IA. 593 pp. l J.13. Witte. 1991. Population dynamics of an Sogard, S M.1990. Parameters of habitat quality for l intertidal 0-group flounder Platichthys flesus epibenthic fishes and decapod crustaceans in New population in the western Dutch Wadden Sea. Jersey estunnes- Ph.D. dissertation, Rutgers Mar. Ecol. Prog. Ser. 73:141 148. University, New Brunswick, NJ. 195 pp. (Not Veer, H.W. van der, L. Pihl, and M.J.N. Bergman. seen, cited by Sogard and Able 1992). Sogard, S.M., and K.W. Able. 1990. Recruitment mechanisms in North Sea 1992. Growth plaice Fleuronectes platessa. Mar. Ecol. Prog.

                                                                                                                    ]

variation ' of newly settled winter flounder Ser. 64:1-12. (Pseudopleuronectes americanus) in New Jersey l Wigley, S.E., and W.L. Gabriel. 1991. Distribution estuaries as deta: mined by otolith microstructure,  ; of sexually immature components of 10 northwest ' Neth. J. Ses Res. 29:163-172. Atlantic groundfish species based on Northeast Southwood, T.R.E. 1978. Ecological methods.  ; Fisheries Center bottom trawl surveys 1%8-86. l Halstead Press, New York. 523 pp. NOAA Tech. Mem. NMFS-F/NEC-80. 17 pp.  : i 150 Monitormg Studies,1996 l l

Williams, P.J., and J.A. Brown. 1992. Development - changes in the escape response of larval winter flounder Pleuronectes americanus from hatch through metamorphosis. Mar. Ecol. Prog. Ser. 88:185-193. Witherell, D.B., and J. Burnen. 1993. Growth and maturation of winter flounder, Pleuronectes americanus, in Massachusetts. Fish. Bull., U.S. 91:816-820. Witting, D.A., and K.W. Able. 1993. Effects of body size on probability of predation forjuvenile summer and winter flounder based on laboratory experiments. Fish. Bull., U.S. 91:577 581. Witting, D.A., and K.W. Able. ' 1995. Predation by sevenspine bay shrimp Crangon septemspinosa on winter flounder Pleuronectes americanus during settlement: laboratory experiments. Mar. Ecol. Prog. Ser. 123:23-31. Zijlstra, JJ., R. Dapper, and J.11. Wine. 1982. Settlement, growth and mortality of post-larval plaice (Pleuronectes platessa L.) in the western Wadden Sea. Neth. J. Sea Res. 15:250-272. Zijlstra, J.J., and J. IJ. Wine. 1985. On the recruitment of 0-group plaice in the North Sea. Neth. J. Zool. 35:360-376. (Not seen, cited by van der Veer and Bergman 1987). Winter Flounder 151

1 1 i 1 l l l

                                                 )

l l 1 i 152 Monitoring Studies,1996

1 Lobster Studies Introduction.. . ... .. . .. .... . .. .. . .. . . . . . .155 Materials and Methods . .. .. ...... . . . . . . . .. . . .. . . . . .155 Results and Discussion... .. . . . . . . . . . . . . .... . . . . 158 Water Temperature. .. ... ... . . . . . . . . .. . . 158 Abundance and Catch-per-Unit-Effort . . . . . . . . . . . . ... 159 Population Characteristics.. . .. . . . . .. . . . 162 Size Frequency.. . .

                                                                                                                                                                     .              .162 Sex Ratios..                .                      .               .                       .                   ..                         . . .          .            .163 Reproduction.             .             .            . .             .                       ..... .                 . . . . ..                      .         . .163 Molting and Growth. . .                   ...                      . . .                                .         .              ..             . . . . . .           .165 Culls..              . . . . . .          .                     .                            ..            .                  ..                           . . 166 Tagging Program . ..                       ..       .          . . . . . .                 . . . . .            ..                    .. .. . . . .                          .168 Movement.. .. .. . .        . . .           . .               . . .                            . . .        .                     .. ...                 . ..              . 169 Entrainment... ..       .     ..       . . .          ..            . . . .                        . . .                                     .. . ..                         .170 Conclusions.. . . . . . . ....... ...... .. ..                                   ..             . . . . .          ..        .         .                   . . .           .        .173 References Cited... .. .. .. ..... . . . . . . .               . . . .                 . . .                    ..    .                ..              .              ..... 173 Lobster Studies 153

1 1 l I l l l I i . l l 4

l i i l

l s 154 Monitoring Studies,19%

I Lobster Studies Intraluction licenses, quotas, or closed seasons and areas to protect the lobster resource. Lobsters in the The principal fishery for American lobster. Millstone Point area are heasily exploited. with over Homarus americanus, in the United States extends 90% caught n the first year after molting to legal from coastal Maine through southern New England size. The local lobster population has been studied and long Island Sound (LIS). The American lobster extensively since 1978 to determine if operation of supports the most valuable single species commercial the Millstone Nuclear Power Station (MNPS) has fishery in the Northeast United States. In LIS, caused changes beyond nose expected fran natural , annual landings have ranged from 0.8 to 2.7 million variability and the high level of fishmg. pounds since 1978 and yielded between 2.4 and 8.4 The potential irnpacts of power plant operations on million dollars to lobstermen employed in the fishery se local population of lobsters include entrainment (Blake and Smith 1984; Smith et al. 1989; oflarval lobsters through the cooling water systems. Connecticut Department of Emironmental impingement of juveniles and adults on the intake Protection CT DEP, Marine Fishery Statistics). traveling screens. and effects of the heated discharge. Between 25 and 30% of the total Connecticut Entrainment and impingement contribute additional landings during 1996 were made in New London mortality to the local lobster population and could county, which includes the Millstone Point area. alter recruitment patterns. Also, heated effluent may Fishing effort is intense throughout the range of the affect the distribution or behasior of lobsters in the species and recent stock assessments have warned discharge area. that the lobster resource is overfished and vulnerable The objectives of the lobster program are to to collapse (NMFS 1996). The intense exploitation evaluate year-to-year, seasonal, and among station cf lobsters throughout their range has raised changes in catch-per-unit effort as well as population concerns over possible impacts of increased fishing characteristics such as size frequency, growth rate, mortality rates on egg production and recruitment to sex ratios, female size at sexual maturity. coastal populations (Anthony and Caddy 1980). In characteristics of egg-bearing females, and lobster response to concerns raised by fishery managers, movements. Since 1984. studies have been biologists, and lobstermen, the New England Fishery conducted during the hatching season to estimate the ' Management Council recommended an increase of number of lobster lanse entrained through the , the minimum legal size oflobsters to improve larval cooling water systems. Impacts associated with production and subsequent recruitment. Another Plant operations on the local lobster population were  ; regulation was implemented throughout the lobster assessed by comparing results of the 19% study to l fishery in the mid-1980s to improve lobster sunival other 3-unit operational study years (1986-1995) and by requiring escape vents in wire lobster traps which to data collected during 2 unit operations (1978-allow escape of sublegal-sized lobsters As a result 1985). Results from the 2-unit period were also

 - cf these management policies, the sustainability of                        compared to combined 3-unit operational data the lobster resource should be enhanced The new                            (1986-1996) to assess impacts associated with the lobster fishery regulations implemented in                                 addition of a third unit at Millstone. These results Connecticut increased the minimum legal size                               were compared, when appropriate, to other studies (carapace length) from 81.0 mm (3 '/u in) in 1988 to                       conducted in LIS and throughout the range of the 81.8 mm in 1989 (3 '/n in) and to 82.6 mm (3 '/4 in)                       American lobster.

in 1990. Federal regulations required lobster Producing states to increase the minimum legal size 5 to 84.1 mm (3 /w in). Lobstermen successfully Materials and Methods lobbied to hold the minimum legal size at 82.6 mm. However overwhelming scientific esidence indicates that fishing mortality rates throughout the range of Description of methods used to conduct lobster lobsters should be reduced. Fishery managers, Population studies can be found in NUSCO (1982, regulators and the lobster industry are currently 1987a). Four pot-trawls, each consisting of five examining other fishing effort reduction techniques, double entry wire pots (76 x 51 x 30 cm; 2.5 cm2 such as trap limitations, moratorium on new mesh) equally spaced along a 50-75 m line buoyed at Lobster Studies 155

  .~.           .              .     . -        . . _ .   -        ~.-           - - _     .        . -    _       ..- - . - ~ _ . . _ - .

l 4 l

                                                                                                    )

b W# % c +FO M f Son,q

                                                              ) %er            Norm             P 0:     Um I' mi
                                                                     \                        /

l'

                                              $         [                              [                                                   i Mon'k Go, JC g

p.  % Poins y'g m g s

g point ( 7 a.mm.,, ] k 4 Fi t t. I a=haa TT-Tweeree. of the Milkaane Nuclear Power Station (MNPS). mui the tlwee lebener sempims statums (ek JC=Jarden Cove. IN= inta I both ends were used to collect lobsters from May lobster removal, pots were rebaited and reset in the l through October. Pot-trawls were set in three rocky same area. On Fridays, lobsters caught that week areas in the vicinity of MNPS (Fig.1). Pots set in were examined and the following data recorded: sex. l Jordan Cove (average depth 6 m) were 500 m cast of presence of eggs (berried), carapace length (CL) to the Mi'Istone discharge. The intake station (average i the nearest 0.1 mm, crusher claw position, missing I depth 5 m) was 600 m west of the discharge near the claws, and molt stage (Aiken 1973). Lobsters were power plant intake structures, and the Twotree tagged with a serially numbered international orange station (average depth 12 m) was located south of sphyrion tag (Scarratt and Elson 1965: Scarratt Millstone Point, about 1600 m offshore near Twotree 1970), and released at the site of capture. island. Beginning in 1984, pots were individually Recaptured tagged lobsters and severely injured or numbered to determine the variability in catch } newly molted (soA) lobsters were released untagged l among pots, and to prmide more accurate values for aAer recording the above data. catch-per-pot than an average catch-per-pot based on Beginning in 1981, the size at which females a total of 20 pots per sampling location. Pots were become sexually mature was estimated by measuring hauled on Monday, Wednesday, and Friday of each the maximum outside width of the second abdominal week, weather permitting; on holiday weeks or segment of all females to the nearest 0.1 mm. during periods of inclement weather the pots were Female size at sexual maturity was estimated by hauled twice per week. On each sampling trip, calculating the ratio of abdominal width to carapace surface and bottom water temperatures and salinities length and plotting that ratio against carapace length were recorded at each station. Lobsters were (Skud and Perkins 1%9; Kroias. 1973). removed from the pots, banded to restram chelipeds, Lobster larvac have been sampled from 1984 to trausported to the laboratory, and kept in a tank , supplied with a continuous flow of seawater. AAer 1996 during the period of their occurreise (May  ! through July) at the discharges of Units 1. 2, and 3. 156 Monitoring Studies,1996 l

l { Samples were collected with a 1.0 x 6.0 m conical (McConnaughey and Conquest 1993). Annual plankton net of 1.0 mm mesh. The volume of geometric mean CPUEs were calculated for all cooling water sampled was estimated from the lobster sizes. The annual abundance (CPUE) of average readings of four General Oceanic flowmeters legal-size lobsters in the MNPS area was estimated located in the mouth of the net; about 4000 m' of by using the 4-mean. The A-mean was a more l cooling water were filtered in each sampic by appropriate statistic for describing the CPUE of deploymg the net for 45-60 minutes. From 1984 to legal-size lobster, since a large number of zero 1993, eight lobster larvac entrainment samples (four observations were present in the data (i.e.. many pots day and four night) were collected each week; contain no legal-size lobsters). Both geometnc beginning in 1994, the sampling frequency was means of all lobsters and A means of legal-size reduced to six samples per week (three day and three lobsters were used to compare annual variation in night). Each sample was sorted immediately, or CPUE. In the following Results and Discussion placed in a 1.0 mm mesh sieve and kept for less than l section, the geometric mean abundance of all 24 h in tanks supplied with a continuous flow of seawater. Samples were sorted in a white enamel lobsters is called "mean total CPUE" while the A-mean abundance of legal-size lobsters is referred to pan: larvae were examined for movcment and as "mean legal CPUE" The distribution-free, classified as live or dead. Lobster larvae were also classified by stage (1-IV) according to the enteria Mann-Kendall test (Hollander and Wolfe 1973) was used to determine presence of significant trends in established by Herrick (1911). The abundance of the time senes of annual CPUE data, and of several 1:rvac in entrainment samples was standardized as the number oflarvac per unit-volume. The seasonal other selected population charactenstics. Slopes of (May through July) mean density was calculated as significant trends were calculated using Sen's estimator of the slope (Sen 1968). the mean of the assumed

• delta" distribution. The influence of water temperature on lobster referred to as 4-mean (Pennington 1983; NUSCO molting was examined by estimating the time when 1988a). To estimate the total number of larvae lobster molts peaked each year and correlating the entrained, the A-mean density was scaled by the total annual molt peaks with water temperature. Molting volume of water pumped through the plants during the sampling period. peaks were derived using the inficction point of the Impingement studies wre conducted at Unit I and Gompertz growth function fitted to data reflecting the cumulative percentage of molting lobsters at 2 intakes from 1975 through 1987; results weekly intervals during the molting season. This summanzed in NUSCO (1987a) included estimates growth function had the form:

of total number of lobsters impinged. as well as mean size, sex ratio, proportion of culls, and C , = 10 0 e "e-n survival probabilitics for impinged lobsters. Possible where C, = impacts associated with impingement of lobsters at cumulative percentage of molting lobsters, Units I and 3 were mitigated by installing fish return t= time in weeks. systems in the intakes, which return impinged p= inflection point scaled in weeks organisms to LIS (NUSCO 1986; 1987b). from May 1st, and Subsequently, NUSCO and the CT DEP agreed to k= shape parameter. discontinue impingement monitoring (NUSCO 1988b). Catch-per-unit-effort (CPUE; i.e., the number of The derivative of the Gompertz function with respect to time yields a " molt frequency" function which lobsters caught per pothaul) was used to describe the describes the distnbution of annual molts. Annual annual abundance of lobsters in the MNPS area. " molt frequencies" were then correlated with mean Because these CPUE data are ratios, which are not additive and have an asymmetric distribution about bottom water temperature during May to investigate the arithmetic mean, the geometric mean was a possible relationship between water temperature and molting. computed to analyze trends in CPUE. The geometric mean is better suited for constructing asymmetnc confidence intervals for skewed data Lobster Studies 157

    . - ,.- - . .                        -      - - . _ _ - - . . ,     . - _ _ . _ . . -            - ~ - , . -                  -.      .      - - - - _ -           .

1 a i Results and Discussion TABLE 1. Mean monthly surface and bottom water j temperatures (*C) measured at each station durmg 2-unit (1979-85), Sunit (19s6 95) and 1996 studies. WaterTemperature j Surface Bottom Mean monthly surface and bottom water temperatures measured at each station during 2-unit, 2-Unit Sunit 1996 2 Unit SUnit 1996 I

3-unit. and 1996 studies are presented in Table 1.

iordan cove 4 Water : .ws-w during the 1996 lobster study were among the colden observed since the Mudy MAY 10.2 12.0 9.4 9.2 9.8 8.9

                                                                                                                                                                          )
                                                                                                                $$ $ l'j began in 1978. Overall mean surface and bottom water temperatures from May through October in                     fu[                               7          3$           f,$    lh)     I Auc           21.2      223     19.2         19.9        20.1    is.s

! 1996 were consistently lower than overall means i SEP 203 21.6 18.8 19.2 193 1RJ reported for each station during previous 2- and 3 ocr 16.8 17.6 15.8 16.0 16.2 15.8

unit studies Average May October surface water 1 - -

during 1996 ranged from 15.8"C at merall 173 18.6 u.0 n.1 us us 4 Twotree to 16.l*C at intake, which was lower than l the ranges reported in previous 2-unit (16.317.3*C) inlais I and 3-unit studies (16.4-18.6*C; Table 1). Bottom MAY 10.1 10.9 9.7 93 10.0 9.0 I[ } j water temperatures were also below average from May to October 1996 (range of overall averages Auc l'$ 20.7 l( 21.1 IN 19.2 I'j

                                                                                                                                               ,          ls' 15.6-15.8*C), when compared to previous 2-unit                                                                  20.1        203     19.0     I SEP          19.8      20.1    18.9         19A         19.5    18.8 (15.9-16.3*C) and 3-unit studies (16.016.6*C).                     ocT            16.1      16.7   15.8         15.9        163     15.8
                                           . at                       ns, & mo Q

mean surface and bouom water temperature values Overall 16.9 17A 16.1 163 16.6 1sJ for 1996 were lower than the monthly means j Twotree reported during the period of 2-unit and 3-unit '

operadon. On average, surface water temperatures MAY 9A 9A 9.0 8.9 9.4 87 j i in previous 2- and 3-unit caudies reached 20.0*C or more at each station in August, however in 19% the [ AUG l4j, 20.0 l,4j 20.2 3j' [7 18.9 l$ l'j 19.6 7 19.8 I t monthly mean surface water temperature reached 18.7 SEP 193 193 18.9 19.1 19.1 18E [ only 19.2*C at Jordan Cove and Intake and 18.9*C at ocT 16.0 16.1 16.0 15.9 16.0 16.0 ] Twotrec Historically, surface water temperatures were man 163 16A u.s u.9 164 15.6 j highest at the stations closest to the discharge (Jordan Cmc and Intake) and were slightly warmer during 3-unit studies when compared to 2-unit natural coohng and hting; meraH Mace wam studies. No MNPS Units were in operation during temPeratums from W to Octh wem smHar i the 19% lobster study and, as a result, the monthly during 19%. 2. and 3-unit studies (15.8,16.3 and and metall surface water temperatures at Jordan 16.4*C, respectively). Field temperature data from Cmt and Intake were lower than those observed thg lohnu mudy are consiment with usults of 4 previously. At Jordan Cove, surface water 1 U"""*"".1 mudh which %, cated that a 2.2 T

                           'a.ww.s averaged 17.3 and 18.6*C from May to                      isotherm resuhing from 3-umt operation could i

October during previous 2- and 3 unit studies, exand into the Jadan Cme ama dudng an c2 dde. , respectively; in 1996 surface water temperature At 600 m from the discharge, a 0.8,C isotherm I averaged 16.0*C (Table 1). At Intake, previous 2- extends to a depth of 3 to 5 m (NUSCO 1988c), and 3-unit surface temperatures averaged 16.9 and c mach & Wn at some pot 4rawls m, , ! 17.4*C, respectively, while those during 19 % the adan Cme and Me stes. i averaged 16.l*C from May to October. The Twotree station, located 1.5 km offshore, is not influenced by the MNPS thermal plume and is less subject to ,i 158 Monitoring Studies,1996 i ) 4 e

                  - - . -                                   ~       -                     .-                                                               -.

   ~. . . .                . . .       . - .        -          .     . - - - . .                       ......~ ~.       ,       - . - ~ .                       . _ ~ - -

1 .i j TABLE 2. Cats antatscs orlobeers caught in wire' pots from 1978 to 1996. i Totalmanher Nanber pots Geomeinc 959. C.I. 64neen legal CPLT' 95'. C.I. caught hauled mean totalCPUE E 81.0 2 81.8 2 82.6 2 81.0 mm 11 1978 1824 1026 1.600 1.454 1.761 g,,122 0.118 0.096 0.144-0.202

!                           1979                3259         2051                   1.404 4                                                                                                  1.302 1.513    3,.123 0.101 0.079             0.107 0.148 1980                2856         2116                   1.103 i                           1981 0.997 1.221    gjg* 0.076 0.063               0.092 0.126 2236         2187                  0.904 i

0.839 0.974 9 3 0.079 0.069 0.083 0.II3 1982 9109 4340 2.006 j- 1983 6376 1.925 2.089 gJig 0.126 0.106 0.144-0.186 4285 1.331 1984 7587 1.250 1.418 gJR 0.109 0.093 0.128-0.168 4550 1.607 j 1985 7014 1.540 1.677 gJg 0.120 0.104 0.140 0.I79 4467 1.352 1.252 1.460

3J33 0080 0.068 0.090 0.I20 1986 7211 4243 1.585 I.501 1.673 1987 7280 4233 3 3 0.060 0.049 0.074 0.097 1.633 1.562 1.707 gg 0.054 0.046 1988 8871 4367 0.070 0.089 1.929 1.846 2.015 gg 0.052 0.047 1989 7950 4314 0.068 0.091 i 1.729 1,645 1.817 0.112 gg 0.053 1990 7106 4350 0.097- 0.126 1 1.531 1.455 1.610 1991 7597 4404 1,542 0.161 0.102 9 07f 0.143 -0.179 1.437 1.654 0.183 0.117 9 3 0.159-0.206

, 1992 ll438 4427 2.457 1993 2.352 2.565 0.208 0.ll4 g33 0.186 0.229 10195 4194 2.301 1994 2.198 2.408 0.197 0.111 gag 0.175 0.220 9849 4256 2.199 1995 2.104 2.298 0.200 0.108 g3.71 0.178 0.223 6435 4317 1.261 1996 1.152 1.380 0.180 0.106 3 3 0.158 0.202 7531 4249 1.587 1.466 1.718 0.152 0.091 gigg O.!34 0.170 2-Unit (1978-85) 40261 25022 l.364 i 1.337 1.403 0.134 0.100 0.085 0.127 0.141 3-Unit (1986-96) 91463 47354 1.762 1.728 1.796 0.148 0.089 0.068 0.143 0.154

                ' 10 fromwire       pots used May simough     rwh= at     and eastson from August through Odober 1978, and from May through October 1979 81; 20 wire 1982-96.

l

                ' 7he    - legni eias from 1978 to 1988 was 81.0 nun (3 % in), nunsmum legal sue was increased in 1989 to 81.8 m to 82.6 nun (3 % in).

l Abundance and Catch-per-Unit-Effort mm in 1990 (A-mean range =0.0710.091; Table 2). j legal CPUE in 1996 was also less than the values The total number of lobsters caught during 1996 during other years of 3-unit operation, when the j was 7,531, which was within the range of previous legal size wu 81.0 mm (1986-88 A mean j  ! 3-unit studies (6,435-11,438); this total was also range =0.079-0.086), but was higher than the 1989 i within the range of 2-unit studies, when 20 wire mean (0.065) when legal size was 81.8 mm. The j traps were used at each site (6,376-9.109; Table 2). ) majority of the annual legal CPUE values during 3-The geometric mean total CPUE for 1996 of 1.587 l unit studies, including 19%. were lower than any j lobsters / pot was also within the range of presious 3- legal CPUE reported in 2-unit studies (1978-85 A-i unit (1.2612.457) and 2-unit studies (0.904-2.006). mean range =0.098-0.173). Legal catches have l Relative to the 1995 lobster catch, which was the steadily decreased since 1978, showing a significant lowest observed in more tien 10 years, the number t declining trend (slope =-0.004, p=0.001). The of lobsters caught and total CPUE increased in 19% decline in legal CPUE is most likely due to the

            . and reversed a trend of declining catches observed fourfold increase in fishing effort since the 1970s                         '

! since 1992. Although total CPUE during 3-unit (ASMFC 1996) and. more recently, to the increases cudies was higher (1.762 lobsters / pot) than during in minimum legal size in 1989 and 1990. Our i 2-unit studies (1.364), no significant trends were )

studies have show7

that the magnitude of legal identified in the time series of total CPUE data catches in any year is highly dependent on the i collected since 1978.

abundance of sublegal-sized lobsters the year before: l The A-mean CPUE of legal-size lobsters (2 82.6 more than 90% oflobsters in the legal size-class had mm) was 0.067 in 1996, which was the lowest value recently molted from the sublegal size-class. Since

 ,           reported since the legal size was mercased to 82.6
;                                                                                            total CPUE in 1995 was the lowest observed in more j                                                                                                                                      Lobster Studies         159
)
                                                                     -                                                        . = , -                       -          .-

. _ _ _ _ . - - ~ 3 ,, o so s oo s o m covt 0.880-1.908). At Jordan Cove, CPUE during 19%

                        -   2"                                                                  was higher than the range of CPUEs reported during o.o E          2-unit studies (0.153 1.402), whereas at Twotree and
                        - Joe,                                                   on$           Intake,1996 total CPUEs were within the range of o=y E'"

y ,' " p on; ato E 2-unit studies. No significant trends nre identified in the time-series of total CPUE data at the

                        & ion                                                                  nearshore Jordan Cove and intake stations, although N

od at Twotree total CPUE significantly increased from h" on

                                      }. I.         I f 1.J } -} .} .} .} .p .) *n'a j     1978 to 1996 (slope =0.062 p=0.011).
                           ""                                                   ""               During 19%, legal CPUE (lobster 2 82.6 mm) was n==...m..n..,              n ..ruuna                         highest at Twotree and Jordan Cove (0.076 and us                                                     aw no                       "*C                                        0.073, respectively) and lowest at intake (0.052; Fig.

2 2). Legal CPUE at Twotree during 19% was the

                                                                                  . .. E      lowest value reported in 3-unit studies (presious an $        range 0.082 0.133), while legal CPUE values for
                       ~ 2m E

E'" \ i om$ 1996 at Jordan Cove and intake were within the y'" an range of previous 3-unit studies. Since 1978, legal and catches have significantly declined at each station 5 tan

                                                                            ]    o is T       (Jordan Cove slope =-0.002, p=0.013; Intake slope =-

[= g$75.L [.lg.l y H 1.p.gi-1.t.).g o io l oot 0.003, p=0.005; Twotree slope =-0.006, p=0.002). om The monthiy patierns of total and legal lobster o ao u n =...m uesp.va neo. . u w.s" almadaa~ during 2-unit. 3-unit and 19% studies are

                          '"                                                                  presented in Figure 3. Total lobster CPUE was twovatt
                       'in                                                      [             generally highest in June or July, and lowest in 2*                                                        'e         October during previous 2- and 3-unit studies. In a["   ,
                                         ]
                                           \        J-
                                                                                   "I        contrast, the 19% total CPUE was highest in pin v'"

y j'}l j ll,J[ October (1.871) and lowest in July (1.299; Fig. 3). Values for total CPUE during the summer of 1996 2 (June, July, and August) were among the lowest

                     $i' ,"       A  n 5 an                                                                    observed since the study began in 1978. Legal-size lobster CPUE peaked in July 19%, which was the W **

h'd.h}.}T}1,,,,,,1 same month that legal catches peaked during

                                                                                ,,          combined 2- and 3-unit study years. The reason for n n e..e m uesuo m a.o.i.ru u n

• the shiA in the pattern of monthly total CPUE is Fig. 2. Mean total CPUE (geonwtric mean
• 95% C.I.) and unclear, although it may be related to cooler water

(([g$N ""

                                                                         ,'% [,,[           temperatures observed during 19%.                  The catchability oflobsters is directly influenced by water from 81.0 nun to 81E mm in 1989 and to 82.6 mm in 1990).                      temperature. When water temperature rises above than 10 years, fewer legal-size lobsters were                                  10*C, lobster activity (e.g., feeding movement, and expected to be caught in 1996, because fewer                                  molting) increases (McLeese and Wilder 1958; Dow sublegal-sized lobsters were available to molt to legal 1966,1%9,1976; Flowers and Sails 1972; NUSCO size.                                                                          19%).        Cooler sea temperatures during 19%,

Annual CPUE values for all sizes of lobster and relative to previous studies (Table 1), may be legal-size lobster are presented for each station in responsible for the lower summer catches and Figure 2. The highest total CPUE during 19% delayed peak in total lobster abundance observed in 1996. occurred at Jordan Cove (2.036), followed by Besides lobsters, pots oRen catch other organisms, Twotree (1.830) and intake (1.073). Total CPUE values during 1996 at each of the three stations were which have been shown to influence lobster CPUE in within the range of previous 3-unit studies (Jordan previous years (NUSCO 1987a,19%). Incidental Cove 1.289 2.642; Twotree 1.627-2.957; Intake catches of all species at each station were used as 160 Monitoring Studies,1996 I

TABLE 3. Total number of lobsers and incidental catch or other 3 00 species caushe in traps. Range (19:4-1995) 1996 2'2', - w '"

                                "-                                           I'bster                    7014-1143s                  753I g2M                                    - .-                                 Rock, Jonsh crab              79-2033'                 s48' e

j135 e Spider crab 1344 314:0* 24572' .so ,.o '

                                         \.             .. - .

Hernut crab 217-72]* 306 {' e Blue crab 21 14: 22 O "' 3'" 2 039 s\ Wumer flounder sununer nounder Skates

-45' 4 40*

12 14 14 54 30 om Ovster toadfah 7 76 s Scup 27-238 c.25 21 Cunner 4s-207 41 0M Tautog 39-250* 167 WAY JUN JUL AUG $[P Sea reven OCT 0-20 0 Whelks 27 17s' 37 0# (*) Covanance analysis identified these catches as sassificant factors afracting lobster CPUE (p<0.05). o yo h Pot catches of tautog increased substantially from I 8 / \ 1995 (n=48) to 19% (n=167), particularly at Jordan 1 O O 'S f

                             !             \                            Cove, where 117 tautog were caught in traps during                                       I 1996.

f '/

                              ~~~~"
                                               'N                                    Most were caught in August (n=25)-

September (n=35), and October (n=52). Although 3 o io

's, N covariance analysis did not identify this species as d . - having a significant influence on the catcha' u ility of

" ' ' , lobsters, the presence of tautog in traps containing lobsters had a profound effect on lobster survival at noo- Jordan Cove from August to October. The

                           ^"      #"

u,'w------3-o- " 5"'

• percentage oflobsters that were dead or not tagged

                                                 .a    a is96          due to injuries and damages was 16,12. and 37% at                                         I g g gg, rc 3. Monthly mean total CPUE and Mosen legal CPUE sharing 2-unit (1979-es). 3 unit studies (19:6-96) and dunng l996.

respectively. Typically, in any given year, the percentage of lobsters that are not tagged due to covariates to identify species which significantly injuries or damage is highest ( but still<5%) during the June-July molt when animals are soft-shelled. (p<0.05) influenced lobster catch (Table 3). During 1996, lobster CPUE was influenced by catches of Tautog are known predators oflobsters (Bigelow and spider crabs at all stations; the total spider crab catch Schroeder 1953; Auster 1985; Cobb et al.1986) and of 24,572 was the second largest observed in the the reason for the increased abundance in Jordan studies (range 1,344 31,480). Rock crabs were also Ccvc during 19% is unclear, although a large year. class of age 2 fish was observed in trawl catches in found to influence the catch of lobsters at Twotree during 1996 These species have had a significant 1994 95. The aMh of mussels, a common food influence on lobster CPUE in previous study years- item of tautog, in the discharge area was greatly The reason for the substantial increase in spider crab reduced during 19% due to MNPS shutdown. The abundance over the past 6 years (>10,000/yr) is reduced availability of prey for tautog may have unknown, although similar increases in spider crab forced fish to search for food in nearby Jordan Cove. catches have been observed in the trawl monitoring and could have been responsible for the increased program The incidental catches of rock crabs and predation observed by tautog on lobsters in Jordan spider crabs have been reported to significantly affect Cove during 1996. Furthermore, with no thermal lobster catch in other studies (Richards et al.1983; discharge entering Jordan Cove due to the shutdown Richards and Cobb 1987). of MNPS during 1996, temperature conditions were favorable for tautog to inhabit the rocky outcrop where pots are set in Jordan Cove. Lobster Studies 161

   - - - . . . ~ . . .                  .-         .- - - ~ ~ ~ . -                  .       - ~ - .               - . - ~ . . . - ,                      . - -

l. i TABLE 4. Summary of lobster carapace length statistics in wire pot catches from May through October,1978-1996. N. Carapacelength (mm) Percentage of Range Meant 95% C1 1 legal sizes

• l 2 81.0 t 81.8 2 82.6 i i

s 1978 1508 53 111 i 1979 71.42033 LE 5.9 4.8 2846 44-100 } 71.2 t 0.26 L6 6.6 5.1 1980 2531 40-% j 70.7 t 0.27 M 5.0 4.1 1981 1983 43-% j 71.0 t 033 M 7.6 6.6 1982 7835 45 103 4 1983 70.820.15 M 5.7 4.7 ' 5432 40 121 1984 71.7 0.19 M 7.4 - 63 6156 45 107 1985 5723 38-101 71.8 t 0.18 M 73 ~ 6.4 1986 713

• 0.17 M 5.1 43 5961 36 107

1987 70.1 t 0.17 M 3.6 3.0 5924 36-99 j 1988 70.2 0.17 ,M 3.2 2.7 7145 21-97

1989 69.520.16 M 2.6 23 6715 34 107 1990 69.920.17 4.5 M 2.9 6040 36 102 70.2

• 0.20 1991 6449 31-101 70.2 0.20 7.9 5.9 M 4

1992 9594 20-103 70.1 t 0.15 8.5 6.5 M i 1993 8487 30 102 70.8 0.15 6.4 43 M 1994 7841 34-100 703 0.17 6.7 4.6 M

1995 5472 - 37-101 7t 4.7 M i 71.9

• 0.20 5 1996 6634 16 % ~10.01 0.5 10.0 73 M 7.1 5.0 M 2-Unit (1975.a5) 34314 38 121 j 3.Unu (1985-%) 713 0.07 7.5 63 53 792ti2 16-107 703
• 0.05 i 63 4.6 - 3.5 '

• Receptures cot included.

!

• andThe minumura in 1990, to 82.6 mm legal (3 8/4 stae in). from 197E to 1988 was 81.0 mm (38/u in), minimum legal size was increased i

? Population Characteristics since 1978 (slope = 0.275, p=0.002), which can be q g 77 9 q , attributed to increases in minimum legal size and to ! increased fishing effort, which has more than mean carapace hgth (CI.) of all h doubled since 1978 (Blake 1991; NMFS 1993). during 1996 was 70.0 men, which was within the When the three stations were compared. the mean range of values reported in previous 3-unit studies CL of lobsters caught during 2996 was largest at + (69.5-71.9 mm), but smaller when compared to Twotree (70.9 mm) and smallest at Jordan Cove

mean CLs reported during 2-unit studies (70.7 71.8 (68.8 mm; Table 5). The value at Jordan Cove was

mm; Table 4). The oversll mean CL during 3-unit h sMen pd in our lobner audies (pmious

studies was snialler (70.3 mm) than during 2-unit range 69.0-71.3). while those at Intake (70.3 mm) studies (71.3 inm). The percentage of legal-size and Twottee (70.9 mm) were within the range of 3-

lobsters (2 82.6 mm) during 1996 was 3.8%. which un t studies (68.9-71.5 mm and 70.0-72.5 mm,

was within the range reported in previous 3-unit respectively). Mean sizes at all stations d ring 1996 q

studies (3.1-5.7%; Table 4) when the legal sizes were smaller than the range of mean sizes reported were 2 82.6 mm, llowewr, due to the increases in in 2-unit studies. In contrast to previous study years, j minimum legal size in 1989 and 1990, the when Twotree yicided the highest percentage of percentage of legal size lobsters during 1996 was legal-size lobsters, the percentage during 1996 was lower than the range reported in 2-unit studies when higher at intake (4.2%), followed by Twotree (3.8%) , legal size was 2 81.0 (5.9-9.1%). The percentages of and Jordan Cove (3.5%). Percentages of legal-size

legal lobsters in our catch have significantly declined at latake and Jordan Cove were within the range of

 <                                                                                     both 2.and 3-unit studies (Table 5). At Twottee the 3

162 Monitoring Stadies,1996 l

   . -.~ -                --          .          , ,     . - . _ - . - .              .                       - ~..          . . - - . - . -              ... . _ .            . - . - - -

i l TABLE 3. Sianunary oflaineer carapear leasth -= in wire paa TABLE 6. Female to male seu rauosa of lobsters caught in

]               coedas at each stessan frian May thraush Ocamber, dunng 2-Usut 4                                                                                                        wtre pots from May through October,19781996.

(1978-19s5). 3-Unit (1986-1995) and 1996 studies s Mean carapace Personnase of 4 lordan intake Twotree All lanath(nunf leaals (2s2 6 nun) Cove Stations 1 JORDAN COVE 2-Usus range 69.s 71.1 2.55.9 3-Unst range 69.0 713 1978 0.79 0.97 1.02 0.92 2.15.1 nuan 1979 0.68 0.83 1.15 0.82 6ts 33 1980 0.66 0.90 1.15 0.a8 IFTME 1981 0.70 0.71 1.19 0.86 2-Usut rans: 69.2 71.s 2.95.7 1982 0.62 0.66 1.09 0.79

.I              3-Unit range                68.9 71.5                     1.s - 5.g                          1983            0.72              0.67       1.25         0E7 1996w:c -                   703                           4.2                              1984            0.60              0.71       1.22         0.82 1985            0.70              0.67       138          0.97 TWOTkFS.                                                                                     1986 I                                                                                                                             0.65              0.73       1.26         0.87 2-Usus range               713 73.7                      4.4 10.4                            1987 4

3-Unst range 0.71 0.63 1.24 0.88 i 70.0 72.5 2.66.2 1988 0.68 0.72 1.15- 0.85 1996 nimen 70.9 3.s a 1989 0.64 0.65 1.08 0.79 1990 0.60 0.65 0.90 0.71 5 ***P'""" 1991 0.51 0.57 1.13 0.74 1992 0.43 0.47 1.45 0.73 1993 0.47 0.59 1.59 0.84 percentage oflegal-sia lobsters caught during 1996 1994 0.5 i 0.67 1.24 0.79

was within the range of 3-unit studies, but lower 1995 0.53 0.61 0.93 0.71 than the range reported in 2-unit studies. 1996 032 037 a79 a48 a

i SexRatias 2-Unit 78-85 0.67 0.72 1.21 0 86 3 Unit 86-% 0.53 0.60 1.17 0.76 l The sex ratio oflobsters collected during 1996 was

,           0.48 females per male, the lowest ratio ever observed                                       .R capturce not eded.

j in these studies (1978 95 range 0.71-0.97; Table 6).  ;

Female to male sex ratios at each of the three More recently, sex ratios of lobsters caught in  !

i stations were also the lowest observed in 19 years of eastern L1S commercial traps were higher, ranging i study. Twotrei., which typically has yicided see between 2.61 and 6.29 females per male (Blake

females than males, had only 0.79 female per male, 1988). Throughout the range of lobster. many researchus have reponed sex ratios close to 1

1 for

, compared to the range of 0.90 to 1.59 reported  ! i previously. Even lower ratios were found at Jo dan sublegal (< 81.0 mm CL) populations of lobsters  ! s Cove (0.32) and latake (0.37) during 1996 and each (Hemck 1911; Templeman 1936; Ennis 1971.1974;  ! of these values were well below the range reponed Stewart 1972; Krouse 1973; Thomas 1973; Cooper } et al.1975; Briggs and Mushacke 1980). since 1978 (0.43 0.79 and 0.47 0.97, respectively). I The occurrence of more females at Twotree than at other stations has been consistent since 1975 (Keser g,ggj,cfjgy i i stal.1983). The overall female to male sex ratio ) l 1 during 3-unit studies (0.76) was lower than during 2 nera mahods wm used to dadne the su at unit studies (0.86) and has significantly declined

• I*"*3*' "*'**#"**"'*'

7 " 'I since 1978 (slope = 0.01, p4.002); significant "' #*O" # "*N '.' declines were also noted at Jordan Cove (slope =- Presence of external eggs (berried). The smallest O.016, p4.001) and intake (slope = 0.019, p4.001). m in was 60 mm aM The cause for the decline in female to male sex r nat on of the sin diarhution oNmed ratios is unclear, although it may be related to males indicatM that Me of the berried females

• increased fishing. In the mid-1970s, Smith (1977) were smaller than 76 and 78 mm during 2- and 3-reported female to male sex ratios in the LIS und 8tudies. TMPectively (Mg. 4.). Anoser nammarcial fishery ranging between 1.06 and 1.81. le ur, Ant dacnbed by Templeman (1935) is

? Lobster Studies 163

_ _ . _ . __ _ __ _. . . __. _ _... _ _ _ _ __ _ _ .. _ _ _ _ _ __ m..._. u o

                                                                     ' ,,                                             in 2-unit (62 mm) and 3-unit studies (60 mm) oe supponed the results of the morphometric relationship between the abdominal width and c

q n.x carapace length. These individuals were between f n' 50-55 mm CL when miposition first occurred g"c5 a-o o (assuming 14% growth per molt). Briggs and

                                  "                                                                                  Mushacke (1979), using the same morphometric
                                                        /

technique, found that females in western LIS begin j to mature at 60 mm CL and most are mature at about 80 mm CL. The New York State Department of ss so es to vs so es so es ion ,os no m no ca-act uncw--' Emironmental Conservation has found berried females as small as 56 mm CL in their monitoring Fis. 4. Proponian ors =rn.d r. mala si veriouniza con.co.d ening studies of the LIS lobster fishery (K. Graulich. 2 una (-) and 34nus muen t. -). NYDEC, pers. comm.). Blake (1994) used a technique desenbed by Aiken and Waddy (1982) to g o. estimate sexual matunty of females in LIS and found l i that half of the females in LIS could extrude eggs at yem S

                                                                . - !Q
                                                                          ...V'
                                                                                           .bP."                    about 73 mm CL in contrast to LIS. female lobsters in the Gulf of Maine seldom become sexually mature f"t g "'
                                          'Y" " ~~~~ . .". /                                ,

at less than 81 mm CL, and only a small percentage r

                                      ~

I#y . are mature between 81 and 90 mm CL'(Krouse 1973). Earlier maturation of females in LIS than in the Gulf of Maine was attributed to warmer LIS a ., water temperatures (Smith 1977; Aiken and Waddy

                                      "     "                                                           .           1980).
                                                                                                'a     "
                                                       "c.,lcc unc'im ,y                                               The percentage of berried females collected during 19% was 8.6%. which was within the range of                ,

previous 3-unit studies (3.812.2%) and higher than 2-unit y-l.2s43.13'lo4*+(4.40*lodd.(1.e5'10 9. r' .30 4 the range during 2-unit studies (3.1-6.2%; Table 7). 4 3-unit y-i.osq2 47 10 p+(3 :s'iodg'.<t.77'to*W, r' .32 Since 1978, the percentage of berried females has ris. 5. Mc 9 relasiandup ben n ew .hdanunal wish so been hishest at Twottee, during 19%,13.3% of the  ! carapace longih reiio (y) and the carapeor lengsh (s) for f. male females Collected at Twotree were berried, followed nahman durins 2 mui(~)and unis muh (. -j and emas 15 I ** by Jordan Cove (4.3%) and Intake (3.3%). The  ; percentages at each of the three stations during 19% based on abdominal width measurements of females, Stre n t c un8C Pmvias 3dt sides, hn which markedly increases during maturation. 8 IC mn888 NPoned during 24mt Calculating the abdominal width to CL ratio and ' * **'*'**" comparing it to CL prmides an index of female six t during 3d operada (8.2Y.) than they at sexual maturity (Skud and Perkins 1%9; Krouse wem ng i opemum (4.Wh h mean C1, 1973). Mean ratios of abdominal width to carapace ""**" "8 *** length were calculated for each 5-mm CL and mm, which was within the range of means reported plotted against the carapace length of lobsters in Presious 3-unit studies (75.3 78.1 mm) but s er collected during 2-unit (1981-85) and 3-unit (1985- un8e of mean I-s NPond kring

                                                                                                                    -u          es             un).         auge su d

96) operation and for 19% alone (Fig. 5). Dunng .

1996, females began to mature at about 55 mm CL, ne males was lower during 3-unit studies and all females larger than 90 mm CL were mature. (76.4 mm) than during 2-unit studies (79.4 mm) and The close correspondence between the 2- and 3-unit 2 Pmpoe d segaMad curves in Figure 5 indicates that female sim at rned Wes muected snce IN Only W. M sexual matunty was similar during both operational the berried females collected during 3 unit studies penods The small sim dberned females collected were above the legal sim compared to 32% during 2-unit studies. The reduction in percentage of bern,ed 164 Monitoring Studies,1996 2

   - .. -              -       . _ . - . . . - - _ _ - _ _ _ -                        . _                              - .= .- -          . - -            . . . . - .

t 4 , k i 4 TABIE 7, Percentage of berrled footales caught at each station and annual carapace length statistkm from 1978-%. 4-Perc-" of %.. IF ' r.,-c .eh imm) i All Jordan intake Twotree N' Range Mean

• 95% C.I. Percent sublega?

stat 6ans Cove l, -

1978 3.4 1.4 2.6 53 58 74 88 a0.1 t 1.04 73 1979 3.1 1.9 2.7  ;

7.2 70 64 93 80.521.28 59 19a0 33 3.5 1.8 5.6 71 66 93 79.1 t 1.27 19st 70 4.2 1.6 2.7 7.1 82 69 97 j 1982 81.2*135 55 3.1 - 0.8 0.9 6.1 108 64 99 80.0 t 1.08 60 1983 4.7 2.1 3.2 8.5 123 66 103 80.5tl.04 63 ) 1984 6.2 3.6 3.5 10.6 173 62 95 79.1

• 0.87 69 l

;               1985         6.2                    3.5              4.5      8.5        1 71        63 94              77.0 t 0A1               82 j                1986        4A                      3.0              23       8.0                                                                                      l 135         65 94              78.020.95                77 1987        5.7                     3.2              1.9      9.6 i                                                                                         158         62 90              76.5 t 0.67              92 1988        3A                      2.4              1.9      6.4        124         63 90 1989                                                                                                     76.9 t 0.82              89 i

5.4 2A 33 8.2 161 65-98 773 0.78 85 i 1990 6.6 2.7 4.0 11.2 165 65 -102 78.1t0A2 87 1991 8.2 3.2 1.5 13.5 226 62. % 75.0 0.75 82  ! 1992 12.1 3.4 1.7 193 491 60 93 753 2 0,44 94 1993 12.2 3.1 2.7

• 19.4 476 62 90 75.620.43 93 1994 10.8 6.1 4.7 16.9 372 61 - 91 75.920.52 93

1995 9.6 5.9 5.9 13.4 218 64 91 i 1996 8.6 763 t 0.61 94 43 33 133 185 63. 91 76.6

• 0.78 89

                                                                                                                                                                       ~

2. Unit 71085 43 2.0

• 2.2 7.1 856 62-103 79.4 a 039 68 3-Unit 86-% - 8.2 3.6 2.9 133 2711 60 102 76.4 s 0.19 91 4 j
• Receptures not included.

6 'Ihe minimum legal size from 1978 to 1908 was 81.0 oun (38/  ; 6 in), minimum legal size was increased in 1989 to SI A mm (3 7/m in), j and in 1990 to 82.6 own (3 8/4 in). j females above the legal size is due primarily to the i spring to early summer (i.e., the end of May to early

. increases in minimum legal size in 1989 and 1990, July).

4 In previous years, when sampling was although the high rates of fishing in LIS remove conducted through November, a second peak in the i most females shortly aAer they reach legal size or catch of molting lobsters was obsened in autumn j after berned females release eggs. The apparent ' (Keser et al.1983). The frequency and timing of stability of the LIS lobster population, despite lobster molts were examined using the Gompertz current high exploitation rates, is due to females that growth funcuon fitted to cumulative percent-molt become mature and spawn at sizes well below the I data for 2- and 3-unit studies and during 1996 (Fig. legal size, thereby providing a buffer against 6). The inflection points of the growth curves were i recruitment failure (Graulich 1996). However, the  ! used to estimate annual dates of peak molting. fecundity of the stock may be lower, as a result of Annual molting peaks were significantly (p<0.05) , relying on smaller berned females to sustain . correlated with mean May bottom water recruitment, which could affect the long-term health of the LIS fishery, temperatures, suggesting that molting occurred earlier in the years when May water temperatures were warmer than average. Conversely, peaks i MoltirqgandGrowth i occurred later when water temperatures were colder than average. The average bottom water i inhawr growth is a function of size and weight temperature dunng May of 2-unit studies was 9.2'C

increment per molt and molt is cy, water and molting peaked on 27 June (Fig. 6), which was a

' temperature is one of the most important factors that week later than the date of peak molting during 3-regulates these . processes (Aiken 1980). The unit studies (20 June) when the average bottom

mayority of molting lobsters were caught from late water temporaice was warmer (9.7'C). During i

Lobster Studies 165 1 i i

1%2; Mauchline 1976). Regression plots and 1 [* g* , parameter estimates of growth for males and females caught during 1996 and in 2-unit and 3-unit studies m 2-.,ao > are presented in Figure 6. During 1996, average E Us"Yo o7l growth increments of both sexes were less than the ' I, /#[ M w\ average increments reported in previous 2- and 3- (( b E,

                       /

4. unit studies. Males grew an average of 6.6 mm per 3 N*i molt during 1996, compared to 8.9 and 8.6 mm 5 p k during 2- and 3-unit studies, respectively. Average hi  %% female growth of 7.7 mm was slightly higher than a

males during 19%, but below the average increment

                                         % '"          5"'     

of 8.7 mm reported during both 2. and 3-unit Fig. 6. Molt frequency curves based on the Compertz studies. The mean growth increment a( Twottee of function fitted to data on lobsters caught dunng 2-unit (1978- 8.1 mm during 1996 was within the range of values 85;-) and 3-unit studno t1986.%; - . ) and dunng 1996 ( o reported in previous 2- and 3-unit studies (8.011.1

         . .),

mm and 7.2-10.2 mm. respecuvely; Table 8).

               .w           **                                              However,1996 growth increments at Jordan Cove (6.2 mm) and Intake (5.6 mm) were the lowest y[

g .w

                    \        w ',* ,,,,

lfa

                                                            "               observed since the study began (previous range 7.3-
               ~                                                            9.1 mm and 7.0-9.5 mm, respectively).            The r "
          .                    '"' N        .'=

below-average growth increments were most likely

          ) rw                   -- Q"" " *

related to the cooler seawater temperatures observed 3 ,',*" throughout the 19% study period. The average

                                              "',,,;,4   >=

percent growth of males and females during 3-unit y n. * '"2

                                                                  ,;,       studies of 12.9% and 13.4%. respectively, was
         !a-
         "   "~

u u o ., '.'"'N ., N similar to that reported by other researchers in LIS. i ut.u uu -na nunnuun < c) which ranged from 11.6% to 15.8% for males and between 12.0% and 15.4% for females (Stewart 1972; Briggs and Mushacke 1984; Blake 1994).  ; Fig. 7. Relationehtp between the date of peak molting of Although growth increments were lower during nabater. (paranweer a from the compertz function) and j 1996, average meremental growth during 3-utut annuai nwr.n bottom water temperature during May. ] studies was similar to growth values reported for LIS ' lobsters and appears to be more related to natural 1996, molting peaked on 25 June when bottom water temperature during May averaged 8.9'C. The variability in seawater temperature than to power plant operation. earliest molting peaks (first and second weeks of June) occurred when bottom water temperatures during May averaged >10.0*C (Fig. 7). Templeman Culls (1936) found molting was delayed I week for enry ' 1*C reducuon in water temperature. Aiken and The percentage of culls, lobsters missing one or Waddy (1980) described the influence of varying both claws, uns 12.2% of the total catch during water temperature or. the molt cycle and found that 3996, which was within the range of values reported at 10*C lobsters quickly entered the premolt stage n previous 3-unit (9.812.2%) and 2-unit studies and progressed to ecdysis. (10.8 15.5 %: Table 9). Claw loss percentages

          . Growth per molt was determined from lobster                   during 1996 were lowest at Twottee (7.3%).

tagging studies by comparing CL measurements at nicrmedsate at Intake (14.9%) and highest at Jordan the time of tagging with those from "r=d Cove (15.0%). The 1996 percentages at Twottee lobsters I tear later. Simple linear regressions of and imake were within the range of previous 2- and pre-molt (tag-size) and post moit (recapture-size) 3-unit studies; however, the percentage at Jordan sizes best describe growth for the size range of Cove was higher than the range of other 3-unit years lobsters caught in our studies (Wilder 1953; Kurata (10.914.5%). most likely due to increased predation 166 Monitoring Studies.1996 l l

                            .-                                                        . - . . ~ . - . . ~ . ~ . . .

I a 4 b I WALES f' rEuALES

                                                                              ;P'

' g

                                                                                                                      }                                               ,'

f es- '

                                                                                                                      .,E  es.

e o

                                                                   /j'                                                w                                         .t
                                                                 /

4 g es. /, 2 .- v g as .

 ;               5 y n.

g

• t*

g u W **./..s

                                                    , ,,g.                                                           jn                                                                               i s

e .,. 9 A,y* .. 2 w e .s. .,.- w .> V k

• a a g ts<

ss 6 v as. as. as ss as n es e ios as u as w e- *

icn CARAPACE LENGTH AT TAGGWG (mm)

CARAPACE 1,[NGTH AT TAGT.ING (mm) MAtB FEMALE 4 i N Growth model R2 N Growth model R2  ! , 2-Unit 300 y=22.168+0.805(x) 0.70 587 y=12.678+0.942(x) 0.79 3-Unit 970 y=18.816+0.851(u) 0.71 1077 y=16371+0.883(x) 0.72 1996 70 y=14.662+0.882(s) 0.76 , 36 y=9.840+0.%7(x) 0.72 F6g. 8. Linear regnesions and parameter estimates of carapace lengths at tagging and recapture thnes of male and female lobsters c::erght during 2-unit studies (1978-85;-),3-unit studies (1986 95;---), and during 1996 (O). t y= size at recapture, u size at tagging (mun). l 1 TABLE s. Sumniary ofichster powth (in nun and as a percentase < initial sue) at endt sation in wire pot catches for the peri oceeber durins 2-Unit (197s-19:513-Unit (19s6-1995) and 1996 mh 6 2:L!Ril 2:L!Dil 122ft . =ans. < mm < =an < =an .f . N mesans (nun) (%) N- means(aun) (%) N Mean(nun) Percentase 4 a Jordan Cove 29-s3 7.3-9.0 10.6-13.s 4s-107 7.3-9.1 10.9 14.3 44 6.2 9.3 Intalte 21 55 7.s-9.5 12.0 14.2 22-72 7.0-9.$ 10.7-15.2 14 $.6 s.4 e j Tweerse 21 96 8.0-11.1 11.7 16.3 58-113 7.2 10.2 10615.3 4s t s.1 12.5 i 2 i by tautog (see Abundance and Catch-per-Unit Effort opening to allow escape of sublegal-sized lobsters,

section). Average claw-loss during combined 3-unit 2 and thereby reduces injury and mortality associated studies was lower (11.0%) than during the 2 unit with overcrowded pots (Landers and Blake 1985).

l study period (12.1%), and may have been related to '; Since 1984, reported claw-loss (11.0%) has been the implementation of the escape vent regulation in lower than losses reported before the implementation 1984. This regulation requires that pots contain an

of escape vents (12.7%). Pecci et al. (1978) reported Lobster Studies 167 F

W

I i I l i l TABLE 9. Parcamisse of cutis dobsiers nuning one w both clawe lobsters during 19% (6.7%) than in previous 3-unit causl8 * *ir* Pou during 197s-lM l (13.5-21.6%) and 2-unit studies (21.1-47.6%). In general, the number of tags returned by commercial Jorden intake Twoorw All lobstermen has declined during 3-unit studies due to cov, sian "' the implementation of the escape vent regulation in 1984. Because most of the tagged lobsters are 1978 21.5 14.7 9.s 15.5 sublegal, fewer Were retailled in Commercial traps lE 19s: b$ 13.4 16.7 lo 4 7.1 I!4 12.1 with escape vents. In contrast. the number of recaptures made in NUSCO traps during 3-unit 1982 13.9 14.1 7.0 113 19s3 14 6 studies has increased. because they do not have 15.3 8.2 12.4 escape vents and retained more tagged sublegal i Ni !3$ $ ii lobsters. . The mean CL of lobsters recaptured in  ; 19s6 1o.9 14.7 6.s 10.6 NUSCO traps was 72.4 mm during 19%, which was

     !9s              $       13 ]        jj          lj           within the range of previous 3-unit studies (72.0-1989         14.4        14.3        s.s         12.2         75.0 mm), but lower than the range of 2-unit studies 1990         123         16.2        s.1         11.9 1991         14.5 (73.0-75.7 mm). Overall, the lobsters recaptured in 14.0        s.2         11.s NUSCO traps were smaller (73.0 mm) during 3-unit
     $!           l1$        '!

t $ l[ studies than during 2-unit studies (73.9 mm). In 1994 t i.s 11.5 6.9 9.s 1995 14.2 15.5 Contrast, Commercial jobstermen recaptured largeT 7.s 11.9 1996 15.0 14.9 7.3 12.2 lobsters during 3-unit studies (78.9 mm) than during i g , g ,p73 gggggg < lobsters recaptured in NUSCO and commercial traps l$$ ldj ,lj 3 7j 3j 12 were also related to the escape vent regulation. Prior to the regulation. commercial lobstermen recaptured many of the sublegal-sized tagged lobsters; currently, with the regulation in force, many of the sublegals  ; that trap related injuries were associated with water escaped from the vented commercial pots, but were temperature, fishing pressure (i.e., handling by retained in unvented NUSCO pots. In castern LIS. lobstermen), trap soaktime, and shell hardness. Of landers and Blake (1985) noted a substantial these factors, Krouse (1976) reported a positive reduction in the number of sublegal-sized lobsters correlation between fishing pressure and the rcained in vented pots, without a corresponding incidence of culls along the coast of Maine. Other decrease in the catch oflegal-sized lobsters benersts ofincorporating escape vents in lobster traps Since the tagging study began, commercial have been noted by many res: archers (Krouse and lobstermen have consistently caught a higher Thomas 1975; Fair and Estrella 1976; Krouse 1978; percentage of tagged legal-size lobsters than Pecci et al.1978; Fogarty and Borden 1980; Krouse NUSCO. During 1996. 24.8% of the recaptures in et al.1993). commercial pots were legal-size (2 82.6 mm) compared to only 4.2% in NUSCO pots (Table 10). Tagging Program The overall percentage of legal sized lobsters isymd in both commercial (24.7%) and NUSCO "Ihe total number of lobsters tagged during 19% (3.8%) pots was lower during 3-unit studies than (6.221) was within the range reported in previous 3- during 2-unit studies (27.9 and 11.0%. respectively). unit (5,307-9,126) and 2-unit studies when 20 wire The declines in percentage of legal-sized recaptures traps were tased at each site (1981-85,4.246-7,575; are attributed to the increase in minimum legal size Table 10). The percentage of lobsters recaptured in and to increased fashing effort, which has more than l NUSCO pots during 1996 was 14.4%: lower than the doubled since 1978. I range of values observed in other 3-unit studies (18.126.2%) and among the lowest values obsened  : during 2-unit studies (14.4-23.9%; Table 10).  ! Commercial lobstermen also recaptured fewer ' 168 Monitoring Studies,1996 I i l ____________i

1 1 j I TABLE 10. Lobster tag and recapture statistics in NUSCO pots (May-Oct.) and commercial pots Gan. Dec.) from 1978 to 1996. J NUSCO c. -al I  ! Number Number Percentage Percentage Mann Number Percentage Percentage Mean tagged recaptured recaptured legala CL(mm) recaptured recaptured legala CL(mm) 4 4 1978 2768 498 18.0 16.7 75.5 884 31.9 43.6 1979 3732 722 81.1 19.4 11.5 75.1 1776 47.6 27.2 77.6 1980 3634 522 14.4 18.8 757 1363 37.5 27.5 1981 4246 707 76.4 167 12.0 74.8 1484 35.0 l 1982 7575 1282 16.9 25.9 763 10.4 73.2 2518 33.2 1983 5160 23.0 75.5 932 18.1 11 3 73.6 I 2266 43.9 27.6 76.9 1984 5992 1431 23.9 8.4 73.0 1290 21.5 34 3 ! 1985 78.8 5609 1216 2t7 7.7 73.2 1189 2L1 2 1986 5797 293 78 3 1194 20.9 47 723 1188 20.4 27.5 1987 5680 1356 23.9 78.2 l 5.5 72.8 1167 20.4 1988 6836 1727 253 78.9  ; 253 43 72.0 1387 1989 6436 1235 20.2 267 78.0 l 19.2 4.4 (9.2) 72.9 1183 1990 18.4 20.7 (24.8) 78.2 5741 1066 18.6 5.5(127) 73 3 1007 17.5 793 s 1991 6136 1109 18.1 26.5(32.8) ' 7.4 (13.9) 73.4 1228 20.0 1992 9126 1842 20.2 33.9(41.5) 80.8 3.9(93) 72.4 1559 17.1 23.4 (28.6) 79.5 1993 8177 1708 20.9  ; 3.6 (8.8) 73.4 1768 21.6 ' 1994 7533 1974 27.4 (47.4) 79.4 26.2 3.1(93) 73.4 1020 1995 13.5 20.0 (28.6) 773 5307 963 18.1 l 5.4 (13.5) 75.0 1116 21.0 1996 6221 897 27.1 (34.1) 80.0 14.4 4.2(10.5) 72.4 419 l 6.7 24.8 (26.4) 79.0 l

3. Unit 7885 38716 7310 18.9 11.0 73.9 l

12770 33.0 27.9 77.1 3 Unit 86-96 72890 15071 207 3.8 (8.9) 73.0 13042 17.9 247 (32.1) 78.9 { o The minimum legal size from 1978 to 1988 was 81.0 mm (3 %. in), nurumum legal size was increased in 1989 to 81.8 cnd in 1990 to 82.6 mm (3 % in). Parenthetical values for percentage legal from 1989 to 1995 and for 3-unit studies (1986-95) corsepond to lobsters 2 81.0 mm carapace length. Movement predominance of localized movement is typical for nearshore coastal lobster populations and agrees with Results from the lobster tagging program were also results of other tagging studies conducted in castern used to assess lobster movement During 1996. most North America (Templeman 1940; Wilder and Gi the lobsters recaptured in NUSCO pots were Murray 1958; Wilder 1%3; Cooper 1970; Cooper et caught at the same station where they were released al.1975; Fogarty et al.1980; Krouse 1980, 1981; Oordan Cove %% intake 89% Twotree 94%). Of Campbell 1982; Ennis 1984). the lobsters that moved from the release sites, most Although the tag and recapture studies indicated moved from Intake to Jordan Cove (10%). This that most lobsters were nonmigratory and remained pattern of short-range movement was also observed in the local area, some lobsters made significant in previous 2 and 3-unit studies and ir, the recapture migrations. Most of the lobsters that moved farther information provided by commercial lobstermen than 5 km from MNPS during 2- and 3-unit studies Ninety-five percent of the tagged lobsters recaptured traveled to the southeast. Since 1978, only 18 tagged in commercial pots were caught within 5 km of lobsters were caught by commercial lobstermen in MNPS (Fig. 9). Stewart (1972) demonstrated a western LIS, e,...,,. dad to 807 caught in The Race strong homing behavior for the nearshore castern and 106 caught in Block Island Soudd (10 km and LIS lobster population Because lobsters are 25 km southeast of MNPS, respectively; Fig.10). terntorial and nocturmi, they have a limited home The large number of tagged lobsters caught by range; individuals leave their burrows at night and commercial kbstermen in The Race suggests that return to the same shelters before dawn. The Lobster Studies 169 I l l i

i l this deep water channel between Long Island and Block Island is a migration route for lobsters that NIANTIC f

                                   .a exit LIS. Once out of the sound, lobsters moved easterly and were recaptured along the Rhode Island
                                 """        \WATNORD
                                            =

coast (33). Buzzards Bay, Massachusetts (6), and in o- "L . waters near Martha's Vinyard and Nantucket Island i

                                                     )

(Fig.11). Only two individuals were reported being fhow f*= ' caught in nearshore waters along the south shore of

                            =

c' Long Island. compared to 28 lobsters which moved

                                     ., [8"                                        farther than 150 km to deeper offshore waters on the
      %*                                        '**[               "

s edge of the continental shelf, where they were caught ' o q ..

          ,,,       (,                           cmr s.*y                          in subrnatine canyons (Hudson. Block, Atlantis, and               ;

Veatch). Other researchers have observed similar exchanges between inshore and offshore lobster Fis. 9. Lacehen and number of lobstars recaptured in the MNPS

 '". by canumerci.f ieben n from 197s io 1996-populations (Saila and Flowers 1968; Uzmann et al.

1977; Cooper and Uzmann 1980; Campbell and { u Stasko 1985,1986). Entrainment Connecticut RI sampies am MNPS h! g /* BIS discharges from mid-May through July during 19%. Because of the MNPS shutdown and limited  ; {

             /

s g operation of cooling water pumps, only 44 lobster yg err 4 S lanw samples were collected during 199A by j is __ cf contract, between 66 and 104 samples were collected

          ' -tiDL4 j MpW              , /                                          annually in previous 2- and 3 unit studies, in addition to fewer samples collected in 1996, the                     ,

gr average volume of entrainment samples was also { lower (2,822 m'), when compared to the range of Fis 10. wien and musiiber or nahmers g t.y easimii.rci.1 Previous lobster lanse entrainment studies (3.513-labeerman >$ kna frosn MNPs in the vicinity ofIAng laiend and 4,556 m'). Black Island sounds (197s-1996). As a result of the MNPS shutdown only 19 lanze were collected during 1996 which was well below the range reported in previous 2- and 3-unit studies (102-625). Only 10 Stage I and 9 Stage 11 lanz g p asA , were collected during 19% (Fig.12). Stage I larvae

               ~                                 -
                 #           ins                                                predominated in previous 2-unit (86 and 87%) and
                                 '            .# d

• i 3-unit studses (38-90%). In general, with the d[lif** *s excepuon of 1988 and 1992, Stage 11 and III lanz were rarely collected; Stage IV lanu were the

{ l

                                           ?         s           a second-most abundant lan21 stage collected in (NJ                ..

previous 3-unit studies. The fact that no Stage IV  ! f 6% lanz were collected during 1996 may also be

                  !   [k                                                        related to the limited operation of cooling-water pumps due to the MNPS shutdown Stage IV larvac are strong swimmers; Cobb et al. (1989) reported an Fig. II. Nusuber of tag resums at "- >$0 km from MNPs (l'78 l"'k                                                                       average swimming speed of 18 cm/s for wild free-ranging Stage IV lobster larvae in an early study of 170 Monitoring Studies,1996 i

                -        - -         - , . .       . _ _ . _        . - -        -. ..                 - . - _ . .            - .           ..._c.-    . -._-

l 1 TABLE 11. Annual 6maan density (number per 1000 m'

• 95*.

n ~ C.lj oflobeser larvae collected in day and night entramment samples i d from 19s4 through 1996.

                             ~I                                                                                                                               l Year       Tune               ocean                                    l f

g 4* I'l ofdev densav* 95*.. C.I. i

                             ]                   s W                                                                                                                                                          -

f3"

     '                                                       '" ~

19s4 Day 0.158 0.061 0.236 Night 0.737 0.138-1336 [ 1985 Day 0390 -0.041 0.s20 l l Night 0.620 0.290-0.951 e4 es se o se e. oo es u u e4 es es 1986 Day 0324 0.063 4 585 1

                       ,              ,            , ,                                           Night               1399*          0.556-2.242 1987      Day Fi6.12 Annual number of lobster larvae and their stage                                                            0.791          0.040 I.542 composition (Stage I-IV) collected in samples taken at the                                    Night                0.667          0.205 1.129 MNPS discharges from 1984 through 1996.

1988 Day 0.727 0.199-1.653 Night 0.688 0.271 1.106 water currents at the intakes of Unit 1, velocilitEs of about 25 cm/s were measured immediately outside '"' the Unit 1 intake during full power operations (4 $ iM3 8 [$. $ cooling water pumps; VAST 1972). In previous 1990 Day 0341 0.10145sl entrainment studies, sampling occurred at one of the NW l.16f 03691.765 three MNPS discharges during full power operation. leal Dey 0.2:7 0.131 0 442 However. during 19%. lobster larvae sampling took Night 0.756' O.502 1.010 place when only one or two cooling water pumps 1992 Dev I.299 were operating. With fewer pumps operating, intake 0.043 2355 Night 1369 0.530 2.209 velocities wre presumably four and Stage IV larvac could have avoided entrainment through the 1"3 cooling-water system. ] 129 , 2 132 Day and night entrainment samples have been 1994 Day 0.268 0.0:5 0.452 collected since 1984 to examine diel variation in NW 1305' O.7 % 2303 lobster lanae M-* During 1996, the density g995 Dev 0.594 0.122 1310 cf lobster larvac in day and night samples was Nigle 2.189* 0369-4.009 similar (0.329 and 0.242 per 1000 m', respectively; Table 11). Significantly higher larval densities were 1996 Dey 0329 1399-2.057 Observed in night samples in six years of 3-unit Nishi 0.242 0.731 1.215 i operation (1986,1989,1990,1991,1994,1995). Number per 1000 m,. This observed diel nriability was similar to results 8 j found elsewhere. Earlylaboratory studies on lobster ""$%o 'I '" " # ""' ' " larvae behavior demonstrated that Stage I lanz  ! cxhibit positive phototaxis and disperse from surface found significantly more lanx (Stage 1) during waters during darkness (Templeman 1937, 1939). nightime than during daytime neuston tows made l la contrast, field surveys conducted by Harding et al. within a protected coastal embayment along the 1 Nova Scotlan Shelf. (1987)in Canadian waters indicated that most Stage I larvae were collected at depths between 15 and 30 The annual A mean density of lobster lan'ae m during the day and were rarely found below 10 m collected in entraimnent samples during 1996 (0.364 at night. Boudreau et al. (1991) concluded that per 1000 m') was the lowest value reported since the thermal gradients influenced the vertical migration entrainment studies began in 1984 (presious 3-unit ' of lobster larvae; all four stages were found to seek range of 0.525-1.385; 2-unit range of 0.409 and warm water above a thermocline regardless of time 0.504; Table 12). An estimated 20,106 lobster of day. More recently, DiBacco and Pringle (1992) larvac were entrained through the MNPS cooling-1Abster Studies 171

-. - .. . _ . - - - - . - . - - ~ - - - . - - - ~_-..~.- - . - . - - . . .~ . _. - - -- - TABLE 12. Ansual Aensen density (number per 1000 m')ofloluser larvae in entraanment sampias dunn8 afwir esason of occurrence an enersuunens estamates with 959. C.L for MNPS from 1984 through 1996. Year Time pened Number Mean included larvae Cool vol. Entramment denany' 959. C.I. (m' x 0*) essunate 959 C.I. 1984 21May-10Jul 102 0.409 0.184 0.635 189.4 77.458 34.s47-120.259 1985 15May 16Jul I42 0.504 0.258 0.749 255.1 128.550 19866 14May-14Jul 232 0.857 0.418-1.297 65.806 191.040 666.2 566.619 1987 18May-30Jun 184 0.943 278.457-864.017 0.274-1.613 423.8 399.608 1988 16May l Aug 571 0.717 0.296-1.137 116.111 683.529 837.6 600,573 247.935 952.372 1989 22Mov 28Jul 237 0.701 0.358-1.044 1990 14May 30Jul 562.5 394.518 280 0.748 201.480 587.556 0.436 1.060 779.1 582.738 1991 7Mey 22Jul 157 0.525 339.671-825.805 0.365 0.685 564.1 296.173 1992 19May 14Jul 625 1.334 205.910-386.435 0.652 2.016 461.2 615.285 300,724 929.846 1993 24May 25Jun 218 1.081 0.273 1.889 360.6 389.767 1994 25May-4Aug 257 0.908 98.433-681.101 0.4451.371 745.2 676.639 1995 30May-21Jul 254 1.385 331.613 l.021.66$ 0 470-2.300 495.9 686.826 1996 6Jun lJul 19 0.364 233.075 1.140.578 0.194 0.535 55.2 20.106 10.716-29.551

                      **Unit Mean         densities 3 besan            are based conunsresaloperation     on the Aensen (NUsCO 19886 and Pennmston 1983).

water system in 19%. which was the lowest estimate reported during both 3-unit (2%,173-686,826) and less than 1% in Canadian waters (Scarratt 1964  ; 1973; Harding et al.1982)1: more than 50% in LIS ' 2-unit studies (77,458 and 128,550). IAbster larvae (Lund and Stewart 1970; Blake 1901). Mechanisms entrainment is related to both the annua! larval of lobster lanz dispersal in coam raters may be density and the performance of the MNPS units related to surface water circulation p.ierns (Fogarty during the hatching season. When all units operate 1983); surface currents, regulated by the wind and at full capacity, cooling water demands are at a tide, converge and are sisible on the surface waters maximum and resulting entrainment estimates are higher. Conversely, entrainment estimates are low as " slick" or " scum" lines. These convergence areas delineate zones of upwelling and downwelling and when one or more units are shutdown for were reported to contain high densities of planktonic maintenance or refueling. The low entrainment organisms. including lobster lanze (Cobb et al. estimate during 1996 was primarily due to MNPS 1983; Blake 1988). " Slick" lines were often seen in shutdown; combined 3-unit cooling water volume of the MNPS area stretching from near Twotree Island 55.2 m'x10' was well below the range of values reported previously (189.4-837.6 into Niantic Bay and could explain the patchy m'x10'). distribution oflobster larvae in entrainment samples. Nonetheless, since Unit 3 began operating in 1986 Furthermore, based on the short duration of the first entrainment estimates have been substantially higher because the cooling-water demand of Unit 3 alone is larval stage (3-5 days), the source of Stage ! larvae approximately the volume required by Unit I and 2 collected in Ibe MNPS cooling-water was probably combined. from the locr.1 spawners (e.g. from Twotree. where The impacts on the adult lobster population due to over 13% of females have been berried during 3-unit operation). Stage IV lanne, however, are in the entrainment of lobster larvae we difficult to assess due to the high variability in lobster larvac water column between 4 and 6 weeks. Therefore, based on water circulation patterns in LIS. it is abundance and stage composition (Bibb et al.1983; Fogarty 1983; Lux et al.1983; Blake 1984,1988) unlikely that Stage IV larvae originate locally. Lund and Stewart (1970) indicated that the large number and the lack of reliable estimates of larval and post-of berried females found in western LIS (27%; Smith lar' val sunival rates (Phillips and Sastry 1980; Caddy and Campbell 1986; Cobb 1986; Blake 1991). 1977) may be responsible for recruitment of Stage IV Disagreement among' researchers on the source and larvae in middle and castern LIS. More recent studies have shows that Stage IV larvae exhibit dispersion mechanism of lobster lanic, and on egg and larval mortality, has led to a range of sunival directional swimming behavior and moved tens of kilometers from the spawning grounds (Cobb et al. estimates during the larval life history phase from 1989; Rooney and Cobb 1991; Katz et al.1994). 172 Monitoring Studies,19% i

1 1 i N 1 4 4 i Conclusions an increase in the proportion of twried females collected. The objective of increasing the minimum

The American lobster fishery is an extremely legal size oflobsters was to enhance recruitment and I

important fishery for New England and mid Atlantic to sustain the lobster resource; lan'al production a states. Scientists and fishery managers are should increase as a larger proportion of berried concerned about the health of the lobster resource, females will be able to spawn before reaching legal ( and are focusing on the intense fishing pressure size. 1 applied to the resource and potential long-term Due to the MNPS shutdown during 19%. the l cffects on recruitment. Several management density oflobster larvac and estimated total number i

'               measures have been implemented to improve lobster                         of larvac entrained through the cooling-water recruitment and survival, including requirements for                      systems were the lowest values reported since the escape vents in lobster pots and increases in                             entrainment studies began in 1984.             Overall 2                                                                                                                                                  t minimum legal size. Results from our studies
'                                                                                         entrainment levels have been substantially higher indicated that more than 90% of the lobsters larger                      during 3-unit operation than during 2-unit operation, j               than the minimum legal size are removed by fishing                        due to the higher cooling water demand of Unit 3.

each year and the trend in legal lobster abundance j The potential effect of higher larval entrainment on has significantly declined since lobster studies began subsequent legal lobster abundance is difficult to in 1978. This decline is due, in part, to increased assess due to the uncertainty concerning the source

;              fishing rates, which have more than doubled since                         of entrained larvae, their sunival rate, and the j                1978, and, more recently, to the increases in relatively long period of time between larval minimum legal size in 1989 and 1990. While the                           settlement and recruitment to the fishery.          The

{ total number of lobsters caught (all sizes) and total impact of lobster lanae entrainment is not CPUE in 1996 were within the range of previous immediately observable and its effect on the fishery i years, the CPUE of legal lobsters in 19% was the may not be seen for several years after the impact lowest observed in 10 years of 3-unit operational occurs Continued monitoring of lobsters wil! ! studies. The magnitude of legal catches in any year 4 demonstrate the relationship of MNPS operations on is highly dependent upon the Waha~ of lobsters the local lobster population and its development. one molt smaller than legal size the year before. Legal catches were expceted to decline in 19%,

based on low total CPUE during 1995, which was l the lowest reported in almost 20 years, with fewer References Cited sublegal-size lobsters available to molt to legal size Aiken, D.E.

m 19%. 1973. Proccdysis, setal development. j One of the most important physiological factors and molt prediction in the American lobster,

regulating lobster actnity is water temperature, and (Homarus americanus).1 Fish. Res. Board Can.

30:1337 1344, l May to October water temperatures during 19% Aiken. D.E. 1980. Molting and Growth. Pages

were amor.g the coldest observed in our lobster 93163 in IS. Cobb, and B.F. Phillips, eds. The studies. As a result of the below-normal water j temperatures, lobster catches and molting peaked biology and man 1gement of lobsters, Vol. I, later during 1996. Overall water temperatures Academic Press,is New York.

j during the period of 3-unit operation were, on Aiken, D.E., and S.L Waddy 1980. Reproductive average, warmer than those observed during 2-unit biology. Pages 215-276 .n IS. Cobb, and B.F. operation and subsequent peaks in lobster catch and Phillips, eds. The biolog and management of mIlting occurred earlier. lobsters, Vol.1. Acaderr.sc Press, Inc., New York. Other changes were 1 observed in the population characteristics of local Aiken, D.E., and S.L Waddy. 1982. Cement gland lobsters during 3-unit studies, but were related to development, ovary maturation, and reproductive , mercased fishing rates and to implementation of new cycles in the American lobster Homarus

fishery regulations rather than to power plant americanus.1 Crust. Biol. 2:315-327.

impacts. The most notable change in lobster Anthony, V.C., and 1F. Caddy 1980. Proceedings

 ;          population characteristics during 3-unit studies was                            of the Canada-U.S. workshop on status of i

assessment science for N.W. Atlantic lobster Lobster Studies 173

 . . - -                 _ - --           - - - . - .               . .-- -.    .       --               - - ~ ~ - - -                .. -

(Homarus americanus) stocks (St. Andrews, Briggs, P.T., and F.M. Mushacke. ! 1984. The N.B., Oct 24 26,1978). Can. Tech. Rep. Fish. American lobster in western Long Island Sound: I Aquat. Sci. 932.186 pp. Movement, growth and mortality. NY Fish ASMFC (Atlantic States Marine Fisheries Game J. 31:21-37. Commission).1996. A review of the population ! Caddy, J.F., and A. Campbell. 1986. Summary of dynamics of American lobster in the Northeast. session 9: l summary of research Special Report No. 61. 48 pp. recommendations. Can. J. Fish. Aquat. Sci. Auster, PJ. 1985. Factors affecting catch of Amencan 43:2394 23 %. lobster, Hamarus enericarus in baited traps. Univ. Campbell, A.1982. Movements of tagged lobsters Conn. See Grant Prog. Tech. Bull. Ser. CT-SG45-1. released off Port Maitland. Nova Scotia, 1944-80. 46 pp. Can. Tech. Rep. Fish. Aquat. Sci. No. I136. 4i Bibb, B.G., R.L. Hersey, and RA. Marcello, Jr. pp. 1983. Distribution and abundance of lobster Campbell, A., and A.B. Stasko. 1985. Movements larvae (Homarus americanus) in Block Island of tagged American lobsters. Homarus Sound. NOAA Tech. Rep. NMFS americanus, off southwestern Nova Scotia Can. SSRF-775:15-22. J. Fish. Aquat. Sci. 42:229-238. Bigelow. H.B., and W.C. Schroeder 1953. Fishes of Campbell. A.. and A.B. Stasko. 1986. Movements the Gulf of Mame U.S. Fish Wildl. Serv. Bull. 53:1 577. of lobsters (Homarus americanus) tagged in the Blake, M.M. Bay of Fundy. Canada Mar. Biol. 92:393-404. 1984. Annual progress report Connecticut Cobb, J.S. 1986. Summary of session 6: ecology of lobster investigations, population structures. Can. J. Fish. Aquat. Sci. January-December 1983. NOAA-NMFS Project 43:2389-2390. No. 3-374-R. 47 pp. Blake, M.M. Cobb, J.S. T. Gulbransen. B.F. Phillips, D. Wang. 1988. Final Report Connecticut and M. Syslo. 1983. Behasior and distribution lobster investigations January 1,1983-December oflarval and earlyjuvenile Homarus americanus.

31. 1987. NOAA-NMFS Project No. 3-374-R. Can. J. Fish. Aquat. Sci. 40:2184-2188.

103 pp. Cobb, J.S., D. Wang. RA. Richards, and M.J. Fogarty. Blake, M.M. 1991. Connecticut lobster (Homarus 1986. Competition among lobsters and crabs and its americanus) population recruitment studies January 1, possible effects in Narragansett Bay, Rhode Island. 1988-December 31, 1990. Pages 282 290 m G.S. Janueson and N. Bourne, eds. NOAA-NMFS Project No. 3U4. 87 pp. North Pacific Workshop on stock assessment and Blake, M.M. 1994. Connecticut lobster (Homarus management ofimertebrates. Can. Spec. Publ. Fish. americanus) population recruitment studies April Aquat. Sci. 92. 1.1991-March 31,1994. NOAA NMFS Project No. 3114.174 pp. Cobb, J.S., D. Wang, D.B. Campbell. and P. Rooney, Blake. M.M., and E.M. Smith. 1984. A marine 1989 Speed and direction of swimming by postlarvae of the American lobster. Trans. Am. resources plan for the state of Connecticut. Fish. Soc. 118:82-86. Connecticut Dept. of Emiron. Protection, Mar. Cooper, R.A. Fish. 244 pp. 1970. Retention of marks and their effects on growth, behavior and migrations of the Boudreau, B., Y. Simard, and E. Bourget. 1991. American lobster, Homarus americanus. Trans. Behavioural responses of the planktonic stages of Amer. Fish. Soc. 99:409 417. American lobster Homarus americanus to Cooper, R.A., R.A. Clifford, and C.D. Newell. thermal gradients, and ecological implications. Mar. Ecol. Prog. Ser. 76:13-23. 1975. Seasonal abundance of the American Briggs. P.T., and F.M. Mushacke. lobster, Homarus americanus, in the Boothbay 1979 The Region of Maine. Trans. Amer. Fish. Soc. Amencan lobster in western Long Island Sound. 104:669-674. NY Fish Game J. 26:59 86. i Cooper, R.A.. and J.R. Ur.mann. 1980. Ecology of Briggs, P.T., and F.M. Mushacke. 1980. The juvenile and adult Homarus americanus. Pages American lobster and the pot fishery in the 97-142 in J.S. Cobb, and B.F. Phillips, eds. The inshore waters off the south shore of Long Island, biology and management of lobsters, Vol 11. New York. NY Fish Game J. 27:156-178. Academic Press, Inc., New York. 174 Monitoring Studies,1996

1 i i

DiBacco, C., and J.D. Pringle. 1992. Larval lobster York State Department of Em'ironmental (Nomarus americanus H. Milne Edwards,1837) Conservation 22 pp.

I distribution in a protected Scotian Shelf bay. J. Harding, G.C., W.P. Vass. and K.F. Drinkwater. [ Shellfish Res. I1:81-84. 1982. Aspects of larval American lobster Dow, R.L. 1966. The use of biological, (Homarus americanus) ecology in St. Georges i environmental and economic data to predict Bay, Nova Scotia Can. J. Fish. Aquat. Sci. supply and to manage a selected marine resource. 39:1117 1129. The Amer. Biol. Teacher 28:26-30. Harding, G.C., J.D. Pringle. W.P. Vass S. Pearre ! Dow, R.L. 1%9 Cyclic and geographic trends in Jr., and S.J. Smith. 1987. Vertical distribution l acewater temperature and abundance of American and daily movement of larval lobsters Homarus

lobster. Science 164

1060-l%3.

Dow, R.L. americanus over Browns Bank. Nova Scotia. 4 1976. Yield trends of the American Mar. Ecol. Prog. Ser. 49:29-41. lobster resource with increased fishing effort. Herrick, F.H. 1911. Natural history of the i Mar. Technol. Soc. 10:17-25. American lobster. Bull. U.S. Bureau Fish. Ennis, G.P. 1971. Lobster (Homarus americanus) 29:149-408. i fishery and biology in Bonavista Bay, Hollander, M., and D.A. Wolfe. 1973. Newfoundland. 1966-70. Fish. Mar. Serv. Tech. Nonparametric statistical methods John Wiley Rep. 289. 46 pp. l j Ennis, G.P. and Sons. New York. 503 pp. 1974. Observations on the lobster Katz, C.H., J.S. Cobb. and M. Spaulding. 1994. 1 fishery in Newfoundland. Fish. Mar. Serv. Tech. ! larval behavior, hydrodynamic transport. and Rep. 479. 21 pp. potential offshore recmitment in the American

Ennis, G.P. 1984. Small-scale seasonal movements lobster. Homarus americanus. Mar. Eco. Prog.

!' of the American lobster Homarus americanus. Ser,103:265-273. ' Trans. Am. Fish. Soc. 113:336-338. Keser, M.. D.F. Landers, Jr., and J.D. Morris. 1983. Fair, J.J., and B. Estrella. 1976. A study on the Population characteristics of the American effects of sublegal escape vents on the catch of lobster, Homarus americanus, in castern Long

lobster traps in five coastal areas of

! Island Sound, Connecticut. NOAA Tech. Rep. Massachusetts. Unpublished manuscript, Mass. NMFS SSRF-770. 7 pp. 1 Div. Mar. Fish. 9pp. Krouse, J.S. 1973. Maturity, sex ratio, and size l Flowers, J.M.. and S.B. Saila. 1972. An analysis of i composition of the natural population of temperature effects on the inshore lobster fishery. American lobster, Romarus americanus, along 3 J. Fish. Res. Board Can. 29:1221-1225. the Maine coast. Fish. Bull., U.S. 71:165-173. 4 Fcgarty, M.J. 1983. Distribution and relative Krouse, J.S. I abundance of American lobster, Romarus 1976. Incidence of cull lobsters. j Homarus americanus, in commercial and americanus larvac: New England imestigations j during 1974-79. NOAA Tech. Rep. NMFS research catches off the Maine coast. Fish. Bull U.S. 74:719-724. l SSRF-775. 64 pp. Krouse, J.S. 1978. Effectiveness of escape vent Fegarty, M.J., and D.V.D. Borden. 1980. Effects of shape in traps for catching legal-sized lobster. l trap venting on gear selectivity in the inshore

Nomarus americanus, and harvestable-sized Rhode Island American lobster, Homarus crabs, Cancer borealis and Cancer srroratus.

americanus, fishery. Fish. Bull., U.S. Fish. Bull., U.S. 76

425 432.

77:925 933. Krouse, J.S. 1980. Summary of lobster, Homarus Fogarty, M.J., D.V.D. Borden, and H.J. Russell. i americanus, tagging studies in American waters 1980. Movements of tagged American lobster, (1898-1978). Can. Tech. Rep. Fish. Aquat. Sci. Homarus americanus, off Rhode Island. Fish. 932:135-140. Bull., U.S. 78:771-780. Krouse, J.S. 1981. Movement, growth, and

Graulich, K. 1996. A model to assess egg mortality of American lobsters, Romarus production and the impacts of fishing mortality americanus, tagged along the coast of Maine.

j on total egg production in Long Island Sound 1 NOAA Tech. Rep. NMFS SSRF-747.12 pp. lobsters. Division of Marine Resources - New

Krouse, J.S., and J.C. Thomas.1975. Effects of trap selectivity and some population parameters on the Lobster Studies 175 4

                                         +,                          -        -

n . ,<

size composition of the American lobster, marine emironment of Long Island Sound at Homms americanus, catch along the Maine Millstone Nuclear Power Station, Waterford, coast. Fish. Dull., U.S. 73:862-871. Connecticut. Resume 1%81981. Krouse, J.S., K.H. Kelly, D.B. Parkhurst Jr., G.A. NUSCO.1986. The effectiveness of the Millstone Robinson, B.C. Eully, and P.E. Thayer. 1993. Maine Departmeri of Marine Resources Lobster Unit I sluiceway in returning impinged organisms to Long Island Sound Enclosure to Stock Assessment Project 31J-611. Annual letter Doll 85 dated May 27, 1986 from R.A. report April 1,1992 through January 31, 1993. Reckert. NUSCO. to S.J. Pac. Commissioner, 61 pp. CTDEP,18 pp. Kurata, H. 1%2. Studies on the age and growth of NUSCO. 1987a. Lobster population dynamics. Crustacca Bull. Hokkaxio Reg. Fish. Res. Lab. 24:1 115. Pages 1-42 in Monitoring the marine environment of Long Island Sound at Millstone Landers, D.F., Jr., and M.M. Blake. 1985. The Nuclear Power Station, Waterford, Connecticut. effect of escape vent regulation on the American Summary of studies prior to Unit 3 operation lobster, Romarus americanus, catch in eastern 1987. Long Island Sound, Connecticut. Trans. 4ist Annual Northcast Fish Wild. Conf. 9 pp. NUSCO.1987b. The etfectiveness of the Unit 3 fish Lund, W.A., Jr., and L.L. Stewart. return system 1987. 20 pp. 1970. NUSCO. 1988a. The usage and estimation of Abundance and distribution of larval lobsters, DELTA means Pages 311320 in Monitoring i Romarus americanus, off southern New England. the marine emironment of Long Island Sound at Proc. Nati. Shellfish. Assoc. 60:40-49. Lux, F.E., G.F. Kelly and C.L Whecler. 1983. Millstone Nuclear Power Station. Waterford. Connecticut. Three-unit operational studies Distribution and abundance of lan>al lobsters 1986-1987. (Homarus americanus) in Buzzards Bay, NUSCO.1988b. The effectiveness of the Millstone Massachusetts, in 1976-79. NOAA Tech. Rep. Unit 3 fish return system. NMFS SSRF-775:29-33. Appendix ! to Enclosure 3 to letter D01830 dated January 29 Mauchline, J. 1976. The Hiatt growth diagram for 1988 from E.J. Mroczka. NUSCO, to L. Crustacea. Mar. Biol. 35:79-84. Carothers, Commissioner, CTDEP. 21 pp. McConnaughey, R.A., and LL. Conquest. 1993. NUSCO. 1988c. Hydrothermal Studies. Pages Trawl survey estimation using a comparative 323 355 in Monitoring the marine emironment approach based on lognormal theory. Fish. Bull., c.f Long Islan1 Sound at Millstone Nuclear Power U.S. 91:107 118. StaSon, Waterford. Connecticut. Three-unit McLeese, D.W., and D.G. Wilder. 1958. The operational studies 1986-1987. activity and catchability of the lobster (Homarus NUSCO. 1996. Lobster studies. Pages 9-32 in americanus) in relation to temperature. J. Fish. Monitoring the marine emironnent of Long Res. Board Can. 15:1345-1354. Island Sound at Millstone Nuclear Power Station, NMFS (National Marine Fisheries Senice). 1993. Waterford, Connecticut. Annual report 1995. Report of the 16th Northeast Regional Stock Pecci, K.J., R.A. Cooper, C.D. Newell, R.A. Assessment Workshop. Northeast Fish. Sci. Cen. Clifford, and PJ. Smolowitz. Ref. Doc. 93-18. 1978. Ghost NOAA/NMFS Northeast fishing of vt.r ui ano u:nented lobster. Romarus Fisheries Science Center, Woods Hole, MA. I18 pp. americanus, traps. Mar. Fish. Rev. 40:9-43. Pennington, M. 1983. Efficient estimators of NMFS. 1996. Report of the 22nd Northeast abundance, for fish plankton sun'eys. Biometrics Regional Stock Assessment Workshop (22nd 39:281-286. SAW). Stock Assessment Review Committee Phillips, B.F., and A.N. Sastry. 1980 (SARC) Consensus Summary of Assessments. Lanal ecology. Pages Il-57 in J.S. Cobb, and B.F, NOAA/NMFS Northeast Fisheries Science Phillips, eds. The biology and management of Center, Woods Hole, MA.135 pp. labsters. Vol II. Academic Press, Inc.. New NUSCO (Northeast Utilities Senice Company). York. 1982. Lobster Population Dynamics-A Review Richards, R.A., J.S. Cobb, and M.J. Fogarty. 1983. and Evaluation.. Pages 1-32 in Monitoring the Effects of behasioral interactions on the 176 Monitoring Studies,19%

1 catchability of American lobster, Romarus Stewart, L.L. 1972. The seasonal movements. americanus, and two species of Cancer crab population dynamics and ecology of the lobster. Fish. Bull., U.S. 81:51-60. Homarus americanus (Milne-Edwards), off Ram Richards, R. A.. and J.S. Cobb. 1987. Use of Island, Connecticut. Ph.D. Thesis University of avoidance responses to keep spider crabs out of Connecticut. Storrs CT 112 pp.

 '            traps for American lobsters. Trans. Amer. Fish.      Templeman, W. 1935. Local differences in the Soc.116:282-285.                                           body proportions of the lobster. Romarus Rooney, P.. and J.S. Cobb. 1991. Effects of time of americanus. J. Biol. Board Can. 1:213-226.

day, water temperature, and water velocity on Templeman, W.1936. Local differences in the iife swinuning by postlarvae of the American Lobster, 4 history of the lobster (Homarus omericanus) on Homarus americanus. Can. J. Fish. Aquat. Sci. the coast of the maritime provinces of Canada. J. 48:1944-1950. i Biol. Board Can. 2:41-88. Saila, S.B., and J.M. Flowers. 1968. Movements Templeman, W. 1937. Habits and distribution of i and behavior of berried female lobsters displaced 1 larval lobsters (Homarus americanus). J. Biol. from offshore areas to Narragansett Bay, Rhode Board Can. 3:343-347. Island. J. Cons. Int. Explor. Mer. 31:342-351. Templeman. W. 1939. Investigations into the life Scarratt. D.J. 1964. Abundance and distribution of history of the lobster (Homarus americanus) on j lobster larvae (Homarus americanus) in the west coast of Newfoundland. 1938. l Northumberland Strait. J. Fish. Res. Board Can. Newfoundland Dep. Nat. Resour. Res. Bull. 21:661-680. (Fish) 7. 52 pp. Scarratt. D.J. 1970. Laboratory and field tests of Templemaa W.1940. Lobster tagging on the west modified sphyrion tags on lobsters (Homarus coast of Newfoundland,1938. Newfoundland americanus). J. Fish. Res. Board Can. I 27:257-264. Dep. Nat. Resour. Res. Bull. (Fish) 8.16 pp. ) Thomas, J.C.1973. An analysis of the commercial Scarratt. D.J. 1973. Abundance, survival, and lobster (Homarus americanus) fishery along the vertical and diurnal distribution oflobster larvae coast of Maine August 1966 through December in Northumberland Strait 1%2-63, and their relationships with commercial stocks. 1970. NOAA-NMFS Tech. Rept. SSRF-667. 57 J. Fish. pp. Res. Board Can. 30:18191824.

• Uzmann, J.R., R.A. Cooper, and K.J. Pecci. 1977.

Scarratt, D.J., and P.F. Elson. 1965. Preliminary Migrations and dispersion of tagged American trials of a tag for salmon and lobsters. J. Fish. ! lobsters, Homarus americanus, on the southern Res. Board Can. 22:421-423. j Sen, P.K. New England Continental Shelf. NOAA Tech. 1%8. Estimates of the regression Rep. NMFS SSRF-705. 92 pp. ] coefficient based on Kendall's tau. J. Am. Stat. VAST.1972. A study of water currents at intakes Assoc. 63:1379-1389. Skud, B.E., of Umt No.1. Prepared for Millstone Point Co. and H.C. Perkins. 1%9. Size Rept. No. 67-83 06-72A. Ilpp. composition, sex ratio and size at maturity of Wilder, D.G. 1953. The growth rate of the cffshore northern lobsters. U.S. Fish Wildt. Spec. Sci. Rep. Fish. 598.10 pp. American lobster (Homarus americanus). J. Smith, E.M. 1977. Some aspects of catch / effort, Fish. Res. Board Can. 10:371-412. Wilder, D.G. 1%3. Movements, growth and biology, and the economics of the Long Island survival of marked and tagged lobsters liberated lobster fishery during 1976. NOAA-NMFS, in Egmont Bay, Prince Edward Island. J. Fish. Commer. Fish. Res. Dev. Act, Project No. 3-253 R 1. 97 pp. Res. Board Can. 20:305-318 . Wilder D.G., and R.C. Murray. 1958. Do lobsters Smith, E.M., E.C. Mariani, A.P. Petrillo, L.A. move offshore and onshore in the fall and spnng? Gunn, and M.S. Alexander. 1989. Principal fisheries of Long Island Sound, 1%1-1985. Fish. Res. Board Can. Atl. Prog. Rep. 69:12-15. ' Connecticut Dept. of Environ. Protection Div. of Conservation and Preservation, Bureau of Fisheries, Mar. Fish. Pro. i Lobster Studies 177

1 I l 1 1 l l l 178 Monitoring Studies,19%

Rocky Intertidal Studies Introduction...........................-..................................................................................................181 Ma terials a nd Method s .................... ............................................................................................ 181 Quali ta tive Sam pling.... ........ .................... .. . .......................... ...................... ...... 181 Abund ance Measurem en t .......................... ...................................................................... 182 AswphyIlum nodasum Studies ............. ....... .. .................... .................................. 183 Da ta Analys is .. ....................................... .. ...... ....... ........ ... ........ ...... . . .... ..... ..... .......... .. ... 183 Resul ts and Discussion .................................................................................................................. 184 Qua li ta ti ve Stu d ies........................................................................................... ................. 184 A bunda nce Meas urem en t ................ ....... ................................... ........................... ........ 188 BarnacIes.................................................................................................................190 Fucus.....................................................................................................................190 Omndrus and comm on epiphytes ....................................................................... 193 Comm uni ty Analysis ................................................................................... ......... 195 AswphyIfum nodasu m Stu d nes ........................................................................................ . 197 Growth....................................................................................................................197 Mortality..............................................................................................................1% Conclusions.....................................................................................................................................200 References Ci ted ............ .................... .................... ................... .. . .. ................ ..................... .. . ... .. 200 t l RockyIntenidal 179 i i

4 j i 1 4 e j 180 Monitoring Studies,19%

_. _ . . _ - - ~_ ._ . _ _ _ . _ _ _ _ _ _ _ . _ _ _ _ _ _ Rocky Intertidal Studies i  : Introduction Materials and Methods 9 Rocky intertidal habitat is extensive along most - 1

• of New England's coastline. and supports a rich and {

diverse community of attached algae and animals. Qualitative Sampling '

• Owing to its location and relative immobility, this community is exposed to a number of emironmental Qualitat ve algal collections utre made during

} stresses, including the heated effluent from coastal odd numbered months at four rocky intertidal i

power plants. Studies of rocky shore communities stations (Fig.1). These stations are, in order of most '

i are commonly included in ecological monitoring to least exposed to prevailing winds and storm Programs designed to assess the impacts of those forces: Fox Island-Exposed (FE), Millstone Point power plants (Vadas et al. 1976, 1978; Wilce et al- (MP), White Point (WP). and Giants Neck (GN). 1 1978; NAESCO 1994; NAl 19%; NUSCO 19%). The MP station was added in September 1981: FE, Rocky intertidal studies at Millstone Nuclear WP and GN have been sampled since March 1979.

Power Station (MNPS) are part of an extensive Prior m 1998. qualiative collections were made environmental monitoring program whose primary monthly, but as current procedures call for sampling objective is to determine whether differences (e.g., in only n odd-numbered months, only those months

abundance, distribudon or species composition) that from historical data are included. A year of

exist among communities at several sites in the quhutive sampling is determined to be from March Millstone Point area can be attributed to construction to the following January, /.c., the latest year of and operation of MNPS, in particular since Unit 3 qualitative algal data (1996) comprises collections began operation in 1986. To achieve this objective, from March 19% to January 1997. The 1985 studies were designed and implemented to identify sample year (3/85 - 1/86) terminated the 2-onit anached algae and animal species found on neshy operational period; the 1986 sampic year (3/86 -

rocky shores to describe temporal and spatial 1/87) was the first in the 3-unit operational period. Panerns d occunence and abundance d these The FE station, approximately 100 m cast of crganisms, and to identify physical and biological the MNPS discharges, is directly exposed to the 3-factors that induce vanability in the local rocky un t thermal plume (during part of the tidal cycle); intertidal communities. This research includes MP and WP are 300 and 1700 m from ' the qualitative algal sampling, abundance (percentage discharges, respectively, and potentially impacted by cmtr) measurements of intertidal organisms, and the plume. The GN station is about 6.5 km west of growth and mortality studies of Ascophyllum Millstone Point, and unaffected by MNPS operation. nodosum. The following report discusses results of Qulitadve collections were used to sampling and analysis in the most recent study year, characterize the attached flora at each site during mi compares these results to those d 2-unit each sampling period. Algal samples were identified operational studies (March 1979-February 1986), fresh or after short-term freezing. Voucher and 3amit operational studies to date (March 1986 specimens were made using various methods in to present). The program has undergone saturated Nacl brine, as dried herbarium mounts, or considerabic modification over the past 17 years, as microscope slide preparations. with the most recent change invohing the reduction Tim quliutive species list includes all in number of sampling sites from mne to four; this attached. macroscopic algal species recorded from report emphasizes results from the currently sampled MNPS sampling stations. Excluded from these lists stauons, but refers to previous studies (e.g., NUSCO are diverse diatom ta.u, cyanobacteria and sonw 1996) where appropnate. crustose, endophytic or endozooic algal species. These elements of the microbiota are present but too Rocky Intertidal 181

 .    -     -                          _ ,_~ .                ~        . _~       __      __        -               . . . . . .      _ -.

t l l ? c LIS

                                                            \
                                                             )_f \

g.d ,' a n,a ~.f. A .~8) - 1 North y 1km l 0: ' 1'mi l 5 V v

                                                                                              /

l Niontic Boy MNP) w D

                                                                               # E                   White Point
           -g g,                                                                                 WPTvNj kN          q eicek 0           po;ni                                                  (

C Fig.1. locauon of the MNPS rocky insertuial samplang sites: ON=Oiants Neck MP= Millstone Point. FE= Fox is N g difficult to consistently collect. and, for many s' ( species, to identify as components of a large-scale

                                                 **"           ?
                                                  '"'

• cmironmental program. Also included in our lists are taxa that may be conspecific or subspecific g

h\ "'7,',,, won d

                                                                      /s forms. or alternate life history stages of erect macroalgae. For simplicity, we refer to each of these h3 '

N [j s g entities as a species throughout this report. Except where noted, nomenclature follows that of South and

                            \                                   ,              Tittley (1986), as updated by Villalard Bohnsack w-en      (1995).

g,g a

                                     ~~

Abundance Measurement Ah8m of rocky intertidal organisms was expressed as percentage of substratum cover. At Fig, 2. Detail qua a ofthe MNPS vmnity: FN=new expenmensal n MM h pmnt Axophyuma siis (19s5pment1. MP and FE as in Fig.1. 8tr3P transects were established perpendicular to the water-line. 0.5 m wide and extending from Mean High Water to Mean Low Water levels. Each 182 Monitoring Studies.19%

transect was subdivided into 0.5 m x 0.5 m quadrats Data Analysis and was non-destructively sampicd six times per year, in odd numbered months. The total number of Analysis of qualitatist algal collections quadrats in each transect depended on the slope of includes a calculation of a frequency of occurrence the transect. The percentage of substratum cover of index, based on the percentage of collections in c11 organisms and remaining free space in each which each species was found out of all possible quadrat was subjectively determined. Understory collections (e.g., at a station, in a month. during 2-organisms, species that were partially or totally unit or 3-unit operation). This index was used to obscured by the canopy layer, were assigned a calculate similarities among annual collections. percentage value that approximately corresponded to using the Bray-Curtis formula (Clifford and their actual substratum coverage. Each quadrat was Stephenson 1975): assigned to a zone based,on its tidal height: Zone 1 (high intertidal), Zone 2 (mid intertidal), or Zone 3 (now intertidal). 2 min (X,, Xa) Ascophyllum nodosum Studies s~s (yu, yay m Growth and mortality of Ascophyllum nodosum, a perennial brown alga. were studied at where S is the similarity index between collections / two reference stations (GN. 6.5 km west of the and k, X, is the frequency of occurrence index for discharge and WP,1.7 km cast of the discharge, Fig.

1) and a potentially impacted station (FN, ca.150 m species i in collectionj; Xa is the index in collection from the quarry discharges, northeast of the Fox k; and n is the number of species in common. A Island-Exposed sampling site Fig. 2). Ascophyllum flexible-sorting (ct= 0.25), clustering algorithm was populations at GN and WP have been monitored applied to the resulting similarity matrix (Lance and since 1979, and those at FN since 1985. Williams 1%7).

Ascophyllum had been monitored earlier, at a site ca. Quantitative analyses included determination of 75 m cast of the original Millstone quarry cut (FO), abundance of intertidal organisms as percentage of from 1979 to 1984. This Ascophyllum population substratum covered by each taxon. Substrata not was climinated in the summer of 1984 by exposure occupied by macrobiota were classed as free space. to elevated temperatures from the thermal plurne Cover values of selected species were plotted against discharged through two quarry cuts (NUSCO 1985). time. Similarities of communities (represented as Upright shoots, or fronds, ofAscophyllum were annual collections et each station) were calculated measured monthly, aAer onset of new vesicle using the Bray-Curtis coefficient formula cited formation. from April to the following April. At above, substituting untransformed percentages for each station, ftAy fronds were marked at their bases frequency of occurrence indices. The same with a numbered plastic tag, and five apices on each clustering algorithm was used to form station / year gy,,p ogs, individual were marked with colored cable ties. Lincar growth was determined by measurements A Gompertz growth curve was fitted to made from the top of the most recently formed Ascophyllum length data using non-linear regression vesicle to the apex of the developauxis, or spices if methods (Draper and Smith 1981). The Gompertz branching had occurred Monthly measurement of function form used (Gendron 1989) has three tagged plants began in June; in April and May, parameters, related by the formula: vesicles were not yet sufficiently large to be tagged, tnd five tips were measured on each of 50 randomly chosen individuals.. Tags lost to thallus breakage bt - Njkt ros were not replaced, and the pattern of loss was used as a measure of mortality. Loss of the entire frond was assumed when both the base tag and tip tags where 4 is the predicted length at time t. (I is the were missing. Tip survival us based on the number asymptotic length (estimate of length at the end of cf remaining tip tags the growing season), k is the rate of decrease of specific gromh (shape parameter) and to denotes the Rocky Intertidal 183

 -- .-- - . . - . _ _ _ ~ - - -                    --         _ .  - - ~ .                  .- - . - . . -                       _      -- -

i l time at which the inflection point occurs (time when the year. Therefore, it is necessary to first identify length is increasing most rapidly). The et parameter components of the flora which exhibit this type of was compared among states and between puiods , then use this information as a using 2-sample t-tests' (P=0.05) based on the baseline from which power plant-induced changes asymptotic standard errors of the parameter estimates. Growth data represenung the latest characteristic suite of species. typical of cold-water growing season (1995-19%) were plotted for all period (January-May) collections, includes Dumontia i stations together and for each station separately, with contorta Polystphonia stricta, Spongonema ) summanes of 2-unit (19791986) and 3 unit (1986- tomentosum, Desmarestia viridis, Chorda tomen-1996) operational data. Because the FN stauon was g ,, established in 1985 2-unit operational data from this Afonostroma grevillcl Protomonostroma undulatum site included only ti$c 1985-86 growing season. and Spongomorpha wta Fabk 4 h equa% distinctive group of species characteristic of warm-water (July-November) collections includes Champia Results and Discussion Parvula. Lomentaria baileyana. Callithamnion roseum, Ceramium diaphanum, Grinnellia ameri-cana. Dasya baillouviana, Hincksia mischelliae, Qualitative Algal Studies Enteromorpha clathrata, Bryopsis plumosa and B. hypnoides. Macroalgal communities in the vicinity of Shifts in natural occurrence patterns related to MNPS may be exposed to elevated water thermal plume exposure desenbed above (decreased temperatures resulting from the thermal effluent occurrence of cold-water species resulting from an discharge. Because temperature is important in abbresiated season, or increased occurrence over an determining macroalgal species occurrence and extended season for species with warm-water i distribution (Loning 1990), alterations of patterns of affinities) can be detected by comparing operational ' spatial and temporal species occurrence are likely. period frequencies (2-unit vs. 3-unit) at stations The current qualitative algal sampling program is potentially exposed to the thermal plume. The only  ; used to monitor these patterns by applying various station where such shiAs have been and continue to floristic analyses to data compiled from periodic be evident is the study site nearest the discharge FE. algal collections. For example, two cold-water red algae, Dumontia Qualitative algal sampling results are presented ) contorra and Polysiphonia stricta, were common ' in Table 1 as percent frequency of species occurrence components of the winter / spring flora at FE during by month, and during 3-unit and 2-unit operational 2-unit operation, occurring in 26% of collections periods by station. The total number of species there prior to Unit 3 start up. However D. contorta  ; collected and identified in 1996 was 101. This total has not yet been collected at FE during 3-unit was within the range of annual totals for 2-unit (81- l operation (Table 1), and P. stricta has occurred in 111; period total of 136) and 3.u iit (88-105; period  ! only 2% of the 3-unit collections. Other cold-water i total of 141) periods. No new species were noted in species (Ulothrix flacca. Afonostroma grevillei, 1996. By restricting our analyses to only the four Protomonostromo undulatum and Spongomorpha currently sampled stations,150 algal species have arcia) occur occasionally at FE during the 3-unit been collected since 1979; this total is only slightly period, but much less frequently than during the 2-lower than the 160 species reported last year unit period. By contrast. a number of warm-water I (NUSCO 19%), when all nine stations were seasonal species have become more common at FE sampled. Of the ten ' lost' species, only Laminoria during 3-unit operation. Among these species are digitata had occurred as anything other than a trace Callithamnion roseum. Grinnellia americana. Dasm f=nt of our flora; L. digitata had been baillouviana, Hmcksia mischelliae, and Br>vpsis relatively common, but only at the Twottee Island hypnoides. ('IT) sampling site. Occurrence patterns of perennial species at FE One effect of a warm-water discharge on the have also changed during 3-unit operation. local macroalgal community may be seen as a shiA Specifically, we have docununted the establishment in seasonal occurrence of species characteristically of populations of species with geographical found in either warm-uter or cold-water periods of distnbutions which extend into warm temperate and 184 Monitoring Studies,1996

l TABLE 1. Qualitative algal collections (Mar.1979 - Jan.1997) by month, and by station during 2-unit (3n9-1/86) and 3-unit (3/86-1/97) operating periods. Values represent number of times found, as a percentage of possible times found. A dash before a species indicates that it was included in collections made in the latest report year. Taxa enclosed in quotes are, or may be, conspecific or subspecific forms, or alternate life-history stages; see text for additional detail.

                 *Ihe last three columns represent 2-unit. 3-unit, and overall study summanes ('T" =present, but <l%).

by month 2-unit 3-unit Summanes Rhodophyta 2MM 2 E N E M h2 E E E ME E MM i! j

  .Stylonema alsidii                        4 3 0 7 23 7                  12 12 0 10               9 8 0 9                  9 6 7 Ny;.,. i vudum ciliare                29 14 14 10 36 38               31 29 17 29             35 27 3 21              27 22 24 Erytl e..dda carnea                      3 3 3         1 12 6           0 7 0 2                11    5 5 3              3 6        5 Erythrocladia subintegra                 1 0 0 0            1 4        2 0 0 0                  3 0 3 0                  1    2     I Erythropeltis discisera                                                                                                                  i 3    1   0     1   6 7        to 5 0 0                  6 0 2 3                 4     3    3
  -Bangia atropurpurea                   70 86 36 12 30 49               33 43 50 50             48 50 55 44             43 49 47
  -Porphyra leucosticta                  71 75 58 19 10 30               31 36 46 29             50 52 56 41             34 50 44
  -Porphyra umbilicalis                  55 80 88 51 30 41               52 69 46 62             76 33 74 44             59 57 57
  -Porphyra linearis                       6 6           0 0 0 1                     0 0 0 0                  0 0 11         3        0 3        2 Audouinella purpurea                    3 0 3 1 1 3                   14 0 0 2                  0 0 2 0                 5  .T     2
 -Audouinella secundata                  28 23 25 20 16 14               24 33 33 24             23 14 21 1I             28 17 21 Audouinella deviesii                   4 3      3 4                    2 7 4 1    3                      5        3 2 2 3                 5 2 3
 -Audouinella saviana                    10 16 19 4 13 17                12 17 21 10             18 17 3 14              14 13 13 Audoumella sp.                          0 0 0              0 0 1                0 2 0 0                 0 0 0 0                  1    0 T
 -Gelidium pusillum                      30 20 16 20 23 32                0 7 0 0                45 83 0 15               2 36 24 Nemalion helminthoides                  0 0 0 6 0 0                     0 0 0 0 Bonnemaisonia hamifera 0 0 6 0                 0 2        1 1  4 9 13 0 1                  0 0 0 19                0 2 2 15                5     5    5
 -Agardhiella subulata 22 17 12 20 26 23               26 5 8 21                82 2 0 6                16 22 20 P:lyides rotundus 3 3 6 10 9 7                    5 2 13 21               5 2 5 6               10 4         6
-Cystocloniumiw.ipu m..                 71 58 71 49 17 49               57 62 67 64 Gracilana tikvahiac 5 62 55 68            62 . 47 53 13 6 3 1 10 9                    2 0 0 0               41     0 0 2              1 11 7
-Ahnieltia plicata                      35 35 33 36 23 30
-Phyliophora pseudoceranoides 71 24 50 52               3    3 42 41          49 22 32 25 9 4 10 4 14                   10 14 8 40               3 3 5 15              19 6 11
-Coccotylus truncatus                    7 13 10 6 6 9
-Chondrus crispus 5 10 13 21              2 2 5 18              12 6 8
                                        %% % 97 % %                     79 100 100 100          88 100 100 100          94 97 %
-Mastocarpus stellatus 57 51 49 52 51 65                19 36 % 69               0 29 97 100           50 56 54 Rhodophysema georgii                   0 0            0- 0 0
-Com!!ina officinalis 1                      0 0 0 2                 0 0 0 0                  1    0 T 74 67 70 70 72 72               98 5 100 83             95 2 95 97             68 72 71
-Dumontia contorta                      23 70 75 9 Olaisiphoma capillaris 1     1    26 40 38 40              0 44 30 32             36 27 30 i   I   4 0 0 0                2 0 4 0 Choreocolax polysiphomme 3 0 0 2                  1    1     1 12 12 6 9 1               0      5 14 0 2                0 20 6 2                6 7 7 Hildenbrandia rubre                     0 0 0

-Palmaris palmata 1 1 0 0 2 0 0 0 0 0 2 1 T T 14 22 22 22 6 14 10 36 13 40 5 11 12 18 26 11 17 -Champia parvula 26 12. 4 54 72 59 33 38 8 64 41 23 33 52 39 37 38

-Immentaria baileyana                    3 0 0 9 52 7
-Lomentaria clavellose 19 21 4 19              15     9 0 11           17 9 12 6 7 6          1 3 4            2 7 0 7                 0 2 2 15                5    5     5

-lAmentaria orcadensis 0 0 4 0 -Antithamnion cruciatum 1 1 0 2 0 5 2 0 0 2 2 1 1 30 1 10 49 43 45 29 48 33 64 20 17 17 33 45 22 30 Antithamnion pectinatum 54 29 23 41 54 55 Cdlithamnion corymbosum 0 0 0 0 73 45 85 64 0 67 43 0 0 0 0 3 3 0 0 0 7 2 0 0 0 2 T 1 -Callithamnion roseum 4 3 9 32 19 1 -Callithamnion tetragonum 1 21 10 17 10 30 3 2 5 14 10 11 36 19 13 12 23 33 48 40 54 48 2 2 20 I4 47 9 23 Callithamnion byssoides 0 0 0 4 0 0 -rallithammon baileyi' 0 2 0 0 0 3 0 0 1 26 12 32623 38 0 2 0 2 5 39 50 36 1 1 33 21 1 -Caramium desiongchampu 3 0 1 0 4 3 -Ceramium diaphanum 0 5 0 0 0 5 0 $ 1 2 2 Ceramium nodulosum 1 0 0 25 38 9 5 17 4 24 6 9 5 26 13 11 12 86 81 87 93 86 86 88 88 88 95 71 91 80 94 90 84 86 Ceraminin cimbncum 0 0 0 0 1 0 0 0 4 0 0 0 0 0 1 0 T Rocky Intertidal 185

 ~ - .-.            .      - - . ._. -_      - - . - - - - .- - - - - - _ - - -                                 _                     -. - - .- ...

1 ( TABLE 1. (cont.) by month 2-unit 3-unit Summaries Rhodophyta IMM J S H E E ME E E M ME E EEM { i

               -W'                con repens
               -Spyridis filamentona 52 26 20 38 43 57                          17 50 33 69        8 52 26 64         43 37 39 l

1 0 0 3 19 9 0 14 0 0 0 20 0 5 4 6 5

              -Scapelia pylsissei                          0 3                      I 3      0 2 4 7 1                       1 2   0 0 2            3       1       2 Griflithsia globulifera            0 0 0 0 4 0                               0 0 0 0           0   0 0 5            0       t        1
              -Onnnellis amencanum                 3        1       0 7 13 19                2 0 0 12
              .Phycodrys rubens 26     0 0 11          4 9 7 0 4 7 3 4 3                               2 0 0 19         0    0 0 9            6 2 4
              -Dasya baillouviana                12 0 0 17 46 36                            12 10 4 17       42 23        8 18    11 23 19
             -Chondna seddolia                     0 0 0 3 7 0                               0 5 0 0
             -Chondna baileyana 0 6 0 2               1     2 2 1       1        1       1 19 1           0 7 17 5         2     8 5 0          6 3 4 Chondna capillaris                 0 0 0                            6 0       0 7 0 0
             -Polysiphonia denudata 1

0 3 0 0 2 I I 4 0 1 3 3 3 2 0 0 5 8 0 2 2 2 3 2

             -Polysiphoma harveyi               36 16 14 46 51 39                          64 52 58 57      32 12 20 17          58 20 34 Polysiphoma lanoes               78 72 74 70 70 71                          45 88 100 79       0 85 100 95        75 71 72
             -Polysiphoma nigra .                 6 14 20 0 0 1
            -Polysiphoma fuccides 00 0 14          11     9 0 15         4      9 7 26 14 16 28 30 23                           14 19 4 55      II 24 2 50           25 22 23
            -Polysiphome stricta                28 42 58 13 6 3 Polysiphoma elongata 26 43 29 38        2 18 23 35         35 19 25                 ]

1 0 0 0 0 0 0 0 0 0 0 0 0 2 0 T T J Polysiphonia librillosa 3 0 0 , Polysiphonia flexicaulis 1 1 0 2 5 0 0 0 0 2 0 2 T I 7 0 0 0 0 1 0 0 0 0 2 2 2 5 0 2 1 Polysiphonia novae angliae 91 61 45 91 88 94 l 69 64 71. 71 100 77 77 82 69 84 79 '

            -Rhodomela confervoides               3 13 6 0 0 0                              7 10 0 10         0 0 3          3    7 2 4 by month                        2-unit Phaconhvta 3-unit        Summaries Ectocarpus fasciculatus 2MM i S H                               E M ME E        E E ME E MMM 4 14 33 12 22 22                          26 12 38 24       3 20 15 21         23 15 18
           -Ectocarpus si?iculosis 17 46 52 48 29 32                          40 50 25 43    21 45 29 45           41 35 37 Ectocarpus sp.                      3 4 0 3 3 0 HincLsia granulosa 5 2 13 2         0 2 2 0              5      1       2 3 6 13 3                         1   4     2 5 8 5        14      5 2 2          5      5       5 j
           -lhocksia mitchelliae               16 720224326                                                                                               l 36 12 8 19    .47 21 9 18            20 24 22 Pila >ella littoralis             12 20 42 10 9 7
           -Spongonema tomentosum 7 57 8 21        2 33 0 12          25 12 17 12 39 25 0 0 0                              14 19 17 7     11 12 9 15            14 12 13 Acinetospora sp.                     0 0                     0        1 0
          -Ralfsia verrucosa 1                         0 0 0 0          0 0 0 3             0       1 T
                                               $5 52 49 68 71 67                          64 71 50 69     11 77 67 76           65 58 60 Elachista fucicola                57 70 81 83 68 57                          60 71 63 64     59 76 73 79           65 72 69 Halothrix lumbricalis                0 4 4 0 1                            0
          -Leathesia difformis 0 2 0 5         0 3 2 2              2 2 2 0 0 49 57 0 0                             14 10 33 14       8 27 21 18         16 19 18
          -Chordena flagelliformis              0 1 23 29 16 1 Sphaerotnchia divaricata 5 14 29 21       3 6 18 11           16 9 12 0 0 3 0 0 0                                0 0 0 2          0 0 2 0              1 T T Eudesme virescens                    0 0 1 0 0 0 Pogotrichum filiforme                                                           0 0 0 0          0 0 0 2              0 T T 0 4 3 0                          1 0       0 5 0 7          0 0 0 2              3 T 1 Phaeosaccion collinsii               0        1      0 0 0 0                    0 0 0 0
          -Pimeteria tenuissima                 0 12 3 0 0 0 2               0 T T 1      0 0        0 5 0 2         3 5 0 5               2      3        3 Punctaria latifolia                 6 10 6 1                        0 Punctaria plantaginea 1    2 0 4 7         8 3 2 6               3      5 4 1        3 4 7 3 3                         0 7 8 7         0 9 0 2               5 3 4 Petalonia fascia .

4cytosiphon lomentaria 75 86 93 48 3 41 55 55 54 76 44 59 58 62 61 56 57 33 97 % 71 0 13 45 74 50 62 42 48 52 48 59 48 52

         -Desmarestia sculcata                  3 6 9 10 3 12 C--..          is vindis 2 0 4 17        0 9 2 20             6 8 7 0 25 48 3 0 0                              5 10 21 14      3 12 15 23          11 13 13 Chorde filum                          0 0 9 14 0 0
         -Chorda tomentons 0 2 0 14        2 0 0 12             5      3 4 0 4 10 3 0 0                               5 0 0 to        0 0 2 8             .4      2 3
         -Lanunana longicruris                14 6 13 9 7 7
         -Lanunana sacchanna 0 10 17 12      2 8 8 23             9 10 9
         -Sphacelaria cirrose 49 52 84 83 58 52                            60 74 63 64    39 59 74 74           65 62 63 54 29 39 39 45 52                            57 48 13 17    83 67 9 29            36 47 43 Sphacelaria rigidula                  0 0 0 0                          1    3     0 0 0 2        0 3 0 0               1      1      'l 186 Monitoring Studics,19%

__- ... - .- . . _ _ _ _ _ - . - .~ .

                                                                                                    - ~ ~                  . -.          .. - _. .

l i TABLE 1. (cont _) by month 2-unit 3-unit Summaries Phaeophvta

            .Ascophyllum nodosum 2MM                     2 S H     E E ME M         E E ME E M M 121 91 94 94 94 91 91                    79 100 100 100    68 100 100 100        94 92 93 Jucus distichus s edentatus          7 13 6 1 4 0                       12 5 13 0          2 2 12 3               7 5 5                 l
           -Fucus distichus : evenescens        3 12 9 1 0 3
          -Fucus spiralis 10 14 0 to         0 2 5 2                9 2 5 1
          -Fucus vesiculosus 9 4 9 9 10 6                         0 7 4 2        -0     5 36 0            3 10 8                )

97 99 99 100 97 97 81 100 100 100 100 100 100 100 95 100 98

          -Sargassum filipendula 13 12 12 12 13 13                       0 0 0 0        77 0 0 0                0 19 12 by month                2-unit                                                      ;

Chlorophvta 3-unit Summaries

         -Ulothrix flacca 2MM 2 S H E M ME E EEMM                                                  M M 121 l
         -Urospora pemcilliformis 41 68 36 6 7 20                        38 40 29 45      14 32 30 21           39 24 30                 !
         -Urospora wormskjoldii 68 75 25 3 6 32                        31 36 42 38      30 33 33 39           36 34 35
           'Urospora collahens' 14 19 10 9 1 3                        10 2 4 7         23 8 5 11               6 11 9 4          3       3 3 0 0 Acrochaete viridis                                                        5 5 4 2           3 0 0 2               4      1   2 1         0 0 ~l            0 0      0 0 4 0          0 0 0 2
         -Protomonostroma undulatum           9 74 74 1 0 1                                                                  1 T T
         -Monostroma previllei                                                    26 40 25 36        8 29 27 29          33 23 27 7 48 41 0 1 3                       24 29 21 19        2 15 18 17
        -Monostroma oxysperma                 0                  0 0 0 0                                                 23 13 17
        -Spongamorpha arcta 1

0 0 0 0 0 0 0 2 0 T T 7 38 64 9 0 1 19 26 33 19 8 17 27 20 23 18 20

        -Spangamorpha aerugmosa 3 6 46 3 1 0                          7 12 17 7        2 12 17 9 todiolan gregarium'                 0 0 0                        0 0                                           10 10 10 Capoosiphon fulvescens               0 0 1

0 2 0 0 0 0 0 0 1 0 T 1 1 1 0 0 2 0 5 0 0 0 0 2 0 Capsosiphon groenlandicum 0 4 0 1

        -Blidingia numma 1    0 1     2 2 8 2           0 0 0 0               3 0         1 Blidingia marsmata 70 72 84 67 72 61                     71 62 67 57      85 56 85 74         64 75 71
       -Enteromorpha clathrata 1         0 0 6 0 1                  5 2 0 2          0 3 0 0               3       1 Enteromorpha flexuosa 4          1 10 35 41 6              29 24 4 29       11 15 3 20           23 12 16 1
       -Enteromorphs intestmalis 54 52 55 58 68 72                     57 36 33 60      88 58 58 64          48 67 60 17 30 42 49 38 16                     38 52 38 60      17 27 26 23          48 23 32
       -Ents eImza                          68 62 87 83 86 75                     64 50 75 71 J
       -Enteromorpha prolifera                                                                     89 76 88 83          64 '84 77 Enteromorphs torta 22 20 25 20 29 23                      31 48 38 48       8 17 5 23           41 13 23 l

Enb.,csdia ralfsii 1 0 4 6 0 0 0 5 0 2 0 3 0 5 2 2 2 Percurnana percursa 0 0 0 10 4 0 2 5 0 5 2 5 0 2 3 2 2 0 0 1 I 3 0 0 2 0 5

       -Ulva lactuca
                                           % 81 91 97 % 97                                          0 0 0 2              2 T 1 Presiola stipitata                                                      98 98 % 95        83 92 94 94          97 91 93
      -Chaetomorpha linum 16 14 14 19 19 17                      0 48 4 0         2 71 0 0            14 18 17 46 22 38 84 80 67'                    71 50 88 74       14 64 61 58         69 49 56 1

Chaetomorpha melagonium 0 0 1 0 0 0 0 0 0 0 0 2 0 0 i Chaetomorpha seres 65 45 49 57 58 61 0 T T Cladophora albida 71 24 42 50 71 48 62 61 47 61 56 0 1 7 9 10 0 Eladophora flexuoss' 7 5 4 17 2 5 0 3 9 2 5 Cladophora glaucencens' 12 1 22 49 33 7 14 2 29 24 17 21 27 29 16 23 21 0 0 1 'l 0 0 0 0 4 2 Cladophora lastevrons 0 0 0 1 0 0 0 0 1 0 T 1 0 0 0 4 2 0 0 0 0

      -Eladophora refracts' 9 4 7 13 16 4                                                                 1     0 T Cladophora aernea                                                        17 72521          6 2 3 8             17 5          9 13 12 51 43 29 25                     48 24 17 40      26 26 24 27 Cladophora crystallina' .

0 0 0 0 3 0 34 26 29 Cladophora hutchinsiae 0 0 0 0 0 2 0 2 0 1 T

     -Cladophora rupestrts 6 4 7 7 7 12                         12 5 8 12         3 8 11        3       9 6 7 Cladophore ruchingeri 1         1        6 9 4 4            2 5 0 2          9 2 6 5               3 5 4 1         !        4 16 14 6
     -Rhuoclonian riparium                                                        0 0 0 0         23 2 6 15             0 11         7
      'RhimcInnimn kemeri' 9 16 17 36 22 4                      19 31    4 26     6 32 5 17          22 15 17 O 3                1'   1 0 0         0 2 0 0
      'Rhuacionium tortuosum'               O O O O O 1                                            0 3 0 2               1      1    1 hopsis plumosa'                                                             2 0 0 0          0 0 0 0                      0 T
     -Bryopsis hypnades 12 0 3 7 14 12                        21 10 0 5        20 2 2 5 1

10 7 8 4 0 9 19 -6 9 5 5 4 0 18 9 2 12 Derbens marma 6 3 1 1 1 6 3 10 8 Codium fragile 24 5 0 0 2 0 0 0 8 T 3 94 81 86 93 97 93 100 83 92 81 98 82 88 98 89 92 91 Rocky Intertidal 157

tropical regions, and are therefore tolerant of 3-unit plicata and their associated epiphytes (e.g. , , temperatute regimes at FE (e.g. , Gracilaria Ceramium nodutosum, Protomonostroma undulatum, likvahtae, Agardhiella subulata and Sargassum Elachista fucicola Cvstoclonium purpureum and filipendula; Taylor 1957; Lening 1990). Similarly, Polysiphonia lanosa) were eliminated. while some species near the southern limit of their normal opportunistic species had become dominant. geographical ranges, such as Mastocarpus stellatus Elevated temperature conditions at FE were more and Polysiphonia lanosa, experienced population consistent in subsequent years comprising collections elimination at FE during 3-unit operation. in the fourth group (1987 %). These conditions Other changes to the murall flora that occurred allowed for more long-term development of the during the 3-unit period appear unrelated to power unique flora now obsensd at FE, characterized by plant operation. Increased occurrence of Gelidium shifts in temporal and spatial species occurrence pusillum has been observed at FE (absent from patterns described abmt. Similar floristic shifts collections during 2-unit operation, but present in have been obsened by other researchers studying 45% of the 3-unit collections). This increase of G. attached algae near thermal effluents (Vadas et al. pusillum also occurred at sites farther from the 1976; Wilce et al.1978; Schneider 1981). It is discharge, including a pronounced increase (from particularly interesting to note that the algal 7% to 83%) at our control site GN. The area-wide community that has developed at FE in response to introduction of an exotic species, Antithamnion elevated water temperatures has persisted. ct least pectinatum, also occurred during 3 unit operation; it qualitatively, through 19%, despite the extended was not collected prior to 1986, but has been found outages of all three units, in 67% of the 3-unit collections. Community analyses, based on annual collections at each station, also reflect both site-i Abundance Measurement specific and area-wide changes to the algal flora. For instance, groupings of collections at GN, MP and WP (Figs. 3a, b and c, respectively) separate into Along with the changes of patterns in species

                         'carly and later sampling years, with the point of          occurrence discussed abmc, exposure to the MNPS separation around 1986-1988. This separation was          thermal plume can change patterns of species certainly influenced by the increasing contribution in    ,w.ma.ae and community dominance hierarchy in recent years of species like Antithamnion pectinatum,    rocky shore communities. Changes in species mentioned above. However, even with such floristic        abundance and altered patterns of zonation at separation, all annual collections at these three           mpacted sites may not be evident when descriptors stations (GN, MP, WP) clustered at greater than 50

• of qualitative community characteristics (as 60 % similarity, indicating a high degree of presented in the presious section) are used.

consistency in the year to-year floral assemblages at Therefore, quantification d species ah==d=* and q these sites. distribution pauerns, through determination of - percent substratum coverage, complements in contrast, the overall similarity of annual qualitative algal studies. collections at FE (Fig. 3d) was only 35%; four groupings are apparent at about the 60% similarity Abundance measurement studies were t'esigned level. The first group represents collections made to sample species abundance mer an area sufficiently during 2-unit 1-cut operational years (1979-82), large as to accurately describe large scale patterns of when the unimpacted flora at FE was similar to that abundance in each intertidal zone (high, mid and obsened at other exposed stations. Temperature low) at each sampling site. Among-station conditions were severely altered when the second differences in abundance patterns are then related to quarry cut was opened in 1983, and account for the site-specific physical and biological controlling characteristic disturbed, or early successional stage, mechanisms including for stations near the MNPS flora (e.g., opportunistic Enteromorpha and discharge, exposure to elevated temperature regimes. Polysiphonia spp.) collected at FE in the second Following subsections describe abundance patterns group (1983 and 1984). By 1985 and 1986 (the of important intertidal organisms, i.e., barnacles, third group), populations of Chondrus crispus, Fucus Fucus, Chondrus and common epiphytes, along with vesiculosus. Ascophyllum nodosum and Ahnfellia analyses d overall community structure. 188 Monitoring Studies,1996

_ . . . . _ _ .- ._ ___ _ . . - . _ _ . _ _ , . . _ _ _ . . _ _ _ .. ..__._.m _ .._.m . _ . . . _ y, .

                                                                                                                                  - 30
         . % 'n
                                                                                                                                  - 40
         +E 50- -a) GN
                                                                                                                                  - 50 60 -                                                                                                                  '
                                                                                                                                  - 60 c 70                              {
                                                                                                                                  - 0
           $                         N                    !                                                 '

g 80 - r i 30 90 -

                                                                                                                                 - 90 9.),    6, 96, 9, 6 S    e ,6# 6, SG, 6      S,#     *&g #s, 6       #, #s, #sg         9 #9 , S, 9(

he

• I.

40 - MP -' 8 SO - - 50 60 -

                                                                                                       - 60 c 70 -

8- 1

                                '             l
                                                                                                       - 70 g 80 -                                               ,          i                  1
                                                                                                       - 80 90 -
                                                                                                       - 90
                  % % % % % % % % % % % '"s, % % %

30 -

                                                                                                                              - 30 90 c) yp                                                                                                               - 40                                                '

E 50 - E

                                                                                                                              - 50
       -f0-
                                                                                                                              - 60 g 70 -            l                   ]                                     ,                                         - 70 8 80 -                                          I
                                                            ,          I                             "

i - 80 90 -

                                                                                                                              - 90
                 's,,  s 's,, 'se , 's, 's,, 's 's, 's,, % 's, %, % 's,, 's,                                's, 's, 'sg 30 -
                                                                                                                                ',0 40 -

l

                                                                                                                            - 40 '

E SO - E - %0 s .60 -

     ~                                                                                                                     - 60 g 70 -

l l - 70

       $,60 -               1 l                                   1 80 90 -
                                                                                                                           - 90
                   >,%,%,'D        e      ,%,%,%,%, Q %,'%,%,%,%,%,%,%,%,

Fh 3. Chanenes ' " . _.. af peroam anularey ofqualaative algal h ley year, at My M WWh Rocky Intertidal 189 F B __ w - -

1 1 l Barnacles However, absence of barnacles from Zone 3 at FE in September 1996 was not the result of thermal stress, i Barnacles (primarily Semibalanus balanoides) are the dominant invertebrate on local rocky shores, as all three units were shut don. Rather. this loss. as well as the reduced recruitment in Zone 3. and are most abundant in the mid intertidal zone particularly since 1989, is related to long-term (Zone 2). Barnacles exinbit an annual pattern of community development during MNPS operation. abundance mated by reproduction and settle.nent in Specifically, emironmental condition at FE since early spring, rapid growth and surface cover Unit 3 start-up have allowed for establishment of an- ' increases in summer, pad decreased abundance extensive low intertidal Codiumfragile population at through autumn and winer due to competition for FE. This population, which persisted through the space, predauon and physical disturbance (Connell 1995-% sampling year, competitively excludes 1961; Menge 1976; Bertness 1989; NUSCO 1993). barnacles through preemption of habitat space The barnacle annual abundance cycle described (Underwood and Denley 1984; NUSCO 1993). abow was observed at all Millstonc study sites, in all

     . three intertidal zones in 1995 96 (Fig. 4)J Maximum                                                                  l Fucus                       1 barnacle cover in the high intertidal (Zone 1) during 1995 96 ranged from 11% (GN) to 49% (FE).                                                                           I De dominant alga on local shores, the Minimum coverage in Zone I ranged from 1% at                                                                         i rockweed Fucus vesiculosus, forms an extensive GN to 10% at FE. In the mid intertidal (Zone 2),                                                                     s canopy over barnacles in the mid intertidal zone, and maximum barnacle cover wm lowest at WP (43%)also occurs in high and low intertidal zones. Other and highest at MP (80%); minimum cover was species of Fucus included in our abundance lowest at FE (4%) and highest at GN (38%). Low                                                                        i estimates are found occasionally at our study sites, intertidal (Zone 3) maxima during 1995-% ranged but contribute relatively little in terms of percent from 5% (FE) to 64% (MP). The annual minimum               substratum coverage.          These species include F in Zone 3 was lowest at FE and MP (0%), and highest at GN(5%).                                        distichus subsp. edentarus, F. distichus subsp.

evanescens (both occur mostly subtidally) and F. Seasonal barnacle abda~ patterns at all spiralis, which occurs in the high intertidal. study sites, except FE, have been relatively Intertidal Fucus distribution patterns and consistent, and are the result of the temporal stability seasonal abundance cycles in the MNPS area are of emironmental conditions at these sites. similar to those reported elsewhere in New England Variability in these conditions among stations is (Lubchenco 1980, 1983; Topinka et al.1981). At considerable, however; natural site-specific factors most study sites, Fucus abundance typically peaks such as degree of site exposure to wind and waves annually in late summer or autumn. reflecting high - and slope of available substratum appear to be the recruitment and growth rates prior to and during this most important mechamsms controlling barnacle period (Fig. 5). Maximum abundance during 1995-cycles and patterns of zonation (NUSCO 1993).

                                                                % in Zone I was greatest at FE (57%), with maxima in addition to natural factors, thermal plume at the other stations ranging from 1 to 20% (Fig. 5).

effects accounted for temporal and spatial changes in barnacle +=h= Highest Zone 2 cover during 1995-% occurred at Due to the influence of tides on MP (74%); maximum cover was least at WP (39%).  ! the thermal plume, effects of thermal increase are in Zone 3, maximum Fucus cover during 1995-% most notable in the low intertidal (Zone 3). Zone 3 was greatest at GN and WP (17*/.); the lowest barnacles are exposed to elevated discharge abundance peak occurred at FE (9%).  ! temperatures for 9-10 hours each tidal cycle during Fucus *=6= 3 bit operation, whereas barnacles in Zones 1 and 2 patterns varied among study populations, reflecting emironmental conditions experience a tidally-indumd refuge from maximum unique to each site. In general.. Fucus is most 3 unit thermal plume incursion, as they are exposed to air during most of this time. Previous reports abundant on moderately exposed shores, common emironments at most of our study sites. Fucus (e.g. NUSCO 1996) suggested that elevated temperatures directly unpacted low intertidal abundance is limited at highly exposed sites by physical stress from wave shock, while at sheltered barnacles by causing complete population mortality sites these species are often outcompeted for space by in late summer every year since Unit 3 start-up. another fucond, Ascophyllum nodosum (Schonbeck 190 Monitoring Studies,1996

               . . .                                                          ~                       -            - _ .         .-                      - .                         _.                                 . . _ .

I 100'

                      '                                                      2nd cw,                    Umt 3 s' ort vF
                      ,                                                      opeMd 70-
                                        '                                                                                                                                 ?
                           *'..             =

I 1, R .O. > ,'Q r

                                                               \ g\                                   I
                                                                                                                            '\

• fili &Aikih3%h,Wks tkM b' \.K(k uA P5[ RAuR(5RAuP($RAu($A u( $A u( A 5 uA $ [uA$ (uAsu su[ A su( su su (su 5( A u (5Au (su s(

R k R P( R (A R PA R {V A P PA R PA R RP &RPRP RR RP PP RP 979 70 80 8 t81828rone 283183848e85858 68 6rone 8 78 78 88 88 98 98 09 09 19 19 29 29 39 39 49 49 59 59 69 69 2 ror e 3 pre dd ers 100-40 GN Umt 3 stort-wp 80-a

                                                                                                              /\ W             g                                                                               ,

s 2 I 10- f \ g/ ( ' l 4- + . ' \r

                                                    ~         V
                                                                        , s;          ff \y( ' ,\jf /,,I
                                                                                                    \
                                                                                                                    \jv            j  \            \           'y'^ v                         l \           \       l y       2-g.

y ,fj... - - . , . . * ** a

                                                                                                                                                                             %\gl I

• i,l I \ /-

                                                                                                                                                                                                       \

O ^

                                                                         ^
                                                                                  -*' ^                                                         *            *
                                                                                                                                                                           * * ' * - *             *- --*             '=

u A Psusu (R A O( A 5 (uA5 (uA5(uA5( uA5( u 5 u $ u 5 u 5 w su su su 5&u (sn su s( R P R P R P R PAR(PAR(PAP[PAR( PAR( PAP( PAR( PAR[ P RA F R[ A 7 ' P 9 0 08 08 18 RP FRP 8 8Fone 88888888888888899999999999999 1 2 2 13 3 4 4 5 5 6fone 6 '27 8 8 9 9 0 0rone 1 31 2 2 3 3 4 4 5 5 6 6

                                                                                                                                                                                           - PrecoM t 100<

i 90-80-Mp Unet 3 atort-wp 10f ' 60' f l

    =

so-40,

                                                                                                                                    \                                /

A l ps \ { 30 jj s .,{ . gg liF $ I ' f. j g f ( ,g

                                                                     ,k.     ,,,,h y                                                                                                    -l                             [ ;7 L                  ll 4
            '                                          fA                               j g,         I i                 -I \ l i i
                                                                                                                            \..}

I*fy .I

                                                                                                                                                                             *l
                                                                                                                                                                                ../\.,.'f

( g ,,['\ ,II 0- . OR), . .jI \e'r! .- -D .'d .gL.i1lfz-lQYlplkJg!lQg g j j u5 u 5 u su 5 u 5 u 5 u su su - w- s"u su 5 u !--u -suL su - ' tu5us l AF R ER A PE RA PERAPERAPf R AP ER A& I AR C GA R (AAREPARCPAP [PAR( P AR E PA RE PAR( A PR CF A RE P 7? a 8 0 80 81 81 82rone 9----- 82 8 1 3 83 88 84858586868787888889899090919 19 ;9; 93 93 94 9d 9* 9S 9F 9C f one 2 rone 3 - Pro dotore 100< Unit 3 stori-wP 80- ! TO-l 60-50-m 40- [ 30- f\ [ u 20- * /\ t, f\ \\ r ./ f 10- A l.1 f, \ l\ l? g l, \ l ;g ' .g 3 ,y fy ? >,I \\ s p r

                                                                                                                                                                ) .,'

4 ,\ .ll\ e-l\ / r-9.\ P 1s s *t

                                     'N' Y                                                    ?j                          '

Y 'I h u AP

                        $C uA ~su   ER su 5 u 5 u su $ u 5 u 5 u $ u su su 5 uA5(uA$CuA5t uA sEuA'sT A E A PE RA PERAPCPAPERAPERAPCRAPERAP ER AP ER A& E R

9 9 R 0P 0 EPRPRPRAC&RP 7 7 8 8 8 81 82rone

                          --- -                            82 8         888888888888899999999999999 13 3 4 4 5 5 6 6 7 7 8 8 9 9 0 0 1 1 2 2 3 3 4 4 5 5 6 6 rone 2                                         tone 3            -

Predators F5 4. Abundance of bunacles in each zone, and of predatory mails m Zone 3, of uniksturtud tranm.ts, from 3/79 to 9 % Rocky Intenidal 191

       - .              .~             .. - _ .                     . . . . _ .          - , _ . . .                      _ _ - -                    .-             __         . . . - . . .                        . . - . . _ . _ . _

100-2nd cut un*t 3 stort-up

                   ;  M                                             openeo Q. . -

n 40 , " ,

                                                                                                                                                                                                    / . .
                                                                                                                                                         .~~                4
    'bh h: .,_                    '

5, r 10 N ' D ) -p }y. ,\7\ /A c-

             ,.         e) %.) %                                              9    ..
                                                                                                                         ! t "\ _a. \/\ !~'                                                    \l, u 5 u5 u 5 u 5 u5 u 5 u5 u 5 u 5 u 5 u 5 u 5 u 5 u 5 u 5 u 5 u 5 u ,

AP (A P tPAP(PARE&AD( PA PCPAPCPAREP AR E &A RE PA EW AC AP E RA P( A R fPAWE&AREPAPEP P PP PR 7 97 08 08 08

                               ----~   18     18   28       8 38 2rene1    J *8 48 58 58 68 8une             8 78 78 88 88 98 98 09 09 19 19 29 29 39 39 49 49 54 59 69 00 2                                rone 3             -

qvareas 1005 wv- un t 3 start-up 80- @ 70 80-50 fg / ^ ,ou )'  ?

                                                                                                                                                                                                             /s N

20- [\ jh \',A -1

                                                          .WY
                                                                                                                                                                                           ^/g
                     /v'QY I

Y 102 v e '" s " i e * * '/pg 'sN . . -/ \. g s

                                                                                                                                                                               .            \.        **   ..

O- [ *y .  %* *

                                                                                                                                                           , w
                                                                                                                                                                     ~
                                                                                                                                                                                 %          .,           %, ' e u

R 5 A [ u A 5 [ u A 5 [ u A 5 ( u A 5 [ u & 5u r, A5(Fu AP 5 (P u A& 5 [P uA9 5 ( D uA & 5 ( Pu [5 Au [5 u 5 u 5 u 5 u 5 u 5 A VPPRPRRRPRPW 5 5'

                                                                                                                                                     & P P      A ( A [ A ( A ( A (

P P P & P P P P 5' 9 9 0 0 1 1 7 2 3 3 4 4 5 5 6 6 7 7 8 e9 9 0 0 1 1 2 2 9 7 7 8 8 8 8

                              * *" - *

• rene I 8 8 8 8 8 8 8 8 8 8 rone 2 8 8 8 8 8 8 9 9 rone 3 9 9 9 9 3 93 94 94 95 95 9t E 9

                                                                                                                                                                              + 9torers 100-90 M.

yp un.t 3 stort-up TO-f.0 - 50- T/g V

                                                                                                                                                                                       ~

10- j' ,3 g Ar / # '\ % #

                                                                                                                                                                       /
                                                                                                                                                                                                       , h" c         !!
            ..                                       L' Q v . r J N # g.x w c o - W - : e.

u AP5f R u 5[ A A u5u5 u 5 u 5 u 5 uA5[ u5 u 5 u 5 u 5 u 5 u 5& u 5( A u 5A u5 u5 P R (PA R (PAR[PAp(pAp C p p D pAp[ pAp[p&p (pAP [PAP[RA&( R P P P( R F( A { j p RP i 7 ' 8 8 8 W ** O O 1 1 2s one 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 2 13 3 4 4 5 5 6 6 7 7 8 8 9 9 0 0 1 1 2 2 3 3 4 4 5 5 0 0 8 8 9 9 9 9 9 9 9 9 9 9 9 9 9 9 )

                                                                                                                                                                                                                                          )

tone 2 tone 3 ' grarers 100. ) 90 80-wp un,t 3 stort-wP i 70-MO-50-l ^ w

                                                                                                                                                   $\                                                 lW 3g                                   .g        i
                                           '        -'                      b                              '
                                                                                                                   ; t,.

0- 9 ' u e A& 5( A upp 5 {A u& p (5AR u (5PAup(5PAuPE5PAupE5PA up5 rFp au rE 5Au pEp 5A u [5

                                                                                                                       &          PAu pt5AupC5:AuP (5PAuh E5p A                   u (5Au E            Su5 F-                                   p p & r#Ar C 79 9    7 08 08 18 18 28 28 38 38 48 48 58 58 8 86 87 87 88 8 t,

88 98 99 00 01 91 9292 9 39 39 40 d9 5u 5o 9 9, rene 1 rene 2 f. C ron* 3 geare*= F4 5, Abundance off=cas in auch zone, and of yazing snails in Zone 3, of undamurbed transacts, from 379 to 9% 192 Monitoring Studies,19%

i I I and Norton 1978, 1980; Keser and Larson 1984). established at MP under 3-unit operatsag conditions, i At our sampling sites, Ascophyllum is only abundant occupying almost 75% of available substrata in Zonc ia transects at WP and GN. Vertical distribution  ! 2. patterns ofinterudal Fucus are also controlled by the i degree of wave exposure, as well as slope of Chondrus and common epiphytes available substratum. More detailed description of  ; the role these natural characteristics play in Perennial populations of the red alga, Chondrus l determining Fucus zonation patterns at each study crispus, form an extensive turf on most low intertidal ' l site is prmidad in previous reports (NUSCO 1992, 1993). rock surfaces in the MNPS area. Several seasonally abundant algal taxa coexist as epiphytes on { in addition to these natural site-specific Chon&us (e.g., Monostroma spp. (including } characteristics, physical stress in the form of heat l Protomonostroma) and Polvsiphonia spp.) instead of from the MNPS discharge is an important competing directly for primary space. Because low

• mechanism controlling Fucus abuwtance in Zone 3 interudal habitat is more susceptible to power plant at FE. Elevated temperatures during periods of >

thermal plume incursion resulted in virtual impacts (as mentioned in previous sections),  ! documentation of abundance patterns of Chondrus elimination of Fucus in Zone 3 each year since the and its associated epiphytes is critical to our , opening of the second quarry cut in 1983 and , ecological monitoring program. throughout 3 unit operation (Fig. 5), except for 19% (the extended outage of all three umts permitted Stable Chondrus abundance has been - documented at three of the four study sites (all but near ambient conditions for :nuch of the most recent FE) during the study period. Abundance maxima at sampling year). In other years since the opening of these sites during , 1995- % ranged from the second cut. thermal stress was most severe at FE approximately 65 to 75% Chondrus abundances { in Zone 3, because organisms there were submerged observed during 1995-% at these three sites were and exposed to elerated temperatures for much of the i within historic ranges (Fig. 6). ' tidal cycle. AAer Unit 3 went on line, thermal stress The Chondrus population at FE now consists of at mid and upper intertidal levels was substantially scattered individual plants, with abundance estimates reduced due to increased discharge velocity, and never exceeding 3% during 1995 %. Low Chondrus Fucus populations in Zones I and 2 returned to abundance has been typical for this site since 1984; abundance levels similar to those observed from estimates for that period have ranged from 0% to 1979 to 1983. Fucus has exhibited long-term abundance 14%. but have generally been <2%. Prior to 1984. cycles at other study sites more distant from the abundance estimates were much higher (40-75%) discharge than FE that are likely unrelated to MNPS and comparable to those at other exposed sites. This extensive Chondrus population was eliminated in operation. Most notable of these is the protracted 1984 by clevated water temperatures from the 2-cut decline / recovery cycle at MP discussed in previous reports (e.g., NUSCO 1994, 1995, 19 % ). Increases 2-unit discharge (NUSCO 1987). Since that time, in Fucus abundance continued during 1995-% to only a few scattered Chondrvs thalli have been observed in upper Zone 3 study quadrats. These historic highs in Zones I and 2 at MP (Fig. 5); thalli are present during cooler months, and are Fucos coverage in Zone 3, although dowa slightly typically climinated each summer by clevated water from last year, remained high relative to coverage prior to 1991. Due to proximity to the MNPS temperatures from the 2 cut 3-unit discharge. This discharge (ca. 250 m to the east) and the moderate consistent scenario indicates that any suc.cessful re-temperature increases measured at MP (2 3*C above establishment of the Chondrus population at FE ambient during slack tides; NUSCO 1994) the during the MNPS operational period is unlikely, possibility of a power plant impact has been Even with near-ambient water temperatures during the 3-unit shutdown, Chondrus did not establish or investigated, but no direct evidence exists, linking the pattern ofFucus abuwtance at MP to power plant maintain appreciable substratum coverage in 1996.  ! oPeratson De present Fucus population at FE De low intertidal community at FE is now recovered relatively rapidly aAer Unit 3 start up, composed primarily of an extensive Codmm fragile , even under much greater temperature extremes than population, persistent populations of Sargassum  ! those at MP, Fucus now appears to be well filipendula, Gracilaria tilvahlac and ephemeral algae including Ulva lactuca, Enteromorpha spp. RockyIntertidal 193 1 i I I

 . . _ . .                  _               _ . . . _ . _               .._..~m                                  _ . _ _ _ .. _ ..                                     -                 .m._..              ..        .m     _.              _          m..       _ . . _ _ _ _ _ _

i i t00' U"'t 3 start- up g FE 2ndn c.ut op a 70< .' 3 7,. EO- ' ' ' , ' ,,, ,

c. * '

50, , g; t / f\ e 40 f '

l.  ;\

                                                                                                                                                                                              \

• I\

0

e. I e

l\ i

                                                                 \

j\

                                                                                            \' l \? /\\l'(%              f j ^
                                                                                                                                                  'l l       b                                 \   r~               tl 1               >
                                                                                                                                                                                                                                                         } 1I 2-          I
                                         \

I \ \ l \ '

                                                                             \ l                         *
                                                                                                                    \
                                                                                                                     '   ll I ;y
                                                                                                                                                                                                  \1                   4 \ f\/;                           t                            t

[j t]_ . V \ _f . I 9 ' t_- t

                                                                                                                                                                                                                                                          'f)

A_ --

                                                                                                                                                                                                                                                            \h u

A 5 u 5 M 5 W 5 u 5 u 3 u 5 u 5 u 5 u 5 u 5 u 5 u 5 u su su 5 u 5 u [5 R P[ A (A [A R (P A R [9 A R (p A # P[ R A p( pAA[ RAP( RAO{ E A P( R A S[ A (A (A $ P( R A V[ A  ! RP RP RO EP RV 7 7 8 8 8 8 8 8 8 e84848S8S86868?87888889899090 91 91 92 90 93 93 94 94 9 w w 0 0 1 1 2 2 3 3Cnonorus 5506 999 Monostrome Polysionon o 100' So' Dnd 3 start-up 80- GN ' 70-60 ,

• bb. , ,

a e

                                     .    ' f f, u,
                                                                                                               =                                       , ' ..                      '

40 ' 30 , * { 20 14 f 10- A 0 Il l\ h4 p

                                                                                                                                                                 /                       ri 11                                        'l l\                       l 0                                                                I
                                                                                                       \f.- -              ?               ,$ V\ f -- -                              --

1 l - u5 A W $[ A M 5AM( 5Au5 u S u 5 u 5AM[ S A Vu( R 5 u( 5 u{ 5 A uV{ 5R AuV( 3R AuP( 5G AuV( 5R Au5 u5[

                        & P[ A           R P k P[ R                   V R D[ A          R p( pAV( RAP[ R                                PR                 A   VR    AV          R                                                                V[ A  RV                              6 79 9        7 08 08 18-----          18 28 28 38 38 48 48 S8 S                          88    68      67    87      88     88     8 98 98 09 09 19 19 09 9 93 93 94 9a 95 95 96 96 Chondrus                                                   Monostromo                                            Pchsiphonso 100:

g MP unig 3, tori.wp g ,

                                                                                                                                                           ...                             .....                   ... r . . .. - -

t so. > a 40 h f 3000-- j\ l l' 10 e lt i l\ A II [n  ;\ g S I lg 1 I\ it 7 4 \ ) f y f g Ig i \ f \ /1 7 \ I l\ j'4\

                    $                                                                         I g               g                    f 5 g,                                              , 1m .           \               \
                                                                                                                                         \         q                                                       \ l i I
                                                                                                              -\                           v                 y

__f___, m_..\. a\ _ . . . J _ ,J _ \j ___; b A 5 u R V W ( A [5Au(5Au[5Au5 { W5u5u5 u5 R P R V R ? R AV{RAD [R A [ R A (WA 5[ uA5( uA5[ uA5{ u&5(W RRRPRPRVRVRPRPRVRVRV A S( A u 5[ A u 5[ A u 5[ 7 ' 9 9 0 0 1----- 8 8 8 8 8 8 8 1 2 2 3 3Cnonorus 8 8 8 8 8 8 4 4 S S 6 6 7 7 u8onostromo 8 8 8 8 88 89 89 90 90 91 91 92 92 93 93 9 445506 99999 pai s.phwo y 100-90; 80 Kp Unit 3 stort-up 70- , 60- . ~ "

                                                 ,.,~,           ' '

i.=~~,'

                                                                                                                            '^                           '   '

50- "

          =       en.
                                                                                   'i'
                                                                                                                             " ~ .
                                                                                                                                  .                                 s
                                                                                                                                                                                                           ~*                        #     ' 

30: { 20 ik 10- / >

                                                                                                                                                                                                       \                            It          l tu I\             /
                                                                                                                                                                                                                                                           *\

l$t ik AMAI'f'\0$\A!' W',!\,l'd\,\Y l 'd u A [5 Au[5A u'[5 u 5 u 5 h 5 u 5 u 5 u 5 u 5 u 3 u 5 u : u 5 u 5 u au v u : R P R V R PA R (p A R [DAR[pAp [pAp (p A R {p A p p[ A R p( A RP [A R (VA & (RAR [pAp [pA pV ( AE V [. 7 7 9 9 0 0 1----- 8 8 8 1223344SS667788990011 O 3 3 4 4 5 9 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 9 9 9 9 9 9 9 9 9 9 9 S G 6 99 Choncrus Monostromo Polysiphc neo Fi S 6. Ah of Clummhw and ausar spaphytas in Zone 3 of undisturbed transsas. from 3'79 to 996. l 194 Monitoring Stu&cs,19%

and Polysiphonia app. and occasionally (e.g.,1995), winter /carly spring at most stations, and suitable for a seasonally heavy set of the blue mussel, A&tilus Monastroma, have rarely occurred at FE under 2 cut edults. Operating conditions. However, it is likely these Tbc presence of both warm-water and cold- condation will occur at FE in spnng of 1997. as all water samaaani epiphytes in the local low intertidal three units s;c expected to be shut down. zone prmides opportunity to document potential temporal shifts in abundance of tlase species in Community Analysis response to altered temperature regimes, typical of those at FE. Polysiphonia spp. (mostly P. novat- Local rocky shore communities are composed anglise and P, harvepl) are common warm-water of over one hund/ed species of attached macroalgae epiphytes on Chon&us, Ascophyllum and Codeum; and sessile o4r.iow-moving animals. The abundance they may also grow attached to rock. The annual and distnbut on of these species are influenced by abundance cycle of Polysiphonia spp. is complex int:ractions between physical processes characterized by a late summer peak, with cover (e.g., tidal height, exposure to waves. water declining to near 0% by winter at most study sites temperaturv) and biological processes (e.g., inter-(Fig. 6). Peak abundances during 1995-% varied and intraspecific competition for light, space and considerably from station to station (3% at GN,11% nutrients, grazing and prehtion (inclushng that by at MP,15% at WP, and 36% at FE). The annual species not normally considered intertidal cycle in Polysiphonia spp. aburutane has been organisms, such as fish and shorebirds), nowth and consistent at all stations except at FE throughout the reproductive cycles). Characterizarbn of these study period. Elevated temperature regimes at FE communities may be descriptim Lg., abundance of since the opening of the second quarry cut (1983) species whose populations are st*le or predictably produced favorable conditions for these species by variable may be representen as time-series of extending the season of occurrence and increasing percentage of substratum covr. rage, as in previous the levels of peak abundance These temperature sections. However, comper'. sons among stations, or regimes at FE have also allowed Polysiphonia spp. among years at a given s'.ation, may also be made to persist through cold water months, when such using multivariate techaiques, similar to those species are typically absent from other sites, desenbed in the Qualitative Algal section, using the including IT prior to 1983. Interestingly, during the abundance of all species found in the transects, even 3-umt shut down in the past year, Polysiphonia those that are rare or unpredictable in their abundance 'in March and May was the lowest occurrence. recorded at this site since the opening of the second Previous analyses (e.g., NUSCO 19%) have cut, and was similar to this species' abundance at the other rocky shore stations. shown that of the rocky shore stations, only at Fox Island-Exposed was the community sampled during The annual abundance cycle of Monostroma the 2-unit operational period appreciably different spp. (M. grevillel and Protomonostroma undularum) from that sampled since Unit 3 start-up. At each of can be described as being out of phase with that the other sites, similarities were highest between desenbed for Polysiphonia spp., i.e., peak abundance operational periods at the same station, indicating a is observed during cold water months (late relatively consistent species composition throughout winter /carly spnng) and virtual absence is noted the study period. dunng warm-water months (July-December; Table 1. Fig. 6). These findings are supported by the most recent This annual abundance cycle occurred sampling data as well, where Bray-Curtis similanty consistently over the study period at all study sites matrices, using annual average abundances of all except FE. Monostroma was virtually absent from taxa found in mid and low intertidal zones at each FE in 1995-% (only 0.2% in March); peak station, are illustrated as clustering dendrograms abundance elsewhere ranged from 17% (GN) to 28% (Fig. 7). At all sites but FE (i.e., GN, MP and WP; (WP) to 44% (MP). Since 1984, Monostroma has Figs. 7a, b and c). all years grouped together at a 50-been observed in FE study transects only rarely, and 70% similanty level, with most of the differences its emer has never exceeded 1% Prior to 1984, attnbutable to changes in Fucus cover, peak annual Monostroma cover at FE was similar to Annual samples at FE (Fig. 7d) were much other exposed sites, ranging from 17% to 48% more dissimilar than those at other stations; at the Prolonged low water temperatures, typical of late 50% similarity level, years grouped into three Rocky latertidal 195

i i 1 es> . ..,, jm- n) (;g . y, d) Ff: --*

                                                                                                                                                                                                                                                                              - -:0 E,,-
                                                                                                               - 3>
,            a                                                 l
                                                                           ,                                                              - >a -                                                                                                                                                -io l;                                                    I                                                         -N                       xo-                                                                                                                                         -0 E. *'     .   . _                                                                                                         =
                   '~
                                                                                                               - %)                         I0 -                                                                                                                                     - 30
                                                                                                                                       ?^                                                                                                                                            ~'

hth '*?, Ng, hs, q '*es 6, Ng, '?y '9a,s9'98, 9s ,98, *s, '9, 8 m k- -M E o- l 11  !!! - *) i l'o -w 3 ,_ _ ,,o D 48 - - 60 so - - so UI'

4. - _ 70 70 -

                                                                                                                                                                                                                                                                                    - 71
                                                                                                                                               ~--                                                                                                                                  ~

I gR)- ]1 - 80 -- J-- -

 ,            e                                                                   l                                                         w-                                                                           l                                                          -w 0 90                  L
                                                ' )'

3

             }        ,%f.!%' % %,,% % '%                              .

I ,-90

                                                                                       '% E% %. !s,I
                                                                                                                                                 'g%4%%%%%%4%%%%%%%

3 %#%s og i TABLE 2. Avarage percana naimerasuas coverage of sama wnh mean overall(Fou Islead Fvpnse a ? only, all yean) abundance >2% in youpenp deseramaed by -% analysis, youp i muinbers correspund to slause in Fig. 7d. 1 [ I saxon Orcup I Group II Group Ill Group tila Group liib s 3, . 4 T-c) WP - ,o CoJramfragde 0 95 24 98 33.41 12.21 36.ts6

'            '                                                                                                                        Focus wesacadosus                            24.33                                        2.76   25 15   27.50                      24 86
                                                                                                              - 80
                    ~
            @                                                                                                                         s,,,,6alamis 6ataao Jes                      18.94                                   15.29         9.29    5.27                           9.80 f
             +-                                                                                                                       rod                                          16.91                                   16 04         3 82    5 25                            3 64                               -

[T'- . l - 91 Chonens cnspus 30.28 4 88 0.39 0 18 0.42 g , o L ((l l3] [ Pdyssphossa h 31 Enterosearphaficamosa 5 90 6 03 II.8) 5.34 4.77 5 68 12.72 5 29 g l l , 0.92 14.54 E Yv Ed O 4 Si d, !94 gd 9 #gS S, p; ?g 9 f as 3 1 3 a - 4.c,.ca 3 09 3 00 5.n 3 54 5 99 ~g Fig 7. Classering deedseyasm of percess amuslanay d undisturbed ra======= a - by year, at rady insersmial samplang seasaous: al Giants Ned, b) klalisstuis Pons, c) Wluse Point, d) Fox laland. Espuerd 2 8 i

i I I i 4 distinct clusters. Group 1, comprising 1979 to 1983, even slight changes in temperature, which makes represents the mid and low intertidal commumty tius species a entical biomonitoring tool for studies prior to the opening of the second quarry cut. This of the ecological effects of thermal effluents. An

 '                    commumty (Table 2) was characterized by high                                extensive review of phenological, ecological and coverage by Chondrus. Fucus and barnacles, with an
 .                                                                                                applied monitoring studies of Ascophy//um was              j appreciable amount of available free space (rock); it                        presented in NUSCO (1993). The results of 1995-96 was typical of the communities found at nearby                               growth and mortality studies are compared with unimpacted sites. After the opening of the second

results from overall 2-umt and 3-unit operational cut (August 1983), elevated water temperatures periods, and are presented below.

resulting from 2 cut 2-unit operating conditions drastically akered the community at FE (NUSCO - Growth 1985). This ahered commuruty is represented by j Group II (Fig. 7d, Table 2), characterized by sharply The Gompertz growth model (Gendron 1989), i reduced populations of Chondrus and Fucus, and i when fitted to monthly Ascophy//um tip length data increases of Codium and Enteromorpha spp. i (Fig. 8), provides useful indicators of Ascophyllum Following start up of Umt 3 in 1986, further i population growth charactenstics. Annual growth in 1 changes to the rocky shore community were noted at i FE. Conditions existing during 2-cut 3-unit 1995 96 (Fig. 8a). as indicated by a. parameter of the j i operation (Group Ill; Fig. 7d, Table 2) permitted re- model, was significantly lower (P<0.05) at FN l , establishment of an extensive Fucus population in during 1995 % (78.0 mm) than growth at both GN I i the mid intertsdal, although Chondrus in the low (99.4 mm) and WP (100.2 mm). The difference j intertidal remained scarce, due to competitive between Ascophyllum growth at GN and WP in i exclusion by Codium. Polysiphonia 1995-96 was not significant. The inflection point. a and Enteromorpha spp. (cf presious sections). Most parameter of the model which identifies the time of years since Unit 3 began operation (Group Illb) show maximum growth rate, was earlier at FN in 1995-96 a high degree of within-group similarity (>70%), (6 July) than at GN and WP (1 and 2 August, i indicating respectively). l a relatively consistent species  ! composition. The exception is 1994 (Group Illa). Annual growth at GN during 1995-% was distinguished by the anomalously large, although significantly higher than growth over 2-unit temporary, setticment of Mytilus noted last year operational periods (90.1 mm), but not significantly (NUSCO 19%). different from the 3-unit period growth (%.9 mm). As the sample year for these quantitative The difference between growth estimates during 3 community analyses begins in March, the most unit and 2-unit operation at GN was significant. recent year (3/95 - 1/96) does not include ilata from inflection points for 2-unit and 3-unit periods were the period since April 1996, when all tlsee units within a day of each other (27 and 26 July, were shut down. Future reports will cocument the respectively). Growth during 1995-% at WP was effects of this extended outage on the rocky shore significantly higher than growth dunng both commuruties of nearby sites. operational periods (Fig. Sc); 90.2 mm (2-unit) and 87.8 mm (3-unit). The difference between operational period estimates at WP was not Araphy//um nodasum Studies significant, and operational period inflection points were within three days of each other (31 July and 28 July for the 2-unit and 3-unit periods, respectively). The status of three local populations of the brown alga Ascophyllum nodosum has been acceceM At FN, growth dunng 1995-% was significantly lower than dunng the 1985-86 2-unit year (90.5 since 1979 through monthly monitoring of plant mm) and the 3-unit operational period (116.5 mm; growth and mortality. Ascophyllum is a key species Fig. 8d). The 3-unit mean was also significantly within the MNPS ecological monitoring program higher than growth during the 2-unit year. The and these studies, as elsewhere, document the value inflection point for the 2-unit year was 18 July, and cf this species as a sensitive indicator of local 22 July for the 3-unit period. environmental conditions. la particular, The among-station relationships Ascophyllum exhibits easily quantifiable responses to for Ascophyllum growth were atypical in 1995-96, Rocky Intertidal 197

i i i3o relative to previous 3-unit study years. For the first i

           '"                                                                                                time in the 3 unit period. growth at FN was                  ;

significantly lower than at the reference sites WP i *) , I oo and GN. In most previous 3-unit years, growth at  ; {" '

                                                          $j',- d,                                           rN h,d h,e, si,nificanoy higher than growth at wp and GN. This growth enhancement was attributed to            l
      'f 50                                                                                                  incursion of the MNPS thermal plume to FN. which Jn                               I                                                                    elevated water temperatures up to 3-l*C for 3-4
                                        ;                      1995-96 hours each tidal cycle. These conditions in previous a,       u                                      w r u - - -- cu - - - - -r.

r aug oci ap, e years created favorable conditions for Ascophy//um  ! growth by: 1) extending the period of " normal" or l

                                                                                                            " ambient" peak growing conditions for local iso                                                                                               populations (18-21'C; Kanwisher 1966; Chock and
          '"                                                                                                Mathieson 1979); 2) more closely synchronit.ing
         'oo      I')                                                    '~~                                these periods of optimal growing temperatures with            j the period of maximum daily solar irradiance (June);

I3 - r. :, - and 3) eievatins iemperatures in late summer above j{5 3n 4

                               / 'M. .                '                                                    normal maxima but below stress levels (22-25*C),

increasing plant respiration and growth rates without exceeding photosynthste production (Brinkhuis et al. 1

             ,  f                j.,
                                   ;                       cane 1976; Stromgren 1977,1981; Vadas et al.1978).

l an, a oc r.o

         ----- rw ises-es av- ,- -                oci 3-onie                   sees-es ao, Conditions for Ascaphyllum growth at FN during 1995-% were not favorable. even though power plant operational status was similar to iso previous 3-unit years. Ciaracteristics of the annual
        ,3                                                                                                pattern of growth at **N observed in 1995-% (early CI                                                    ~~,                               rapid growth in cari, er followed by a decline

_'m ~ in growth rate in autumn) were similar to those

    !      3                                      ,

4',, - observed at the original experimental station (FO) {5 j during 1983-84 (NUSCO 1992). The FO population

                               /                                                                          that year was stressed by temperature increases of 7-
    }n                  ,.

9*C following the opening of the second quarry cut.

             ,,-                 :l                       htte Pomt it is unlikely that the thermal plume alone caused a,n       a              ,          oci         o.c                     .                    stress to the FN population in 1995-%. The summer
        - -- ru i9es-es - -- .                        3-onie             r.. ,9i 3- , ',

of 1995 was one of the warmest observed during this study. Ambient monthly average surface w1 ster iso temperatures in late summer /carly autumn 1995 s

       ,~                                                                                                 were 1-3*C above operational period averages (see j

dI Lobster Section in NUSCO (1995)). These unusually

    - 'oo                                                y'                                               high summer temperatures (up to 22*C) probably l'

I 3 ,4',.. - contributed to higher growth observed at references

• l j}5 sites in 1995-96, and coupled with 2-4*C thermal j

                               /                                                                         incursion, may have created stressful temperature jn g        .,
                                !!                        Fa= l=laad conditions that reduced growth rate . of the FN I

n, population in late summer. Kanwisher (1966) and o a ao, oci o,e r.o a

       - --- rw s ees-as -- --                       3-un.i                   i ss s.,n, 6 Vadas et al. (1978) both reported gradual demise of               l Ascophy//um once temperatures exceeded 26*C.                      I Fig. s.         Ascophyumm growth: a) darins 1995-1996, b<l) presan year,3-unit and 2 unit operational periods at each station.                                                      Mortality curws are sw comparu powsh madel smed io iap -

lensih deia,inchides unsaian points. Error bars repressni manddy mesn ianses

• 2 sE.

Emironmental stress to local Ascophyllum populations can result in breakage and loss of the 198 Monitoring Studies,1996

i fQ g . l .

                .         I            x
                                                                                                                                     %.D 1

i = j *(,N N ,

                                                                                                                              =                '

N < j, a N f( = b s' I ' *. Y m,,, u j, 3 - N- , m. m7,.e - ,J d

                .                                                                             4          --        -       ,.

d U 3 E fi

                =                                                                '                                                                                    .

j 88 asente und caente u sk ' e as e no em me sen = me em em se =

                               .--. 3 -we -- - pe                                            sans-ee                                     .--. 3-ye - - kwe                                 seen-en G                                                                               l Mg'                                                                                          +b.

m i l ' m M f*= .g i , I"

                                                                     %~~h                          m.ll o

2 '=

                                                                                     ,             u               .                                                          .  .

a '

                                                                                                                            'im g

j ,.

                                                                                                                           <g                                                    \;        D%d.q,

j.  ; W pJ s-

a. e ne as m. een a, m. me en au se =

                              . - . 3-we - - Fue                                            ions-es                                     .--- 3-we - - Fue                                     sees-es
             =-

m -- s

                                                                                                                             -             N__         ,

o a m \g ja i 4N

                                                                                                                          )m                                 ;\
              *                                        \'                                 .

i . *= n ( 1 l i. , I. i g

              *                                                       ,  b ~: .N                                        s*

n'*

                                                                                              .q
                                                                                                        '- - - i          n=                                     U. . .,, '~~             k. ,.. & ,   ,

e ree w - ree m '

a. e ne em n. e. m,

                                                               ,                                                                m.            me             em                       n.       e.             =
                       .--. re ises-as --                                                                                                                                 ,

a-we ines-es . --. ru seu -- Swa ises-es Fis.9. Ascophyllum mortalny, as number of rema'uung tesged Fig.10. plants. at each stauan. Ascophyllum mortahty, as number of remaming tagged upa, at each sianon, upright shoots or fronds, commonly referred to as of plants at GN occurred between July and August, mortality. Population mortality.is monitored by at WP between October and November, and at FE cxamining patterns of frond base tag loss (referred to between August and September. None of these as plant loss; Fig. 9) and apical tag loss (tip loss; losses was associated with a major storm event. Fig.10). Plant loss at GN during 1995- % (72 %) Temporal tip loss relationships at each station were was higher than both 2-unit (52%) and 3-unit (51%) similar to those described above for plant loss. Tip operational means Plant loss at WP during 1995-% loss at GN during 1995-96 was 84%. which was (60%) was also higher than operational period higher than the 2-unit and 3-unit means of 75% and means of 55% and 56% for 2-unit and 3-unit 72%, respectively. Similarly,1995-% tip loss at WP periods, respectively. Plant loss at FN (68%) was (80%) was high compared to the 2-unit (75%) and 3-lower than the 1985-86 2-unit year (80%) and the 3-unit (72%) means. At FN, tip loss for 1995-% unit mean (70%). During 1995 %, the greatest loss Rocky Intertidal 199

. (86%) was interPWl*, relative to the l985-86 2- of many inteftidal species at FE. e.g.. presence or l umt year (90%) and the 3-unit period (81%). extended season of occurrence for species with Ascophyllum mortality study in 1995-96  ! warm-water affinity (Codium. Sarpatsum. { I revealed no evidence of power plant impact. Our Gracilarna. Myrilus) and absence or abbreviated sampling site nearest the discharge (FN), has season for species with cold-water affinity generally had higher mortality rates than at reference ,' sites. However, these higher mortality rates do not (Choneus. Monostroma. Dumontia. Littorina). This

1. ahered community at FE has exhibited such appear to be related to proximity to the discharge, consistency and resilience to change that many but rather to the higher degree of population components appear able to persist, even in the exposure to wind- and wave-induced stress at FE. absence of a thermal addition. as during an extended

compared to the more sheltered reference sites. power plant shut down.

Furthermore, while population suess was indicated incursions of water with temperatures elevated 4

by growth studies in 1995-%, mortality estimates 2-4'C abow ambient impacted the Ascophyllum 4 were well within the range of previous years. An population nearest the discharge (FN). When such j area-wide seasonal pattern of mortality has been addition to ambient temperature did not exceed the } observed throughout our studies which further tolerance of Ascophyllum, this resulted in increased ! implicates wave-induced stress as a major cause of growth relative to that at more distant stations; this

j. mortality. During both 2-unit and 3-unit operational pattern was seen in most 3-unit years. However in i periods, mortality rates were highest during the the 1995 % growing season, ambient water months of August through November, when strong temperatures were unusually warm, and the thermal i

storms and high energy waws were frequent. Many addition contributed by the MNPS discharge may l 4 studies elsewhere point to the strong relationship have been sufficient to produce unfavorable ' between mortality and degree of site exposure to conditions for growth at FN. prevailing winds and storrns (Baardneth 1955,1970; in summary, impacts to the rocky shore Jones and Demetropoulos 1968; Vadas et al.1976, associated with operation of MNPS are restricted to 4 1978, Wilce et al.1978; Cousens 1982,1986; Vadas within 150 m of the discharge. Current rocky ! and Wright 1986). ' intertidal studies have been sufficient to detect and The status of Ascophyllum population recovery document these ecologically significant changes to l at FO. our original potentially impacted site, the local shore communities, and will allow following power plant-induced elimination of the assessment of any further changes, should they entire population from the site in 1984, has not occur. j changed from that reported in recent previous years (NUSCO 1995,19%). Some indisidual plants have settled, grown and persisted at FO during 3-unit

References Cited operation; however, no significant recovery has occurred to date. Environmental conditions at FO Baardseth, E.

i created by 3-unit operation. although less stressful 1955. Regrowth of .lscophyllum nodosum AAct Harvesting. Inst. Ind. Res. Stand.. than those during 2 unit 2-cut operation, are outside Dublin. 63 pp. i the extremely limited range of conditions required Baardseth E. 1970. Seasonal variation in l for successful widespread Ascophyllum recruitment-Ascophyllum nodosum (L.) Le Jol. in the Trondheims0ord with respect to the absolute live j ' and dry weight and the relative contents of dry ' Conclusions matter, ash and fruit bodies. Bot. Mar. 13:13-22. Bertness. M.D.1989. Intraspecific competition and ! Rocky shores in the vicinity of MNPS continue acilitation in a northern acorn barnacic i to support a rich and diverse community ofintertidal Population. Ecology 70:257-268. i plants and animals. Impacts of the power plant to s, .H., mempel aM E Jones. IE i these local commumties are limited to those areas Photosynthesis and respiration of exposed calt 1 exposed to the thermal effluent for at least part of the marsh fucoids. Mar. Biol. 34:349-359. Chock, J.S., and A.C. Mathieson. tidal cycle. Effects noted since Unit 3 began 1979. j operation include shins in occurrence and abundance Physiological ecology of Ascophyllum nodosum i j 200 Monitoring Studies,1996 3 k

 ,       (L.) Le Jolis and its detached ecad scorpioides                                         Menge. B.A.       1976. Organization of the New (Hornemann) Hauck (Fucales, Phaeophyta). Bot.                                               England rocky intenidal community: role of                     -

Mar. 22:21 26. Preda6on, compeddon and emironmental Cinord, H.T., and W. Stephenson. 1975. An heterogeneity. Ecol. Monogr. 46:355-393. Introduction to Numerical Classification. NAESCO (North Atlantic Energy Senice Co.). Academic Press, New York. 229 pp. 1994. Seabrook emironmeistal studies,1993. A Connell, J.H. 1%I. ENects of competition, characterization of environmental conditions in predadon, by Thais lapillus and other factors on the Hampton-Seabrook area during the operation r natural populations of the barnacle, Balanus of Seabrook Station. balanoides. Ecol. Monogr. 31:61-104. NAl(Normandeau Associates, incorporated). 19%. Cousens, R. 1982. The efect of exposure to wave Seabrook Station 1995 cmironmental studies in action on the morphology and pigmentation of the Hampton Scabrook area. A characterization Ascophyllum nodosum (L.) Le Jolis in south- of environmental conditions during the operation ' castern Canada Bot. Mar. 25:191 195. of Seabrook Station. Cousens. R. 1986. Quantitative reproduction and NUSCO (Northeast Utilities Senice Company). , reproductive cKort by stands of the brown alga 1985. Rocky Shore. Pages 1-41 in Monitoring Ascophy//um nodosum (L.) Le Jolis in south- the marine emironment of Long Island Sound at castern Canada Est. Coast. Shelf Sci. 22:495 Millstone Nuclear Power Station. Waterford 507. Connecticut. Annual Report,1984. > Draper, N., and H. Smith, 1981. Applied Regression NUSCO.1987. Rocky Intertidal Studies. Pages l-Analysis. John Wiley and Sons, New York. 709 66 in Monitoring the marine emironment of pp. Long Island Sound at Millstone Nuclear Power Gendron L. 1989. Seasonal growth of the kelp Station, Waterford Connecticut. Summary of Laminaria longicruris in Baie des Chaleurs, studies prior to Unit 3 operation. Quebec, in relation to nutrient and light NUSCO. 1992. Rocky Intertidal Studies. Pages availability. Bot. Mar. 32:345-354. 237 292 in Monitoring the marine emironment Jones, J.E., and A. Demetropoulos. 1968. Exposure of Long Island Sound at Millstone Nuclear Power > to wave action: Measurements of an important Station Waterford Connecticut. Annual Report, ecological parameter on rocky shores on 1991. Anglesey, J. Exp. Mar. Biol. Ecol. 2:46-63. NUSCO.1993. Rocky Intertidal Studies. Pages 49-Kanwisher, G.W. 1966. Photosynthesis and 92 in Monitoring the marine environment of respiration in some seaweeds. Pages 407-420 in Long Island Sound at Millstone Nuclear Power H. Barnes (ed.) Some Contemporary Studies in Station Waterford Connecticut. Annual Report, Marine Science. George Allen Unwin Ltd., 1992. , London. NUSCO.1994. Rocky Intenidal Studies. Pages 51-Keser M., and B.R. larson. 1984. Colonization and 79 in Monitoring the marine environment of growth dynamics of three species of Fucus. Mar. Long Island Sound at Millstonc Nuclear Power Ecol. Prog. Ser. 15:125 134. Station, Waterford Connecticut. Annual Report, Lance, G.N., and W.R. Williams. 1%7. A general 1993. theory of classificatory sorting strategies,1. NUSCO. 1995. Rocky Intertidal Studies. Pages I Hierarchical systems. Comput. J. 9:373-380. 171 201 in Monitoring the marine environment Lubchenco. J. 1980. Algal zonation in the New of Long Island Sound at Millstone Nuclear Power England rocky intertidal community: an Station, Waterford Connecticut. Annual Report, experimentalanalysis. Ecology 61:333-244, 1994. Lubchenco, J. 1983. Littorina and Fucus: efects of NUSCO.19%. Rocky Intertidal Studies. Pages 41-herbivores, substratum heterogeneity, and plant 66 in Monitoring the marine emironment of l escapes during succession Ecology. 64:1116- Long Island Sound at Millstone Nuclear Power 1123. Luning, K. Station, Waterford Connecticut. Annual Report, 1990. Seaweeds: Their Emiromnent, 1995. Biogeography, and Ecophysiology. John Wiley Schonheck, M.W., and T.A. Norton. 1978. Factors and Sons,Inc. New York. 527 pp. controlling the upper limits of fucoid algae on the shore. J. Exp. Mar. Biol. Ecol. 31:303-313. RockyIntertidal 201 j l

Schonheck, M.W., and T.A. Norton. 1980. Factors Studies. Pages 307 656 in Benthic Studies in the mntrolling the lower limits of fucoid algae on the Vicinity of Pilgnm Nuclear Power Station,1%9-shore. J. Exp. Mar. Biol. Ecol. 43:131-150. 1977. Summary Rpt. Boston Edison Co.  ! Schneader, C.W. 1981. The effect of elevated temperature and reactor shutdown on the benthic marine flora of the Millstone thermal quarry, Connecticut. J. Therm. Biol. 6:1-6. South, G.R., and I. Tittley. 1986. A checklist and distributional index of the benthic mariac algae of the North Atlantic Ocean. Huntsman Marine Laboratory and British Museum (Nat. Hist.) St. Andrews and landon. 76 pp. Stromgren T. 1977. Short-term effects of temperature upon the growth ofintertidal focales. J. Exp. Mar. Biol. Ecol. 29:181-195. Stromgren, T. 1981. Individual variation in apical growth rate in Ascophyllum nodosum (L.) Le Jolis. Aquat. Bot. 10:377-382. Taylor, W.R.1957. Manne Algae of the Northeast , Coast of North Amenca. Univ. Mich. Press, Ann Arbor. 870 pp. Topinka, J., L. Tucker, and W. Korjeff. 1981. The distnbmion of fucond macroalga! biomass along central coastal Maine. Bot. Mar. 24:311-319. Underwood, A.J. and E.J. Dealey 1984. Paradigms, explanations and generalizations in models for the structure of intertidal communities of rocky shores. pp. 151-180 in D.R. Strong, Jr., D. Simberloff, LG. Abele and A.B. 'Ihistle, eds., Ecological Communities: Conceptual issues and the Evidence Princeton University Press, Princeton N.J. 611 pp. Vadas, R.L., M. Keser, and P.C. Rusanowski 1976. Influence of thermal loading on the ecology of intertidal algae. Pages 202 251 in G.W. Esch and R.W. MacFarlanc (eds.) Thermal Ecology II. ERDA Symposium Series, Augusta, GA. Vadas, R.L., M. Keser, and P.C. Rusanowski. 1978. Effect of reduced temperature on previously stressed populations of an intertidal alga. Pages 434-451 in J.H. Thorp and G.W. Gibbons (eds.) DOE Symposium Senes, Springfield, VA. (CONF-771114, NTIS). Vadas, R.L., and W.A. Wright. 1986. Recruitment, growth and management of Ascophyllum nodosum. Actas 11 Congr. Algas Mar, Chilenas:101-ll3. Villalard 3tal-% M.1995. Illustra'.ed Key to the Seaweeds of New England. Rhode Island Nat. Hist Sury., Kingston R.I.144 pp. Wilce, R.T., J. Foestch, W. Grocki, J. Kilar, H. Levine, and J. Wilce. 1978. Flora: Marine Algal 202 Monitoring Studies,1996

1 i I Eelgrass 1 1 i Introduction. .... .... ... . ... . . . . . . .. . . .205 Matenals and Methods . . . .. ... .. .

                                                                                          . . . . .                        .       .       . . .      .         . . .           .205 Results . .. . . .. . . .. . . . . . . . . . . . .. .. .. . . . . . . .                 . . .      . . . . .             ..            ...                                    ..207 Temperature... .. . . . . . . . . . . . . . . . . . . ..                                 .          . ..               ..       . . . . . ..             .          ..207      ;

Sediments.. ... . . . . . . . . . . . . . . . . . . . . . . .. . . .. . .. .. 207 Shoot Densit ..... .... .. . .. . . . . .

                                                                                      . . . . . . . . . . . . . .            . .. .               . . . . . .         .    .. .. 207 Shoot Length ...            . . . . . . . . . . . . . .
                                                                              . . . . . . . . . . ....... . . . . . . . . .                             ..        ..         ..210 S W Sta k.........                       . . . . . . . . . . . . . . .                   . . . . . . .        . . . . . . . . . ......              .        .       .210 Reproductive Shoots.... .                      . .           . . . . . . . . . . .              . , .          .       .
                                                                                                                                                 . . . . ..                .. .. 210      :

Discussion .... .. .. .. . ... .. .... .... . ... . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . ... . . . .. 210 Conclusions.. . . . ... ... n na

                                        . . . . . . . . . . . . . .      . . . .    .       . . . .          . . .      . . . . . . . .           . . . .               . ... ..'16       i n NM Litw....
                                                                                                                                                  . . . . . . . . ..          .. 9.
                                                                                                                                                                                 . I 6    i 1

I i i l i l 1 l Eelgrass 203 4

.I i n 1 4 i 1 i i i e t a 4 J 1 I a 1 4 i .1 J i, 4

;     204 Monitoring Studies,19%
                                "Y

Eelgrass  ! Introduction Materials and Methods Ecigrass (Zostera marina L.) occurs widely in estuaries and lagoons of temperate and warm boreal Three eelgrass study sites in the vicinity of MNPS coasts in the Atlantic and Pacific Oceans (Setchell were sampled during 19% (White Point-WP, Jordan 1935). Such a geographically broad distribmion is Cove-JC, Niantic River-NR) (Fig.1). The WP and the result of this species' wide tolerances to JC stations, located 1.6 km and 0.5 km east of the temperature, salinity and water depth (Osterhout Power plants discharge, respectively, are within the  ; 1917; Scachell 1929; Uphof 1941; Burkholder and area potentially influenced by the 3-unit thermal t Doheny 1%8; Dillon 1971; Thayer et al 1984). Pl ume (ENDECO 1977; NUSCO 1988). The NR AAer the disappearance of most eastern North site, located about 3 km from Millstone Point. is American and European populations in the 1930s unaffected by power plant operation (Fig.1). Water  ; (Tutin 1942; Rasmussen 1973, 1977), the depths (at mean low water) were 2.5 m at WP,1.5 m importance of celgrass to coastal ecosystems at NR and 1.1 m at JC. The WP and JC sites have (described in previous reports, e.g., NUSCO 1994) been sampled since 1985. The NR site has been  ! has become widely recognized. Following the relocated several times since 1985, due to changes in destruction of Zostera populations in the 1930s, distnbutional patterns of eclgrass in the river. The  ! shoreline crosion resulted from increased wave scour original sampling site (#1) located midway between and changes in current patterns. Habitat alteration Camp Rowland and the navigation channel (Fig.1), also occurred within the subtidat zone, e*idenced by was sampled in the summer of 1985 and June 1986. declines in abundance of many commercially and A substantial population decline at site #1 occurred recreationally valuable species (Stauffer 1937; in July 1986, resulting in the establishment of Dexter 1947; Milac and Milne 1951; Orth 1973, another NR sampling site (#2) 50 m to the south. , 1977; Rasmussen 1973, 1977; Thayer et al.1975; nearer the migation channel. Site #2 was sampled ' Stevenson and Confer 1978; Zieman 1982). for the remainder of the 1986 season; however, by , Eelgrass beds on the north shore of Long Island September 1986, the eclgrass population at this site Sound (LIS) are concentrated in shallow protected had also disappeared In June 1987, a new NR treas cast of the Connecticut River (Koch and Beers sampling site was established at the nearest siable  ! 19%). Extensive meadows of eclgrass are common Population, located in the lower river (#3). A 1 in the vicinity of Millstone Nuclear Power Station slower, but steady, decline of the eclgrass population (MNPS). Temperature changes have been at site #3 has been documented since 1987 (NUSCO j demonstrated to affect celgrass populations by 1993), and by June 1993, no eelgrass shoots were 1 reducing growth rate, lowering resistance to disease, observed at this site. However, the recovery of the and reducing the production and germination of celgrass population at the old NR site (#1), noted in seeds (Burkholder and Doheny 1968; Phillips 1974, 1993, permitted NR samples to be taken again at this 1980. Orth and Moore 1983; Evans et al.1986; station during the 1993 and 1994 sampling periods Zimmerman et al.1989; Taylor et al.1995; Vergeer (June September). Again, no plants were observed et al.1995) Because of the ecological importance of at site #1 in September 1994 or at the beginning of eelgrass and the prediction that the 3-unit thermal the 1995 sampling year (June), and a new sampling Pl ume could reach to the nearest population in site had to be established on the east side of the hrdan Cove (ENDECO 1977; NUSCO 1988), the channel, opposite Smith Cove (site #4). Monthly Present study was initiated in 1985 to monitor this observations of NR#1, NR#2 and NR#3 have population and others nearby Objectives of the continued since their population disappeared, Present study are to identify temporal patterns of however, no eclgrass recolonization was obsened at celgrass distribution, abundance and reproduction in any of these sites through the 1996 sampling year. the vicinity of MNPS and to determine the extent to Samples were collected montluy at each site from which changes in these patterns are the result of June through September, the period of maximum natural variability or MNPS operation. standing stock and plant density. At each station,16 samples were collected by SCUBA divers from randomly placed quadrats (25x25 cm, 0.0625 m2 ) within a 10 m rad us of the station marker. The upright shoots from plants within each quadrat were harvested, placed in a 0.333 mm mesh bag, and I l Eelgrass 205

         -      .    -     . . . . .      .-       .            - . . .    .-                    ~_     _     . -     . . -        ~_     .     ~ _ _ _

I-a t i a {

                                                          %  Niemue                                                                                     .

y*...,k Neeth t j I tem s as 2 "

                                                         ^J E
                                                                        \

i

                                                                          ,,s,/~                                                                        i

' l Jc 2  :

                                                                                             \
                              )
                                  /

(4'ib ) we,me rosas  ! m v-rA / s h N aLAC PotNT Fig. 1. Map of the Minissans Point area showing the lacesian of solyass sampling stauans, JC=Jerden Cove, NR=Niantic River (l = sampled 19s5-June; 19s6 and 1993-1994,2=sangeled July 19s6,3= sampled 19871992,4=ammpled 1995-1996), WP= Wlute Poet. ' taken to the laboratory for processing. A 3.5 cm throughout the study; their weights were not diameter x 5 cm deep core was taken concurrently recorded. Ecigrass standing stock was estimated as with celgrass samples for analysis of sedimentary the weight of the shoots taken from each quadrat.  ; characteristics at each station. Ten.perature in - From 1985 to 1987, shoots were weighed, then dried Jordan Cove was measured by submerging an in an oven at 800C to constant weight. Dry weights ' encased thermistor-recorder Temperature from 1988 to 1996 were estimated from the wet- i measurements have been recorded in Jordan Cove weight / dry weight relationship obtained above.  ; since 1991. All Millstone units were shutdown Nonparametric methods were used to examine i during the 1996 sampling period, so there was no trends in the time series of eclgrass shoot density and ' possibility of thermal addition to Jordan Cove. This standing stock. The distribution-free, Mann-Kendall

• provided an opponunity to more closely examine test (Hollander and Wolfe 1973) was used to natural factors that affect temperature at JC, such as determine whether the time-series of mean monthly insolation and tides. To assess daily cycles, standing stock biomass or shoot density exhibited temperatures are reported this year (1996) as hourly significant trends. The slope of the trend, when averages (rather than daily averages) for 2 discrete significant, was estimated by Sen's estimator of the one-week penods in June (when air and water slope (Sen 1968). Ecigrass shoot length was not l temperatures are coolest, but daily solar irradiance is statistically analyzed, because growth occurs at the  ;

at a maximum) and August (when warmest water base of the shoot (from a basal meristem) and tips 1 temperatures typically occur, NUSCO 1996). continuously erode, and because leaf turnover rate is  ! All shoots collected were counted in the laboratory highest during the summer (Roman and Able 1988). and the longest blade of each shoot (up to 20 plants Mean sediment grain size and silt / clay content were per sampic) was measured t<J the nearest centimeter, determined using the dry sieving method (Folk The number of reproductive shoots in each sample 1974). Sediment samples were heated to 5000C for was used to estimate the percentage of ied 64 24 h to determine organic content, ec:imated as the shoots in the population. Shoots were rinsed in difference between dry-weight and ash-weight. Both fieshwater to remove invertebrates and epiphytes. silt / clay and organic content were recorded as a Epiphytes on eclgrass shoots were minimal percentage of the total sediment sample weight. l l 206 Monitanng Studies,1996 i l l

l i Results silvclay and organic contents (Fig. 3). In general, sediments at stations nearest MNPS OC and WP) have been less variable than those at NR. Variability Temperature at NR has possibly been caused by frequent , relocation of the sampling site within the Niantic Average daily water temperatures at the MNPS River. Sediments collected during 19% wre intakes and discharge for the period June through coarser at JC (mean grain size,0.22-0.25 mm), than September 19% are presented in Figure 2a. Because those at WP (0.10 0.14 mm) and NR (0.09 mm in all MNPS was not discharging heated effluent, intake months). Silvclay content in 19% was highest at and discharge temperatures were always within 1*C NR (monthly range: 34.8-40.3 %) relative to WP of each other. Temperatures during this period (8.6-25.5%) and JC (1,1-3.6%). Similarly, sediment ranged from 12*C in early June to 21*C in late organic content was higher at NR (6.7-7.6%) than at Augusvearly September. Water temperatures at the WP (1.4-5.4%) or JC (0.9-1.6%). All sediment JC eelgrass station and at the MNPS intake and parameters measured at JC and WP in 19% were discharge are presented as hourly averages over two within the ranges for presious years. Monthly mean six-day periods in 19% (June 16-21 and August 4-9; grain size and organic content estimates at NR Figs. 2b and c, respectively) to better illustrate during the 19% sampling period were also within natural daily temperature cycles in Jordan Cove. As histoncal ranges. although mean grain size was with daily average temperatures, intake and smaller relative to most presious samples. Siluctay discharge hourly temperatures were generally within content at NR in 19% ranged from 34.8 ( August) to l'C of each other, and fluctuation over the daily 40.3% (September). Monthly silt / clay content cycle was minimal at these two sites. Conversely, estimates at NR in 19% were high compared to most ' considerable temperature fluctuation was observed at previous years, with the September 1996 value being JC over the daily cycle, with increases of up to 4 5*C the highest obsened at any site in the Niantic River , observed on some days. Lowest temperatures were s nce the beginning of this study in 1985. l similar to those measured at the MNPS intake and discharge. The highest peaks obsen'ed (e.g., June 16 and 18, August 8 and 9) were related to solar Shoot Density insolation, as all occurred in mid-afternoon of sunny days; lower peaks were noted on days with more Annual mean shoot density in 19% was highest at (440 '" " "" '* " Some o the lower peaks occurred at other times of 2 the day such as night or early morning. These m ) an I west at M OM Mmh Me , secondary peaks were the result of flushing of more II' ^"""*I***"*""*****N'" " ' * * ' ' protected (and presumably warmer) waters of upper MontMy sh &nsWu ( n in hig)torical in Im rangedrangu. fmm 313 (August) to J:rdan Cove out to the JC eelgrass site, as they Gune) at , fmm Ouly)t 6 Wne) at typically followed ebbing tide. Additionally, heat trInsfer from sediments may have caused additional NR, and from 115 (July) to 215 (September) at WP. warming, particularly during slack tide. Similar Monthly mean densWu um also wWn the ranga of previous years (Fig. 4). daily temperature patterns observed in Montsweag Bay, Maine were also attributed to the effects of Trend anaWs apphed to qme-seria of montW direct solar heating, indirect solar heating via mud shoot densities mdicated significant decreasing flats, and tidal effects (Thompson 1978). trends at JC (slope =-3.912 shoots /m*/ sample period; p<0.01) and WP (slope =-2.985 shoots /m / sample period, p<0.01). Trend analysis was not performed gg on NR data due to lack of a consistent time-scries for any given site resulting from frequent sampling station relocations following localized population i Sediments at celgrass sampling stations have been disappearanca' characterized since 1985 through monthly Oune-September) measurements of mean grain size, and Eelgrass 207 l

        '25 A.

D sachars *

                                                                             .         +

by

                                                     'akI
                                                                   ~
                                                                                         D G 15 >                -

~f c/ v

               ,,s 10-J'JN96                     JUL96                AUC96                 $(996             Mi96 25-B.

Jordan Cove 20 n E as w E 15 v- ]p I tw 7#hq,.3~yg j i taaske Discharg e IO-16JUN96 17JUN96 18JUN96 19JUN96 20JUN96 21JUN96 gpJyNgg 25 C. 8 's e

~

7 Jorden

           ;] Cove-
                .c jwyessW
 %                  l 8               latak e 15 1

i 10 OAAUC96 OSAUC96 06AUC96 07AUC96 06AUC96 094UC96 10AUC96 Fig. 2. Mean daily weser temperstwas at the MNPS intake and &acharge imm June through September (A.) 1996 howly everage water empersawee at the Jorden Cove selgrass seasian and MNPS issake and encharge temperasures from June 16 23 (B.) and August 410 (C i l 208 Monitoring Studies,1996 I

20 JORDAN CM

  ?
 ?    40.0 t'. zo O'

• 20.0

         ,   e- s. *A't:   + tR< ~.',,$'0'f,5< WA'* Y *~* Y $ -*                               **    'Y**          '1 "* . D g                         y    ,',    ,

s 's, ; * * *

 $ as                                 *

• T 0.6 1

w ee A U m : ; ; ; h;% wy;  : : %#* 6 c.o h-f f $' $ $ f f h f f f i f f f f f f f f f f f 20 NlANilC RNEP i 5* e e no */o %, o \ 9 i /, - g -

                                                                                                                         /        \;
  & 10.0                 ,
                               '\

e 4 \e p.) Y* A ,/, /' , ,! ', ,

                                                                ' : \ )j, y \' s^ y* s , j
                                                    ~               _*       ,

a

• i* . , , /,3/'*j,y' ,

v i.....* . ,. , . y io. , . *

                                                                                                              $ . /*
  $                          V                                                                                  'o
  $ 08 i    0.6 1

w G4

       I

[yO gA' %6 \ba#8 NM 'SH A  % b 00 fl'fff$fffffff$fffff/!$$N S8 WHf1E POINI F

 ! 40 0 t;' mo "J                                                                                                                                            e
  '                                                                                                                                            s Nl',',
  .          .  , ,' s J , / * ^ \ , < / '. + / ' W . , ,,-. , ,'^ ~ ^ ! 'w ]

y ,o e s s.e ^ ** ~ er*. e 's- o.s - < ^

• s s e***e.se-s .e-e"~. -*<****s ,e o'

  $ , cs-
 }.

w 06-04-E b; ; ; ; ; i 1- ; ; ; ; d . I;2 O h _ ^ ^ -- - ^' ^^ * - M 4 u 9.0 < iffff$ffffffififfff/lfi/

• FC 3. Mean pain sue (4), ergenic comes (* Jand sih/ clay content (0) of sedemens at MNPS selyans stauons . Jorden Cove, Niantic River i

and White Pass ensupled dwing the pened June-September from 1985 through 1996. Eelgrass 209

l TABt11. humaal and menshly average shoot density (noJm') length (cm) and dry weight stanchns mesi (pwm') of selsrass j sampled naar MNPS durms the pened June to sepasaber anse 1985. l ANNUAL MEANS 19 % 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 Jun Jul Aus Sept shoot Demasy Jordan Cow 572 713 542 468 411 338 603 630 484 282- 450 440 537 414 313 414 Niantic River 413 72 294 307 240 225 249 233 385 132 239 310 476 253 269 244 white rout 2s6 218 227 161 335 185 242 204 310 141 237 17I 178 115 167 215 shaat tagsh Jentan Cove 57 57 77 75 74 38 48 53 54 35 37 51 58 45 57 44 Niamaic River 50 39 81 88 94 73 51 48 58 28 23 66 82 70 58 48 Whne Poet 107 116 126 86 110 106 87 72 107 95 92 79 98 110 64 58 , M Jonian Cove 243 276 258 238 202 105 169 210 160 60 104 121 127 110 139 110 Niantic River 156 32 184 181 183 143 81 79 125 18 29 90 135 103 74 48 White Poet 265 260 201 90 236 180 148 110 275 100 180 89 " 97 84 81 95 Shoot Length Reproductive Shoots Average shoot lengths during 19% were longest at Annual and monthly percentages of reproductive l WP (79 cm), shonest at JC (51 cm) and intermediate shoots are presented in Table 2. The highest annual at NR (66 cm); annual means were within cverall percentage of reproductive shoots at any station since ranges observed since 1985. Monthly shoot lengths 1985 occurred at NR in 1996 (15.0 %); in presious in 1996 were highest in June or July and lowest in years, annual percentages of reproductive shoots at September. Shoot lengths in 1996 were 44-58 cm at NR have ranged from 0 to 8.7%. At JC the 19% JC,48-82 cm at NR, and 58-110 cm at WP, and fell annual percentage (2.0%) was within the range of within historical ranges at each station (Fig. 4). previous years. At WP,4.1% of the shoots collected in 1996 were reproductive; this was also within the Standing Stock range of annual percentages in previous years (0.4-10.4%). Monthly percentage of reproductive celgrass Average eclgrass standing stock during 1996 was shoots in 1996 was highest in June at NR and WP higher at JC (121 g/m') than at NR (90 g/m') and (29.6 and 9.9*/., respectively) and July at JC (3.7%). 2 WP (89 g/m ; Table 1). The annual standing stock The July 1996 sample at NR also contained a high estimates at JC and NR during 1996 were within the percentage of reproductive shoots (17.8%). No historic range; however, standing stock at WP in reproductive shoots were collected at NR after the 1996 was the lowest recorded over the entire study July 1996 sample period. In August 1996. 0.3 % period. and 3.6% of the plants sampled at JC and WP. Monthly standing stock estimates in 19% ranged respectively, were reproductive; no reproductive 2 from 110 g/m (July and September) to 139 g/m* plants were collected at any site in September.

                    ) at JC, from 48 g/m2 (M^cmber) to 135 (Auf(June) g/m                   at NR. and from 81 g/m* (August) to 97 2                                                                                                   Discussion 3/m (June) at WP. Monthly mean standing stock estimates for 19% were within histo *ic ranges (Fig.

Considerable fluctuations in population parameters 6). Trend analysis, performed on monthly dry-mon tored for celgrass have been noted over the past weight estimates over the entire time-series, 12 ym Fluctunions have been mon pronounced indicated that standing stocks have significantly 2 n the Niantic River, where a general declining trend declined since 1985 at JC (slope =-4.079 g/m / sample has been evident since early study years. While a I period, p<0.05). Here was no significant trend in patchy eclgrass population continues to inhabit the  ! standing stock over the study period at WP. Standing ' Niantic River, health of the overall Niantic River stock estimates have declined at NR over the study population remains questionable when compared to period; however, trend analysis was not performed

                                                                                             ,,,g 4. years (1985-1990) and to historic on the NR standing stock because of sampling observations (Marshall 1994). Since 1985, we have mation relocatena observed only patchy transient populations within 210      Monitonng Studies,1996

JDPC>AN COVE 1200 1985 19M 1987 1988 1989 1990 1931 1992 1993 1994 1995 199f 1000

   ^

E

   -f Ca wo I

y.- }}1 1 200 0

            $   ,>f' O l 0 / $ S E l [ h l l $ h I h f l tY!t?l NtANTIC RIVER 1200 1985     1986     1997    1988   198a      1990       1991      1992        1993   1994    1995     199f-1000
 ,7
  .E e' M'                                             ,

3 , A v on 400 200

                                                  }          }}                   }                           }

Y 4 0 - * - [ltY $ $lfl$ fl$V $ s' t t f6 t v ft WHfTE POINT 1985 1986 1987 1988 1989 1990 1991 1992 1993 199f, 1994 1995 1000

 ^

E 1 o M S 9 S0 t I { l

       >=        44 4I l    I ll  i ;4 4         i+ f+1       4             11 4              p4        II   4  14 1 4

0

            $h$h$hlhEh$h9 h(l$hOh$hih FO 4. Mean number of eelgrass shoots per m's 95% C.I. at Millstone eelgrass sianons sampled dring the pened Juno September from 1985-1996.

Ecigrass 211

i l JJROAN COVE

               '"                                                                                                                   i99+            1995        '996 i9e3      i9es    in7      i9es     ise9          1990         i99i           193:               1993                                                                -!

t i '"  !' 3 2 'S i G 5 s in

        - gim                                                                                                                                                                                 ,

a ,,. . +4 4+. 44

                        ,'+ +*

{* 2

                                       +
                                                         ,       I
                                                                          + ,

44 +4

                                                                                                                              ++
                                                                                                                                      +,+        ,
                                                                                                                                                               *+

I i n

• fh//f/f/f/////f/flflfhfh NIW1( RWER 1985 1986 1987 1988 1969 1990 1941 199? 1h3 1994 i495 19 4

        . { 14 f iso G

h IM i I

         $i*                             i 4    4 i                I4 4

i 4 4 i  : 4 + +Y 4 0 4 +

                                                                                , 4 4              i                        4                                   +
                     +

i +4 4i

                                                                                                 +
                                                                                                                                +

4i .. . ... D

                    $hfh/hhh//lll///h//hh/h/

WHIT [ POINI 1985 1986 1987 1988 i969 1990 1991 1992 1993 1994 1995 #996 j i?5 3 y iso E O '" I {I Y { } f f i= 4 44 4 11 4+ 1 4 I il II 4 I

         *n                                                                                                                     Y        j               I i                                           I                          4                    {4                                      i y so 4                                        i                                                                  4+

4 n 0-lltt $ $ $ $ $ $ $ ,t $ $ $ ( $ g $ $ $ ; j g J. Fig. 3. Mean shese longih (ann) 393% C.L st Minstone selymns sessions dunas the pened h - - frain 1985 W im 212 Monstonag Stubes,1996 e -'-w-- M yy y w y -t e=

pomen 1981 1985 1989 tm H91 IM2 IW IW N

• 1985 1956 42-400

    -   u 3ao 3

l F

    ; m. 11 i

1 Ym 4-o sw

                                                                                         }
                                                                                                                     }
        -                                                                                                               1 if           {                        4 13 so                                                                                  i        i fI O

lj'fh/$$h////f$flfl/lllll imu mte 1988 1989 1990 1991 1992 1993 1994 1995 1996 1985 1986 1987 42 400

    ,- u
    .e j   u Y=

i. '"

t i i

        ... l                            f i        1                                                      4 S                                                               i f                              +}-           f
          ,             . 4                                                       .        1 +f    ..

f$Ilf///fEf/fll$$//Of$i$

        ,                                                war mi 1985      M86   1987     1988    1989     1990       1991     1992     1993     1994       1995     1996 du 400
    ; a j :w
    ?m G

iw 12

         "                                  '}                                                                 i     I l                                                                       +

o llfhlllllhlhf$f$fhlll$$h Fi36. Mean dry weight (yams per m') 395% C.L st Milhsone esipass manions sampled & mas du period June % freen 1985 hp 1996.

s Ecigrass 213 l

i TABII 2. 1985 Numberthrouth of reprodusve September 1996. shoots, total numtwr of shoots and percentage of reprodunve shoots at selgras Year Annant June Jaly Augen Septeenber W'  % # Totaf  % W Total  % 8 Total  % W Total  % Jorden Cow I 1985 44 1.9 10 561 1.8 23 591 3.9 l 11 514 2.1 0 622 00 1986 70 2.5 23 756  : 3.0 21 585 36 13 1046 1.2 13 464 2.8 1987 72 3.3 18 581 3.1 24 537 4.5 19 496 3.8 11 555 2.0 1988 58 3.1 20 469 4.3 I 11 502 2.2 2 415 0.5 25 487 5.1 1989 30 1.8 16 534 3.0 12 526 2.3 2 356 0.6 0 228 0.0 1990 9 0.7 2 167 1.2 7 365 1.9 0 395 0.0 0 424 0.0 1991 24 1.0 14 448 3.1 10 647 15 0 654 0.0 0 1992 662 0.0 17 0.7 9 558 1.6 8 643 '2 0 708 0.0 0 611 0.0 1993 93 4.8 56 493 11.4 36 5l0 7.1 1 516 0.2 0 417 0.0 1994 2 0.2 230 1 0.4 1 383 <0.1 0 251 0.0 0 1995 261 0.0 3 0.4 6 256 2.3 2 343 0.6 0 646 0.0 0 554 00 1996 35 2.0 16 537 30 18 488 3.7 1 323 0.3 0 414 00 Mantic River 1985 53 3.2 33 414 8.0 19 308 6.2 1 398 0.3 0 532 0.0 1986 15 5.3 3 1 33.3 14 170 8.2 0 95 0.0 0 18 0.0 1987 21 1.8 4 401 1.0 11 242 4.5 6 239 2.5 0 294 1988 44 0.0 3.6 19 356 5.3 17 309 5.5 0 290 0.0 8 273 2.9 1989 68 7,1 36 333 10.8 21 288 7.3 Il 187 5.9 0 150 0.0 1990 53 5.9 19 225 8.4 32 266 12.0 2 189 1.1 0 218 0.0 1991 12 1.2 5 197 2.5 7 276 2.5 0 296 0.0 0 227 0.0 1992 5 0.5 1 229 0.4 4 442 1.0 0 181 0.0 0 81 0.0 1993 134 8.7 94 607 15.4 38 387 9.8 2 350 0.6 0 196 1994 0.0 3 0.6 3 182 1.6 0 340 0.0 0 0.0 5 0 0 0.0 1995 0 0.0 0 149 0.0 0 373 0.0 0 334 00 0 200 0.0 1996 186 15 0 141 476 29.6 45 253 17.8 0 269 0.0 0 244 00 White Point 1985 27 2.4 8 394 2.0 17 290 5.9 2 222 0.9 0 238 0.0 1986 79 9.1 Si 293 17.4 14 161 8.7 6 234 2.6 8 182 44 1987 53 5.8 20 305 6.6 12 238 5.0 13 180 7.2 8 184 4.3 1988 30 4.7 3 186 1.6 13 161 8.1 5 133 3.8 9 164 1929 5.5 61 47 31 461 6.7 32 480 6.7 0 194 0.0 0 204 00 1990 77 10.4 47 199 23.6 25 212 11.8 5 186 2.7 0 144 0.0

  *1991       28        2.9       16       441        3.6      12      308        3.9      0       112     0.0     0     105      0.0 1992        4        0.4        I       270        04        3      194         1.5    0       195      0.0     0     155      0.0 1993       48        3.9      20        403       5.0       17      313        5.4     11      368      3.0     0     156      0.0 1994       43        7.6      29        181       16.0      14      152        9.2     0       108      0.0     0     122      0.0 1995       35        3.7      24        314       7.6       11      234        4.7     0       237      0.0     0     161      0.0 1996       28        41       16        176       9.9       6       12$        48      6       167      3.6     0

• Total number of reprevee-- 2 215 00
• Total number of shoots (vegetauve + reproducuve).

214 Monitoring Studies,1996

i l l the Niantic River. 1-N short-term celgrass impact, owing to the high natural variability of recolonizadon has been reported in the Niantic River emironmental conditions at this site. Jordan Cosc is previously (at NR #1 from 1989-1993; NUSCO shallow, with large sand flats that are exposed to 1994). However, while all previously sampled areas summer heating and, during extremely low ddes. to in the river (NR #1, NR #2 and NR #3) continued to freezing in winter. A continuous temperature show no signs of population recovery in 19%, no recorder deployed on the sediment surface within the indicadons of population decline have been noted at JC study population measured higher water the current sampling station in the Niande River temperatures, relative to ambient temperatures at (NR #4), sampled in 1995 and 19%. the MNPS intakes in 19% (Fig. 2). Temperatures long-term decline / recovery cycles have been were elevated up to 5*C abeve intake temperatures r: ported throughout the distribudon of celgrass since during afternoons on sunny days, and smaller the 1930s. Loss of celgrass has been attributed to a temperature increases appeared related to tidal var ety of causes, ranging from natural, e.g.. ' wasting flushing of warmer water from upper Jordan Cove. disease' (den Hartog 1987), severe storms (Patriquin Regardless of the exact mechanism, since MNPS was 1975), or grazing and uprooting by swans (Marshall not operating and not discharging heated emuent 1994) and other waterfowl (Vermaat and Verhagen during the summer of 1996, the elevated temperature 1996) to human activitics, e.g., eutrophication was a natural phenomenon. Therefore. it is possible (Bulthuis 1983; Orth and Moore 1983; Cambridge that the disparity between JC and ambient and McComb 1984; Neverauskas 1985; Burkholder temperatures observed in presious years was 1993; Taylor et al.1995), decreased light penetration primarily related to these natural factors. (Fletcher and Fletcher 1995; Koch and Beer 19%) Setchell (1929) first stressed the importance of cad changes in near-shore land use (Kemp et al. temperature in regulating celgrass growth and 1983). Short (1988) suggested that a decline in development. It was later shown that celgrass is water quality and the presence of Labyrinthula sensitive to even small temperature variations contributed to declines of celgrass in the Niantic (Thayer et al.1984). Eelgrass fails to produce seeds River in the late 1980s. It is likely that these factors, at temperatures above 15-200C (Burkimider and perhaps further exacerbated by waterfowl grazing, Doheny 1%8; Orth and Moore 1983). Higher water were largely responsible for more recent population temperatures, e.g., from heated power plant losses observed during this study. The high number emuents, could climinate eelgrass from nearby areas cf reproductive shoots in the 19% samples from NR (Phillips 1974; Thayer et al.1984). For example, make h reasonable to expect small transient haA= of another seagrass (Thalassia populations to reappear in the river in the future testudinum) in Florida (Roessler and Zieman 1%9; through seed production and dispersal. Regardless Wood et al.1%9; Zieman 1970; Roessler 1971) and of the nature of emironmental conditions, none of l of a salt marsh grass (Spartina altermfora) in the factors affecting this celgrass population were Maine (Keser et al.1978), declined significantly in related to the operation of MNPS because thermal  ; the vicinity of power plant emuents. These studies l emuent does not reach the Niantic River. indicated that elevated water temperatures increased Some indication of moderate populadon decline respiration beyond levels that could be supported by was also noted at the other eelgrass study sites. plant photosynthesis. Increased water temperatures Given the proximity of the JC study population to in Jordan Cove, regardless of the cause, could stress 3 MNPS, and because modeling predictions indicated I tids population for reasons described above. possible exposure of this population to the MNPS 3-However, the apparent decline of the Zostera unit thermal plume (ENDECO 1977; NUSCO 1988), populadon in JC may be related to factors other than population characteristics and temperature temperature; similar to the Niantic River, shallow conditions at this site have been closely monitored. water in Jordan Cove allows for high rates of Fellowing decline in 1994, the population at JC celgrass grazing by brant, geese and swans. A improved in 1995 and 1996. However, analyses of general decline in shoot length has also been the eleven-year time-series still showed declining observed at JC, possibly indicating that water depths trends in two important population parameters: shoot are decreasing at JC, perhaps due to movement of density and standing stock biomass. This trend is sand into the celgrass bed from nearby sand bars. likely due to relatively high shoot density and The dynamic nature of nearshore sand flats in Jordan standing stock levels in early study years and fluctuating levels in more recent years. Cove have been documented historically (OMNI it is dimcult to associate the general decline of the 1995). If water is becoming shallower at JC, depth related stress mechanisms may have also become hrdan Cove celgrass populadon with a power plant Ecigrass 215

 .___. - _._.._.                              - -..              -             .-  -. . - .                         - - - .- -. - _ ~ _ .

more praaaaa~d accounting f:r population declines previous yea:s when MNPS was operating. I 4 in recent years. indicating that the MNPS thermal plume has, at ( A decline in shoot density, but not standing stock, most, only a minor influence on water temperatures ' was noted at WP. The reasons for this apparent at the JC study site. Thermal plume modeling and decline are unclear. The population at WP is in field studies indicated temperatures increases of up deeper water, and therefore less susceptible to the { to 1-2*C at JC. Increased natural environmental i natural environmental stress mechamsms potentially stress related to sand shoaling within the eelgrass ' affectag the JC populauon. The MNPS thermal bed may have caused declines in shoot density and plume most likely does not affect the population at standing stock at JC. Reduced water depth over the i WP, based on hydrodynamic modeling and thermal bed may cause this population to be more susceptible 4 plume studies. Furthermore, MNPS did not produce to solar heating and grazing by waterfowl a thermal plume during the eclgrass growing season in 1996, when shoot density and standing stock were  ; low relative to many past sampling years when the Refestnces Cited { plant was operating. Therefore, this apparent short-  ! Bulthuis, D.A. 1983. Effects of in situ light term decline of the eclgrass population at WP in

reduction on density and growth of the scagrass 1996 was due to natural variability, and not to  !

MNPS operation. lleterosostera tasmanica (Martens ex Aschers.) , den Hartog in Western Port, Victoria. Australia. . J. Exp. Mar. Biol. Ecol. 67:91 103. 9ecitasions Burkholder, P.R., and T.E. Doheny. 1968. The  ! I biology of eclgrass. Contribution No. 3. Dept. Conservation and Waterways. Town of Shallow water habitats near MNPS at JC and WP Hempstead.Long Island. Contribution No.1227 continued to support relatively stable celgrass populations, while similar habitats in the Niantic Lamont Geological Observatory, Palisades, New York; 120 pp. River have, for the most part, become unsuitable for maintenance of populations. Since 1985, the Niantic Burkholder, J. 1993. Botanist im>estigates impact , of Nitrate on Scagrasses. Coastlines 3:6. ' River eclgrass popultation has declined, and has been Cambridge, M.L., and A.J. McComb. 1984. The characterized by small transient patches. Patch loss of seagrass from Cockburn Sound, Western expansion through rhizome spreading, typical of a healthy population, has not been noted in the Niantic kastralia.1. The time course and magnitude of River over the study period, and the only observed scagrass decline in relation to industrial development. Aquat. Bot. 20:229-243. recolonization event through seed germination was short-lived (<2 years). ha- the Niantic River is den Hartog, C. 1987. " Wasting Disease" and other located well away for any influence of the MNPS dynamic phenomena in Zostera beds. Aquat. Bot. 27:3-14. thermal plume, declines there were related to other Dexter, R.W.1947. The marine communities of a emironmental factors such as water quality, disease or waterfowl grazing. tidal inlet at Cape Ann. Massachusetts: A study Data collected in 1996, when MNPS was not in bio-ecology Ecol. Monogr. 17:261-294. Dillon, C.R. 1971. A comparative study of the operating, demonstrated that natural emironmental variability factored strongly in less extreme primary productisity of estuarine phytoplankton population fluctuations observed historically at WP and macrobenthic plants. Ph.D. Dissertation. and JC. While the WP population is on the fringe of Univ. North Carolina, Chapel Hill. 112 pp. the predacted areal extent of the thermal plume, ENDECO (Emironmental Devices Corporation). temperature monitoring there has never indicated 1977. Postoperational Units 1 and 2, thermal incursion. Therefore, recent declines were preoperational Unit 3 hydtdermal survey of the attributed to natural variablity rather than power Millstone Power Station. Rpt. to Northeast plant operatsort Utilities Senice Co. Elevated temperatures at JC, relative to ambient MNPS intake temperatures, have Evans, A.S., K.L, Webb, and P.A. Penhale. 1986. been awa= red directly in the past, and were possibly Photosynthetic temperature acclimation in two icsponsible for periodic population declines observed coexisting seagrasses, Zostera marina L. and Ruppia maritima L. Aquat. Bot. 24:185-197. at that site. Elevated summer temperatures of up to 4-5'C were observed at JC in 1996, and were Fletcher, S.W. and W.W. Fletcher. 1995. Factors attributed solar warming. These temperature affecting changes in scagrass distribution and increases in 1996 were similar to increases in diversity patterns in the Indian River Lagoon 216 Monitoring Studies,19%

1 complex between 1940 cnd 1992. Bull. Mar. Sci. Orth, R.J. 1973. Benthic infauna of eclgrass, 57:49-58. Zostera marina, beds. Chesapeake Sci.14:258-Folk, D. 1974. Petrology of Sedimentary Rocks. 269. Hempshill Pub. Co., Austin, Texas. 192 pp. Orth, R.J. 1977. The importance of sediment Hollander, M., and - D.A. Wolfe. 1973. Non- stability in scagrass communitiet pp. 281-300 in parametric statistical methods John Wiley and B.C. Coull (edL Ecology of Manne Benthos. Sons, New York. 503 pp. Univ. South Carolina Press. Columbia. SC. Kemp, W.M., W.R. Boynton, R.R. Twilley, J.C. Orth R.J., and K.A. Moore. 1983. Chesapeake Stevenson and J.C. Means.1983. The decline of Bay: An unprecedented decline in submerged submerged vascular plants in Upper Chesapeake aquatic vegetation. Science 222:51-52. Bay: summary of results concerning possible Osterhout, W.J.V. 1917. Tolerance of fresh water causes. Mar. Tech. Soc. J. 17:78-89. by marine plants and its relation to adaptations. Keser, M., B.R. Larson, R.L. Vadas, and W. Bot. Gaz. 63:146 149. McCarthy. 1978. Growth and ecology of Patriquin. D.G. 1975. ' Migration' of blowouts in Spartina alterniflora in Maine after a reduction scagrass beds at Barbados and Curacao. West in thermal stress. pp. 420-433 in J.H. Thorpe and Indies. and its ecological and geographical J.W. Gibbons (eds). Energy and Environmental implications. Aquat. Bot. 1:163-189.  ; Stress in Aquatic Systenut DOE Symposium Phillips, R.C. 1974. Transplantation of scagrasses Series (CONF-771114). Nat. Tech. Infor. Ser., with special emphasis on eclgrass. Zostera l Springfield, VA. marina L. Aquaculture 4:161-176. Koch, E.W. and S. Beers. 19%. Tides, light and Phillips. R.C.1980. Responses of transplanted and the distribution of Zostera marina in Long Island indigenous Thalassia testudmum Banks ex Konig I Sound, USA. Aquat. Bot.. $3:97107, and Halodule wrightil Aschers. to sediment Marshall, N. 1994. The Scallop Estuary: The loading and cold stress. Contrib. Mar. Sci. Natural Features of the Niantic River. The 23:79-87. Anchorage Publisher, St. Michaels, MD. 152 pp. Rasmussen E. 1973. Systematics and ecology of Milne, L.J., and M.J. Milne. 1951. The celgrass the Isefjord marine fauna (Denmark). Ophelia catastrophe. Sci. Am. 184:52 55. 11:1-495. Neversuskas, V.P. 1985. Port Adelaide sewage Rasmussen, E. 1977. The wasting disease of treatment works sludge outfall. Effect of eelgrass (Zostera marina) and its effect on discharge on the adjacent marine emironment. emironmental factors and fauna. pp.1-51 in C.P. Progress report, July 1982-May 1984. EWS Rpt. McRoy and C. Helfferich (eds). Seagrass 85/6. , Ecosystems: A Scientific Perspective. Marcel ' NUSCO (Northeast Utilities Sersice Company). Dekker Inc., New York. 314 pp. 1988. Hydrothermal Studies. pp. 323 355 in Roessler, M.A. 1971. Emironmental change Monitoring the marine emironment of Long associated with a Florida power plant. Mar. Poll. Island Sound at Millstone Nuclear Power Station. Bu!L .2:87-90. Waterford. Connecticut. Three-Unit Operational Roessler. M.A., and J.C. Zieman Jr. 1 % 9. Studies 1986 1987, The effects of thermal additions on the biota of NUSCO.1993. Eclgrass. pp. 33-48 in Momtoring southern Biscayne Bay, Florida. pp. 136-145 in the marine emironment of Long Island Sound at Proceed. Gulf and Caribbean Fish. Inst. Contrib. Millstone Nuclear Power Station. Waterford. No. I165,22nd Ann. Sess. Connecticut.1992 Ann. Rpt. Roman, C.T., and K.W. Able. 1988. Production NUSCO.1994. Eelgrass. pp. 35-49 in Monitoring ecology of celgrass (Zostera marma L.) in a Cape the manne emironment of Long Island Sound at Cod salt marsh-estuarine system, Massachusetts. Millstone Nuclear Power Station, Waterford, Aquat. Bot. 32:353-363. , Connecticut.1993 Ann. Rpt. Sen, P.K. 1%8. Estimates of regression coefficients NUSCO.19%. Ecigrass. pp. 69-79 in Monitoring the marine emironment of Long Island Sound at based on the Kendall's tau. J. Am. Stat. Assoc 63:1379-1389. Millstone Nuclear Power Station, Waterford, Setchell, W.A. Connecticut.1995 Ann. Rpt. 1929. Morphological and phenological notes on Zostera marina L. Univ. OMNI (DDL OMN1 Engineering LLC). 1995. Jordan Cove Study. Calif. Publ. Bot. 14:389-452. Setchell, W.A. 1935. Geographic elements of the marine flora of the North Pacific Ocean Am. , Nat. 69:560-577. j Eelgrass 217

Short, F.T. 1988. Ecigrass-scallop research in the Zostera nolt/l Hornem.: coupling demographic Niantic River: Final report to the Waterford-East and physiological patterns. Aquat. Bot. 52:259-Lyme Shellfish Comnussion. November 15, 281. 1988. 12 pp. Vergeer, L.H.T., T.L Aarts. and J.D. deGroot. Stauffer, R.C. 1937 Changes in the invertebrate 1995. The ' wasting disease' and the effect of community of a lagoon aAct disappearance of the abiotic factors (light intenstiy. temperature. celgrass. Ecology 18:427-431. salinty) ant infection with Labyrinthula sosterne Stevenson, J.C. and N.M. Confer. 1978. Summary on the phenolic content of Zostera marma shoots of available information on Chesapeake Bay Aquat. Bot. 52:35-44. submerged vegetation. US Fish Widl. Serv Wood, E.J.F., W.E. Odum. and J.C. Zieman. 1%9 FWS/OBS-78/66/ InDuence of seagrasses on the productisity of Taylor, D.I., S.W. Nixon, S.L. Granger, B.A. coastal lagoons. Pages 495-502 in A. Ayala Buckley, J.P. McMahon, and H.-J. Lin. 1995. Castanares and F.B. Phleger (eds). Coastal Responses of coastal lagoon plant communities to Lagoons. Univ. Nac. Autonoma de Mexico, different forms of nutrient enrichment -a Ciudad Univ., Mexico, D.F. n===n experiment. Aquat. Bot. 52:19-34. Zieman, J.C. Jr. 1970. The effects of a thermal Thayer, G.W., S.M. Adams, and M.W. LaCroix. effluent stress on the scagrasses and macro-algae 1975. Structural and functional aspects of a in the sicinity of Turkey Point. Biscayne Bay, recently established Zostera marma community. Florida. Ph.D. Dissertation, Univ. Miami. Coral pp. 518-540 in L.E. Cronin (ed). Recent Gables Fla.129 pp. Advances in Estuarine Research. Acad. Press, Zieman, J.C. Jr. 1972. Origin of circular beds of New York. Thalassia (Spermatophyta: Hydrocharitaceae) in Thayer, G.W., W.J. Kenworthy, and M.S. Fonseca. Southern Biscayne Bay, Florida, and their 1984. 'The ecology of eelgrass nudows of the relationship to mangrove h====h Bull. Mar. Atlantic coast: A community profile. FWS/OBS- Sci. 22:559-574. 84-02.147 pp. Zieman, J.C. Jr. Thompson. V.S. 1982. The ecology of the 1978. Final Report. scagrasses of South Florida: A conmmnity Environmental Surveillat.cc and Studies at the profile. U.S. Fish and Wild. Sersice. FWS/OBS-Maine Yankee Nuclear Generating Station 1969- 82/25.124. 26 pp. 1977. Section 5. Estuary Water Temperatures. Zimmerman. R.C., R.D. Smith and R.A. Alberte. Tutin, T.G.1942. Zostera. J. Ecol. 30:217-226. 1989. Thermal acclimation and whole-plant Upbof J.C.T.1941. Halophytes. Bot. Rev. 7:158. Veimaat, J.E., and F.C.A. Verhagen. carbon balance in Zostera marina (celgrass). J. 19%. Exp. Mar. Biol. Ecol. 130:93-109. Seasonal variation in the intertidal seagrass 218 Monitoring Studies,1996

BenthicInfauna In trod ucti on ..... ....................... .................... .....--..--- - --- -..-- ---.--.------- 221 Ma teria ls and Method s .................................................~................................. ~........--....~. 221 DataAnalysis.....................................................................................................................~..........222 Sediments...................................................................................................................222 Tren d Analysic .... ............................... . .......... ..... . ........ ...... ......... .. .... .... . .. .............. . ... ... .... 222 Comm uni ty Analysis ........................................................................................................ 222 Resuts........................................................................................................................................223 Sedimen ta ry En vironm en t ............................................................................................... 223 General Comm unity Com position.................................................................................. 225 Fa una: A b un d a nce ........ . ..... . . .................. .......... .. ................... . ... .. . .... .... ........ .... .... . ... ..... . 227 Num bers of Species ......................................... ... .......................................................... 227 Com m uni ty Dom i na nce ................................................................................................... 227 Do mina n t Ta xa ....... .... .... ................ ..................... ........... ... .......... .......... .. . ... . .............. . . ... 232 s Classifica tion and Cl uster AnalYsis ................................................................................ 240 Discussion...............................................................................................................................244 Conclusions..................................................................................................................................245 References Ci ted .. ................................ .......... ..... ...................... . .... .... . ......... ... . . ........ ......... .... . .. . . 2 BenthicInfauna 219

 >                                  r                                                                                                                                ,

                             ,_ .m- a ~ ' ' -^ '

s l l l i I 1 l l I i

                                                       )

l t f l 220 Monitoring Studies,1996

1 Benthic Infauna latroduction The MNPS ecological monitoring program has l i included studies of soft bottom subtidal infauna in long laland Sound benthic habitats in the the vicinity of the power plant since 1973. This ! vicinity of Millstone Nuclear Power Station (MNPS)

!'                                                                               monitoring program was designed to measure support rich and diverse infaunal invertebrate                  infaunal species composition and abundance, to i

communities. These communities are important as a identify spatial and temporal patterns in community i source of food for numerous invertebrate and structure and a8=aha~ and to assess whether i vertebrate species, including lobsters and demersal observed changes might have been the result of

               ~ fishes (Richards 1963; Moeller et al.1985; Watzin               construction and operation of MNPS. To date, j

1986; Hom and Gibson 1988; Commito and Millstone studies have identified impacts to infaunal i Boncavage 1989; Franz and Tanactedi 1992; i communities that were attributed to Unit 3 intake Commito et al.1995). De burrowing and tube. construction (NUSCO 1987) and to 3-unit operations building activities of infauna also promote nutrient (NUSCO 1988a), as well as to regional shifts in ! recychng from sediments to the water column j species composition and abundance that apparently (Goldhaber et al.1977; Aller 1978; Gaston and were the result of natural events. This report i Nasci 1988), and promote the passage of oxygenated 5 water deeper into the sediments. Presents results from the 19% sampli9g year, and compares them to results summarized from 2-unit The close association of benthic communities (1980-85) and 3-unit (1986-19 % ) operational 4 with the sediments. where most pollutants ultimately Periods at MNPS. accumulate, also makes them effective indicators of { acute and chronic emironmental conditions (Diaz ! Materials and Methods and Schaffner 1990; Warwick et al. 1990; I Somerfield et al.1995). Documented changes in ! benthic communities following disturbance (Boesch Subtidal infaunal communities in the vicinity of i MNPS were sampled twice per year (June and and Rosenburg 1982; Young and Young 1982; j September) from 1980 through 19% at four stations Gaston and Nasci 1988; Regnault et al.1988; Rees j (Fig.1). A sampling year is comprised of collection

'              and Eleftheriou 1989; Warwick et al. 1990; made in June and September of the calendar year.

NAESCO 1994; Prena 1995; Somerfield et al.1995) The Giants Neck station (GN), located 6 km west of prmide a framework of baseline studies that may be l MNPS, is outside the area potentially affected by used to evaluate impacts on benthic marine systems. power plant operations. This station was used to l Emironmental variability is inherent to coastal i identify possible region-wide shifts in infaunal benthic systems (Holland 1985; Nichols 1935; { Holland et al.1987; Warwick 1988; Rees and conununity structure and composition occurring

independently of power plant operations. The intake Eleftheriou 1989; Boero 1994). Natural variability, l station (IN), located 100 m seaward of MNPS Unit 2 LIgether with an incomplete knowledge of how j physical and biological factors influence the and Unit 3 intake structures, is exposed to scour produced by inflow of cooling water and the effects structural and functional ecology of benthic l communities (Diaz and Schaffner 1990), hinders the of penodic dredging. The effluent station (EF),

! located approximately 100 m offshore from the i ability to describe those communities. Dus, long-station discharge into Long Island Sound, is exposed term monitoring studies are necessary to assess

!                                                                            to increased water temperatures and scour, and to changes in marine environments (Thrush et al.1994; l             Prena 1995). Such studies are the principal means               chemical or heavy metal additions to the cooling of characterizing changes in species composition and water discharge. The Jordan Cme station (JC) is

{ fluctuations in abundance, which occur in response located 500 m east of MNPS and is potentially - impacted by 3 unit - operations. The area to acute or chronic climatic conditions (Boesch et al. encompassing this station experiences increases in i 1976; Flint 1985; Jordan ant; Sutton 1985),

surface wa er temperatures of 0.8 to 2.2*C above variations in biological factors, such as competition l and predation (e.g., Levinton and Stewart 1982; ambient during certain tidal stages (primarily ebb Woodin 1982; Kneib 1988), and human activities. tide) due to the 3-unit thermal discharge of MNPS l

i Benthic Infauna 221 1

 - -. - -             . ~ . -          ..- -.               _.    -- . . . . -                        _ _ -              . ~ . .    ~-. _-

i 1 i (NUSCO 1988b). At each station, ten replicate given particle size ($). and p and Kate the location 2 samples (0.0079 m cach) were collected by SCUBA and shape parameters, respectively (Draper and divers using a hand-held coring device 10 cm in diameter x 5 cm deep. Each sample was placed in a Smith 1981). This function was fitted to data separately for 2-unit and 3-unit operational periods 0.333 mm mesh Nitex bag and brought to the and for the current year using non-linear regression surface. Samples were taken to the laboratory, where they were fixed with 10% buffered formalin. methods Two-sample t-tests were used to test for I AAer a minimum of 48 hours, organisms were differences (a=0.05) between the and Kparameters floated from the sediments onto a 0.5 mm mesh sieve of curves, based on data collected during each and preserved in a 70% ethanol solution with Rose operatioid periM. Bengal added to facilitate. sample processingc Samples were examined using dissecting Trend Analysis

            ;;ds% (10x); orgamsms were sorted into major gmups (anneh ankmpods, moHuscs, and others)                               Tbc nonparametric (i.e.. distribution-free) Mann-for later identification to the lowest practical taxon                Kendall test (Hollander and Wolfe 1973) was used to and N OHgwheetes and Anchocals e                                      determine whether the 2-unit and 3-unit time series each treated m, aggregate because of the difficulties                 exhibited      significant     trends,    and    Sen's associated with identifying these organisms.                           nonparametric estimator of the slope (Sen 1%8) was Organisms too small to be quantitatively sampled by                   used to test for tre,nd differences. These two tests our methods (meiofauna; e.g., nematodes, ostracods,                                 M by Gilben (1989) as panicularly well vi+ =. and foraminifera) cre not sorted. Gram                         suited for analyz'ing environmental moraitoring data size and silt / clay fraction were determined from a 3.5 cm diameter x 5 cm core, taken at the time of bm m diWM m@ m 5d and because relatively short time series (n<10) are
                                                                                   ~~

infaunal sampling. Sediment samples were analyzed e. In this repon, plots of the original using & dry sieving meeM describM by Folk m nthly data (June and September), and a graphical (1974)' representation of the linear trend are provided for community abundance, numbers of species, and for Data Analyses '*I********' Sediments Community Analyses Sediment sieve fractional weights were used to Comparisons of annual coHecdons at d construct cumulative curves for 2-unit (1980-85) and N#" wm ma e by calcu a ng the BmWs 3-unit (1986-%) operational periods by pooling the similan.ty mdex between each pair of years, using the June and R T _ _ weights from each sieve used formula (Clifford and Stephenson 1975); for grain size analysis within each operational period, with years serving as replicates. ShiAs in sedimentary enviromnents over the 2-unit and 3 unit 2 min [X,, L,t) operational penods were then quantitatively ==ami using the Gompertz function. This function has a g= '"# sigmoid shape and describes cumulative data (e.g., growth data) that are not necessarily symmetrical IXv + Xa) ' e-I about the midpoint of their range (Draper and Smith 1981). This feature provides the flexibility to fit where 4 is the similarity index between yearj and cumulative data with or without an inflection point year k;X, is the log transformed (In+1) abundance of (s-shaped versus parabolic) within the observational taxon i in yearf: Xa is the abundance in year k; and range. The form of the Gompertz function used was: n is the number of taxa in common, for'which, on average, at least two individuals were found per year. C, = 100exp(-pe*) A nexible-sorting (p -0.25), clustering algorithm was applied to the resulting similarity matrix (lance where C is the cumulative sediment weight at a and WiHiaans 1%7). 222 Monitoring Studies,19%

                                                       \

i1.J/ F North p

1. k m 0:

3 mi

                                             ,r
                            ~P         }

W Niontic Boy p [ \, 5

                                                                       ,              a IN             JC
                                                                              ,               ,,;,, p,;n, ET
                                                                                                                     ?

.o a :::

GN s Fig.1. JC-Jordan Cove) established as part of the long<erm morutoring prograrn for Millstr. . Results and silt clay evntents observed at all stations during 1996 wr6 within the ranges of previous years of 3-Sedimentary Environtnent unit operation. Mean grain sizes at EF in 1996 were above the 2-unit period range, and silt / clay content Parameters used to characterize the wm w tk 2-unit range. Connneh, at JC in sedimentary emironment at infaunal sampling ' ***" E" " ' # # # # ~"" ""E ' strtions in the vicinity of MNPS included analysis cf '"I *I*I **' * * * *" I '*"8*' mean grain size and silt / clay content (Fig. 2). Based muu athe mes W on Munt sh a we g on mean grain size, sediments in 19% were finest at g. mu to chracch JC (0.17 mm and 0.26 mm in June and September, sedimentary emtronments, and allowed statistical respectively) and coarsest at EF (0.52 mm and 0.54 C mPanson of sediments collected during 1996, and mm). Mean grain sizes were intermediate at IN t (198485) aM 3dt (IM) operadonal (0.29 mm and 0.31 mm) and GN (0.40 mm and 0.44Peh Bad on ses of Gompenz paranten mm). Silt / clay contents of sediments collected in nsatM h fining Gompenz cums to me data. 19% were highest at JC (16.1% and 24.9% in Jime s sant Mmnm kween 2-st and 3-umt and September, respectively), lowest at EF (1.3% Penods a noted at de EF W JC stadons, ne end 1.8%), and intermediate at GN (12.2% and s , at U rhd se declimng s Wclay fraction 12.8%) and IN (9.3% and 9.6%). Mean grain sizes and 6e slightly larger grain size since Unit 3 began operation (Fig. 2). Comersely, at JC, increased BenthicInfauna 223

_ - - _ - _ - - - - - - - - - - ~ (NUSCO 1988b). At each station, ten replicate given panicle size ($) and p and Kare the location 2 samples (0.0079 rn cach) were collected by SCUBA and shape parameters, respectively (Draper and divers using a hand-held conng desice 10 cm in Smith 1981). This function was fitted to data diameter x 5 cm deep. Each sample was placed in a separately for 2-unit and 3-urut operational periods 0.333 mm mesh Nitex bag and brought to the and for the current year using non-linear regression surface. Samples were taken to the laboratory. methods. Two-sample t-tests were used to test for where they were fixed with 10% buffered formalin differences (a=0.05) between the and Xparameters After a minimum of 48 hours, organisms were of cunes based on data collected dunng each Doated from the sediments onto a 0.5 mm mesh sieve operational period. and preserved in a 70% ethanol solution wiJi Rose Bengal added to facilitate sample processing.

 ,                                 Samples      were       examined   using    dissecung                                                                           Trend Analysis microscopes (10x); organisms were soned into major groups (annelids. arthropods molluscs. and otherst                                         The nonparametne (i.e.. distribution-free) Mann-for later identification to the lowest practical taxon                 Kendall test (Hollander and Wolfe 1973) was used to and counted. Ohgochaetes and rhynchocoels were                         determine whether the 2-unit and 3 unit time senes each treated in aggregate because of the difficulties                  exhibited                                           significant     trends,  and Sen's associated with identifying these organisms.                           nanparametne estimator of the slope (Sen 1968) was Organisms too small to be quantitatively sampled by                   used to test for trend differences. These two tests our methods (meiofauna: e g., nematodes, ostracods.                   were suggested by Gilben (1989) as panicularly well copepods. and foraminifera) were not soned. Grain                    suited for anahzing environmental monitoring data size and silt / clay fraction were determined from a                because no dis'tribudonal assumptions are required.

3.5 cm diameter x 5 cm core, taken at the time of and because relauvely short time series (n<10) are mfaunal sampling. Sediment samples were analyzed acceptable. In this'repon. plots of the original usmg the dry sieving method described by Folk monthly data (June and September). and a graphical representation of the linear trend are provided for community abundance. numbers of species. and for Data Analyses '****' Sediments Community Analyses Sediment sieve fractional weights were used to mpansons of anmial coMons at each construct cumulative curves for 2-unit (1980-85) and stanon were made by calculating the Bray-Cunis 3-unit (1986-96) operational periods by pooling the si any inh been each pak ohears. using tk June and September weights from each sieve used f rmula (Clifford and Stephenson 1975): for grain size analysis within each operational period, with years sening as replicates. Shifts in sedimentary emironments over the 2-unit and 3-unit operational penods were then quanutatively assessed N 8v'Idl using the Gompertz function. This function has a S# g = '" sigmoid shape and describes cumulative data (e g.; Xu ~ X1) growth data) that are not necessarily sy1nmetrical e-/ about the midpoint of their range (Draper and Smith 1981). This feature provides the flexibihty to fit where S3 is the similanty index between yearj and cumulatia a with or mtiaut an inflecuon point year t Xg is the log transformed (In+1) abundance of (s-shaped versus parabolic) within the observadonal taxon / in yearj;X, is the abundance in year t and range. The form of the Gompenz funcuon used was: n is the number of taxa in common. for which. on average. at least two indniduals were found per year. C, = 100exp(-pe**) A flexible-soning (p = -0.25). clustenng algonthm was applied to the resulting similanty matnx (Lance where C. is the cumulative sediment weight at a and Williams 1967). 222 Monitoring Studies.1996 __ _ _ _ _ _ _ - _ _ . - - - _ - - - - - - - - - - - ~ ' - - - " " - __ - '~

_ _ - - _ - - - - - - - - - - - - - ~ - O t ii

                                                                                    \\

I t

                                                                                                  ]1) i i J. \
                                                                                     \
                                                                                                                \

s, [ North P

                                                                                      ]\
                                                                                       <~                l 1km                                           \

1' n i 3 f) s D i M f h

                                                                               , '^ N;                                        j'

( .I Q ~ N p-J l

                                                                                                                        %. s y

f 6 v.h< bh ,. [ (i j '^3. Vi

• l

                                                  /(                                                    IN Y, JC                       % g,,

L }

                                                                                                                                         % g" N        .

f

                                                    .r        So,-> \,                                                    <

eV Q pc,n, j ( GN  % I l Fig 1 Map of the Mi!! stone Pome area showing the locanon of subudal infaunal sampimg statmns (EF-Efnuent. GN-Giants Neck IN=lntake. JC= Jordan Cove)estabhshed as pan of the long4enn niorutonng program for Millstone Nudear Power Station. Results and silt / clay contents observed at all stations during 1996 were within the ranges of previous years of 3-Sedimentary Environment unit operation. Mean grain sizes at EF in 1996 were above the 2-urut period range, and silt / clay content Parameters used to charactenze was below the 2-unit range. Converselv. at JC m the sedimentary em'ironment at infaunal sampling 1996. mean grain sizes were below the 2-unit range. stations in the vicinity of MNPS included analysis of and silt /clav was above the 2-unit range. mean grain size and silt / clay content (Fig. 2). Based Cumt$lative curves based on sediment sicyc on mean grain size. sediments in 1996 were finest at fraction weights (Fig. 3) were used to charactenze JC (0.17 mm and 0.26 nun in June and September. sedimentary em1ronments. and allowed statistical respectively) and coarsest at EF (0.52 mm and 0.54 companson of sediments collected dunng 1996. and mm). Mean grain sizes were intermediate at IN 2-umt (1980-85) and 3-unit (1986-96) operational (0.29 mm and 0.31 mm) and GN (0.40 mm and 0.44 periods. Based on t-tests of Gompertz parameters mm). Silt / clay contents of sediments collected in esumated h fitting Gompenz cums to de data. 1996 were highest at JC (16.1% and 24.9% in June sighnt Merences kuten 2-umt and bunit and September. respectively). lowest at EF (1.3% Penods were notd at tk EF W JC Mons. h and 1.8%). and intermediate at GN (12.2% and s at EF reh ee &dinsg sWela) Mon 12.8%) and IN (9.3% and 9.6%). Mean grain sizes an e s 18My larger grain sh smce Unit 3 kgan operauon (Fig. 2). Conversely. at JC. increased Benthic Infauna 223 _J

ih 6 So o y (FFLU(NT j 30 o f 20 0

               *           '"      ~         '
                                                 ,~ ,_           ..".-K to f       s
                                                                                                                        ,           -            Jm .                        --
             - oe 1 oo h os
                                                                                                  ' p'      %=

N g s ,

= v. N%

                           ,,      g                        ;r
                                                                = #N#lW
                                ; / ,f                    f///////////t                                                                                 '        ,' //

Sa n y GaANTG NCC# b "* F. Y 20 0 w

         & ,, ,                                  5 --. " -,o ~ e e, og ',\ ' a                       *e              ,'R    .

9* s ,G

                                                                                                                              ,,           t-.,              e        e-o 1o
         , os o8

p. \

                                                                                                                                                 8, b                                                    ?N 00 w" 3                                               ~
                             /           t                          t             t      F       ?            t'          t     1

• s
• 5 t So o -

y sNTAvl

      ;                .% o                                                  y\

j f 200 m / f, b , so o f wY QS \,' *~ ~e a e- -e e,. N, N e p- me N y -e y " to et

     ,h                  00 m                   e.

f a2- t1 ' - g,.

                                                                                           ,A p N oo.
                            / / / / / / / / / / / / / / / /                                                                                                   J'        ? .'

So.0 - 7

so JORDAN COVE f

    $ ,, ,o                                                                                   \

5 ,oo f\ f ," N ,a

                                                                                                                                                          ," "g

• Q ,e* , m

                                                                                  ~e    ,)              "*~~*                     *
                                                                                                                                                      'y'          k~
   ,                   08 04 h                  o4-       #'

l o_

                                                                                               < w _= == -

ag  % ,.am'

                                                                                                                                                                          ,o ao y      t      1-                      t       t      1      W      t

• n 4 Fig 2.to Mean 1996. grain size (nun) and siluciay content (%) of sedimeras at Millstone subtidal sand statums for Jurw and September 224 Monitoring Studies,19%

___ _ _ - - - - - - - - - - - - - - - - - - - - - - - - - - - ~ ioo

              =                        '"""'                                                               ?,r silt / clay content and decreased average grain size
              $ ,,                                                                                                                              over the same time period distinguished 3-unit E                        $3ES$5,IdiE,/                                               a              [                          sediments from 2-unit sediments. This trend of finer sediments at JC during 3-unit operation continued

( / / through 1996, as these sediments (based on F** / / j ' cumulative curves) were also significantly different d :s 7 " ,. . ] / from those collected during 2-unit operation. V General Community Composition 4 oo 2 oo i oc o.So O 25 0.12 o o6 o o3

                                                                             ~ Sm "

ioo Parameters used to monitor general benthic y community composition included numbers of species 3 yn Y"2. ".' kr.%~.'?ypogg, j f and of individuals in major invertebrate groups

                                                                                                                  ,y" j

im collected during 1996. and means of these

                                                                                                              .g                              parameters for 2-unit and 3-unit operational periods r                                                                                       '          "

(,o ,_ / (Table 1). Number of species in 1996 was highest at EF (94), lowest at JC and IN (80) and intermediate g < -

            " #'             6 >' ,'                                                                                                          at GN (86). At both GN and JC, the numbers of species in 1996 were lower than 2-unit and 3-unit operational period means (86 vs.107 and 92
                  " oo              2 oo     i oc o so c, , 5'a (-)

o.23 o >2 e os o o, respectively at GN; and 80 vs. 100 and 93, ioo respectively at JC). At EF, the number of species in 1996 was the same as the 3-unit mean (94), but less 3 '"',",'..i n ,, $ti. ,*.T.'?pogs im [ than the 2-unit mean (110). The 1996 number of 5 species at IN was greater than the 2-unit mean (68). 48 h but less than the 3-unit mean (81). Among-station [ ,, //,8 relationships were similar between operational periods; mean number of species was highest at EF, g ' d52 lowest at IN. and the intermediate numbers of

                                         ,j[-                   #                                                                           species at JC and GN were similar to each other.

Total number of organisms collected in 19%

               " . oo              2 oo i on o .o o- se -

o as o >2 o o, o e, was highest at JC (6,31I), lowest at IN (1,948), and ico intermediate at GN (4,581) and EF (2.889). The

                                   *** c o*                                                                                  p              relationships among sampling sites over both 3                                                                                                                c operational periods were similar to those in 1996:

e ,, $37,0.~'lYdl# ,c 8 highest abundance at JC (7.115 and 7.578. during 2-E / .n" [ F ,, [,-# ..# unit and 3-unit periods, respectively). and lowest at IN (1,565 and 3.048, respectively). Total abundance N* in 1996 was low at EF, GN and JC relative to l2' v [

'f

• operational period means. At IN. 1996 total abundance was lower than the 3-unit mean, but y# higher than the 2-unit mean.

4 00 1 00 1.oo o so o- se W 0 25 c i? o oe o os Most of the irivertebrate species identified in 1996 were polychaetes; the numbers of polychaete species ranged from 46 to 50 (Table 1). Polychactes Fig. 3. Cumula6ve curves based on frsenonal weights of sedamma had also been the dominant taxon in the 2-unit and conected durine the 2-unit (19801985) and 3 unit (1986- 3-unit operating priods (species number ranges of 1996) oper:6 anal penods. and dunns 1996 at Mustone

              'ub6 del stanons.                                                                                                           35-58      and 44 51, respectively). Polychaetes were also most ab.ndant in terms of numbers of individuals, accounting for more than half of the total individuals it all stations bel ?F Mollusc and Benthic Infauna 225 l

l

 , _ _ - _ _ _ _ _ -                   - - - - - - -                                                                                                             ~      ~

TABII 1. Total number of species (S) and number of axhviduals (N) for each major taxon collected at the MNPS infaunal stations during 1996, and annual means and Coefficient of Vanation (CV) during 2. unit (1980-1985) and 3-unit (1986 1995) operational years. 1996 2-Unit Penod(19801985) 3 thut Period (19861996) (5) (N) MEAN E MEAN G MEAN G MEAN G (S) (N) (S) (N) Effluent Polychaeta 50 624 58 3.7 2431 16.9 51 2.8 il8A 13.0 Oligochaeta - 1816 . 1809 8.9 . . 2120 8.I Mollusca 21 230 21 3.1 303 22.6 20 5.7 253 16.0 Arthropoda 23 176 31 5.7 520 19.7 23 6.5 251 140 Rhynchococia . 43 . - 104 31.5 . . 92 34 4 Total 94 2889 110 $167 94 3900 Otants Neck Polydseta 50 3034 58 4.4 3654 1.8 51 2.9 3745 10.3 Ohgochaeta . 1870 . . 1000 14.0 . . 1134 8.7 Mollusca 18 113 20 9.5 188 15.0 19 4.9 153 13.3 Arthrupoda 18 221 29 4.7 517 32.4 22 8.7 615 40.8 Rhynchocoela - 43 . . 39 26.4 . . 37 15.4 Total 86 4581 107 $398 92 5684 10L42 Polychaeta 46 1060 35 4.1 838 19.3 44 4.2 1761 19 6 Ohgochaeta - 435 . . 133 18.7 . . 30g 13.9 , Mollusca 14 180 12 18.0 156 31.9 15 6.2 235 14.4 Arthropoda 20 256 21 9.9 432 14.1 22 5.5 732 30.2 Rhynchocoela - 17 . . 3 26.2 . . 12 14.9 Total 80 1948 68 1565 81 3048

2. gala!L921's Polychaeta 47 4407 55 4.1 3972 24.1 53 1.9 5076 10.6 Oligochaeta .

1405 . . 2540 16.1 . . 1413 7.2 Mollusca 18 386 22 14.1 338 25.2 21 4. I 441 7.4 Arthropoda 15 102 23 12 4 219 19.7 19 5.6 605 38.2 Rhynchncosta . II . . 46 27.4 . . 43 16.7 Total 80 6311 100 7115 93 7578

• C.V. of the mean asumate = (Standard Error /Mean) x 100 t

226 Monitoring Studies,1996

arthropod species were not as numerous as (1986-96) was similar to that observed during the 2-polychaetes; ranges for numbers of species in 1996 unit operational years. However, analyscs of long-were 15-23 (Arthropoda) and}}