B12700, Corrects Error in 821029 Environ Rept OL Stage Re Mortality Coefficient of Winter Flounder Used in Second of Two equivalent-adult Methods.Revised Calculations Encl

Corrects Error in 821029 Environ Rept OL Stage Re Mortality Coefficient of Winter Flounder Used in Second of Two equivalent-adult Methods.Revised Calculations Encl

ML20235U676
Person / Time
Site:	Millstone
Issue date:	10/07/1987
From:	Mroczka E NORTHEAST NUCLEAR ENERGY CO., NORTHEAST UTILITIES
To:	NRC OFFICE OF ADMINISTRATION & RESOURCES MANAGEMENT (ARM)
References
B12700, NUDOCS 8710140194
Download: ML20235U676 (26)

Text

--

F ..

.o q ?Y l

?@ h. w b

. UEES

,,,, umou sa., as m. cww General Offices

• Seiden Street, Berlin, Connecticut 9 meia .=amn atomc (*= P.O. BOX 270 m.ou weuoma c""" HAnTFORD, CONNECTICUT 06141-0270 k k ] C $,,",I',C.~', (203) 66s-5000

{l' October 7,1987 Docket No. 50-423 B12700 Re: 10CFR51

. U. S. Nuclear Regulatory Commission Attn: Document Control Desk Washington, D. C. 20555

References:

1. Millstone Nuclear Power Station Unit No. 3 Environmental Report, Operating License Stage, Vol. 2, Section 5.1.3.3, October 29,1982.

2. Final Environmental Statement related to the operation of Millstone Nuclear Power Station, Unit No. 3, U. S. Nuclear Regulatory Commission, December 1984.

Gentlemen:

Millstone Nuclear Power Station,~ Unit No. 3 Unit No. 3 Environmental Report - Operating License Stage In Reference (1), Northeast Nuclear Energy Company (NNECO) described three methods of assessing three-unit impact of the Millstone Nuclear Power Station (MNPS) on the winter flounder. These findings were included in Reference (2) by the NRC. The first two involve so-called " equivalent-adult" calculations and the third is a deterministic impact assessment model developed specifically for the {

determination of MNPS impacts.

During recent discussions between NNECO and te Connecticut Department of l Environmental Protection Marine Fisheries Office, the latter noted an apparent  ;

error in a mortality coefficient used in the second of the two equivalent-adult i methods. The purpose of this letter is to correct that error using the proper j mortality coefficient with more current data than available at the time i References (1) and (2) were completed. The revised calculations are included in Attachment 1. An evaluation of the equivalent-adult method of impact assessment is also included (Appendix to Attachment 1) to provide a perspective on its reliability.

NNECO believes that the deterministic model used in References 1 and 2 I provides the best and most appropriate measure of impact and the conclusions reached on the basis of this model remain in effect. A conservative potential f

reduction of 5' to 6% in Niantic River winter flounder population size was projected to occur af ter 35 years of three-unit MNPS operation, followed by a recovery to within 1% of the equilibrium population level af ter an additional 65 l years.

8710140194 871007 g Ogl PDR ADOCK 05000423

C PDR g

gr my; - - - - - - - - - -

c .

J .r l '

,' pr *

. L , : f

] y .n r s in , ,

h '

^', ,

'Q ,1 < J U. Si Nuclear Regulatory Commission -

D ' " ~

, , . B12700/Page 2 4 4. i October 7,1987-9, w > -

. , +" If 6 you~ havej any questions .concerning this submittal, please call Mr. Paul M. .

4. /, . , '

, = 3acobson, Manager, Northeast Utilities Environmental Laboratory, at (203) 444-4239.

-3.- '

~

'Very truly yours, NORTHEAST NUCLEAR ENERGY COMPANY j /

f1 ,

EMpr~oczka (/

p; (, Senior Vice President

-cc:- Mr. Charles Fredette 6 *

. Department of Environmental Protection

< , . Water Compliance Unit

'122 Washington Street

Hartford, CT

,4

. Mr. Eric Smith .

Department of Environmental Protection

-Marine Fisheries' Office

,,

Waterford, CT 06385 Mi. Richard Barlow,' Director .

Water Compliance Unit 71 '.. '

' Department of Environmental Protection 4

165 Capitol Avenue

, ,, Hartford, Connecticut 06115

> Mr. Stanley J. Modzelesky s

- Administrative Assistant Connecticut Siting Council

' l Central Park Plaza New Britain, Connecticut 06051 W. T. Russell, Region I Administrator R. L. Ferguson, NRC Project Manager, Millstone Unit No. 3

- > .W. 3. Raymond, Resident inspector, Millstone Unit No. 3-

'a' . ' _

=,

w a: .

t Docket No. 50-423 B12700

.r:

9 Attachment 1 Millstone Unit No. 3 Equivalent - Adult Revised Calculation October,1987 i

4 l Docket No. 50-423 k'

B12700 Attachment 1 Millstone Unit No. 3 Equivalent - Adult Revised Calculation

,\

October,1987

1 y-EQUIVALENT-ADULT REVISED CALCULATION-f E

The potential impact of 3 unit operation at Millstone Nuclear Power Station (MNPS) on the Niantic River winter Counder population was' assessed by three methods in the " Millstone Nuclear Power Station Unit 3 Environment Report Operating Iicense Stage" (NUSCo 1983; pages 5.149 through 5.1-63) submitted to the Nuclear Regulatory Commission in 1983. The three assessment methods were:

the equivalent-adult method of Ilorst (1975), a modified equivalent adult method (Goodyear 1978), and a detenninistic model involving both hydrodynamic and population projcetion submodels (lless et al.

1975). -The second of these methods was the direct conversion of larval entrainment to equivalent adults using the length frequency distribution of entrained larvac classified into 0.5-mm size classes.- The estimated natural larval mortality per 0.5 mm of length increment was applied to each size-class to detennine equivalent juvenile loss due to entrainmenti After estimating the equivalent. juveniles entrained, juvenile and adult mortality rates from the literature were used to calculate the equivalent age 4 adult loss. In doing so, the juvenile monthly mortality rate of 0.30 (Pearcy 1962) was mistakenly used as

. a survival rate in the calculations resulting in the underestimation of the equivalent-adults.

The present document presents the results of a new assessment of larval entrainment using the same

~ method with updated larval entrainment data (1976-85) and recent mortality estimates directly derived from our data on the Niantic River winter flounder population. The equivalent adult method of impact assessment and its assumptions are also discussed in light of what we now know about the local winter flounder population. The methods and infornmtion used for the revised calculation are provided in an appendix.

I,ife table of the Niantic River winter floimder population Ikjuivalent-adult calculations require mortality rates for various life history stages. The most recent estimates of these rates for the Niantic River winter flounder are given in Tabic 1. Mortality rates that were directly derived from our data on the Niantic River population include mortality from hatching through stage 3 larvac (3 mm through 7 mm), juvenile mortality during the first 150 days (May through September), and annual mortality rates of age 2 and 3 fish. The methods and data used I

.a Table 1. Curre.it life table for Niantic River winter flounder population.

Mortality Rates Stage or I) oration Cumulative Percent (7.) (7) Cumulative Age-class (dnys) days I)nily I)aily Total Total Mort.

I gg 15 15 - -

1.628" h

I.nrval stnFen 1.2, & 3 55 70 4,59 .047 2.5R3 G1 (3 to 7 mm) 1,arval singe 4 10 R0 3.25 .033' O.330 (7 to 9 mm) h Juvenile i 150 230 1.9 .019 2.R 50 d

Juvenile il 135 365 1.2 .012 1.639 9.030 Age 1 365 -

.53 .005 1.930' h

Age 2 365 -

.20 .002 0.725 h

Ae3F 365 --

.20 .002 0.725 12.410 Total mortality (7.) from egg to nge 4 = 12.41 , , , ,

2 Mnximum nilowable (7) for n vinble population: -log,  ; = 12.41 J.

Menn fecundity (/) - 490,l00 eggi per female I

• lislimnted as the difl'erence between 7, = 12.41 nnd the total (7) from hatching to age 4 h

lislimated from our datn on the Ninntic Itiver winter flounder (see Appendix)

• I!stimnted by linent interpolation between the daily (/rs corresponding in stage 3 and juvenile 1.

d I:stinmted by linent interpolation between the dnity (7)'s corresponding to juvenile I and age 1.

• I)erived by llean et al. (1975)

I Fstimated fiom our data (see Appendix Tnble 2) 2

to estimate these rates are provided in the Appendix. Rates that were estimated by linear interpolation between the daily instantaneous mortality (7.) of the preceding and sueeceding life-history stages were stage 4 larvae (7 to 9 mm) and juveniles during their last 135 days of development. The annual mortality rate of age i fish was from liess et al. (1975). Total mortality from spawning to the age where the fecundily is equivalent to the mean population fecundity (f) must not exceed the instantaneous mortality calculHed as: , , , ,

2 Zt " " IDEE  :-

J.

for the population to remain viable (llorst 1975). Therefore egg mortality rate or hatching success was estimated as the difference between the maximum allowable total mortality of 7, = 12.41 (Table

1) and the total mortality from hatching through age 3 adults. Ilceause the estirnated population mean fecundity of 490,l00 eggs per female spawner corresponds to the mean fecundity of a 4.3 year ohl fish .i (see Appendix Table 2), choosing age 3 fish as the cut off point in the life table was conservative.

Equivalent-adult calculation for 3-unit entrainment The average annual entrainment of larval winter flounder under 3-unit operation was estimated as 175 million distributed according to the length-frequency distribution shown in Table 2. Supporting .

i information on entrainment estimation, entrainment mortality, and length-frequency distribution are j provided in the Appendix. lintrainment loss in Table 2 represents larval mortality due to entrainment.

This mortality was estimated as 100% for larvac in the 3.0. to 7.0-mm size classes (stages I, 2, and

3) and as 20% for the 7.5- to 9.0-mm size-classes (stage 4). 1.arval instantaneous mortalities per unit time (Table 1) were converted to the following mortalities per 0.5 mm of length increment: 0.287 for the 3.0- to 7.0-mm size-classes and 0.0825 for the 7.5- to 9.0-mm size-classes. The application of these mortality rates to the 175 million entrained larvac results in an equivalent loss of 40.686 million of i newly recruited juveniles, which in turn are equivalent to 456,980 age 1, 66,330 age 2, 32,123 age 3, or 15,559 age 4 fish.

1 i l

\

\

I

Table 2. Estimated winter flounder entrainment anIl equivalent juvenile recruits.

Number l'ntrainment Total Z to - Juvenile Sire- entrained loss juvenile equivalents

'6 class (x 10 ') (x 10 ) recruitment (x .10 ')

3.0 15.312 15.312 2.913 0.832 3.5 13.971. I3.971 2.626 0.794 4.0 12.R26 12.826 2.339 1.237 4.5 14.415 14.415 2.052 1.85; 5.0 . 19.817 19.817 1.765 3.392 5.5 19.604 19.604 1.47R 4.472 6.0 21.l l0 21.110 1.191 6.416 6.5 22.013 22.013 0.904. R.9I 4 7.0 17.661 17.661 0.617 9 529 7.5 11.039 2.208 0.330 1,587 8.0 6.440 1.288 0.24R 1.006 R.5 2.926 0.5R5 0.165 0.496

! 9.0 0.866 0.173 0.083 0.159 l

l 40.686 "

" Juvenile recruits are equivalent to: 456,980 age 1, 66,330 nye 2, 32,125 nFe 3, and 15,559 #Fe 4 lish.

l l

l 4

L__________. s

4 t

L Discussion The assessment method known as equivalent-adult (llorst 1975) providos " worst case' approximated projections of adult fish lost as a result of larval mortality caused by entrainment. The method uses the probability of natural survival of the larvae to extrapolate the number of ' equivalent" adults that the entrained larvae could have contributed to the spawning stock. The three major assumptions of this methodology are that the population is at equilibrium, that recruitment is strictly proportional to the parent stock, and that the probability of natural survival for the entrained larvae is known. A discussion of how well the equivalent-adult method and its assumptions apply to assessing larval

~

entrainment at MNPS follows.

The direct extrapolation of equivalent adult fish from the annual estimates of larvae entrained at MNPS leads to unreasonable predictions of adult fish losses because the assumptions for such extrapolation do not apply to the Niantic Iliver winter flounder. I;irst of all, the method must assume that larvac in the vicinity of the plant intakes (in Niantic Ilay) cither wouhl be able to find their way into the Niantic itiver or would survive to become adults at the same rate as larvac in the Niantic Iliver. The fate of larvae found in the northern portion of Niantic Ilay (likely to be susceptible to entrainment),

was investigated by liess et al. (1975) using a model of tidal circulation and larval dispersal. They found that only 25% of the initial numbers were left after only four tidal cycles and that the loss was due to tidal flushing offshore in a net castward movement through Twotree Island Channel. This compares to 85% of the larvae in the Niantic River remaining there after four tidal cycles. Thm findings suggest that, relative to the Niantic River, the Niantic Ilay has poor larval retention characteristics, g likely resulting in a high loss of larvae from the llay. Although it could be argued that hydrodynamic (

modeling may not apply well to all larval stages, liess et at (1975) validated the model predictions through dye studies and the model assumptions should at least apply to early larvae (stages I and 2),

which account for an average 33% of the total annual entrainment at MNPS. 1. ate larvae (stages 3 and 4) exhibit a diel behavior (NUSCo 1987) which could make them less susceptible to tidal Gushing.

Ilowever, substantial numbers of these larvae may still drift offshore during the time they are in the water column at night. It would also seem reasonable to expect substantially lower survival rates in the flay for late stage larvae that are not flushed ofTshore and for emly juvenile fish than in the Niantie Itiver nursery area. Therefore, the direct extrapolation from entrained larvae to equivalent aduhs, couhl

'W 5

4

L , .'

]

he used to assess entrainment at MNPS only after adjusting total entrainment for potential losses duc -

to flushing out of the Ilay and to higher natural mortality for the larvae remaining there. 3 The estimated 92 million stage 3 larvae (56% of 175 million) entrained annually under 3-unit operation is equivalent throagh backcalculation to about 3 billion spawed eggs. This production requires about 6,200 adult females. Although it seems possible that some of the entrained larvae might have been spawned in areas west of Niantic Ilay, the above number of 6,200 required spawners is not large 1 i enough to invalidate the assumption of the Niantic iliver as the major source of entrained stage 3 larvae at MNPS. This is not inconsistent with the findings of Iless et al. (1975), who indicated that reganlless nf plant operation, almut 4% of the larvac hatched in the river were flushed out of the system in each tidal cycle. Over 80 tidal cycles (about 40 days, the approximate development time from hatching through stage 2), as much as 96.18% of the initial larval production could become available for entrainment or lost ofTshore.

Although not as critical for extremely conservative impact assessments, the assumptions of equilibrium population and recruitment proportional to parent stock are not met either. There is new evidence suggesting that recruitment of Niantic River winter flounder (Fig.1) is very sensitive to water temper-atures following spawning (February and March), and that recruitment rates are also negatively alTeeted by large parent stocks which produce large initial numbers of larvac. This infonnation was acquired recently through the analysis of possible stock-recruitment relationships and it further supports our contention that equivalent adult calculations will not provide valid assessments of larval entrainment at M NPS.

In summary, two equivalent-adult methods and a population and hydrodynamics modeling approach were used in NUSCo (1983) to assess the impact of larval entrainment on the Niantic Iliver winter flounder. An examination of the assumptions inherent of the equivalent-adult methodology suggests that direct extrapolation from total numbers of entrained larvae to adult fish is not appropriate for estimating impacts at MNPS. The most problematic features of the method are that it i: mores the larval dispersion patterns from Niantic Ilay and that it assumes equal probabilities of survival for larvae ,

in the Ilay and in the Niantic Iliver nursery area. Ily contrast, the impact assessment model developed by the University of Ithode Island (lless et al. 1975) accounts for the different larval retention characteristics of the llay and the Itiver and simulates larval losses with and without plant operation  ;

6 I

i

\

25-R = 4.058

• P

• exp(~0.114

• P)

• exp(-0.177

• Temp) l 20- ,

D , -~~~

's, ,'

l ca, s ,

N 16- /

I h /

p ,'

1. A s (5 l [ j / 's *&,, #4 s / ,

• 4,,, 's 0, 10* j ,s** ,5 m m, *  %

~

% 0, 9

e x l/ ,-

l' +c;. . ~.

.;%;r:s,or 's

. tha[o~v,,~ ,-

I r

/e/ ,

/ e,,s.~,,,m~~ s,~%e , ,%o,fe S.  ! , . ,~,'

,t,/ ,',' . potor,'~ ,,, ,

t

• j

~~~..~~~~~~~~

~_

l 0 ,' , , , _ , , ,

o S 10 15 20 25 30 35 40 PARENT STOCK INDEX l'igure I, Prr timinary stock-recruitment relationship for the Niantic River winter ununder population from 1979 through 19R6 data.

' 7

e to produce more realistic impact estimates. This modeling approach was independently reviewed by flatelle Nonhwest I.,boratories (Thomas et al.1978) and found adequate for the problem at hand given the model assumptions and information available at the time. NUSCO believes that the projections of this model, which included the cumulative impact of entrainment losses over the expected

- lifetime of MNPS, continue to be the most reliable assessment. Ilecause the error in the equivalent-adult calculations had no efTect on the impact projections from the model (5-6% reduction in population size after 35 years of plant operation), the conclusions reached in the Environmental Report (NtJSCo 1983) remain unchanged.

A stochastic impact assessment model including the latest information available on the stock-recruitment relationship of the Niantic River winter flounder is currently under development by NtJSCO. This model will be coupled with a Probabilistic Risk Analysis submodel to assess the effect of environmental variability on the model impact projections over the lifetime of MNPS. The model will also simulate j the dynamics of the winter flounder population under various regimes of postulated annual larval losses due to entrainment. The determination of larval mortality rates, which require reliable larval aging methods, continue to be an important research goal at NtJSCO. In addition, we have plans for investigating the feasibility of using new methods for larval stock identification and for assessing the successful metamorphosis and viability of winter flounder larvac in the Niantic Ilay.

REFERENCES CITED I

Goodyear, P.C. 1978. lintrainment impact estimates using the equivalent adult approach. I J.S .

1:ish and Wddlife Service, FWS/OllS 78/65/ July 197814 p.

liess, K.W., M.P. Sissenwine, and S.II. Saila. 1975. Simulating the impact of entrainment of winter I

flounder larvac. Pages 1-30 in S.II. S,ila, ed. Fisheries and energy production: a symposium.

Irxington Ilooks, D.C. IIcath and Co., Irxington, M A.

Ilorst, J.T. 1975. The assessment of impact due to entrainment of ichthyoplankton. Pages 105-118 I in S.II. Saila, ed. Fisheries and energy production: a symposium. lxxington Ilooks, D.C. IIcath and Co., lexington, MA.

8 l

8i' t 8 i

I Northeast Utilities Service Company 1983. Millstone Nuclear Power Station Unit 3 environmental report. ~ Operating license stage. Vol. 1-4.

1987. Monitoring the mariste environment of long Island Sound at Millstone Nuclear l'ower Station, Waterford, Connecticut. Summary of studies prior to Unit 3 operation.

Pearcy, W.G. 1962. Ecology of an estuarine population of winter flounder Pseudor/curonectes americanus (Walbaum). Ilull. Ilingham Oceanogr. Coll. 18(1):1-78.

Thomas, .I.M., J.A. Mahaffey, K.L Gore, and D.G. Watson. 1978. Statistical methods used to assess biological impact at nuclear power plants. J. Environ. Mgt. 7:269-290. 1 i

I l

h I

I 4

I c

s l

9 1

f

APPENDIX i

Larval mortality rates .

Mortality rates of larval winter flounder were estimated from data collected in the Niantic River.

Starting in 1983, three stations in the River were sampled twice weekly during the occurrence of winter flounder larvac. Data frort 1983 were excluded as smaller larvac were undersampled because of net extrusion (N11SCo 1987). Examination of the length-frequency distribution of farvac collected in the River (three stations combined) showed that larvae in the 3.0-mm size class were most frequently collected (Appendix Itig. I). Mean length at hatching for laboratory reared larvac was 2.94 tr.m (SE= 0.017; n = 160) (NUSCo 1987). Therefore, the number of larvac in the 3.0-mm and smaller-size classes were used as an abundance index of newly hatched larvac. The rapid decline in the frequency of larvae in the 3.5. and 4.0 mm size-classes was attributed to both natural mortality and tidal flushing from the River. Tidal studies conducted in the Niantic River suggested that older larvac (stages 3 and

4) utilize vertical migration in relation to tidal How to enter the river and those within the river use a similar behavior to remain there (NUSCo 1987). The increasing frequency of larvae in the 6.0- to 7.0-mm size-classes can be at least partially attributed to this behavior. The number of larvae in the 7 mm size class was used as an abundance index of laivac nearing the end of stage 3 development.

Survival rate from hatching through stage 3 was estimated as the ratio of the abundance index of the 7.0-mm si7c-class larvae to the abimdance ndex of the 3.0 mm and smaller size-classes larvae.

2 The abundance of young larvac in the River is reduced due to tidal flushing to the ik.f. Although fless et al (1975) estimated the loss of larvae from the entire river as 4% per tidal cycle, they also  ;

determined that the loss from the lower portion of the river was about 28% per tidal cycle. Therefore, the 6aily abundance estimates of larvac in the 3.0-mm and smaller size-classes at the station located in I the lower portion of the River were increased by a factor of 1.93 to compensate for the 28% decline l per tidal cycle with two cycles per day. The abundance index for newly hatched larvae, after adjustment j for tidal Gushing, and for larvac in the 7.0 mm si7e class were calculated by summing the mean weekly abundance (three stations combined) during each larval season. Total larval mortality through stage 3 in 1984,19RS, and 1986 ranged frmn 84.59% to 96.91% (Appendix Table 1). These total mortality rates represent a mean instantaneous rate (Z) of 2.583. Since estimated developmental time from j hatching throue.h stage 3 was 55 days, the daily instantaneous mortality was estimated as Z = 0.047. I j 10 l

I f

l- ------------- _ i

Y NIANTIC RIVER 1984 20' Y . '

5 o

i-b  : o U- . A 10- "

, y)4

. . 7 o.,

2.0

,_ 5.

2.5 30 3.5-

_e __

4.0 4.5 50 5.5 60 2_

5.5 7.0 7.5 80 8.5 N90 LENGTH (MM) 40-NIANTIC RIVER 1985

<! 30-g s ,

Q 20-M  : ..,

d . t 10-

,1ggji 20 2.5 30 35 ELEEEEEbEBL_

4,0 4.5 50 b.S 60 6.b 7.0 7,5 80 85 90 1.ENGTH (MM)

Appendix 17igure 1. l.cngth-frequency distribution by 0.5-men sire classes collected in the Niantic River (<tations combined)in 1984 through 1986.

I1 .

i, .

t h

l l

NIANTIC RIVER 1986  !

30-r 20*

Y

,e i b

o '

i 1 - i 10- , L gf l b< . ..

n 2.0 2.5 30 35 4. 0 4.5 50 5.5 60 6.5 7.0 7.5 B.0 8.5 90 y LENGTH (MM) i 1

1 Appendix Figure 1. Cont.

12

4 L\

F Appendix larval winter flounder abundance indices from the Niantic River and estimated Table i rnortality rates from haiching to 7 ram.

Abundance Index Newly 7.0-mm Year hatched size-class Mortality (%) Z 1984 7,005 635 90.94 2.401 1985 13,733 424 96.91 3.478 , . ,

1986 2,459 379 84.59 1.870 ..

mean = 2.583 }.

I 1

l l

l

l These larval mortality rates are only preliminary estimates until additional aging information is available

. and new simulation studies are conducted to better estimate tidal flushing rates. Previous attempts to age larvac by examining otoliths with a light microscope weic not successful, but an improved technique i

developed at 11niversity of Rhode Island '(Dr. A. Durbin, personal communication) may allow the use of otoliths to age winter flounder larvac.

Post-larval mortality rates Survival rate of metamorphosed young-of-the-year winter flounder from May through September was determined using catch curves with estimated abundance at a station in the lower Niantic River during 1983,1984, and 1985 (Appendix Fig. 2). Monthly survival rate estimates ranged from 0.552 to 0.569 during those three years. The mean of the monthly instantaneous mortality rates was used to estimate i the average daily mortality of Z = 0.019 for the ISO-day period covered by the catch curves.  !

Annual survival rate of age I winter flounder was derived by liess et al. (1975) as Si 2 0.1454, which i

corresponds to an instantaneous mortality rate of Z = 1.93. We have used this estimate in all our j calculations because the catch data of Niantic River age i flounder are not adequate for mortality estimation purposes (NtJSCo 1987).

Survival rate of adult (age 2+) Niantic River winter flounder was estimated using two catch curves with age and abundance data taken during the years 1978-79 and 198183 (Appendix Fig. 3). Neither catch curve included data from 1980 because of inconsistencies found during their analysis (NtJSCo 1987). The mean of the two annual instantaneous mortality rates was Z = 0.725, which was equivalent to m annual survival rate of 0.484. This estimate included both natural and fishing mortalities.

i Population nican fecundity i

A length-fecundity relationship for winter flounder was determined in 1977 using 48 females taken in i the Niantic River. This relationship was expressed as:

2 11ccundity = 0.08238 (length in em)* (r = 0.81) l l

l l

14 l

7; j_ ' 's

! a

- - 1983 - LR r --

E ~4-3 2

z

• 1 w

e o 3- i m

• h * .

• 5

c. * .;

$ Z = 0.137 2 *

$ S (weekly) = 0.872

$ S (monthly) = 0.552 5 r2= 0.60 1-MAY JUNE JULY AUGUST SEPTEMBER 1984 - LR Gr 4-2a w

• x

• t U

• 1 o 3-

• O-

• 5 n.

Z = 0.132 *

$ 2 *

$ S (weekly) = 0.876 , , {

[5 S (rnenthly) == 0.564 r2 = 0.79

• 1-MAY JUNE JULY AUGUST SEPTEMBER Appendix Figure 2. Mortality determined by catch curve for Niantic River young winter flounder from 19R3 through 1980.

l l

l I

i i

1 i

15 l

\

l .-. :

l:

1 f

t i

1985 - LR 1

G 4-x W

2 W

x a

k 3'

• x s C s l

C y x

• W C.

6 Z = 0.120 ,

S (weekly) = 0.878 l

5 S (menthly) = 0.569

-E r2= 0.23 ,

1-MAY JUNE JULY AUGUST Sr r idASER i 1

Appendix l'igure 2. Cont.

16

ed 1978 -79 8*

7- ,N ,

6*

n =

5 ca - 5' 1

o .

bC

~

4~

• Z = 0.657 I S = 0.518 l 3- A = 0.482 r2= 0.96 2-3 4 5 6 7 8 9 10 AGE 1981-83 10' 9- .

a

g. .

7-2 "

$ 6-s ,

=

z T._ 5-Z = 0.793 4- S = 0.453 A = 0.547 3' r2 , o,91 2-4 5 6 7 8 9 10 3

AGE Appendix l'i Fure 3. Mortality determined by catch curve for Ninntic River adult winter flounder from 1978 79 and 19RI-83.

17

Population mean fecundity was calculated for Niantic River winter flounder as shown in Appendix Table 2. Annual survival rates for adult flounder were applied to an arbitrary initial number of 1,000 age 1 Osh to obtain the age-distribution for the population at equilibrium. The survival rate of 0.1454 (lless et al.1975) was used through age'l to obtain numbers in age 2 and the annual rate of 0.484 given above was applied to obtain numbers of fish in age classes 3 through 12. Age I and 2 females j were not mature and thus did not produce any eggs. Ilased on length-frequency distributions, an estimated 50% of age 3, 80% of age 4, and all older females were mature. The length-fecundity relationship given above was used with mean lengths-at age for females to estimate mean fecundity by age. Fgg production for each age-class was then determined by multiplying the age-specific fecundity value by the number of reproductive fish in cach age-class. Appendix Table 2 shows that about one-half of the eggs are produced by age 4 and 5 females and another one-quarter by fish 6 and 7 years old. The population mean fecundity of 490,100 was determined by dividing total egg production by the number of mature fish. This fecundity is equivalent to that of a 4.3 year old female fish.

)

Entrainment estimation IIntrainment estimates are based on sampics collected at discharges of MNPS Units 1 and 2 (18 samples per week in 1975-82 and 8 samples per week in 1983-85). Generally, winter flounder larvac are entrained from late February through the end of .fune with greatest densities from mid April through May. The entrainment estimates were obtained by multiplying the median entrainment density per unit volume during the larval entrainment season times the total water volume passed through MNPS during the same period. The larval entrainment season was defined as the period corresponding to 95% of the annual cumulative abundance which was determined by excluding the sampling dates that provided the first and last 2.5% of the cumulative entrainment over the entire entrainment period.

Restricting the season in this manner reduced the number of zero-valued data points in the tails of the time series of data used in the median calculations. The median of the data distribution was used instead of the arithmetic mean because the data were highly skewed and the mean wouhl not adequately describe central tendency. A nonparametric method (Snedecor and Cochran 1980) was used to construct 95% confidence intervals for cach median density and corresponding entrainment estimate. )

t l

Annual entrainment estimates are available from 1975 through 1985 (Appendix Table 3) during the operational period of Units I and 2. The average annual entrainment estimate during this period was IR l

Appendix Table 2. Niantic River winter flounder mean fecundily calculation.

Population I) distribution Distribution of Percent b h age Fraction of mature Fecundity egg production Contribution (eggs x 10') (eggs x 10*)

Age distribution" mature Osh by age-class I 1000.00 0.0 0.0 0.0 0.0 0.0 2 145.15 0.0 0.0 0.0 0.0 0.0 3 70.30 0.50 35.15 173.1 6,084.5 13.17 4 34.05 0.80 27.24 413.9 11,274.6 24.40 5 16.49 1.0 16.49 673.2 11,101.1 24.03 6 7.99 1.0 7.99 880.6 7.036.0 15.23 7 3.87 1.0 3.87 1,216.6 4,708.2 10.19 8 1.87 1.0 1.87 1,507.4 2,818.8 6.10 9 0.907 1.0 0.907 1,849.6 1,677.6 3.63 10 0.439 1.0 0.439 1,908.8 83A.0 1.81 11 0.213 1.0 0.213 2,052.5 437.2 0.95 12 0.103 I.0 0.103 2,160.2 222.5 ( 48 1 l 94.27 46,198.5 100.0 3

Population Mean Fecundity = 46,198.4 x 10 - 94.27 = 490,100 eggs per female spawner .

'For equilibrium population and the annual survival rates from the life table.

l bl )crived from our data.

l l

l 1

f 19

Appendix Yearly median densitics (n per 500 m 3) of winter flounder larvac in entrainment Table 3. samples during their season of occurrence and annual entrainment estimates with 95% confidence intervals for MNPS Units 1 and 2 from 1976 through 1985.

l l Year Median 95% Cl II, stimate (x106 ) a 95% CI (x10 6) 1976 158.0 114.2-185.I 95.5 69.0-111.9 1977 68.3 54.5-96.3 30.9 24.7-43.6 I978 86.6 65.0-106.4 58.4 43.8-71.7 1979 90.3 69.7-108.4 36.6 28.2-43.9 1980 201.5 163.6-234.6 140.I i 13.7-163. I 198l 139.2 99.3-182.6 47.6 33.9-62.4 1982 183.5 147.5-215.I 137.1 I I0.3- 160.7 1983 244.4 'I58.1-314.8 172.9 111.9 222.6 1984 185.5 107.5-226.2 90.0 52.2-109.8 1985 107.1 78.8-149.5 65.9 48.5-92.1

• 1'he average entrainment for the 10 year period was 87.5 million larvac.

20

87.5 million winter flounder larvac. To assess the potential impact of three-unit operation, this estimate was doubled to 175 million because the water passed through Unit 3 is about the same as the combined volumes used by Units 1 and 2.

I!ntrainment data collected from 1983 through 1985 were examined for developmental stage composition and length-frequency distribution by 0.5-mm size-classes (Appendix Fig. 4). The percentages of larvac entrained were: 2% for stage 1, 31% for stage 2, 56% for stage 3, and 11% for stage 4. Almost 70% of the entrained larvac were in the 5-mm and larger size-classes. A comparison of developmental stage and size-class showed that stage I and 2 larvac were primarily in the 3.0- to 4.5-mm size classes, stage 3 in the 5.0 to 7.0-mm size-classes, and stage 4 in the 7.5-mm and larger size-classes.

i Winter flounder entrainment mortality studies were conducted in 1983 on larvae that had passed through i the plant (Nt1SCo 1987). No stage I larvac were collected during the study. All stage 2 and 3 larvat j were dead at the time of collection or within a 24-hour latent mortality period. Of the 24 stage 4 i larvac collected 19 (79%) survived a 96-hour latent mortality period. Therefore, when assessing the impact of winter flounder entrainment 100% mortality was assumed for all stage 1, 2, and 3 larvac, but only 20% mortality for stage 4 larvae, APPENDIX REFERENCES CITED Iless, K.W., M.P. Sissenwine, and S.II. Saila. 1975. Simulating the impact of entrainment of winter flounder larvac, Pages 1-30 i_n S.B. Saila, ed. Fisheries and energy production: a symposium. D.C.

IIcath and Co., icxington, MA.

Northeast titilities Service Company. 1983. Millstone Nuclear Power Station Unit 3 environmental report. Operating license stage. Vol. 1-4.

Snedecor, G.W., and W.C. Cochran. 1980. Statistical methods (7th Edition). The Iowa State University Press, Ames, I A. 507 pp.

21

m;;[

13' 12' E

b /

3 . .,

10-9-

g 8'

@ 7-g 5, ;d

,,#j 8-2 L E- wn.

3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 ' 7.0 7.5 8.0 8.5 9.0 LENGTH (MM)

STAGE. 1 2 3 4 Appendix Figure 4. percernge of winter flounder larvae entrained in each 0.5-mm size-class by developmental stage in 1983 thtough 1985.

I l

1 22

___ _ _ _ _ _ . _ _