ML20247F255
ML20247F255 | |
Person / Time | |
---|---|
Site: | Pilgrim |
Issue date: | 12/31/1997 |
From: | Richard Anderson, Desmond N BOSTON EDISON CO. |
To: | ENVIRONMENTAL PROTECTION AGENCY |
References | |
51, BECO-5.98.036, NUDOCS 9805190236 | |
Download: ML20247F255 (150) | |
Text
_ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ - _ _ - _ _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ - - _ _ - _ _ _ _ _ _ _ _ _
saman an an Pilgrim Nuclear Power Station Rocky Hill Road April 27, 1998 Plymouth, Massachusetts 02360-5599 BECo Ltr. 5.98.036 Nancy L Desmond Regulatory Relations Group Manager Planning and Administration (SPA)
U. S. Environmental Protection Agency P. O. Box 8127 Boston, MA 02114-8127 s
NPDES PERMIT MARINE ECOLOGY MONITORING REPORT
Dear Sirs:
e The Pilgrim Station Semi-Annual Marine Ecology Report No. 51 is submitted in accordance with Part I, Paragraphs A.8.b & e, and Attachment A, Paragraph 1.F, of the Pilgrim Nuclear Power Station NPDES Permit No. MA0003557 (federal) and No. 359 (state). This covers the period from January through December 1997. I hf N. L. Desmond h'
I l
Radmisc/RDA/ecorpt97 I
Attachment:
Semi-Annual Marine Ecology Report No. 51 i
1SOUC1 i
\ I
\
I g
/ .-
9005190236 971231 PDR ADOCK 05000293 R PDR C_________ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
so.a m.m-,
Pilgrim Nuclear Power Station Rocky Hill Road Plymouth. Massachusetts 02360-5599 Nancy L Desmond Regulatory Relations Group Manager April 27, 1998 BECo Ltr. 5.98.036 Mass. Department of Environmental Protection Regulatory Branch - 7th Floor One Winter Street Boston, MA 02108 NPDES PERMIT MARINE ECOLOGY MONITORING REPORT
Dear Sirs:
i The Pilgrim Station Semi-Annual Marine Ecology Report No. 51 is submitted in accordar.ce with i
Part I, Paragraphs A.8.b & e, and Attachment A, Paragraph 1.F, of the Pilgrim Nuclear Power Station NPDES Permit No. MA0003557 (federal) and No. 359 (state). This covers the period from January through December 1997.
lA/lhk h N. L. Desmond Radmisc/RDA/ecorpt97
Attachment:
Semi-Annual Marine Ecology Report No. 51
Reloted to Operotion ofAlgrim/totion SEMI-ANNUAL REPORT NUMBER 51
\
JANUARY 1997 - DECEMBER 1997
' pt. ..,
{}, Ffi?$D
- *].f. < ' **..',[
b
\;./r>-
$vl4
.. % /' '
~
.r-
. c. w
' ' Th r W }{p's -
' k.]k.j,,
op ,p
. -t -
) v.g BOSTON EDISON COMPANY REGULATORY AFFAIRS DEPARTMENT
& Boston Edison
l l
l l
r l
MARlr2 ECOLOGY STUDIES l RELATED TO OPERATION OF PILGRIM STATION SEMI-ANNUAL REPORT NO. 51 REPORT PERIOD: JANUARY 1997 THROUGH DECEMBER 1997 l
DATE OF ISSUE: APRIL 30,1998 l
l Compiled and Reviewed by: .
Nobert D. Anderson I
Principal Marine Biologist l
l.
1 i
Regulatory Affairs Department Boston Edison Company Pilgrim Nuclear Power Station Plymouth, Massachusetts 02360 l
I l l
4{ . vg,j' f /
- h i-gl 2 m3. ~t i) ,.
y
,k
- fl .
t; .'; .
ft .
- l. .
?)' >~
.. _ ,, , .c -
, ( 'l I:
? , ,
ai s ,' N -
[ ' .$1
- s, .
s'! '
,.~
' ,l/ 1 ,
5*"'
,, c %.,',. .,g
]$
Q j' J (). t x e.- .
s- -
14W I y 1 s. .
, m. . i ,
/
4 - .
h.. ' : x
~
. P' , '
".E
.. , h. <
' d, l l ,.
3:; -.
.t3 ,
5 -
., , a
. t,
' i ' V .4.g +-
L l' , RQg -
!<)- kT*
sp n,
N ;
,g ' ,
ily O ;
1 :
3
. ; g c{ .. .
- 9, .
.i.
~
\. l
) , .
(- ;. ,' ., ,
( .i
,[g -
f ,
' i. l .
, .' s '. ,.
-I t !- g, l,
. .s.
6 .. G , [' . ' .
, i,
- .i.
, . -..:,w...,
, ' , ' :* . !j
. T ,. .
'8
.p * * . 's
'1 i
e' . , ' . ,$ ' .' ' .'
TABLE OF CONTENTS SECTION I SUMM.\RY l 11 INTRODUCTION lil MARINE BIOTA STUDIES lilA Marine Fisheries Monitorino and impact Annual Report on Assessment and Mitigation of impact of the Pilgrim Nuclear Power Station on Finfish Populations in Westem Cape Cod Bay, January - December 1997 (Mass. Dept. of Fisheries, Wildlife and Environmental Law Enforcement; Division of Marine Fisheries) tilB Benthic Monitorino and Impact Benthic Algal Monitoring at the Pilgrim Nuclear Power Station (Qualitative Transect Surveys), January - December 1997 (ENSR Consulting and Engineering) lilC Entrainment Monitorino and Impact lilC.1 Ichthyoplankton Entrainment Monitoring at Pilgrim Nuclear Power Station, January - December 1997 (Monitoring) - (Marine Research, Inc.)
111C.2 lchthyoplankton Entrainment Monitoring at Pilgrim Nuclear Power Station, January - December 1997 (Impact Perspective) - (Marine Research, Inc.)
lilD Impingement Monitorino and Impact Impingement of Organisms at Pilgrim Nuclear Power Station: January - December 1997. (Boston Edison Company)
IV Minutes of Meeting 88 of the Administrative-Technical Committee, Pilgrim Nuclear Power Station Radmisc/ML ;ne90. doc ii
SUMMARY
Highlights of the Environmental Surveillance and Monitoring Program results obtained over this reporting period (January -December 1997) are presented below. (Note: PNPS was operating at high power level during most of January - December 1997 with the exception of a refueling outage between mid-February to mid-April).
Marine Fisheries Monitorina:
- 1. In the April- November 1997 shorefront sportfish survey at Pilgrim Station there were 4,143 angler visits and 1,394+ fishes recorded for a catch rate of 0.34.
Striped bass (45%) and bluefish (54%) dominated the sportfish catch. The presence of a strong thermal discharge component during most of 1990 - 1997 resulted in good sportfishery success compared with outage and low power years.
- 2. During late July - December 1997 fish observational dive surveys, fish species were observed in the thermal effluent area. Striped bass and tautog were the most numerous fishes seen, being abundant in the Pilgrim discharge current.
Striped bass observations peaked in late July /early August while tautog were consistent throughout the summer into November.
Mass. DMF recommendation - Data from the dive and sportfish surveys reveal that certain species are attracted to either the elevated water temperatures (spring and fall) or current. This places them at risk of impact from temperature aberrations, chemical releases, and potential gas bubble disease mortalities.
As such, some form of direct visual monitoring should continue for as long as the plant is operational.
1-1
- 3. Several hundred cunner were tagged from 1990 - 1996. Many were recovered in the Pilgrim vicinity indicating that movement of this species is local consistent with its residential nature. Juvenile recruitment studies were continued to assess Pilgrim impact on this species, which in 1995/1997 was minimal and in 1996 was inconclusive because of adverse weather conditions (i.e., severe storms), which naturally impacted recruitment and influenced study results during the Fall.
Mass. DMF recommendation - An evaluation of impact of Pilgrim Station on the local cunner population has been rendered. Minimal power plant impact was indicated from the recruitment studies in 1995/1997, while in 1996 recruitment results were inconclusive due to meteorological events. DMF ascertains that cunner recruitment is likely controlled by natural post-settlement processes and power plant impact is of minimal consequence.
- 4. Winter flounder tagging in the Plymouth Bay vicinity to estimate adult population size and Pilgrim Station impact has accounted for 14,776 fish with 830 (5.6%)
tag retums by the end of 1997. The 1997 population estimates based on an Area Swept Method and various mark-recapture models for the Plymouth Bay area ranged from 115,353-519,751 adult winter flounder (age 3+). This equates to roughly a 9-41% adult population impact from PNPS entrainment of 55,400,000 flounder larvae which makes it difficult to define the true magnitude of PNPS impact on the species.
Mass. DMF recommendation - DMF is not optimistic that continuation of this study will yield a more accurate or precise estimate of population size. More years of study to define the impact of Pilgrim Station on this species may not provide a definitive answer.
1-2
- 5. Rainbow smelt egg restocking of the Jones River (Kingston) to mitigate the high PNPS smelt impingements in December 1993 (5,100 fish) /1994 (5,300 fish) accounted for 1,800,000 fertilized eggs being transplanted in 1994/1995. Once hatched, these eggs supplemented those produced by the river's spawning population of this species. Both of these impingement incidents have the potential of impacting the local smelt population and were further mitigated in 1996/1997 by improving the smelt spawning habitat in the Jones River to j enhance egg survival, +. rough the use of several dozen specially designed egg collecting trays.
Mass. DMF recommendation - Future Jones River and other local smelt spawning habitat enhancements should consider improving water quality until spawning-run smelt numbers, which DMF monitors, substantially increase.
Impingement Monitorina:
- 1. The mean January - December 1997 impingement collection rate was 1.43 fish /hr. The rate ranged from 0.06 fish /hr (June) to 7.38 fish /hr (November) with Atlantic silverside comprising 46.5% of the catch, followed by rainbow smelt 12.6%, Atlantic menhaden 8.6% and butterfish 8.3%. Fish impingement rates in 1989 - 1997 were several times higher than in 1984,1987 and 1988 when Pilgrim Station outages had both circulating water pumps off and reduced pumping capacity for long periods of time.
- 2. The November / December 1997 Atlantic silverside impingement accounted for 59% of this species' annual collection.
1-3 l
L_________________________________._
- 3. The mean January - December 1997 invertebrate collection rate was 0.45+/hr with jellyfish and sevenspine bay shrimp dorninating. Green crabs and common starfish accounted for 32% of the catch. Nineteen American lobsters were sampled. The invertebrate impingement rates in 1989 - 1997 were similar to those recorded at Pilgrim Station during the 1987 and 1988 outage years, despite much lower circulating water pump availability in these outage years.
- 4. impinged fish initial survival in the Pilgrim Station intake sluiceway was approximately 28% during static screen washes and 61% during continuous washes. Four of the dominant species showed greater than 50% survival overall.
l Benthic Monitorina Four observations of the discharge, near-shore acute impact zones were performed during this reporting period. Denuded, sparse, and stunted zone boundaries were indistinguishable during September 1987 - June 1989 discharge surveys as a result of the PNPS extended shutdown. However, these surveys did note impact zone boundaries in fall 1989 - 1997 primarily because two circulating water pumps were in operation most of the time resulting in maximum discharge current flow. The scouring impact area in 1997 varied from 1,662 m' (March) to >3,587 m2 (September). Except for March, the 1997 denuded and total affected zones were the largest ever seen seasonally during this study (since 1983). This may be due to a combination of heavy mussel settlement on Irish moss, good PNPS operating capacity and possibly dredging siltation effects.
1-4
Entrainment Monitorina:
- 1. A total of 37 species of fish eggs and/or larvae were found in the January -
December 1997 entrainment collections: 16 eggs, 37 larvae.
- 2. Seasonal egg collections for 1997 were dominated by yellowtail flounder, fourbeard rockling and winter flounder (winter - early spring); Atlantic mackerel and labrids (late spring - early summer); rockling/ hake, windowpane and labrids (late summer - autumn).
- 3. Seasonal larvae collections for 1997 were dominated by sculpin, rock gunnel and sand lance (winter - early spring); winter flounder, Atlantic mackerel and cunner (late spring - early summer); hake, Atlantic menhaden and cunner (late summer - autumn).
- 4. No lobster larvae were collected in the entrainment samples for 1997,
- 5. In 1997, an estimated 2.390 x 10' fish eggs and 6.949 x 108 fish larvae were entrained at Pilgrim Station, assuming full flow capacity of all seawater pumps.
On an annual basis, eggs were dominated by the labrid-Pleuronectes group and Atlantic mackerel, and larvae by sand lance sp. and cunner.
- 6. Entrainment sampling, net mesh size efficiency comparisons were conducted showing 0.202 mm mesh more efficient in capturing cunner eggs, and stage 1 and 2 larvae than 0.333 mm mesh.
1-5
7 On several occasions in 1997, tJnusually abundant"ichthyoplankton densities i'
were recorded including Atlantic menhaden eggs and larvae for the most extended time period. This possibly reflects strong annual spawning production for the species involved.
l 1 8. The mean annual losses attributable to PNPS entrainment for the adult stage of
{
three abundant species of fish for 1997 were as follows: cunner 498,281; Atlantic mackerel 1,103; winter flounder 3,414-47,087. None of these losses for l cunner or Atlantic mackerel were found to be significant in the context of preliminary population or fishery effects, respectively. Comprehensive population impact studies are presently being conducted for winter flounder in the Pilgrim area with preliminary estimates being inconclusive, radmisc/jandec97 l
s 1-6 i
INTRODUCTION A. Scope and Obiective This is the fifty-first semi-annual report on the status and results of the Environmental Surveillance and Monitoring Program related to the operation of Pilgrim Nuclear Power Station (PNPS). The monitoring programs discussed in this report relate specifically to the Cape Cod Bay ecosystem with particular emphasis on the Rocky Point area. This is the thirty-eighth semi-annual report in accordance with the environmental monitoring and reporting requirements of the PNPS Unit 1 NPDES Permit from the U.S. Environmental Protection Agency (#MA0003557) and Massachusetts Department of Environmental Protection (#359). A multi-year (1969-1977) report incorporating marine fisheries, benthic, plankton /entrainment and l impingement studies was submitted to the NRC in July 1978, as required by the PNPS Appen'dix B Tech. Specs. Programs in these areas have continued under the PNPS NPDES Permit. . Amendment #67 (1983) to the PNPS Tech. Specs deleted Appendix B non-radiological water quality requirements as the NRC felt they were covered in the NPDES Permit.
The objectives of the Environmental Surveillance and Monitoring Program are to determine whether the operation of the PNPS results in measurable effects on the marine ecology and to evaluate the significance of any observed effects. If an effect of significance is detected, I 1
Boston Edison Company has committed to take steps to correct or mitigate any adverse situation. l These studies are guided by the Pilgrim Administrative-Technical Committee (PATC), which was chaired by a member of the Mass. Department of Environmental Protection in 1997, and whose membership includes representatives from the University of Massachusetts, the Mass.
Department of Environmental Protection, the Mass. Division of Marine Fisheries, the National Marine Fisheries Service (NOAA), the Mass. Office of Coastal Zone Management, the U.S.
Environmental Protection Agency, and Boston Edison Company. Copies of the minutes of the ll - 1
(
l Pilgrim Station . Administrative-Technical Committee meetings held during this reporting period are included in Section IV.
B. Marine Biota Studies
- 1. Marine Fisheries Monitorina A modified version of the marine fisheries monitoring, concentrating on indicator species is being conducted by the Commonwealth of Massachusetts, Division of Marine Fisheries (DMF).
The occurrence and distribution of primarily cunner and winter flounder around Pilgrim Station and in adjacent areas are being determined. Population parameters and related life history statistics are being studied to address Pilgrim Station impacts from entrainment of ichthyoplankton and impingement of juveniles and adults.
Smelt spawning habitat was enhanced in the Jones River (Kingston) in March / April 1997 to mitigate the large impingements of rainbow smelt on Pilgrim Station intake screens in previous years.
A finfish observational dive program was initiated in June 1978. SCUBA gear is utilized on periodic dives from May-October in the PNPS thermal plume area.
Results of the marine fisheries monitor lng and impact analysis during the reporting period are presented in Section Illa.
- 2. Benthic Monitorina The benthic monitoring described in this report was conducted by ENSR Consulting and Engineering, Woods Hole, Massachusetts.
Qualitative transect sampling off the discharge canal to determine the extent of the denuded and stunted algal zones was continued four times a year (March, June, September, and December).
11 - 2 1
Results of the benthic monitoring and impact analysis during this period are discussed in Section 1118.
l
- 3. Plankton Monitorina Marine Research, Inc. (MRI) of Falmouth, Massachusetts, has been monitoring entrainment in Pilgrim Station cooling water for fish eggs and larvae, and lobster larvae.
Information generated through this monitoring has been utilized to make periodic modifications in the sampling program to more efficiently address the question of the effects of entrainment. These modifications have been developed by the contractor, and reviewed and approved by the PATC on the basis of the program results. Plankton monitoring in 1997 emphasized consideration of ichthyoplankton entrainment and selected species adult equivalency analyses. Results of the ichthyoplankton entrainment monitoring and impact analysis for this reporting period are discussed in Sections IllC.1 and IllC.2.
- 4. lmoincement Monitorina The Pilgrim Station impingement monitoring and survival program speciates, quantifies, and determines viability of the organisms carried onto the four intake traveling screens.
Sin :e January 1979, Marine Research, Inc. has been conducting impingement sampling with results being reported on by Boston Edison Company.
A new screen wash sluiceway system was installed at Pilgrim in 1979. This new sluiceway system was required by the U.S. Environmental Protection Agency and the Mass. Division of Water Pollution Control as a part of NPDES Permit #MA0003557.
Special fish survival studies conducted from 1980-1983 to determine its effectiveness in protecting marine life were terminated in 1984, and a final report on them appears in Marine Ecology Semi-Annual Report #23.
11-3
-__7______.______.___________
i a
Results of the impingement monitoring and survival program, as well as impact analysis, for this reporting period are discussed in Section lilD.
D. Station Operation History Tne daily average reactor thermal power levels from January through December 1997 are shown in Figure 1. As can be seen, PNPS was in a high operating stage during this reporting period with a 1997 capacity factor (MDC) of 73.4% Cumulative capacity factor -
from 1973-1997 is 54.4% Capacity factors for the past 15 years are summarized in Table 1.
E. 1998 Environmental Proorams A planning schedule bar chart for 1998 environmental monitoring programs related to the operation of Pilgrim Station, showing task activities and milestones from December 1997 - June 1999, is included after Table 1.
II - 4
a1I4lIl1J._
.!!1. 14 11iI iI
.g a.$ gI =e6<* =o*5; Daao l .
am t t
"mg=l i i u :
o *= =*ncE!*m t::
i
=m_
- !! g.$.E5
- :gmE=,m_.
. m , _
. mI
- .: n E'
- .m !i
!s. .m =..=i,E. E : . m s # _
=E E= b =.iE.=*c E == EH. E.
. E !.
i n
_ fn. .
g:.
Nr.* s' o3=
- .=:::m.:m!g.$=Eb_a:
!8
- .u
- !j . $ :18 !.aG. E.:gm nE E i- .
f E _.i i
' -. a:i u
.ii.
E.. =Eg._
E
$: . rx= ,^ u n..;.: !
..EE"::
5m: :,:!.E .: !.
s Ew*n=
Eie .3.== : ni ::~==:x nn . :.g. b E*l=n.rIy i :. :: E "::::_..m i .
g_.** : . E= i2i..
i.
E _=E g. $_.E 5= n:
E
- .i .
- ::==
- . m
$..k=W:t .n
- .;.I: Et *x x= g ! .::E "i . ::::-. -"i : :: . i
! =:_ E - g.:=:. : ,
- ? . i ! :
i= ==_g._= .:i.i.. . .==. EE = =ni.1.==_.= g 5 =E E === :m==: :. =n =
- 2==.=ri i5e: -s.
I ::< _ E2t== g !. $b =.Ef
.2ii : ,
= -
ex+xj . .- ::=
n E t4 e=: ==._ =
- = ..:
E.."...t . : a.= = g b m. E E == =8.
- .ii
- . i El i=!. i i !. = !
8t. El=.::j:. .= *u. n!=. E.Eig.E. s:e.=m:=ni me .
r isa:e . . : e E. !
i..m
$ HE=.mm : .. =i.. :
l
- g _
. mm. 2 i::
E i:
.:: i,
_m2 : .: . .
E' i..* .== u : == : = g E$ .=E=.=:m. ;j m =u2== == m
. . i :
i ! . .
.. . = E t.m.2
"=::E.=E.._E i
:
i.
. Ee. : .$ =! E :.! m=._
= : .
!:..2i.= . =. =!.
~E E : == a :i. n=_. .g !=. !uu E g. E: .g :=. . =:
e
==~*! : :
.! E E == .
i _: = = = :_. :
2== i. g b =. E.:
. g. .
==E3.8. -
~ = =: : .m .=
- 2=_ . .. :=!1 E_m=~==ue=
!E g. b =. EE:Ei== i
! .= ===x= == ::
E~ :
ii ==..i-.
.m
. 8e==U ; r Ec: ::. ==.j. g b =.l
. .g. .
2=:!* .U.:.t : :. . .
E i = ==:== == =
====u l: :. ::
i E i.. . . .i i- .
. . E E2 5= =i. == g. b E== g N:. =:.s i-E m::m=m=::3:t. m:,,
- . = . .
- *32 . .
- ii n
_ . - - :..: i. .g.
- .=l2n !..
l.
_ :t. ::. :.t :
- .t
!EmE-.
E E :==
- . "'=::=.Ee==.=
i .
_ o
- i. .m.:=c_.:.:
i ~i m= g $ . :
=. =! :
~
- =E=~====..
. : : . . = b = E == ====:= 1..
. 5::1 = ... ,
it a
2= = rG s E.:3*
8 t
t := = = = .
t S
- .EN!:E =E E ::.E == :.== i == E =
$. 3: 2:: .==. 3
. =
m ! ggE E b E.
== :e =:. E.f:s i= =.=~. 5"::. 2_m m: ) r ir:: 3 r::
2.
2
- E,2 :====i.==:.:
g.
%ew
- =:.b . - .g;= ! g= E..a u =: ==
d o
- ~:. = z E"
. ii" : : . . .
==.t . n ~
aP m=. ~=i_. E. *=t ==. :g4_ n
. . .' : .: i
. == i .,g!.
=.ii-
. P'g = *=.== =mm := E !
?En.==..m.
m.Eb
= .l.
. i
. ii i
r m: .i-i i
i i-
=.
! E m=_. m==:=.We , t r
==
a =
3.= == == m== s b =E S=._. ==n.~. i u==. ===_. =
i-
. E.E b=E i. . ii :=
7 E :n.3.
. i i-
=g. !. .= 7 Mlc 9 WE . .
2=: :
E i.:= i- .
9 (
- . . !, $ =m i. :=: *m=:= . - ==n =
- 1 : :.
9 ::. 3 : : u:. .*=
- .!. 9 eN l
1 E. :.= 1 v
E: .*:::
R
. ==n ::
i.
- s. .
. E E. . .= . .
.l
- a i
R em
=. :=: '2 === == i = =._ i b==E =f=e == u=. == i= ==_. =,- EB E .
b =!' Li r
- . .:::::. i
===="::
- f ii.
B
- = =m;.:' 'i. i EE E..=:= .
= rg
=
=-i .
ei l M :.
M wP m:
E=c.b.==... ::===.
E
- ==. U ::: :=:m. t .== == g == .E E_ ==r ===:== == =:
t u E C
E
== =:=:~.
. =
- =:
. 2= == g b =. E E == ===== == =
b=EE. E._. .
=
.=
C E Po or f
. . . . . .. l D
= == . . ": =.
= g b =E E.::. :
.:=.=======
=
x D a7
==m m:.:i .:.
- . :.= =: = == . g . b.:: =EE.:
E .x=.==.!====
=
- . - m9 r 9 Y ! *t*:
. . .. .i =.: Y e1 R
==t'tE=== . g .N. .;E = == == :=
- .:4 . .. R hr
.a===
- = Teb E E3:..::m..=w= g== b E == E. g $:: i.Em n_m==2.i
= =2 A !.n. n;.!:= == . . i_ . !. . A E m:
- =:===
.=
U .
- .2*
.E-i-i: .
U r om N . ~E.3m4 E.. . :: .6::==.===.g.b ;. .
.! !E.!.:_ !.E N t e
, == s .= ca ec
. ! i..
A E.
.n . . . :i
! . ! ~3.
A
- i. .:=-:: :s *E E!. :t: !.
- i. i .
J E. =
E:: En=g.N=s"!
i= J t..
!. =:==n:== eD l j. : i
. m : =: E.n E
32 mg . : n m== .
.I F
=E
.E
.! E m= =u==::
.! E . a .. . . . ::
. . == -
R- ey
. . $.! ! g.
. .E _
l_. 3.:===m.g 8 t g. gr
$.=~=i .s' E t n:
EE C.u:!=Eg bb i"ii.= i: ti mng E. x
.! !i .
. i :
- . E : E:=E . . i aa _
. g.== . 3.n Es!ts =g i=k' E5=g:. . E:==
3.
n :
.s
.u r u -
e n
!e t
.E1 s: i i:
g n .
t! .
E- - _
- is .! i
- e. E- .n v AaJ
- E::i:: e: = .
b l5= =g: ,gE 5 E=
. : i
=tE.:::=::n.m:m.g : : j t
,. $ .E
.l 3..
. B3 '..t S=.
.: ==.1 .
5!,! 5 =EE m i.. =E 1i.. g =s:.=
g_ :.. lym i
ao I.
a.::.: u.4 .
E . . .
Df r i : t r
E 3.W:: i! ::
i! ~2m .:isn c E
. .=
j!.!m:.! "g_! E. :E. ! .E
- .:i s=. ::Ee E..=:. ,:= :: =
. im . i::=
- ~ : t-a =!i.
.:it . i .!
- i
.E E
.=
g.Et=
i !
s::
E.=g =p. U= uw,. i nE 3g.::
!E !! E i :=
. in i: ;.:. :: .:: t-it8
.j .!
.!E :: : :.:n .
. Em i
m:E::5g O 3' tEs:i tm5 !.b=ib= ::g . E::%
E =! : ::. =
1 e
i E: =: .: . . .:
g E
- agg.
m r E= R!'!E3tcm.
~ Eeb3=*i=m i :mg :. a u .
- .::n.,m:=: -:a:
E: . : : : =: . .
g E = E'.E : :!U mE'.a Em ;
.:E:::.m i ::im~~
. . .i sb.g_. . .
gn1.mE b Ei: g::m:! : m iF
=E t2 :.:t"::
. : !.i .,ii=iEge.
==$g.b .:,!!
! : i ! : :
8 t -
u g .=t =: t n,.::m:
i
.EE.E: . a= mE:==.m=i=. g.s:. s
- m:
ti .
i
.. Em. .E i = m. n= .
- g. . m.: : i. i =~ui.b i . =g =..E E =e..==.a .
m:: g.::i..i.s ===. . es .
.i_==_.b
- g. = ! !
3::.n
- .8~r:
m= .g b !E_==.
. . E
.! gs. Eu:. 3 . ..m m.i. g.i.. :: n.:
E
.. =::.
i i i.
tr . :
E= :.t =E- . == = = . g.b=!E =:. i . M = = =::.:
n.U =. . EE
. E E.::.
g .E . =.=
l1 E.2*
8!* m:= t ==
. . g gb.i!
.E E E m=. =:.
5 5..i:: m= E=. .=
ir == m
.!= !
5m 3m.:
= -
.ug .a:: .i = -
_ mEm 5 N =E :==: g._.b_E"Ei= 5.==m=
- . . n:
f: . g
_ . E= EE.t *_:::*n_ ==_ g..! $ _Ei E i . . = m mg ,!m m=.= : . :
. i:.n .
.!. E
. i
.g:::i: . S E. . .
i="= ..:E !:=l$. . . ...: . =
l i
- . im
. uE:lE E "t= :
- 7:.::i: E:ulEe5,.EE_.E En:mlin . E i- :.r=gg.
mEsg a!ug !
Ee su s
g EiE-! g i E"n:3~E 1 3'
gp.niEb.._.E i i ii i R ::e.$ = . . i"lE>.:mg
_;n_:e: :a_ m
" m. Et :. x-i :
- Ei.
.."g := ee
- s4: -
=::.$ : g. . E_ am
- t .
2 E :
. =:- f-E2 g
ui B: :5b=ib !
n _* .
i . .
- =E 3Nhb. E. . mau
...E i E_.g .
3
.Ei i
- t : =E $ $ S b.E E_.g.
E E
_:- C:
s
.E - sum F2t-.nE"E l .- r l .
.aE"E $ $ R.sE . mE @.g uM_::
s :
. ~ c.L: ~
,,t1 ! 3 21c,g 5.,.Jn>
l I l 4l )
J 3 0 0 3 7 3 2 5 1 5 0 6 1 3 8
9 8 0 7 9 7 6 0 3 6 9 8 8 0 _
1 9 9 9 8 9 6 8 8 8 7 7 1 8 ly h
g 4
8 9 0 0 0 0 0 0 0 0 0 0 0 7 1 u 1 0 0 0 0 0 0 0 0 0 0 0 0 0 o
R
(
5 0 3 8 8 3 0 9 5 4 4
% 8 1 1 4 T 9 4 9 1 0 4 5 6 6 1 5 8 9 4 E 1 5 5 8 9 9 8 9 9 7 9 8 9 8 N
C 6 5 7 9 9 5 D 8 0 0 0 0 0 0 0 0 M 9 9 7 6 1 7 1 7 9 2 1 0 0 0 0 0 0 0 0 1 7 G 8 4
N) T 9 I
7 1 9 S
U A 8
9 0 0 0 0 0 0 0 0 0 0 0 0 0 4 R
, 9 1
F 1 0 0 0 0 0 0 0 0 0 0 0 0 0 8E 7, Rs Oe 9B 1 E Te 8 M B5 Cg r 8 T,E 8M9 Ae 9 0 0 0 0 0 0 0 0 0 0 0 0 0 S T 8E 19 FD 1 0 0 0 0 0 0 0 0 0 0 0 0 0 U P 9V Y2 G E1 ,O Y, T3 9 U S ENA C= A 8- N6M I
8 7 5 4 6 8 5 1 0 1 9 A% 9 0 0 0 0 64 6 82 05 523 0 6 7 8 3 Y U1 5 P 00 1 0 0 1 1 1 6 7 2 1 R 5R J - 1 A 1 -
A -
Ct 0 H U EH 1 n 9 4 4 0 9 3 1 5 6 7 8 5 3 C RILBC e R BR OR Tm 9 9 7 0 4 7 6 5 4 1 8 6 4 2 I
Nn 1 9 9 3 5 7 9 5 9 2 9 9 9 7 A EP TA Uor MFA CM Nv n i
1 7 84 O 0 9 4 9 6 7 5 4 2 7 1 4 21 1 93 Oe I
9 5 8 4 2 0 0 0 8 6 4 3 8 8 MMMMM Te 1 9 8 8 9 0 0 0 2 9 9 2 9 5 OOOOO Ah t RRRRR T F FFFF So 2 6 4 4 5 8 8 4 4 Rg t
9 1 8 7 2 6 NNNNN Ein 9
1 6 9 0 3 7 7 7 7 4 2 3 5 9 9 8 5 9 9 9 9 9 7 1 6 0
8 %
4 OOOOO I I I I I Wda TTTTT Oo 4 AAAAA Pl l 3 0 7 2 5 3 9 8 0 0 8 5 RRRRR a 9 9 6 3 4 4 7 0 6 4 8 0 4 0 = EEEEE Rm 9
9 9 8 6 0 7 8 8 8 9 8 9 4 -
PPPPP Ar Ee 1 7 )
7 9
OOOOO Lh t 4 9 NNNNN I
I I I I C s 9 8 5 5 3 5 2 6 2 7 2 1
SSSSS Ue 9 .
8 2 9 3 4 7 7 8 0 0 2 7 5 3 7 h s PPPPP Nt a 1 9 7 7 6 9 9 9 8 0 0 0 8 6 9 t MMMMM Mm 1 n UUUUU
( o I
Rx o i
5 3 4 R m PPPPP Gr 9 1 1 7 7 7 3 5 8 4 RRRRR L p 9 .
9 6 4 0 0 5 5 7 6 4 9 8 6 O 2
> EEEEE I p 1 9 9 7 0 0 6 9 9 9 9 9 9 7 T TTTTT Pa C s AAAAA A e 1.
6 9 1 4 3 9 2 3 3 3 4 0 9 7 5 F
Y t g
a WWWWW AAAAA e 9 2 9 9 5 8 4 5 2 1 4 4 7 0 T u o EEEEE l
b 1 9 9 9 7 9 9 9 9 5 9 9 9 9 I C SSSSS T
a A
=
GGGGG 7
9 5 1 4 4 1 5 4 4 7 5 8 4 P NNNNN I
9 2 2 0 A TTTTT I I I I 1 7 8 5 6 7 8 9 8 3 C AAAAA 1
9 4 0 2 9 9 9 9 9 9 6 6 7 E L LLLL 8 V UUUUU 9 r
e r r % T I
CCCCC t r
a y
y r b e e L A RRRRR I
I h
c r b b I I I h
r a
a t s meo me m A L U
CCCCC /
c u urh c U t
n r l
y e u te b g o v ce N M OOOOO s o n a be a p a ir n ly t NNNNN im J F M A M J uJ uA uS eO cN oD e N U M A C * ===== d a
r l11lll l
m g Y
Q o m r M o P
- C l
8 9
_ lu f R F o
- P A
(
C l t
s T u
- AI se
- P r R w
- Q M i e
v e
_. R
- B E
- F N
A S
J
. s M efe0 i
c P
i c
s A C
E a
ml D E m a R
_ D s
e mE u
uL M
C r
g o G
_ i a
s s
m N R P r
g O
V
__ #J R t
,A e
a r
g o
r A
P E
i t
n r
o P o in
- P T r
$ s m g t
s M o L n
A t
C cro a O s) r i r m 9 g o a r
epmpit e l ia o t r
g 9
oItea T
n IPI inI r
o o t) e ss e o P
r i N P e m o ngr M it E
t E
S t
t a i 9 9 n i
r P 9 i m
m o
g o
r p E P
(
it o
n 9
e v
it m
M G s s C )
S MD o o M9 r p
b u N U
Q s
u i
w l
lu F
E D AP 9
p a
S O
a n
(
C P
N E S E C o R L
U J
r Mo m D) o e
mNo t
s no o
)
e no s
T A
P
(
g i n w N ir h
t t D P
p E
B I
V N
P e m e m o e t
el er E
D r r t
t itl rgaIit 1 rg m
a io1 m p n E m o m Co co N o U
J M C e)o mo s m P r m o P r 8 e o 9 E E 9 R
N e m c 9 C -
9 9 B B A
t it a
r g
b u SE 9 b u C 9
f r
o 9 Y P m o S I 1
A m r S
( I ( H d E R T n M
o C o P 7
S E
E H
C I
H N
E m e S P
t l 9 I T
)
S l
u f R S B m O E I N o A
P E
D F
(
C s H S
F r
o E
B f r
o c e N A P N T (lAl s e t
lu I
F f
d d
n e
R P s P n R e m
- m
- A a w m m A o U r g e m c
- o i
v e
o c R e
P r
e R R B
g E i n
F r t
o i
n o
N M A 8 J
s 9 e,%
_ e
_ C
_ C E
T D A P
1 v E
- rNO f
t Z
Y
. A
.M
_ o
_ R s iio P
A 3 w e
mB we 4
s 0goplC
-- M R
A 4
3 w
e 7
wO YO i
O .
m
?
t t
r o
t e
r p
r o
E B
t o
- o p p t r
_ c e r e o b r p
_ B u t E f l
a e F S a r u r
- h d n l a
is n n u a n F o l a n N & t n n a
__ AJ o e i F
8 s m 9 e
it im oN s
o m C
E D
m,_ (
n c,
w e
b S
u i
t r p v o e e p r e t r.
V s r f m
O e a
- N i l a r o
- d u d c u t b t
s s n it u r
o n m S T
C n
o o s
_ p a b u h s
iio s
O it w e S i a s m r F l
u o
m s
sE3 Ol ml _P /
o oO osysF
_ p i w
all c
o 3 m C
_ p t
r n E _ E
- S P
E d.
r e
a C
E o
c t r
o p M M B D
t o
_ B
- u r
B e r y t r
cmc o o t b o
_ p f
G l
f t a p U r t r e e r
r A e o d r p
_ t l a
n p n t u i
e o r W 3 r t o p
nn mm_
6 0/
0 t f n a r L
U 7 7
- 0) a r
e e r i oson 7
J 0 ft d m e io a
r m l
a m e
id t t
n a
,e a
s d
(I t
r iml c b /
ol Fil 8sl us It v r
e o
N o u w 9 n s U p S i e t o b J
e v im i
o r e b ta r s
l a
r u
l u e s
e u o S p at Y o n C o w A n n E p r M o a B r e
- i t
a i
d e d n
v m n U r e u e s o R s p fl P b A o e r r r P t e
_ , G l t
e i n
N a W
_ R I w
_ A R F r e
_MO M d T
I D n N M U
- O o B M E
t
- F S E O' .
I P R 8 E 9
. AN S H e
u J I F s 3 s
_ E 7l m1
_ N I 2 lo 0'.
2
_ 1 O R
- E D A
M 2
- V E
- O O N
N_ Z
i1 ij l Ilfljl llf1 lill1lll A,
R, a
u u Y C v p
A E M B e
_ o s t v C d y -
a E -
R t
a B d.
o P d iv A t n
L a
t a i ivI al us R
B d t iu 4
0 A it n M a s m l a a n b P p u e MNi u
s n
a s
r it m ys B u b E p t u a s 3 F o s 9 u n t 2 ht S 0
, o_
i u r P s t
r N
A c
e s Co PNo u _ o p S e
J ue a o,
l r M w a A
a r
l rl l l u
n C
E uw s ,
n a R D 8 9 G p
e r O v
o N
p o R )
C E P 75 B
T 0 L 5 3
A C 1 ,
0
_ O s e
W 1 g ,O 0 _
m 3u 0!
o T 0 o,q1 iC.
I l l l i
s n
.o .,
4 0p e
r E
B nO N A P
E M#
eg o p m E o S
aN r
e m
t r
ca su e
m mf )t C
E B
o mtr o
p e er r
t orp t
t r
o p
)
dr 3
(
g MT NM I
G o a t e n o
ORE p f l r
- U r t a a i e in a l
A d t r lp
- l r
a
(
t r
r o
p d
n F u nm a R P o n a s u e o n p e r t t I
S L
U J
n a- r t f
a n
e im esc V E i
m la u d r
itI m m e s
8lr ta n N DP -
- I l l l l e nl n o E N -
N 8 s 7 1
am
- b /
c 9 it e (
w itva 8 i
- U J
9 WO C m u v m p
e S
)
e r
d n s r
o b
u i t 9
_ p 2
(
p e
c E
S l
a u 9 .
o 1 Y
A _ C E
g in P r
B Q -
M S
B lp -
m a P -
s N
R s
_ P t P
A c e
I ! l l s n
- a R_ s r
- t t A s O
- U G ,
2 1 t) e N e v R
I R m s it a
I S 1
_ O R s n s, G N (
g it 8 T M e a N E la
_ in u I
E c c F N o I
R o l Q
O t
. O t p M T m T O. i O.. a P N P s N N 8' -O 8 l
s A E 9 M 9 t J
M s m C c
e E e mo u e I a
u s
. G sl H sl n N si s 1T sl e a r
C P I
i N n t E
DM wo E a0 s e v
I B i
- - t a
- 3 E
4 E
it l
a 3
E G
V O N N u A O O Q P N_ Z Z l I l ,l l
Y A
M
_ A o, 1
_ P s
_ A oq 14o
/
9 s
o a o o 9
9 l s o m. ;mlE c1 1
P 4 7
3O 1
r mgop e B r
1 p o A
M 1
C o
.o t r t r
t t
m o o r o p p op B
E F
/-
7 1
2 Ot r
b S
c u
e r
ft a
r l
e r
a n
l e
r a
o ic d i u p F n e h n n N
l r
a t
n e t o
n 8 a S A
J u
n n
B
/mo e 9 t M alClE m imi b A t
o O
E f
a r
B o
c S u
R D d p
e t
t r
o
/w v
r.
G P
r p e m o
O V
O ft r
b c R N
d a
r S
u P it ic L T m h A
t C b n O u e S / B o
T N
P E
S C
E B N o E C
E B
M G M y N U
A t r
b t
n O i
c w
aO op e m R o er I L
U
= C t m
o V m& c E f c
8 J
m B
o d a
r
/
w v
N 9
9 1
I R s e i s
l s
l a
u R E N s s t U
M o ly e nn 8 J y a e.
b n t a- 9 g a s i i
n t a t a a
me 9 Y lp a d s 1 A d 8 m
M a
s a g b it m
9 p S os sO o
n u is S P e
r P A t o C s N naMogI k
P n E a e A
lam+
P lo B c.
to loIl rp c
o P n_ s s R e A x _
e . isy t M lp A G e M
la e y m N n h b a I
R a s p S O
I a s -
R _
t is f T a B I M ly p E
F N o d a m O t
_. & n a
a s
M . _
g T O. n a n N P _ is t
a t o
N E 8 s d k A 9 . e n J M ene . c it a
N l
uo s
. or m l P
k sl3l pl b u
I R s T
si e S C
E D
N ao n
s lp E m 7 - a 9
9 S S 1
E V N O
N O Z
Y
._ AM R 0 1 ,
_ P /
i A s e 4B wO 0
om!
n e 43f 0 p lC o
t
'C 7,m e
r E n R i A s p B a M 4 ! o o t o m n m C r o t aum E p t l
a s r n
u B e o )
B m m m y l r p e
a c s r
E F 3u b a r /
s t o
em t n in l a
t e a mE e f
u n l u
_ mo I n g s )
m R r N
t o
C o f t
a r
m o
M n a ir e
r e
r y
S A 8 M
mN olt J
m E d
( c 9 a b
. B /
B A
o t d r w it
, . s oI oI v m e r
, p R 7 b
R i
2 i
s C , s t e u
/
2 u E
, y e r 1 q a e S e G
l
_D t n h a u
r a s n f
_ O I
(
a a n .
t t a a tp a
R V
_ O d d p e r
. N it e
,o g
n b m I r
p t s
e P
_ m u R n
_ T s o is s S M r e
L
_O C 0 1
f
,O s
r om,C- ,
3 e
E B
e c
ir r A 10 3
gs o o
o i
oI o
r B a
T 0;l 0p o
Cdo E s l
i s It s s
)l p ly e 7
7 N P 1 r 9 E
S t p B o
ly a
t e
e k
e e
lp m
0 E
r o
p t
t r
o n
a a
h is s i w-S a
M
_ GAU r
e p e
t a s B( N O
I l
a d ly r
a R
_ n l
& n M i
f a a RI u
n mr n
g n a b y
R I
M a o is t a g U
L i
- I R
s e d i n V J
m e
s M
c o
r it m
lp m
N 8
9 lbg ylpIbl a e S u s E
_ N l
p n
_ U J
it i n
m t o 8
_ m lp a k 9 b S n u m a S a s
l P
9 Y 1 A n )
o .
M t k
n T
N S P
I a A Pl M R
P S N A T E
'N P
_ R E h
. A V c M I e 1m D
(
T 3 0
r E e G n
_ B E R I n
F A o H t C
S O.
I D P
_ N
_ A 8
__ J L 9 3 A e 9 1
_ M u0
__ 'R s E a 5l
. C H E
O T 2O 1
1 6
_ E N
V
_ O O Z
N_
.ll ll
Y
_ A
_ M z m mO t
u 's o uE 7'! nPm sC
_ UF
_ 7u R l m ,
str utr N A_ o /
iI w*'
L u
kolp p oIP e
E A ' e r D
_ A M 3n O
Lu rE t
l r
a u
n l
a u
n M
A P
p n n E e a a o
_ B r p l f
o t
E a t r
F t r i n l a o
_ o p
f g
v o
r p
e
__ e r
n p l r
N it p a A
i a n a u
_ J u i Pr/ t n n n n n el a
_ a m 8 C S m 9
_ E E o it D D c m P /
b
_ N w u v
_ r S
_ V o
_ O c
_ N i E
_ i m m B m
,l
,u uO T
C iE it 7 n! r C
R
_. O m it o
p N /
r e P
_ i
}
o pI r iE e
l a D 7N ml u M
_ P r E
a n /
S a u n AP e n a c
n i a
mo E
e p i - t G e s t
- _ U r p m e f r
o A
t s o p r l a e o l a v r
_ p in o l
a e
_ L r f r p u U g p n J la n n u a a n i tn/
t -
nl a-iI P
r e me nIi N i m s U m m J e 8 s o 9 it
/c it w m v
m b
_ Y r r
. A e o u p S
. M C
S E E B D
P R N P
A S
,iM A
R R A G
. M O
_. R P
G
. B E N I
F
_ R O
T I
N N
A O J M T
R
.1O C P E E
_. D R 7
V E
_ O N O
N_ Z
1 ANNUAL REPORT ON ASSESSMENT AND MITIGATION OF IMPACT OF THE PILGRIM NUCLEAR POWER STATION ON FINFISH POPULATIONS IN WESTERN CAPE COD BAY i
l Project Report No. 64 (January to December 1997) J By l Robert Lawton, Brian Kelly, Vincent Malkoski, l
John Boardman, Paul Nitschke and Erin Casey l
l April 1998 Massachusetts Department of Fisheries, Wildlife, and Environmental Law Enforcement Division of Marine Fisheries 100 Cambridge Street Boston, Massachusetts 02202
TABLE OF CONTENTS 59Giso P_ age I. EXECUTIVE
SUMMARY
I II. INTRODUCTION 6 III. METHODS AND MATERIALS 8 IV. RESULTS AND DISCUSSION 17 A. Physical Factors 17
- 1. Power Output-Thermal Capacity 17
- 2. Pump Operations 17 B. Finfish Species ofImportance 19
- 1. Cunner 19
- a. Background 19
- b. Eggs and Larvae 20
- c. Juveniles 23
- d. Adults 32
- 2. Rainbow Smelt 37
- a. Background 37
- b. Eggs and Lanae 38
- c. Juveniles 39
- d. Adults 39
- 3. Winter Flounder 40
- a. Background 40
- b. Eggs and Larvae 42 ii i :
I secta Pm f
- c. Juveniles 44
- d. Adults 44
- 4. Other Species 52
- 5. Impact Perspective 55 V. CONCLUSIONS 60 VI. ACKNOWLEDGEMENTS 64 VII. LITERATURE CITED 65 iii
LIST OF TABLES Table Engs
- 1. Important mdicator species ofTthe Pilgrim Nuclear Power Station. '7
- 2. Age-length key for cunner (sexes combined) in the Pilgrim 35 Station area 1997.
- 3. Multi-year (1994,1996, and 1997) age-length key for cunner (sexes 36 l combined) in the Pilgrim Station area. J
- 4. Tag returns by area for wmter flounder (at large at least one year) marked 48 during their spawmng season in Areas 1-3 from 1994-1997.
- 5. Esdmated abundance in numbers of winter flounder (bottom area 49 calculated at MLW), with 95% confidence limits, of winter flounder 2 280 mm (TL) and for pooled lengths estimated by otter trawl density extrapolabons (adjusted for gear efficiency) in the Pilgrim study area, spring 1997.
- 6. Mark-recapture model estimates, with 95% confuience limits, of absolute 51 abundance of winter flounder 2 250 mm TL in the Pilgrim study area, spring 1997.
- 7. Recreational Mxfish catches reported by creel survey over three decades 55 at the Pilgrito % don Shorefront in relation to plant operation .
- 8. A summary of mechanical unpacts of Pilgrim Nuclear Power 56 Station on selected fmfish species and mitigation undertaken in the ofTsite waters of western Cape Cod Bay.
I iv
LIST OF FIGURES Figure Eags I
- 1. PNPS investigative area for rambow smelt, wmter flounder, and 8 cunner, January-December,1997.
- 2. Station locations forjuvenile cunner recruitment study off Pilgrim 10 Station 1997.
- 3. The t-bar survey tool used by divers to estimate abundance ofjuvenile 12 cunnerin the Pilgrim Station area.
- 4. A collectmg unit of the type used to collect and incubate smelt eggs 14 -
(smelt shown above) in the Jones River
- 5. Winter flounder with Petersen disc tag attached (tag not to scale). 15
- 6. Annual means and 25-year cumulative Mean Capacity Factor 17 (MDC Net %) for Pilgrim Nuclear Power Station,1973 through 1997.
- 7. Operational lustory of the two circulating seawater pumps at 18 Pilgrim Station by month for the years,1983 through 1997.
- 8. Expanded number of cunner eggs entrained at Pilgrim 21 Station,1987-1997.
- 9. Expanded number of cunner larvae entrained at Pilgrim 21 Station, 1987-1997.
- 10. Mean cunner recruit densities (the average of 10 transects) per 26 site per day for each year ofrecruitment studies in the Pilgrim Station area,1995 through 1997.
- 11. Mean cunner recruit densities (the average of 10 transects) per 27 day by station for each year of recruitment studies in the Pilgrim Station area,1995 through 1997.
- 12. Smelt egg densities within Zones A & B of the Jones River 38 habitat cdew.st area,1997.
- 13. Division of Marine Fishenes winter flounder tagging area. 46
- 14. Recapture zones of wmter flounder (Pleuronectes americanus) 47 tagged in areas 1 3 by the MA Division of Marine Fisheries in the decade of the 1990's.
v
I. EXECUTIVE
SUMMARY
The follomng are the 1997 highlights of study fmdings for selected species. Additional information can be found in the Conclusions'sectiors of this report.
Cunner O Cunner (Tautogolabrus adspersus) larval abundance in the Pilgrim Nuclear Powet Station (PNPS) area l w.cid dramatically in 1997, with larval entrainment up 486% from 1996.
l l O Impingement of cunner at PNPS was relatively light (39 fish) in 1997. However, the entrainment of cunner eggs and larvae was equivalent to the loss of 498,000 adults from the local population.
O 'Ibe recapture of cunner tagged in PNPS over the past few years confirms that this species has limited seasonal movements and demonstrates a high degree of reef fidelity.
O. A total of I84 cunner (selected specimens) from the PNPS area was aged; fish ranged up to age 8.
l O For the three remnt survey years (1995-1997), there were large differences in annual recruit abundance ,
! dunng the settlement penod in the study area, with 1997>l995>l996. Annual larval entrainment values l l
st PNPS matched up well with recniitment levels observed for those years.
O Over the three years of the recruit study, linear density-dependent mortality was evident in the post-settlement period.
l O A consistent hierarchy occurred annually in recruit densities spatially during settlement, with densities at Discharge > Rocky Point > White Horse Beacit O in the vicuuty of PNPS, a possible alteration in hydrodynames, such as an eddy created by the thermal discharge cummt or other physmal factors associated with the power station may induce higher numbers l of cunner larvae to concentrate and settle at the Discharge site. This was converse to our initial
. ~F na ofloweri recruitment at this site due to entramment effects from nearby PNPS.
OS Any marked effect of PNPS cetramment of pre-settlement cunner larvae on recruitment success is more likely to occur during a recruit season where larval supply and subsequent settlement are relatively low overall, which likely occuned in 1996. In relatively high recruit years (1995 and 1997), post-settlement i
density Aw=d=r ffectse on recruits become major dsni. ants ofcunner recruit success. In retrospect, l muumal power plant impact was indicated from the recruitment studes of 1995 and 1997, while in 1996, the impact of PNPS on recruitment success was inconclusive due to meteorological events.
l l
I 4
1 1 l l
l i
09 The cunner recruitment study was concluded in the fall of 1997. After three years of study, it appears that PNPS is having a limited impact on local cunner recruitment, and, thus, the local population. We felt that fundmg would be better expended on enhancing our flounder tagging project.
Rambow Smelt 9 Rambow smelt (Osmerus mordar) impingements of difTerent magnitudes have occurred at PNPS over the years of station operation. The smelt is considered an "important representative species" due to I abundance and iwael importance in the P!ymouth, Kingston, Duxbury Bay (PKDB) area. Boston Nm Co. (BECo) funded DMF to take remedal measures to compensate for impingement occurrences.
l Our overall goal has been to increase the number of adult smelt in the local population to offset power plant mortality. EfTorts included augmenting natural egg production and enhancing spasming habitat to optunize egg hatchout.
9 During the springs of 1994 and 1995, smelt eggs were taken from two genetically isolated wild anadromous populations: one from the Weweantic River, Wareham and the other from Back River, Weymouth and transplanted into the Jones River, Kingston. Eggs were collected using our portable sphagnum mass-filled trays, which provide ideal habitat for egg development. We stocked a total of ca.
1.8 million smelt eggs into the Jones Rivcr. Larvae were expected to imprint on the waters of PKDB and return to the Jones River and its other tributaries to spawn when sexually mature.
9 To address spawning habitat enhancement, we again used our moss-filled egg collecting trays. Each spawning season from 1994 to 1997, a number of trays were placed in the Jones River on the smelt spawning grounds where spawning activity was greatest in past years. The trays were set in the nyer before smelt spawning v+=uei and were removed after smelt egg hatching was completed.
I 2
_ _ _ _ _ _ . _ ._ i
Unwastd fouling material was regularly removed from the trays to improve water circulation over the eggs.
09 'Ihe enchng portion ofour smelt project was discontinued after 1995 due to reduced egg product. ion in our two " source" streams, with no edur supply of eggs to be found. The habitat enhancement efTort was officially ternunated at the end of 1997. We felt that funding for this project would be better spent on unproving water quahty in the Jones River and other tributaries to PKDB. However, DMF will continue to place a limited number of trays into the river during the spawning period to continue husbandry practices. We also will monitor spawning activity and egg production densities, while conducting periodic checks to make sure the river is free of obstacles, such as fallen trees, that could hinder fish passage.
Wmter Flounder 9 PKDB and the surroundmg coastal waters are important spawning areas for winter flounder (Pleunnectes americanus). In the PNPS study area, winter flounder exhibit fairly high fidelity to natal spawning grounds They also undertake local seasonal movements which appear to be temperature-driven.
O in 1997, an estunated 55.4 million winter flounder larvae were entrained at PNPS, which equates to an equivalent loss of 47,087 adults from the local population. This is by far the highest flounder larval entrainment of the past 11 years.
O An estimated 770 winter flounder - age 0 and age 1 - were impinged at PNPS in 1997. l l
9 No winter flounder reportedly were caught by anglers at the PNPS Shorefront in 1997.
9 In 1997, we tagged 7,487 winter flounder with Petersen disc tags, bringing the study total to 14,776. 1 As of the end of 1997,830 of the total flounder tagged to date had been recaptured (recapture rate of I
3
5.6%). Sixty four pcreent of the 128 tag returns (with known capture locations) at large at least one year by the end of 1997 and captured during the MarchJune spawning season were taken from the general area where they were tagged.
O Density extrapolation using the Area Swept Method provided an estimate of the adult winter flounder
(
population size for the study area portrayed in Figure 1 (fish 2 280 mm TL, i.e., age 3 and older fish).
This estimate was 321,832 adults.
O P=M= estimates of winter flounder (sub-adults and adults) for the study area (fish 2 250 mm TL) obtained from mark-recapture models (excluding the Petersen estimate - 500,000+ fish) were in the 100,00+ range.
09 Due to the low recapture rate of tagged flounder (5.6% to date) which affects the predictive value of mark / recapture population models and estimators, and the variability in our area-swept estimates, we, as yet, cannot define with confidence the magnitude of entrainment impact on winter flounder oft PNPS.
The equivalent adult loss because of entrainment mortality in 1997 is preliminarily estimated to represent bctwwi-9% and 40% of the estimated adults in the local population. The continuation of this study in 1998 will not necessanly yield a more precise or definitive estimate of population size, although it will provide a larger data base.
Other Soecies O Striped bass (Morone saxatills), bluefish (Pomatomus saltatrix) and Atlantic mackerel (Scomber scombrus) were the species reported in the recreational catch at the PNPS Shorefront in 1997.
O Striped bass and tautog (Tautoga onitis) dominated the SCUBA finfish sightings off PNPS, with small aggregations of cunner also obsen ed.
09 Data from the sportfish and underwater visual surveys reveal that some finfish species are attracted to 4
I
the elevated water tap (spring and fall) and/or cwrent of the thermal discharge at PNPS. This places them at risk ofimpact from temperature aberrations, chemical releases, and potential gas bubble disease pulAr . As such, some form ofdirect visual monitoring in the discharge area should continue for as long as the power plant is operational.
l l
l 5
II. INTRODUCTION i The Massachusetts Division of Marine Fishenes (MDMF) power plant team has conducted field mveshgabons to assess anL.. .;al effects of the operaban of Pilgrim Nuclear Power Station (PNPS). In some estances, mitigative or rcmedial measures have been instituted to ofTset negative impact. This work was funded by Boston Edison Company (BECo) under Purchase Order No. LSP007549.
In 1997, we focused on cunna (Tautogolabna adspersus), wmter flounder (Pleuronectes americanus) and rainbow smelt (Osmerus mordar), employing a suite of gear types, equipment and techniques to conduct sampimg and, when appropriate, to undertake restorative cfTorts. Measurements, enumerations, percentages and abundance esumates were used in data quantification. Descriptive statistics are, at times, summarized in tables or displayed in figures. Statistical procedures also were used.
From extensive field studies off PNPS, it is evident that mechanical aspects of station operations, i.e.,
entrammmt of fish cess and larvae, and to a lesser extent, impingement ofjuvenile and adult fish, pose greater envuonmental threats than does the release of waste heat into receiving waters.
'lhe three finfish species of most concern are cunner, winter flounder and rainbow smelt (Table 1). In the environs of PNPS, cunner spawn, use the area as aursery grounds and reside asjuveniles and adults. This species is a temperate reef fish that is structure oriented, territorial and basically sedentary. Cunner have a 1=li=1 distribution, are abundant and appear to be good indicators to assess local environmental stress. At l
PNPS, cunner is primarily affected by early life stage entramment, being the dominant fish entrained there.
Winter flounda spawn in the PNPS area which also serves as nursery and feeding grounds. This flatfish is highly valued as a commercial and recreational species. Like cunner, winter flounder larvae are entrained in l relatively high numbers.
Rainbow smelt is valued as a recreational species in the nearby Plymouth, Kingston and Duxbury Bay (PKDB) estuary. Several incidents of relatively high smelt impingement have occurred at PNPS over the years of station operation.
6
Our objectives in 1997 were: (1) to exanune recruitment dynamics of cunner in the PNPS area and evaluate power station unpact, and age cunner to gmerate catch at-age data which will be used to estimate natural Table 1. Important indicator species off the Pilgrim Nuclear Power Station.*
Background
s w Se Poss&
Most Significant Source ofImpact SPcCE as an Inhta h rces yg;,io,y (Based on Results to Date)
Species ofimpact Cunner g d, r, e I, E, TC Entrainment - is number one in egg collectiori at PNPS (June-Aug)
Rainbow r, s I, TC Impingement targe incidents in December of *78, RIS Smelt '93,'94 Winter d, r, e, s I, E, TC dntrainment - large number oflarvae collected (April-RIS flounder May)
RIS representative importatit species selected in the onginal 316 (a and b) Demoneration Document and Supplement to assess Pilgnm Station impact (Stone and Webster 1975 and 1977).
d a dominant species in the Pilgrim area.
r alocalresident e -commercialimportance s recreationalimportance I impingement E entrainment TC - discharge current cfTects: thermal / current
- Note: Indumfor species sciechon rationale: these three species were selected because they have shown the most potential for impact otT Pilgrim Station and may be indicative of power plant induced stresses to other marine fish species.
mortality needed for the adult equivalent model calculations; (2) in the case of winter flounder, determine dtscreteness of the local population and estimate absolute abundance of the adult segment of the population; and (3) to enhance the quality of spawning habitat for rainbow smelt in the Jones River, a tributary to the nearby PKDB estuary where the local population apparently originates and thus increase the collection of eggs on ideal substrate to improve egg survival to hatching.
This annual report includes a description of sampling design and methodology, together with findings, j conclusions and any inuiwi.cridations. Progress achieved in assessment surveys and ongoing restorative projects was highlighted for the aforementioned indicator species in the PNPS area.
l In 1998, our efforts will focus almost exclusively on winter flounder as the indicator species of concern.
7 1
III. METHODS AND MATERIALS i The study area for 1997 is bounded in Figure 1.
3 . . .
.... b. :
gpg:. ....
ODyj$ ..-
ws sw$p[::{l
'!qjj$$$ f!: :., -
COD'
- 75.w.w.v -
6gyI'.
"5$w.vgmmh xx.: .
mu. xxx..
e:ec - gm
' gl .x.m .
m s. .
.;:grm . - '. -
v.piber:o.e:+.
- 00M. :
. '121TCONTOUlt::
,xxmov.vg 6: Mj!f.@d$$ % i f%:
ppneeee ;uxx .
gpoeea.g m x
. sfsfNO iSIh"2 a',w. . .:.:.3,hf,;.#h ,...
1
- +MH&', C ~.
44wtit:4r
~
xxxxxxn .mxxadv.v.A xxxxx n 4;gggArcc;vx y.v n-
.xxxxm N m $,. c -
32Mdi99%%%
MPE99Hs$a' m;ct,m:e a a . n.. .w.gsg.,.m ,
PII. GRIM > $y:$. ' '
NUCI. EAR ~ N Ny$$"
h'M.
POWER Coastline not to scale STATION' MANOMET PT.i$550 Figure 1. PNPS investigative eres for rainbow smelt, winter flounder,and cunner, January-December,1997.
Cunner Enns and Larvae. Entrainment of the cunner pelagic eggs and larvae is monitored at PNPS by Marine Resenth,Inc. (see Entrainment section in this report). The Adult Equivalent Model is used to equate entrained eggs and larvae to equivalent adults that would be lost from the local population because of power plant mortality.
8
Cunner fecundity by age, length, and weight was n:cently investigated for fish collected off PNPS by Paul Nitschke, a UMass, Amherst graduate student at the time, and working with our cooperation. Funding was provided by BECo.
Information on the hydrodynamics of western Cape Cod Bay, including the residual and variance flow patterns that constitute circulation and which afrect the resultant dispersion ofichthyofauna, was compiled by Eric Adams of the Massachusetts Institute of Technology, Cambridge. The geographical bounds for the origin of cunner eggs and larvae that are entrained at PNPS (primarily June through August) were identified by nwdcling. Cunner recruitment to the PNPS vicinity is likely influenced, for one, by the status of spawning cunner loca+al elsewhere; therefore, we had to define the areal limits of the local population. Our definition of the local population may in actuality be more of an abstraction than a natural unit. From our past capture-recapture <,
studies, it appears that adult cunner in our study area have a relatively small home range; thus, the only substantial recruitment to this temperate-reef population presumably results from settlement of their planktonic larvae, following metamorphosis.
Juveniles. To address the relationship ofrecruits to population size, settlement / post-settlement processes affecting recruitment were investigated for a third consecutive year. A SCUBA census of cunner recruit (juvenile) densities was undertaken as we continued to examine cunner recruitment with the goal of assessing entramment impact at PNPS. locations with similar habitats again were sampled in 1997. We know that habitat is important to recruitment success, which is also affected by other environmental cues, e.g., temperature and demographics. We believe that several years of comprehensive data will be insightful as to local cunner recruitment processes and possible powys plant intervention.
9
The dmssty ofnew seenuts was quantifwd at a si'ejust off PNPS and at two other locations away from the power plant (Figure 2). In that PNPS " crops" young cunner (larvae and new recruits), recruitment at similar habitats may be higher with increasing distance from the power station. As part of our sampling design, we collect information on post *1=wn2 recruit densities by observing age-0, i.e., young-of the-year (YOY) juvendes over time in comparable habitats but at spatially different sampling locations. If early settlement difTers spatially, we can test the cfTects of density ou recruitment at the end of the recruitment period. An assumption of the adult equivalent model is that there is a direct relationship between the pelagic egg and the benthic adult life stages. If this is true, that should exist a direct relationship between the number of recruits at the beginning ofrecruitment with those extant at the cnd of the recruitment penod Restated as a hypothesis - initial settlement is correlated with net recruitment at the end of the recruitment season.
l Three sites, which appeared to have similar habitats, including similar substrates and macroalgae, and which were used in our studies of 1995 and 1996, were resampled in 1997. One station was located about 20 m southeast of the discharge canal at PNPS. The other two locations were 0.5 km northwest of the Rocky Point power plant at Rocky Point and 1.5 km southeast I at White Horse Beach (Figure 2). The center point Pilgrirn m a
cWe station E , White Horse of each fixed station was masted using a moonng block with surface buoy. Painted rocks were
= amoeranseenershs I placed on th: bottom beside each mooring as an _
aca Manomet Pt.
aid to locate the site in case of buoy loss. At each sampling site,10 transect starting points were Figure 2. stationlocationsforjuveniteeunnerrecruitmenistudyoft Pilsrim Station,1997.
established so as to radiate out from the center point like spokes of a wheel. To muumize any bias of our diving on counts along other replicates, the beginning i
of each band transect was located 5 m out from the 10
l center block and was marked with a numbered, pamted rock. Transects were evenly spaced at 36' intervals using an underwater compass, starting with a heading of 000' magnetic At the inception ofour diving recnntment survey, and again at the termmation of the sampling program, the benduc habitat at each sampling location was exammed to characterize the habitat and its stability over the course of the study. Habitat was quantified by visually estunating percent composition of the dominant substrate, including algal types in each square and at each point intercept of a prescribed grid. The substrate categones were: filamentous algae, fleshy algae, crustose algae, sand, and cobble. Cobble was defined as small nxis (s 15 cm) without algal conr. Bouldess were covered with macro-algae and were not considered a separate category. Habitat categories then were grouped into two functional units, referred to as complex structure (filamentous and fleshy algae types) and non-structure (sand, cobble, and crustose algae). The proportion of complex structure to non-structure among reefs for each time period was compared using chi-square analysis.
In addition, algal height measurements were obtained at 1-m intrervals along the survey transect line..
Rugosity (an index of substrate structural complexity) was measured by laying a 10-m fine-linked brass chain on the substrate and allowing the chain to conform to the bottom contours A straight-line measurement of the distance (m) covered by the chain was made, and the process was repeated along the entire length of each transect. The ratio of the sum of measured chain distances required to sample the transect to linear distance (10 m per transect) is the estimate of rugosity for that area. Rugosity measurements were not done at the end of the sampimg penod in the fall because stonns had moved some transect markers which would introduce a bias into the measurements.
Cunner recruits were visually enumerated at each station, while traversing each of 10 band transects 2
(replicates), with each transect covering lx10m, using SCUBA. At each station,100m were censused on 2
each outing, and a mean density value was calculated (mean recruit density per 10m ). To delineate width and length of a transect, a 1-m wide t bar sampling tool (Figure 3), with attached compass and line-reel holding 10 m ofline, was pushed ahead of the swim path by one of the divers acting as navigator. A second diver (the census 11
taker) recorded the number ofcunner recruits while swunming above and slightly in front of the navigator. This position insured good visual coverage of the swath of the transect.
Systemabc sampling for noruits began on 15 July and 1 meemr continued through 31 October 1997. Sampling was usually i I conducted two to three times each week, for a total of 31 samphng days. Enumeraban was completed at all sites within ,
each sampling day to avoid temporal variation. Lateral w-u underwater visibility was measured using a Secchi disk after w n u > g g _ n. ,
w =ma. m, w.= sw. w all counts were completed.
Adults. We have studied movements, home range, Figure 3. The t-bar survey tool used by divers to estimate age, distribution, and abundance (relative and absolute) of abundance orjuvenile cunner in the Pdgrim Station area.
cunner in the vicinity of Pilgrim Station, with a past interest in their behavioral response to the thennal discharge current. For several > ears, we tagged adult cunner with a Floy T-bar anchor tag as part ofour investigabons. We did not tag in 1997 but did record captures of tagged fish from past years. To sample cunner, baited fish traps were fished ovemight to optimize captures, in that cunner forage most actively at dusk and dewn. All cunner were counted and measured for each trap haul, and sex was determined when possible, most often by external exammation. Most fish were released at the site of capture.
These data generated catch per unit efront (CPUE) information, i.e., catch per trap haul, a measure of relative abundance.
1 The age structure of the local cunner population from the Pilgrim area has been examined by aging otoliths taken from selected specimens over the last several years. Aging information was incorporated into the l recently conducted fecundity analysis. Retained cunner were weighed to the nearest gram, measured for total length to the nearest millimeter, and sexed via intemal exammauon of the gonads. Sagittal otoliths were removed, cleaned, labeled, and stored dry in glass vials. Unprocessed otoliths were viewed sulcus side down under a 12 I
.. . . . _ _ _ _ _ _ - - - - - - _ - - - - - - - - u
I l
&ssectag scope with mflected light, and annular rings were counted. This was a feasible approach for most fish through three or four years of age. Annulus formation in cusmer has been validated by researchers in Newfoundland (John Green, personal comnnaucation)', and we are confident that the clear hyaline bands we read are annuli.
Otoliths from older fish or younger fish difTmult to age were mounted in Pro-Texx medium on a macacope slide sukaas side up and allowed to dry at least two days. The mounted otolith then was ground down l
' with 320 grit sandpaper until the annular rings could be differentiated. Each otolith was aged independently by two readers. If concurrence occurred on the readmgs, an age was assigned to the fish. When disagreement occurred between re idings, the otolith was viewed again by both readers, and, if a consensus was not reached, the fish was excluded from the age analysis.
We have contanued to milar# CPUE and length data and to age selected cunner to prepare an age-length key, with which to generate a catch-st-age matrix, which in tum will be used to estimate natural mortality for
, the adult equivalent model.
l Rainbow Smelt Eggs and Larvae. The demersal, adhesive rainbow smelt eggs were collected and cultured on egg mIMag units. Each unit (35.6 x 45.7 cm) was a weighted wooden frame, enclosed with chicken wire, and filled with unptr==I sphagnum moss as substrate for egg deposition (Figure 4). Egg trays were placed into selected rime areas of the upper Jones River smelt spawning ground and inspected every few days, seniced, and momtored for egg deposition, development, and survival Fouling macro-algae were removed from the trays and discarded damii of the spawning area. We endeavored to minimize egg disturbance and destruction on the river bed and on our trays during this process. Following egg hatchout, lan'ae are expected
' John Green, Biology Department, Memorial University, St. Johns, Newfoundland l
13 ;
I l
1 to be carried downstream and out of the Jones River into the waters of Plymouth, Kingston, Duxbury Bay as they develop. When adults they should home back to this estuary, ascending the Jones River and possibly other tributaries to spawn.
Juveniles. Bere have been three unusually large rainbow smelt impingement incidents that occurred at g j Pilgrim Station, in December of 1978, '93, and '94. The majority of smelt impinged were age-0 fish (juveniles). 9 Impingement sampling data are collected by Marine asstis1iisisi\g
". \\\\\\\\\\\\s.
Research, Inc. (see Impingement section, this report). " \\\\\\\\\\\\".
Adults. Adult rainbow smelt (Figure 4) also are sm.n r g a Tr.y impinged at Pilgrim Station (see Impingement section). Figure 4. A collecting unit of the type used to collect and incubate smelt eggs (smelt shown above)in the Jones River.
Winter Flounder Ens and larvae. Data on these two life stages (primarily larvae) are collected by Marine Research, Inc.
in their entrainment sampling program at Pilgrim Station (see Entrainment section, this report).
Juveniles. Juvenile winter Dounder are impinged at PNPS, with monitoring data also collected by Marine Research, Inc. (see Impingement section, this report).
Adults. Our objectives have been to determine the discreteness (fidelity) of the local winter flounder population and to estimate absolute abundance. This information is being used to assess impact of flounder entrainment and impingement at PNPS.
During the winter flounder spawning season north of Cape Cod (ca mid-March to mid-June), some winter flounder may move in and/or out of PKDB (Figure 1), with evidence of spawning both inside and outside this l
estuary Flounder may aggregate in pre-spawning staging areas out in deeper water, with some moving into the estuary at night on a flood tide to spawn in the shallows.
IV
We again contracted a commercial fishing vessel, the F/V Frances Elizabeth, to sample winter flounder, both for tagging purposes and to estunate density. The study area was extended this i
3 ear to include the watcrs from Humarock, Marshfield southeastward ; -g to the Mary Ann buoy, Manomet, from is-d.cie (9.2 m MLW) out to the 36.6 m (MLW) depth contour (Figure 1). 'Ihe trawl gear was different from that used in 1996, in that the net used last year was ~4: M severely damaged during commercial fishing operations and had to Figure 5. Winner nounder with Petersen disc tag be replaced. Thei+/+-- = was a larger Yankee otter trawl (21.9- attached Ons not so suele).
m sweep and 15.8-m headrope, wluch had the same 15.2-cm stretch mesh and a 7.6-cm mesh liner); it was fished with 12.8-m legs and 78.6-m ground cables. The trawl doors (#5 l
Bison doors) were of steel,(1.5 m x 0.9 m and 181 kg each). Warp length varied with depth of water fished, rangmg from 73.8 to 92.3 m. Despite changes in some of the gear dimensions, measurements made while fishing mdicated that the door and net spread were approximately the same as in 1996.
Winter flounder were enumerated, measured (total length [TL]), and assessed for sex, maturity, and reproductive state before being released near capture sites. In addition, flounder 2 250 mm TL were marked with blue Peterwri disc tags (Figure 5). Data were also collected on net geometry and the trawl distance of each tow.
Tow duration and legah averaged 30 minutes and 1.2 km, respectively. We generated independent estimates of population size via mark and recapture and by an area-swept approach (density extrapolation).
We p. creed one estunate of wmicr ikmnder population size (instantaneous abundance) from the 1997 contracted u.... -dal vessel trawl catch data using an area-swept approach, based on density extrapolation. As trawl gear efficiency in our sampling is unknown, we estimated it to be 50%. To estimate density, the number ofwmicr flounder by low (data transformed - In(x+1)) was divided by the area of bottom covered. Tow length was it ....d, and tow width was estimated from the trawl doors' spread on the bottom Door spread is used i
15
as a measure of the width because of the " herding" action caused by the sedunent cloud generated by the doors and legs while towmg Catch per area was calculated for individual tows. Estimates computed for adult winter flounder 2 280 mm TL and for all sizes pooled of winter flounder captured were doubled to reflect the assumed 2
50% catch efrumency Density cshmates (number per m ) wcre muitaplied by the total bottom acreage in the study area to obtain cutunates of population size. Bottom area was averaged using a dot grid and navigational charts Acreage was converted to square meters We also estimated numbers in that segment of the winter flounder population of fish >25cm TL, including subadults and adults, using several mark-recapture methods Closed population models-single episode of tagging and one of recapturing = Petersen method, multiple markmgs - Schnabel method, Schumacher and Eschmeyer method, Mark method, and Capture method, Open population model-multiple census - Jolly-Seber method For w- i-E-- eshmates of absolute population abundance in 1997, except for the Petersen method, we used our blue tag (1997) recovery data for winter flounder at large 22 days (determined by plotting time-at large versus distance traveled) and recaptured during our tagging penod (late March to late April 1997 and for the first additional sampling day after tagging was completed) for a total of 16 sampling outings. For the Petersen method, we grouped the data into two penods for analy sis - the first 15 sampling days (i.e., the tagging period) versus the last 15 sampling events.
Other Fish Species )
Eees and Larvae. Egg and larval information for other finfish species entrained at Pilgrim Station were obtained by Marine Research, Inc. (see Entrainment section, this report).
Juveniles. We also collected data onjuveniles of several finfish species via SCUBA diving and fish potting. Impingement data were obtained from Marine Research, Inc. and BECo.
Adults (Same as forjuveniles) 16
l IV. RESULTS AND DISCUSSION l A. PHYSICAL FACI' ORS
- 1. Power Outout-Thermal Canacity Pilgrim Nuclear Power Station's capacity factor (MDC net percent) is an index of operational status that wvam etes thermalloadmg into the receiving waters of the manne environment. This factor is relevant when l
assessing long-term thermal impact on marine organisms. By permit regulation, Pilgrim Station is 611 owed a mammum discharge temperature of 38.9'C and an effluent A T 18'C above ambient. For the 25-year history 1 \
of plant operations, the long-term mean MDC at Pilgrim Station is 54.4%, with annual averages ranging from 0.0% (outage years) to 90.5% in 1996 (Figure 6). Annual power level had been on the rise for the last three years l
- 65.2% in 1994 to 90.5% in 1996 -
! however,the annual mean for 1997 was tun emeny acron m we 73.4%. This decrease was due to an "
=
i l
outage from mid-February through mid- "
= l 5 m_
l l -um I
April. Average - monthly thermal "
l
= -
capacity values in 1997 rangcxi from 0% " !
l l f ll in March to 98.7% in October. l l
- 2. Pumo Operations """"""""""""""""""""",""""
i
"""""a GMNUAL MEANS
- Once-through, open-cycle cooling at Pilgrim Station induces a Figure 6. Annual means and 25-year cumulative Mean Capacity Factor (MDC Net %) for Pilgrim Nuclear Power Station,1973 through 1997. ;
localized water current flow.
l Two circulating seawater pumps [586.7 kl/ min each] (155,000 gals / min) withdraw water from the intake embapnent. The cooling water circulates through the plant condenser tubes before being discharged back into the waters of Cape Cod Bay laden with waste heat. At ebb tide, efIluent velocities can exceed 2.1 m/sec (7 fVsec) c3 the egress of the discharge canal. This results in scounng of the benthos and concomitant erosion of substrate 17
along the bottom path of the discharge plume.
Throughout the operational history of this power plant, there have been station outages, when one or both circulating seawater pumps wtre not operated (Figure 7). Such periods have occurred periodically and generally are short-lived; however, prolonged outages occurred in 1984 and from 1986-1988. During 1997, both 4
circulating pumps were continuously operated except for the outage period (mid-February throupi mid-April),
when only one pump was in operation.
-4
- c :p a - ', ,E ~,'x;
, : I ': . i T h a T !, ':
9 l I I !
I l
i l ; .. !
n .., :. ,i. . m. e. .i. .s .a : .a
' 9t 'l 1944 ia- ' 199b 1e 19NG i nhos 1990 1 i. 199/ . * .+ 1994 1995 1990 19a+ /
Figure 7. Operational history of the two circulating seawater pumps at Pilgrim Station by month for the years,1983 through 1997.
l 18
. . l
B. FINFISH SPECIES OF IMPORTANCE
- 1. Cunner Ra.ckcround Cunner (Tautogolabrus adspersus) are common and abundant finfish members of temperate reef communities of the western North Atlantic, occurring from the intertidal zone out to depths of over 90m and ranging from Conception Bay, Newfoundland south to NW: Bay. New recruits (juveniles) generally reside in shallow waters and, thus, are especially vulnerable to power plant impact. Relatively site attached and structure oriented, cunner are associated with natural and artificial reefs, includinE bedrock outcroppings, sheltered rock substrate, glacial boulders, pilings, piers, shipwrecks, jetties, and breakwaters.
Cunner are fairly sedentary. Age-Ojuveniles occupy only a few square meters of area. Even as adults, 2
home ranges are relatively small (less than 3,000 m ), with most activity limited to core areas averaging less than 100 m2 (Pottle and Green 1979; Bradbury 1993). An infaunal, labrid, reef species, its primary defense behavior is to seek refuge within a shelter's infrastructure They do not undertake extensive migrations (Green and Farwell 1971; Olla et al.1975) and may overwinta in their summer habitat in a quiescent state (Dew 1976). When water temperatures fall to 7 - 8 C, cunner become inactive, and then as temperatures decline further they become dormant and remain so until spring when waters warm above 6 C (Olla et al.1975)
Often found in highly localized aggregates, members of a cunner population can be spatially separated into subunits, some of which occupy patch reefs. As such, they may be vulnerable to local perturbations, e.g.,
point-source pollution and intensive sportfishing with resultant mortality. Sensitive to stress at night, especially if habitat quality is poor, they enter a sleep-phase, which is characteristic of the labrid family of fishes. This behavioral trait reduces their responsiveness to environmental stimuli after dark. Cunner appear to be a good indicator species to assess stress in inshore areas because of their distribution and territorial life history characteristics (Olla et al.1975; Dew 1976).
19
At PNPS, the intake breakwaters and discharge jetties augment natural structure, providing high-relief, structurally complex habitat. The cunner needs refuge areas from the moment an individual takes to the bottorn aAer metamorphosmg froru larvae tojuvaile, for shelter is impcrative to its sleep phase, md, in general, provides protection from predators and oAen offers foraging opportunities. Eddies are created at the locations oflarge, )
I glacial boulders in the discharge area, which enable cunner to reside in the vicinity of tne PNPS cfiluent, where t
current velocit ies at low tide would normally limit their maneuverability or even preclude their presence
'Ihe cunner at all life stages is a numencally doininred fish in the environs of PNPS. Densities dcunner, j I
bothjuvcniles and adults, vary spatially in the PNPS area, with highest relative abundance found, to date, in the discharge area off the seaward side of the outer intake breakwater. From our extensive tagging of cunner for a !
number of years, we know that even adults have high site fidelity within a year and between years.
I 1
Enes and Larvae Cunner eggs and larvae are.. a at PNPS during spring and summer They are most likely I determinate spawners,in that their eggs are shed during a relatively short time period Having captured large -
numbers ofripe male and female cunner off PNPS in late May and June over several years, we believe that some spawning does occur in the imm~iiate vicinity of the power station. In Cape Cod Bay, cunner spawning is most miense in July and August. The pelagic eggs hatch in 2 to 3 days, while the pelagic, pre-settlement larval stage ranges from 18 to 37 days. In that time, the larvae can be dispersed quite a distance, although their ability to vertically move in the water column can offset some of the horizontal movement.
In 1997, cunner eggs and larvae were entrained at PNPS from late April to September and from June to September, respectively. The Labridae-Pleuronectidae grouping has dominated fish egg entrainment at PNPS over the years of plant operation, oAen comprising over 90% of the fish eggs and larvae collected. Substantially more cunncr eggs than larvae are entrained each year which likely is related to high egg mortality and to a lesser extent the ability of the late-stage larvae to move in the water column and avoid entrainment. The entrained eggs 20
l l
I markedly contribute to adult equivalent values. Cunner reproduction and settlement waxes primarily during a pulse lasting about three to four weeks in the Gulf of Maine.
Over the last four years (1994-1997), an estimated 1.6 to 4 billion cunner eggs and 10 to 99.6 - , , , , , , , , , , . , , , ,
million cunner larvae have been entrained at PNPS, while ,
l
! adult equivalent estimates hrve ranged from 177,813 to ,,,_
540,491 individuals. Large numbers ofcunner eggs and .
larvae are entrained annually at PNPS (Figures 8 and 9).
For analysis ofimpact, it is assumed they all die and are = = = = =======
lost to the local population. In 1997, cunner entrainment Figure 8. Expanded number of cunner eggs entrained at was estunated at 1.7 billion eggs and 99.6 million larvae, Pdgrim Stetim,1987-1997.
which equates to 498,281 future adults potentially lost to the local population via entrainment as estimated by the -
.88 -
Adult Equivalency Model, which utilizes an average -
cunner lifetime fecundity of 21,656 eggs per female (Mike Scherer, personal communication)' In 1996, an estimated 2.8 billion cunner eggs and 17 million cunner
, .. ET-5! $>
larvae were entrained, equating to 318,377 adults. * " " " """""**
Although the number of eggs entrained in 1997 was 39% _
Figure 9. Expanded number of cunner larvac entrained at lower than in 1996, larval entrainment dramatically Pdgrim Station, 1987 1997.
increased 486%. The result ng i equivalent adult estimate increased by 56% in 1997. The underwater visual __.
' Michael Scherer, President, Marine Research, Inc., Falmouth, MA 21
cartsus ofrecruit densities for age-O cunner indicated that recently-settled recruits were noticeably more abundant in the PNPS vicinity in 1997 than in the previous two study years. This agrees well with larval entrainment at the plant for these three years (Figure 9), indicating the pool of pre-settlement larvae in the environs on PNPS was considerably higher in 1997. I Lack of a food supply and predation are implicated as potentially major sources of mortality to marine fish larvae. Even small changes in the duration of the early life stages can lead to substantial increases in modality. Fast growth and rapid planktonic life stage transitions are important, as subtle declines in growth rates, e g., via cooler temperatures, can result in predation mortahy increasing by an order of magnitude (Richards and Lindeman 1987), by prolonging the time individuals spend in the high-mortality larval stage. We kn1w that ambient water temperatures in winter, spring, and summer were noticeably cool in 1996 off PNPS, with canner I
larval supply and recruit densities relatively low.
While the magnitude ofcunner entrainment loss is substantial at PNPS, its biological significance to the local cunner population is speculative. With a larval stage duration of I8-37 days and since the lan ae are pelagic and the adults and juveniles do not move very far, the colonization of reef patches occurs by the settlement of metannphosed larvae from the plankton to the demersal habitat as ajuvenile. Most probable, numbers of cunner larvae arrive in the PNPS area from ofTsite spawning grounds, while a large ponion of the lan ae hatched from eggs produced in the immediate area of PNPS are advected from this area by prevailing currents. It appears that local cunner egg production in the PNPS area was not particularly high in 1997 based on egg entrainment collections; however, the number oflarvae entrained ranked fourth in the hierarchy of 17 years of entrainment I
collections at the plant (19801997). This suggests that large numbers of cunner lan ae drifted into the PNPS vicinity from offsite areas in 1997. A basic characteristic of marine fauna with dispersive pelagic lan ne is that recruitment ofjuveniles into a habitat is generally independent ofoffspring (eggs and lan ne) produced by resident )
adults in that habitat. Because most marine populations are demographically open, local recruitment often is uncoupled from local reproduction by the dispersive larval stage. For most demersal fishes, larval recruitment 22 l
from elsewhere produces the only substantial input of new individuals. If recruitment fails, the local population will decline, regardless oflocal fecundity. Recruitment to open populations can vary temporally oy several orders ofmagnitude.
We have an estimate of the geographical bounds of this population based on recruitment sources, via hydrodynamic modeling ofcunna egg and larval &spersion in the western Cape Cod Bay area conducted by Eric Adams of M.I.T. The model prediction was that 90% of the cunner eggs and lan ae entrained at PNPS comes from within approumately 8.8 km of the power station to the north, from about High Pines Ledge to White Horse Beach (Figure 1). The estimation of adult cunner population size, via tag and recapture, was determined to be unfeasible because of man-power constraints, logistics, and behavioral characteristics of cunner, so comparative juvenile recmitment success at various locations near PNPS is being used to qualitatively assess potential station entrainmentimpact on this species.
Juveniles Juvenile cunner are impinged on the traveling screens at PNPS, typically in smnmer and fall (Lawton and Anderson et al.1984). Only 39 fish (alljuveniles) were estimated to be impinged in 1997. Impingement numbers from 1979 to 1997 have fluctuated but with an overall decline from a high in 1980. Past sunival studies ofimpinged cunner at PNPS revealed initial sunival rates of 24 to 100% (Anderson 1990,1993), with influencing factors including the number of power plant pumps on line and the operational mode of the screen wash system (i.e., static vs. contmuous). Impingement, when combined with sportfishing and entrainment, impact cunner populations.
The Station's escharge, with its high velocity current, waste heat, and periodic chlorine load, can afTect all cunner life stages, particularly the distribution ofjuveniles in the rxeiving waters. In addition to thermal stress, the fast-flowing current can limit cunna mobility and maneuverability. Small age-0 juveniles, e.g.,20-30 mm TL, do not stray far from home shelter and ordinarily avoid the discharge current, which, on an ebbing tide, 23 l
can exceed 2.1 m/sec at the egress of the discharge canal. When the power station is operational, small cunner oficsi are seen by our divers in a control areajust w=wie the thermal discharge. Auster (1987) reported that adult cunner forage further from reef substrate and on current exposed surfaces longer than do thejuveniles. As the cunent dec cases in velocity, smalla cunnar move up into the water column out of the reefinfrastructure arxi onto current exposed surfaces to feed. This process is reversed as current velocity increases.
In 1997,we again surveyed habitat at each sampling site located on three distinct boulder reefs both at the kf--N (June) and end (November) of our field recruitment study to assess similarities between sites and stability over the recruitment season. We enumerated recruits generally two to three times a week, weather permittirig, from 15 July to 31 October for a total of 31 sampling events. We separated the recruit data set into two time penods for analysis - settlement and post-settlement, and considered the dividing point as between 27 and 28 August. Cunner larvae were uncommon in entrainment collections at PNPS after 27 August (Mike Scherer, personal communication)2 While investigating habitat characteristics amongst the three reefs ,a significant difference was found by chi-square analysis in the proportion of structure to non-structure amongst reefs at the beginning of our field study in 1997 (X2 = 20.69, N = 1,659, P < 0.001), with the Discharge site having significantly more stmeture.
'Ihis ehl dificsence was maintaned to the end of the recruitment study, in addition, as in the previous two years, the proportion of habitat structure declined ovcr the recruitment season at all three sites, probably resulting from the loss of macroalgae from storms or grazing. There was no significant difference (P > 0.05) found amongst the three survey sites in substrate rugosity, which reflects substrate relief complexity, when measured at the survey's inception. Whereas, alt .:1 height was significantly different amongst reefs both at the beginning and end of the recruitment season (start: F = 11.8, f = 6.0, N = 329, P < 0.001; end: F = 10.8, TC = 4.6, N = 300, P < 0.001), with the White Horse site having significantly lower canopy height than that at the Discharge or 2 Michael Scherer, President, Marine Research, Inc., Falmouth, MA 24
l Rocky Point sites. Overall, algal height also h-M at all three reefs from the beginning to the end of the season 1
Eponstmet. We define sethusit as the time and process of an individual first taking up permanent residence in the demersal environment following metamorphosis from larva tojuvenile. Within 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> of settlement, cunner acquire the dark pigmentation of adults; small, pigmented fish (~10-45 mm TL, age 0) are referred to s.s recruits. Recruitmmt success is defined as the density (number per area) of age-0 recruits having survived at the l
end ofour sampling pmod (31 October), which in essence was the end of the recruitment season andjust before l recruits went into hibernation On the first day (15 July) of our survey of recruit densities (number 10m-2) in 1997, settlement had l 9--- =91 only at the Discharge site (2 recruits for 100m of2 area surveyed). On the second survey date (18 July), low numbers of recruits were present at each of the three sampling sites. However, recruitment remained relatively low until early August, with increasing densities occumng thereafter through late August /early September (Figure 10). Remutment generally peaked the end of August /beginning of September, with mammal densities averaging 154.5 recruits 10m-2 at the White Horse reef,177.0 at Rocky Point, and 282.0 at the Ducharge. By way ofantrast, peak densities at these respective sites in 1996 were: 41.4,43.2, and 83.2. This we a marked difference in settlement between these two years. For the three survey years (1995 1997),
there were large differences in annual recruit abundance during the settlement period in the study area, i.e.,
1997>l995>l996 (Figure 11). Assuming that PNPS is an objective sampler of cunner larvae in the area as an index orlarval abundance, the entrainment oflarvae at the power station matches up well with recruitment levels:
1997 = 99.6 million larvae entrained, 1995 = 47 million larvac, and 1996 = 17 million larvae. Levin (1996) found that the amount oflarger cunner larvae in the water column determined the amount of cunner settlement in the Gulf of Maine.
It is also apparent that a consistent lucrarchy occuned in recmit densities spatially during settlement each year,with densities at the Discharge > Rocky Point > White Horse Beach (Figure 10). Settlement was by far the 25
1995 meen recna eenney por 10 somre metro 1*"
,ec___-._____ _ _ _ _ _ _ _ _ _ _ _ _ _ _ .
4.c_________ __ ___________.
ix ________
,0a_______ _ .
so____ _
_ _ _ g _ g _ _. _ _ _ .
so____ _ _ _______ __
s0____ _ _______..______
~
n____ _ _ _ _ _ _ _ _ - - _ _ _ _ _ _ .
0......_2,............................................................
SQ7 7/19 740 M M2 8Q2 848 Of7 MS 90610610f131006 Cheervetun Day 8% h
. % .is+m. .iensone.,e..
1996 uses menes senser per te eenses eneene 1%
.o_____________ _ _ _ _ _ _ _ _ _ .
A no____
sums - _ c = . 2._ . . . . _ .
star frie 7me se ma esta stas e/F 9/94 Me the 1M3 tMe Otoserseen Day Gemahg esmer4
. n% .. es
- ane. w . ,e + nom, ..
1997 ween =cna menemy est 30 somm mee,s l m
asa-__-_--- - - - - _ - - - - - - _ - -
20a_______ _ _ _ _ _ . _ _ _ _ _ _ _ .
l 1sa-___--_ Q _-_q__--f-10a______ _
/w (
\9 50 ___ ________
4Wy e
^^
! O l
847 7t19 700 64 M2 802 640 e/7 M6 BQ610410r1310Q6 Observstan Day een,.esanon
. - ,....,. _ , e. . e.,p.e. .,
l Figure 10. Mean cunner recruit densities (ibe average or 10 transects) per site per day for each year or recruitment studies in the Pilgrim Station area,1995 through 1997.
26 l
L
Dedese u.an mena one ey om 30 e ans w
y 2._______ _ _s____.._____. _
isa.______ _
_ _ _\% _ _ _ ** _ _ _ _ _)4 ico _ _ _ _ _ _ _____ ______.
SQ7 7nt 740 84 .Sr12
. . . . . .4G2 . . . . . .808. . . . .t/7 . . . . . . . . . . . .en s 906 im , . . . .0n3,006 Chesneesn Day s= am.
+ on . isms . cm . ism a om .ier wnne Heroe m a , ..
i%
ia_ _ _ _ - . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ .
in-_______ _ _ _ _ _ _ _ _ _ _ _ _ _ .
in____ __ _ _ _ _ _ _ _ _ _ _ _ _ _ .
w-______ _ _ _ _ _ _ _ _ _ _ .
so-_-___ _ ___ _ _.
a _-_- ____ _ - ___
N n.---_
/ _ _-- --- ----
- m. .
v ser rns roe em ena ses ese w one ese som son:vene Oe wee.a Day e- n. ems.
+ ene,n .ms a w m.n .isse a una,m .wr Rodry Point unn=cnae- eyem50e me 2e isc-_______ .. _ _ _ _ _ _ _ _ _ _ _ _ .
i._______ _
Og
'aP_ y 30- . _ _ _E.........dflQ*_. . . . . . . . . . ,.
SG7 7nt 700 M en2 842 SOS trF ent SOS $04 $0r11,006 Cas.Neeen Dey sm
. a ,, . ins + n ,, . isse e n ,, . iner Figure 11. Mean cunner recruit densities (the average of 10 transa:ts) per day by stanon for each year of recruitment l studies in the Ngrim Station area,1995 through 1997.
1 1
i 27
l highest at the Discharge each year, with the Rocky Point and White Horse sites having lower but more similar settlement levels. As for 1997, multivariate repeated-measures ANOVA showed the Discharge site to have significantly greater interim recruit counts than the other sites for the overall settlement period (Discharge vs.
Rocky Point: P < 0.0001; Discharge vs. White Horse: P < 0.0001), and Rocky Point had significantly higher recruit densities during settlement than did the White Horse site (P < 0.02). The magnitude and duration of change in recruit densities observed over the three-year penod was informative. Each of the stations within a year reflected the same trend when comparing among years.
l Distribut;on and abundance of reef fish species, whether in temperate or tropical zones, is afTected by l an interaction of patchiness in reef-related resources, namely, shelter from predators and food supply.
l l Macroalgae are often the major source of structure on temperate reefs, prosiding juvenile reef fish with shelter i
from predators and foraging opportunities. Variation in recruit abundance of temperate-reef species, particularly i
at relatively large spatial scales, often is explained by variability in macroalgal structure. There is evidence of l a strong efTect of algal habitat, including its density, percent cover, and height of individual plants on recruit l
l densities as related to prehnn intensity (Jones 1984; Carr 1989,1991; Holbrook et al.1990; and Levin 1993, 1994). levin (1994) found that when a single form of macroalgae dominated a reef patch, cunner recruit I
densities were relatively low; however, when reef patches were characterized by high algal cover consisting of several forms, densities were high.
As pelagic, pre-settlement fish larvae are advected, there is very likely a probability that they may sample and ultimately select for a particular habitat. Evidence that pre-settlement fish can prolong the duration of their pelagic life stage (Victor 1986), presumably when appropriate conditions for settlement are lacking, supports the tcnet that late-stz,ge larvae are capable of distinguishing among difTerent habitat types at relatively large spatial l scales. However, as benthic juveniles, reef fish do not move over large spatial scales, and, thus, only have the opportunity to select among local habitat characteristics that are patchy at a smaller scale than their home range.
Juvenile cunner (10-12 mm standard length) settle to a range of habitats but generally are associated with i
28 l
4
macroalgal microhabitats (levin 1991,1993).
Cunner remnt desities among reefs for each 3 car at the k-- --g of the post-settlement period difTered, i
with the Discharge reef having substantially higher densities than at the other two reefs (Figure 10), in that differences in larval settlement produced variable recruitment among reefs found at the beginning of post-settlement. Over most spatial scales and differing geographic regions, the most commonly found larval l
estribution pattern is aggregation, i.e., one of p*Eaas. Chunped distributions of reef fish larvae may result l
from the dynanuc interplay between oceanographic and behavioral processes. A number of physical oceanographic (hy&udyrunnic) processes can influmoe espersal or retention of larval fish, including: geostrophic and wind-induced current regunes and related factors - wind mixing of the surface layer; upwelling; temperature, i
! saluuty, and nutncnt &scontmuitics mananasM with frontal zones (gyres), eddies, or coastal plumes; tidal forces; shallow-water intamal waves; and bottom boundary-layer dynanucs In the vicinity of PNPS, a possible alteration in hydrodynanucs,such as an eddy created by the thermal discharge current or other physical factors associated with the power staten may induce higher numbers of cunner larvae to settle at the Discharge site.
i A relationship existed between recruit densities during the settlement period and densities found two i
weeks into the post-settlement penod. Iain (1996) likewise found a significant relationship between settlement and density of recruits two weeks following the last observed settler in late August. In addition, we found l
evidmoe ofdensity4A4 mortality, with a decline in recruit abundance through time evident and attributed l to mortality via predation. Ereig.600 was not the cause h===r age-0 juveniles are site attached, occupying only l
l a few square meters of area. Mortality was linear and gmcrally variable among reefs. During the post-settlement
! penod, post-hoc tests iruhratM that mortality rates at the three reefs ollen differed from each other. For example, i in 1995 recruits on all three reefs experienced significantly different post-settlement mortality (P s 0.005);
l however, in 1997 only the Discharge and White Horse sites difTered significantly (P = 0.009) from each other in mortality rate during post-settlement B- - tag the three study years in more detail as to cunner recruitment success (i.e., recruit density as i
29 j
[ !
l l _- _ _______ __ _
l l
a measure ofrelative abundance at the end of the recruitment period), we found inter year differences in 1995, it is evident that vanaten in mortahty (density dependent) existed among sampling sites, being sufficient to alter the distribution of age-0+ cunner from a non-uniform distribution at settlement among stations to a post-setiement uniform distribution, in that the Discharge reef with the highest initial recruit densities had the same recruit abundance as the other two reefs by the end of the post settlement period (Figure 10). A physical disturbance in the form of storm evcsits may well have ovemdden other mechanisms driving recruitment success in 1996. Severe storm sea conditions experienced during the last half of September,1996 evidently led to a drastic decline in recruit numbers at the three sampling sites towards the end of the recruitment period (Figure 10). There likely was substrate scounng and habitat alteration with a reduction in refuge habitat (e.g.,
macroalgae), which may have caused a displacement of recruits to new sites or more likely outright recruit mortality. By way ofcontrast, recruitment success in 1997 was three times greater at the Discharge site than at the White Horse or Rocky Point reefs (Figure 10).
Spatially among years, recruitment success of cunner at the Discharge site was markedly different each year (Figure 11). However, e.en with initial settlement differences, the Rocky Point and White Horse sites had the same relative cunner recruit abundance at the end of the recruitment seasons of 1995 and 1997 (Figure 10).
This similarity suggests a canying capacity at these two sites, with recruitment success likely mediated by resource limitaten, including competition for available suitable habitat for shelter and food supply, and with post-settlement density-dependent predation mortality also acting upon initial settlement.
Recruitment success for marine fish, which can fluctuate greatly from year to year due to a host of physical and biological factors (Sinclair 1988), has a major influence on population structure. Recruitment success is influenced by habitat structure / quality and how habitat quality (e.g., macroalgal cover) affects sun ival as a refuge from predation. Density-dependent predation can mask initial recruitment patterns (Steele 1997),
uncoupling settlement from recruit abundance at the end of the recruitment season.
A marked effect ofpowcr plant entrainment of pre-settlement cunner (lan ae) on recruitment success is 30
more likely to occur dunng a remmammit samanni what larval supply and subsequent settlement are relatively low ovwall. Ahthough recruitment is a demographic event of relatively short duration, the variation in recruitment, within and among years, influences population size. Initial recruit levels should always be considered when I
cxarmning numbers in a population; however, other factors become meressingly important in high recruitment years, when settlement is large ceiough for post-settlement density-i,wAe effects on settledjuveniles to be a major determmant of recruit success Conversely, in poor recruitment years, recruit limitation, i.e., density-mdgandent flucinahneis in larval supply, is expected to be a major descrnunant explaining patterns in future adult nunbers At any time, one or both may be important Jones (1990) found with a coral reef, sedentary species that with recruitment densities below I recruit m 4, adult numbers were recruit limited; but at densities above I recruit m4 , post-reauitmmt densitye compensation becomes increasingly important. long-term cfTects ofrecruitment on population structure are likely to be known by studies continued over a time scale equivalent to the longevity of the species (Jones 1990).
We used secruit counts at the three study reefs to qualitatively assess the efrects of PNPS entramment on cunner densities. When testing for a power plant disturbance, we assumed mortality was responsible for the decline in recruit abundance over time. During the settlement period, there were an influx of new recruits and a decrease in number ofexisting recruits via mortality. Average recruit abundance during settlement is a measure of the relative settksnent at each reef. We also assumed that the number oflarvac in the water column deternuned the amount of settlement and incipient recruitment at a particular area.
If PNPS markedly impacts cunner larvae through entrainment, it is rt:asonable to believe that the Discharge reef, wtuch is proximal to the power station, should have lower settlement and concomitantly reduced recnut abundance as cu,oy.M to more distant reefs. However, in each of the three survey years, we found the highest relative settlement near PNPS. Possibly, changes in hydrodynamics, such as the occurrence of an eddy produced by the discharge or other physical factors associated with the pow::r plant may contribute to increased l
larval settlement at the Duxharge reef. Furthermore, in the pool of recruits potentially impacted by entreinment 31
I at PNPS, a portion has a reduced probability of contributing to the adult population because of post settlement density s-wg=== tion. Therefore, in general during two of the years (1995 and 1997), for both the settlement
' and post-settlement penods, the data suggest that PNPS had an incaacequential efTect on recruitment success of the local cunner population. However, power plant intervention could play a more important role influencing recruitment dunng a season ofrechaced setticanent. Whcm scellement is relatively low such as in 1996 (Figure 10),
a direct relatxmslup between larval supply and recruitment success may have existed, with entrainment reducing recruit abundance that would have existed without this power plant intervention. It should be noted that it is the sum of the recruitment proce==>< over several years that govern population size.
Adults i
Adults also are impinged at PNPS, but the number has rarely been large. A review of temperature tolerance data on cunner (Kmne 1%9) suggests the presence of an exclusion area within andjust outside the PNPS discharge canal during summer and fall when the plant is fully operational. However, cunner kills from elevated temperatures have not been documented at PNPS. Our SCUBA observations in the thermal discharge area at flood tide dunng prevous summers revealed far fewer cunner inside the mouth of the discharge canal than 60 m immediately seaward. We have measured bottom water temperatures exceeding 30 C at the mouth of the effluent, and it appears that cunner avoid the immediate area at this time.
Cunncr are susceptible to acute, local sportfishing mortality, because of their relatively small home range (Green 1975) and nearshore distribution. Tag returns from our multi-year capture-recapture program clearly mdicate that cunner are relatively stationary, at least during the warmer months off PNPS. In the decade of the 1970's,cunnerled the shore-based sportfish catch at PNPS. None reportedly were landed in recent years, which may be a result ofincomplete reporung during informal creel surveys conducted there.
The objectives ofour 1997 adult cunner work wae: to monitor relative year-class strength by collecting catch data and by aging fish to generate an age-frequency matrix for the local cunner population. In tum, 32 L .
l l
survivalis estunated which is used in an Adult Equivalency Model analysis for cunner From June to July 1997, 5,423 cunner were captured in baited fish traps set in a standard manner overnight at two locations. Of these, 1,897 came from the seaward side of the outcr intake breakwata at PNPS. The remaining 3,526 cunner collected came from the landward side of the outer breakwater within the intake embayment.
Over seven years (1990-1996), we tagged 6,506 adult cunner in the PNPS area. This pool of marked fish provided tag-recapture informatxm that confirmed cunner's site fidelity within a year and for consecutive years. For example, in 1996, our sampling efforts at the outer breakwater netted 161 cunner tag recaptures, with all returns conung from the outer breakwater where the fish were tagged. None had traveled around or through the bn:akwater to the intake side. That were multiple recaptures, e.g.,35 cunner tagged and recaptured in 1996 were taken more than once; one individual was recaptured seven times and another eight. Lateral movement of tagged cunner along the outer breakwater was minimal. Eighty-five percent of the recaptures were taken within 18.3 m of the tagging site; of these, over half did not move from their original capture site. These data provide first-hand evidence that adult cunner in the PNPS area have a relatively small home range, a behavioral trait
< documcrited for Newfoundland fish (Green 1975). In our 1996 trapping efforts, we also recaptured two cunner tagged in 1992,3 tagged in 1993,36 in 1994, and 176 in 1995. All were recovered from the same side of the breakwater where they had been originally tagged and released.
Tag retention can be a problem when external tags are used. Cunner often take refuge under rocky outcrops and within crevices, so there is the potential for snagging and tag loss. In many finfish tagging programs, the percentage of recoveries typically ranges from 3 to 10% (Matthews and Reavis 1990). Our techruque of trapping cunner and recapturing them in fish traps rendered an overall 9% tag return in 1992,31%
in 1993,11% in 1994,26% in 1995, and 25% in 1996 (includes multiple recaptures). By recapturing our own tagged fish, unreported tag returns are not a problem.
In a cooperative venture with UMass Amherst, fecundity of cunner was investigated off PNPS; this
- p. . m was needed for the Adult Equivalent Model. We continued to conduct cunner aging work in 1997 and 33
to develop a synoptic age-length key (Tables 2 and 3). Age was needed to establish an age-specific fecundity I p lhip and survival values for adult cunner We collected 211 cunner (62 to 215 mm TL) in our traps from the PNPS area in 1997 for ageing; of these,184 were measured and successfully assigned an age using collected otoliths (Table 2). About half the otoliths (N=97) were mounted on slides and hand-ground prior to analysis.
Twenty-seven fish could not be aged - 16 had damaged otoliths, two had crystallized, unreadable otoliths, while 9 could not be assigned an age according to our criteria (see Methods this section). Completion of annulus formation, which likely occurs in May in the PNPS area, recently was validated to occur during this time period using oxytetracycline-injected cunner kept in cages in the field under ambient temperatures by researchers in Newfoundland (John Green, personal communication)', This agrees with annulus formation occumng in late May-June reported for fish in Buzzards Bay south of Cape Cod by Serchuk and Cole (1974) and in May for cunner from Connecticut (Dew 1976). Both studies utilized scales for aging.
Annuli for whole otoliths could only be read reliably for fish up to three or four years of age. The benefit ofusing ground otoliths to accurately age cunner is apparent for fish age 4 and older. Some cunner otoliths had eight or more annuli which were difficult to read, but there have been few fish of this size in our samples.
Accurate aging of these larger fish is difIicult due to the crowding of the outermost annuli. There are some large cunner in the PNPS area, but they are excluded generally from entering our traps because of their size. Based on our underwater visual observations, these large fish are uncommon. We did grind otoliths from a few larger fish given to us by lobstermen caught in their traps. We have estimated ages of over 12 years old.
' John Green, Biology Department, Memorial University, St. Johns, Newfoundland A1 B3X9 l
)
J l
34
l l'
l l
l
! Table 2. Age 4ength bey for cummer (seses com'enned)in the PSgrimi fiaallon area,1997.
Tesel b sMs age 1 3 3 . . 6 , .
6848 6 6 4618 6 3 8 73 15 6 5 18 1449 II 4 15 9849 8 5 l) 36 M 4 3 I 7 98-95 3 3 3 Elst 3 3 105.948 3 3 6 SElle 3 3 4 Ill-Ils e I $
336338 9 4 1 6 all-tas 3 3 5 5368M 3 3 5 138.838 3 3 S RElde 4 3 3 1 9 141-345 1 2 3 146tM 3 3 8 1 J
ISS.lSS 4 3 3 8 SElde t $ 4 10 163-168 9 6 7 146 110 2 6 8 9 171-875 3 9 3 9 It&tas 3 3 3 381-909 4 3 7 306tM 3 3 4 198-195 1 4 5 6 IE300 I 1 Tand 884 m* i 1
35
)
Table 3. Meill- (19N.1996, med 1997) ase4mageh key for cauwr (seses cosahamed) la j the stem seastam arum.
W As.
Tassilm aus wm e s-I 3 3 4 s 6 ? e 9 le ela s s nas a s so-as 4 e see s s 6:4 ss is 4670 N 3 23 75-7s 9 le 19 4 49 $3 91 30 esa s3 ss 4 32 aste 6 s4 s as 98-ts 3 ft Il 25 96.100 se 9 39 ses-ses si s4 s w seolse 6 is 3 u Ill-las N s I M 186830 3 19 9 1 32 Ill Its 33 16 3 32 lEl38 9 17 9 27 133-tas 3 le is 33 tElet 3 le se 3 3 31 148 14s 9 13 9 s 27 84898 9 IS s 3 at 158-155 1 14 6 4 3 26 i
inses a se se s as 165-14$ 9 13 3 1 26 l # 919 8 Is 1 1 27 171-175 9 le 7 3 29 IEles 3 3 4 9
! 181-10s 7 6 1 1 15 196190 1 2 4 7 199-l#$ 1 6 3 1 il 36
== i 2 2 201 205 3 1 206 240 1 I 2 188 315 2 1 3 ti+m i i 22122$ 2 2 226 210 i i Teamt dah 657 aond
- 2. Rainbow Smelt Backcround The goal for our 1997 rainbow smelt project was to enhance quality of smelt spawning habitat in the Jones River, the major smelt spawning tributary to PKDB, the presumed origin of the local smelt population (Figure 1). We placed 188 specially designed egg collecting trays in a selected area of the smelt spawning ground on the Jones River to collect naturally spawned smelt eggs on ideal habitat for egg deposition and embryo development. Sphagnum moss fills the trays and provides a substrate that has interstices, which in turn, give the depositional material three dimensions. The moss represents a micro-emironment that ofrers protection for the developing embryos, thus reducing " egg tumover"(loss). Water seeps into the moss, providing a continuous supply of oxygen to the eggs and carrying away metabolic wastes.
We found that our sphagnum-filled trays have consistently collected higher egg sets than natural hard abiotic substrate. The smelt spawning ground in the Jones River is comprised largely of hard bottom (sand, gravel, and cobble). Areas with endenue aquatic vegetation also provide an ideal surface for egg attachment and development, but comprise generally less than a quarter of the bottom area on that spawning ground. Sutter (1980) mported that smelt egg survival to hatching was about 10% on vegetation but only 1% on hard surfaces.
Our trays of sphagnum provided ca. an additional seven percent of plant material to the upper spawning area.
37
l
)
Eens and Larvae I
Total smelt egg producten in the Jones River En so.mo.n in 1997 was down mahan-w=lly from last year. About , a%=
,i .
3 ==== =
90% of the upper spawmag area (Zone A) was not l[
mh=d (Figure 12 ). This included the majority of our une .. .
188 ogg trays, which were spread throughout the .,
W; spawmng area. We considered areas containing more than 50 eggs per square inch as heavy sets,20-50 per I square inch, modante, and less than 20, light. Egg sets A" 3"'s Figure 12.sm.m .as a hy wnhin Zans A & B ordu Jann River on natural substrate and on the egg collectmg trays t=6d* ahm===== ==.1997-was relatively light. Density counts on egg trays (obtamed using square inch counting screens) averaged only 10 eggs per square inch. The upper one third of Zone B had patchy egg sets of light to moderate densities.
Condmons in the Jones Rivcr this spnng wue not especially conducive for successful smelt reproduction. Water tanperatures was unusually cold. Numerous northeast spnng storms not only created a heavy freshet, but also resulted in the deposition of tree trunks and limbs into the Jones River which became obstructions. These envvenmmtal factors may have limited the number of spawning fish entenng and moving up this river system.
'Ihe maao-algae, which had been problematic in past years, were again very abundant this year. However, the majority ofsmelt spawning appeared to occur before the algae proliferated, which we feel did not hinder initial egg deposition.
Although egg sets on the trays were light, many Amencan eels (elver stage) took up residence in the trays, which also prende refuge areas. Wlun the trays were removed from the river aAer smelt egg hatching, the
. moss and elvers were empted into the pond above the dam. The trays have been found to be a good collector of cIvers and offer promise for aquaculture purposes in the future.
38 I
-- )
Dunng the 1997 smelt spawning season, Eel River, Town Brook and Smelt Brook (other tributaries in the PKDB complex) were ia=W weekly for egg praadiaa We sampled areas of known spawning activity based on past observations Town Brook and Smelt Brook had somewhat better egg production than in 1996, howvar, egg sus was generally light and sporadically located over the spawning areas. We did not find any egg dapnadm in Eel River. Spnng storms d==W large amounts of sand into this system which may have covered any egg sets. This poims out the potential problem of sedimentation to habitat quality.
Juveniles
^
Smelt L-; . ------ at PNPS in 1997 was estunated to be 1,579 fish. Since 1993, an estimated 24,118 smelt have been impinged at PNPS. A ig.:stive sample ofirnpinged fish was measured each year. The lengths of these fish were compared to mean lengths of smelt by age-group from an earlier Jones River study (lawton e al.1990). 'Ihe majority ofimpinged smelt wae apparentlyjuveniles (ages-0+ and 1+ fish). The Jones River smelt spawning run has been relatively small fu well over a decade, and these power plant mortality incidents may have impacted existmg and future population growth.
Milla The number of spawning adults in the local population was especially depresse'd throughout the late 1980's and early 1990's. It has been difficult to observe spawning-run smelt in the Jones River during many of our visits to this system even in recent years. The 1996 overall egg density in the Jones River indicated more smelt spawmng than in the past few years. However, very little spawning activity took place in 1997. Egg sets were generally light and sporadically deposited over the spawning grounds Any large impingement incidents will likely have adverse efTects on future stock rebuilding efTorts.
39 I
l
- 3. Winter Flounder
Background
Winer flaimder (Meumnecres americanus) range the northwest Atlantic from the Gulf of St. Lawrence to Chesapeake Bay (Bigelow and Schroeder 1953), being fod b water temperatures betwoon 0 and 27'C and salinities from 4 to 30 . They can form discrete, resident populations which undertake localized seasonal movements (Perlmutter 1947; Sails 1%1; Howe and Coates 1975). Flounder movement and migration are apparently temperstwe diiven (Pearcy 1%2; McCracken 1%3; Scarlett 1988; Powell, R.I. DEM, unpublished data). Some adults emigrate from shoal waters when water temperatures rise above 15 *C and return u waters cool below this level. Other groups of wmter flounder are resident, and, although an avoidance temperature of 24.4 *C was reported by Meldrim and Gift (1971), their year-round occurrence has been documented in some estuaries (Olla et al.1%9; Wilk et al.1977) at water temperatures around 24*C. In addition, Phelan (1992) round adult winter flounder throughout the year in an offshore area of New York and New Jersey, Rament on a menstics' study, Purce and Howe (1977) concluded that estuarme groups of winter flounder do not necessanly constitute separate genetic, biological units. A group may be comprised of an assemblage of adjacet estuanne spewmng units that internux, of which some may be more geographically isolated than others Homing patterns have been in .e;ed to some estunnes (NUSCO 1986; Black et al.1988; Scarlett 1988; Phelan 1992; Powell, unpubbshed data), and several tagging studies (Iobell 1939; Perlmutter 1947; Saila 1%I; Howe and Contes 1975) have provided evidence of high fidelity to specific embayments for spawning following offshore migrations in consecutive years. At the same time, somnvinter flounder disperse to distant locations (Sails 1%1; McCracken 1%3; Howe and Contes 1975; Phelan 1992), and there may be a random search back for the natal spawmng grounds (Saila 1%I), following random food searches (McCracken 1963). Phelan (1992) eperniaW that populations may be discrete only during the spawning penod, with random temperature-related seasonal movaments resulting in an intermix at other times of the year. If the scad for natal spawning grounds has a random component to it, then some winter flounder may be found in non-natst locations during the 40 l
f spawning season From mark and recapture work in the Inner New York Bight, Phelan (1992) purported that winter flounder there formed a dynanne assemblage, consisting of thece reproductively discrete spawning sub-I populations: one that " homes" to natal spawning grounds in the Navesink and Shrewsbury Rivers, another l l consisting of an aggregadan ofgmerally sedentary fish found in Sandy Hook and Raritan Bays, and a third group l
l found ofTshore, with all three capable ofintermixing.
l In Massachusetts, Lux et al. (1970), Howe and Coates (1975), and Pierce and Howe (1977) concluded from meristic and tagging work that, for management purposes, winter flounder consist of three stocks - one l
north of Cape Cod, anothar south and cast of Cape Cod, and the third on Georges Bank. A comprehensive winter flounder mark and recapture progrr.m (more than 12,000 fish tagged at 21 locations) was conducted in Massachusetts during the 1960's by Howe and Coates (1975), who found that flounder migration generally enmmpassed relatively short distances; however, extensive movements of some tagged fish did occur. Flounder dispersal, overall, was greater south of Cape Cod, where many areas are shoal (<l 8.3 m) with waters warming consulerably during the summer. Retums from release sites north of Cape Cod revealed that movement generally was more limited, with many tagged fish recovered in respective subarca release sites, even years later.
Winter flounder spawn prmcipally at night and when water temperatures are at or near the lowest for the year, occurring during late winter and early spring. Spawning occurs in estuaries (bays, rivers, harbors), over ;
shoals outside estuaries, and on ofTshore banks. It usually takes place in the shallows over firm bottom, e.g.,
gravel, sand, celgrass, and pelagic algae. The eggs are demersal and adhesive, and those that fall onto sofl, fine sediments or onto algal mats are less likely to develop. Hatching occurs in about two to three weeks at water temperatures of 3-5 C, Larval stage duration generally is 4-6 weeks, and the pelagic lan ae, which are relatively non-buoyant, can move vertically in the water column, thus somewhat offsetting the effects of a diffusive I
cmvironment. Age-O fish (juvcriiles) are more tolerant of higher water temperatures than are the adults, and they oRen remam in estuarine nursery areas throughout their first year; age-1 fish may do the same (Buckley 1982).
41
____________________J
I The PKDB estuary, not far from PNPS, is a local spawning ground for winter flounder, although spewrang also occurs outsule this estuary (Figure 1). The adult segment of the local population is exploited prior to the spawmng season by a regulated w...,su.1 otter trawl fuhary that is open from i November to 31 January, with a nununum legal fuh siae of 305 mm TL. In past years, this fishery was open into the spring, but declining flounder abundance prompted a mandated reduction in temporal effort.
- Spawmng success, recruitment, and population wA.-.cc are maintamed where physiography and oceanographic circulation enhance larval retention in specific geographic areas. Size of the spawning grounds and larval retention areas are luniting factors to absolute aHa- Winter flounder population size is a function of the size of the physical system underlying larval retention. Large populations generally are found in large boys and on large offshore banks; whaeas, smallar populations are associated with coastal ponds (lagoons) and smaller estuanne river systems (Howell et al.1992). Clearly, the magnitude of impact of a given mortality (power plant related or otherwise) is inversely related to the absolute abundance of the population afTected.
Habitat and wata quahty can be issues on inshore flounder spawning and nursery grounds because these areas are typically subject to anti.4-y=6 alterations and environmental degradation. The various flounder life stages can be affeded by dredgmg, fdling of wetlands, toxicants, disease infestation, hypoxic conditions, and power plant-induced mortality. Direct mortality or the loss of reproductive and growth potential can result.
Along with natural and fishing mortality, impingement and entramment of winter flounder by power plants can maatantially add to total mortality. Losses may be especially problematic when power plant intakes are located in or near spawrung and nursery grounds (Normandeau 1979), e g., at PNPS. All life stages of winter flounder, at least seasonally, inhabit the artificial intake embayment at PNPS, which simulates a cove.
Eens and Larvac The larvae ofwinta floundcr are much more susceptible to power plant entrainment than are their eggs, wiuch are demersal and adhesive. The benthi-pelagic larvae, especially the later stages, are generally more abundant near the bottom of the water column during the daytime and, thus, are vulnerable to entrainment as 42 l
l i___.____--_____--_-----_
bottom water is drawn into the intake structure At PNPS, entrainment of winter flounder larvae has ranged from an em=W 3.5 to 55.4 million annually over the last 18 years (1980 to 1997); the 1997 estimate was, by far, the highest recorded dunng this entire period. Larval entramment was substantially up from 19% (18.6 million larvae). De second highest annual entrainment value of this penod was 29.8 million larvae recorded in 1981.
We questioned whether the large increase in larval entrainment in 1997 was a result ofincreased numbers of spawning adults in the study area. From the Massachusetts Division of Marine Fisheries' spring trawl-survey time senes, we found that the survey biomass index for the Gulf of Maine stock had not significantly increased in 1997, suggesting that flounder numbers were not up this year. Their spring survey has shown record high recruitment of age 2 flounder since 1992. However, we found no correlation between the number oflarvae entrained at PNPS and the number of age-2 fish surveyed, using a two year lag.
By way of companson, the total number oflarval winter flounder entrained in 1996 at the Millstone Nuclear Power Station and the Brapon Point Power Station was estimated to be 53.9 million and i16.4 million, respectively. Entramment at the former was the second lowest since three-unit operation began in 1986, and at the latter, the value was considered to be relatively low in the time series.
Larval mortahty due to entrainment at PNPS in 1997, assuming no survival and using the Adult Equivalent Model, which assumes population equilibrium and no density-dependent compensation, equates to the total loss of47,087 age-3 winter flounder. His Mm=W loss to the local population is more than three times greater than last year's projected loss of 15,395 adults. Entramment losses at the station for the last 11 years have been estimated to range annually from 2,619 adults (1987) to this year's high of 47,087 adults. The equivalent adult estimate (stage specific) for entrainment at Brayton Point Station in 1996 was 32,192 age-3 winter flounder Gibson (1994) exammed data for several winter flounder populations and found that after accounting for adult nxxtality, recruitment rates were lowest in tluce populations (located in Mt. Hope Bay, Niantic River, and off Plymouth in western Cape Cod Bay) that are subject to entrainrdent by nearby power plants.
43 i
I ._.________._.____________________U
rW% the geograpinc exumt of the local populatxm was important to establish the source of flounder larvae entramed at PNPS. His power plant has been shown to entrain larval winter flounder produced in PKDB, but also produced from sites outside the estuary in western Cape Cod Bay (Marine Research, Inc.1988).
l Juveniles In 1997, an estimated 770 wmter flounder were impinged at PNPS. All werejuveniles (age 0 and 1).
Winter flounder were impinged during all seasons of the year, with the highest numbers collected in the month of April. De number impinged this year represents about $6 age-3 adults.
Juveniles tolerate water temperatures up to 27'C, but sublethal effects begin to appear at 20*C, with foodmg inhibition evident at 24-27'C. His should preclude juveniles from the immediate dischrge area in late summer, when temperatures can exceed these values.
1 Adults Ducct mortality ofwmter flounda has been rare in the thermal plume oft PNPS. When exposed to high water temperatures, flounder may vacate an area, if possible, or try to avoid thermal stress by burying into the bottom which would be loww in tanperature than the ovctlying watcr (McCracken 1963); Olla et al.1%9). Adult flounder can tolerate up to 26*C, but above 22.2'C they become inactive and cease feeding. Occasionally during past summers, bottom water temperatures have approached 30'C at the mouth of the PNPS discharge canal.
Stone and Webste. (1977) predicted that adult winter flounder would be excluded by thermal stress from the unmoduse vicinity of the Pilgrim discharge during late summer and early fall, although this impact area is small, at most likelyless than 4,047 m 2.
No winta flounder reportedly were caught by anglers at Pilgrim Shorefront in 1997. In the 1970's and early '80's, this species ranked among the top five sportfish angled in the recreational fishery oft the power plant.
We again contracted the F/V Frances Elizabeth so we could catch and tag winter flounder, recapture tagged fish to descrnune fidelity, and estunate absolute population abundance via density extrapolation and mark and recapture techniques. Between 25 March and 21 May 1997, we successfully executed 195 standard trawl 44
laws withm the study area and caught a total of 29,090 winter flounder for a mean catch of 149.2 fish per tow.
Of the 11,792 flounder sampled durmg the 94 taggmg tows,7,487 (2250 mm TL) or 63.5% were tagged, with a mean number tagged per tow of 79.6. It should be noted, however, that there were probably multiple recaptures of some smaller, untagged fish. In comparison, in 1996 we made 108 trawl tows, sampling 6,708 winter flounder, of wiuch 4,997 (2250 mm TL) or 74.4% were tagged. The mean catch per tow was 62.1 winter flounder; whereas, the average number tagged in a tow was 46.2.
~ From 1994 to 1997, we tagged 14,776 winter flounder at the time of spring spawning fmn Humarock to Manomet Point (Figure 13)in Zones 1-3 (defined study area - Figure 14) of westem Cape Cod Bay. The number and nurumum size of fish tagged by year was: 1994 - 245 fish (a 20 cm TL); 1995 - 2,047 fish (a 20 cm TL); 1996 - 4,997 fish (a 25 cm TL); and 1997' - 7,487 fish (a 25 cm TL). All fish were released in the general vicinity of capture. Tag returns came from commercial and recreational fishermen and our research efforts.
Through December 1997, 830 tagged water flounder have been recaptured at least once during our sampling or by commercial or recreational fishing for an overall return rate of 5.6%. A recapture rate of 2.8%
was reported by Phelan (1992) from 7,346 winter flounder (a 18 cm) tagged in the Inner New York Bight in the late 1980's. In many fish mark and recapture programs, the percentage of returns ranges from 3 to 10%
(Matthews and Reavis 1990). By way of contrast, Howe and Contes (1975), working in Boston Harbor and Plymouth Outer Harbor thirty years ago, reported overall water flounder tag return rates of 58% and 35.8%,
respectively. It should be noted, however, that their returns were compiled for longer periods of time after the flounder were tagged. We believe that under-reporting of recaptures and spatio-temporal bans on commercial fistung in inshore watm mntributed presently to our and Phelan's (1992) lower return rates. Phelan (1992) did not offer a monetary reward for tag returns This year (1997), we awarded a prize of $500.00 for tag return information, with the wmner chosen by lottery from a pool of all retums.
i
~~j h}h : . 8( /
3 e
/ 1 e
8
[
i i , ?
1- i f, .
C) t
]
u,#
2, ;m,gp.
/,?h p?:; .If '
1,.y ll':
'I' 5
5 1
it-o' f> a6% *?
1 y.
y ,(/ .e y }
- {4 -
- y-y4
y
~""
?g u.
5 l*j Y 'o m ll(??5 ' % 9 a
~ "
1..
p
& Q, Llr l I
n i
! al i, c
- W ,, .l
?ll4 yl
/ 5 v
l l. g ,
- - _1 46
f'/// / /
' , ,w ^?- -
-'F{ '
s .r, s-
\
\
\
Y JW gQ s
100 s
- G*
/,
y
&po@*
V ,
,g&f s co0^*$ e** ,g,.*? #
o
I l There were 319 tag returns (from commercial, recreational, and research catches) from fish at large at least one 3 car by the end of 1997, wluch represents 4.4% of the fish tagged through 1996. Of these recoveries, l
we have recapture locations for 265 (83%). For the latter, the returns by recapture area (Figure 14) are found in l
Table 4.
Table 4. Tag naurus by ares for winter flounder (at large at least one year) marlwd desing their spewning seesea la Arves 14 fren 1994-1997.
Ane Number of Pertent of Receptures Total Recaptuns 1 4 1.5 2 164 61.9 3 6 2.3 4 31 11.7 5 12 4.5 6 41 115 7 7 2.6 Total 265 ba=nally, highest returns came in spring (41.7%) and fall (41.1%). Overall,61.9% (164 l sh) of the recaptures with known capture locations came from Area 2, where most of the tagging was done. During the spawning season (March-June),128 recaptures with known locations were obtained, with 82 or 64.1% taken in the tagging area (i.e., the defined spawrung area). Despite the overall low tag-return rate, of the recaptures a high percentage came fan the spawning area, which suggests " homing", or, at least, a fairly high fidelity of the local population.
Nevertheless, geographical isolation during the spawning period is not complete, as some fish have been recaptured at this time in other zones.
A density extrapolation (Area Swept Method) technique was used with data collected from 101 trawl tows made on the FN Frances Elizabeth over the pmod of 26 April to 21 May,1997 to estimate winter flounder t
population size - one for a segment of the flounder population 2 280 mm TL (age-3 and older considered to be 48
adults), the o'.her for the entire water flousuler popuistion (all sizes) (Table 5). Areal measurements were estimated for MLW.
Our estima;=s of water flounder absolute aburdmoe (population estimate) for the study aira (see i Methods section, this report) using an area-swept approach are 160,916 adults { age 3+) and 452,515 total winter flounder (all ages / sizes). 'Ihese estunates att based on a gear efficiency of 100% Trawl gear efficiency is a vanable rather than a caostant; nevertheless, we assumed it was more likely closer to 50%; thus, the adjusted estunates are 321,832 adults and 905,031 total flounder. Precision of the estimates was relatively low (Table 5), with coefficients of variation of 80% and 90%, respectively; thus, comparisons of 1995,1996, and 1997 l-abundance estunates e oflimited value. Even with a standardized sampling nehmialogy, coefficients of variation around density estimates will remain high, owing to the patchy distributions of most benthic species.
l l Table 5. Wh*=d =6-d-la saumbers of winter Amender (botteen eres calculated at MLW),
with 95% confidesee linden, of wiseer fR===d a 288 nue (TL) and for peoled Iregehe seekmetod by asser erewi doesley estrepelseines (adjumeed for geer effleiency)la the Pilgren seedy eres, spetag 1997.
l Teenilleesome Area Numeber of Upper tower (aqante pieters) fleender 95% CL 95% CL Fleender 274,097,240 321,832 372,750 270,914 a 280 num T.L AM Flounder 274,997,240 9a5,a31 - 1,057,a08 75,054 l
l There is gear selectivity,in that we used a 7.6 mm mesh cod-end and fewer small fish were retained; thus, an i f
l l expanded estunate of abudance is biased toward larger fish. There is spatial variation in abundance of this l
species (Lawton et al.1995), and m have not always evenly distributed our sampling effort. It is noted that the adult estimate in 1997 is for a larger study area than in the presious two years. Based on a modeling prediction of the origin of the flounder luvae entrained, we had expanded the study sampling area in 1997, increasing the areal coverage by 76% Based on the model predictions of Eric Adams of M.I.T., we know that winter flounder 49
larvae entramed at PNPS can come from as far away as 17.7 km. In 1997, our study area ran from Humarock, Marshfield south to Manomet Point, Manomet.
To place the water flounder equivalent adult estimate in some perspective, it was compared with population sim estimates for the study area. Population estimates came from two methods - area swept by trav.i and mark-recapture First, a perent loss of adult wmta flounder from larval entrainment was estimated using equivalent aduks (47,087) obtained from entramment monitoring and the Adult Equivalent Model and our ares-swept estimate of the numbcr of adults (321,832) residing in the study area, as dermed for 1997. The estimated adult loss W== of entramment cormsponds to 14.7% of the adults projected by area-swept to reside in the study area during the 1997 winter flounder spawning season Last par's estimated stock reduction due to cetramment, based on a smaller study area, was estimated to be 8.4% of the area-swept estimate of adult stock sim. However, it must be remembered that it would take approximately three years for the entrained larvae to reach matunty. A review by Manne Research,Inc. (1986) of winter flounder early-life studies at PNPS revealed that stock reductions of 0.7 - 2.2% (relative to a larger stock sim back then) were estimated to be possible W== of plant operations Given that coast-wide winter flounder populations have been severely depressed by ovaiidig , power plant entramment could significantly add to total mortality affec;ing a localized discrete population. However, we know from our mark and recapture data to date, the degree of fidelity to the local spawning area is not 100 %, with some mixing likely going on with other nearby spawning units.
Population estimates for 1997, with confidence limits, obtained from various mark-recapture models, are found in Table 6. As to the tagging models of estimating population sim (fish a 250 mm TL), all six !
l cstimates are to the same power, ranging from 1.2x105to 5.2x105subadults/ adults. Five of the estimates were !
in the 100,000+ range. All were hampered as to predictive value because of our overall low number of tag ;
recaptures. ' The Petersen model is the simplest estimator and had a bias in that we did not know how many unmarked fish of 2 250 mm TL were involved in multiple recaptures during the last 15 days of the study, when i
no fish were tagged. With the Jolly-Seber model, we could not generate a confidence interval because of the 1
50 i
l l
l l Table 6. Mast-recapture niedel-ad==w wiek 95% cesAdence lindes, of absolute abundance of winter nounder a 250 men TL in the Pilgrha study arve, spring 1997.
l Medal Numeber of Inwer - Upper Flounder 95% CL 95% CL Pseermee 519,751 443,426 627,820 Scheebel 115,235 101,885 132,610 Schaemacher 122,445 94,922 172,446 Mark 154,934 135,892 177,872 Captuse 164,443 142,187 190,387 Jouy-Seber 150,140 - .
overall low recapture rate, and, based on the chi-square goodness-of fit test, model adequacy to parameterize the population is constrained. The Mark program has flexibility in model selection and design, but because of limited tag retums, the Capture model was preferred because it best fits the data and is the most commonly used model for capture-recapture studies, where capture probabilities vary only by time (although changing environmental conditions can affect capture probabilities).
The estunated adult loss of wmter flounder (47,087 fish)in 1997 becapse oflarval entrainment (55.4 .
mdhon larvae) at PNPS was compared to mark / recapture modeling estimates of present abundance of tie local population. Entramment loss as a percent of estimated abundance of winter flounder 1250 mm TL present in the study area during the 1997 spawning season ranged from 9% (Petersen model) to 41% (Schnabel model).
~ Dree of the values were at about 30% (Jolly-Seber, Mark, and Capture models). Intuitively, when one looks at cetramment numbers for 1997, the equivalent adult estimate, and population estimates for subadults/ adults by area swept and capture-recapture methods (even though they range widely from 100,000+ to 300,000+ to 500,000+ fish), it is reasonable to conclude that PNPS impact at the level of entrainment in 1997 was biologically meaningful to the local winter flounder population. We deem the loss of 47,000 age-3 fish via entrainment in 1997 to be an important source of mortality to an already depressed population from overfishing.
51
In retrospect, we have selected the Capture Model population prediction as best fitting the winter l flounder data and compared entramment unpact at PNPS to that at the Brayton Point Station. Gibson (1994),
, using data collected prior to the wmter flounder stock decline in Mount Hope Bay in the mid 1980's and applied to four methods of estunating population size, generated an average population estimate of 378,957 fish with confidace limits of 40,000 - 718,000. 'Ihe avcrage loss to entrammmt of 32,829 age-3 fish (unstaged approach) amounted to 82% of the Iowar confidence limit and 5% of the upper limit. At PNPS, the population estimate for 1997 via the Capture Model was 164,443 fish with confidence limits of 142,187 - 190,337. The loss to entrunment of47,087 age-3 fish (staged approach) amounted to 33% of the lower confidence limit and 25% of the upper conrh limit.
- 4. Other Species Data on some species were @=iaM from a creel survey MWM at the PNPS Shorefront Recreation Area and by SCUBA divers in the thermal discharge.
The Shorefront was open to the public from 1 April through 30 November in 1997, while the creel survey ofshore-based anglers began there on 20 April and ended on 30 November. A total of 4,143 anglers was micrviewed during 177 sampling days over the 8 months. This included 47 more sampling days than last year.
l The goal was to obtam basic information on sportfishing activity, including fishing efTort and locations, and gamefish catch over time. There were two data collecters, who were seasonal public relations' personnel for BECo; they WM the creel inventory in addition to their other duties. Only weekends were sampled in April, May, and November, while there was daily coverage June through October.
It is clear that most of the fishing efTort was expended in the discharge area from oft the two discharge canaljcaties Some efTort was expended off the outer breakwater, with a minor efTort ofTthe rocky beach located north of the discharge canal. Anglers pnmanly sought striped bass (Morone saratilis) and bluefish (Pomatomus saltatrix), with not much directed elTort for groundfish.
52
'the overall monthly average number of angler trips per day to the Shorefront in 1997 was 23.4, while individual monthly averages ranged from a low of 3.5 in Novembcr to a high of 34.5 in May. Effort was similarly high from June through Sp.ki, rangmg from 24.3 to 29.5 angler visits in a day. The recorded sportfish catch totaled 1,394+ fish [the (+) represents days when the catch of a species was only qualitatively appraised),
comprising three species , viz. bluefish, striped bass, and Atlantic mackerel (Scomber scombrus). The overall mean catch rate (i.e., catch per angler trip) was 0.34+, with a monthly range of 0.0 (April and November) to 0.46+ in July. Reported catches were markedly down from last year, when the overall mean catch rate was 1.5 l
fish per angler trip.
In 1997, the percent composition of the overall recreational catch was 53.8% bluefish,45.6% striped bass, and 0.5% Atlanuc mackerel. Highest monthly catch (pooled spcoes) occurred in July at 389+ fish or 27.9%
of the 8 month total, followed by September (332+ fish - 23.8% of total) and August (21.2% of total).
Atlanuc mackerel, whch were not recorded in the sportfish catch oflast year, were caught in June (4 fish) and July (3 fish).
Stnped bass donunated the monthly totals of May and July, being caught six (May-October) of the eight months surveyed. The first striped bass was landed on 24 May, which is rather late for their appearance in the sportfish catch at the Shorefront. However, it is noted that PNPS was in a scheduled refueling outage in March and April, when only one circulating seawater pump was operating, and little or no waste heat was released. Of the 637+ striped bass reported to be caught at the Shorefront in 1997, most were sublegal (< 86.4 cm TL); about a doan legal bass (> 86.4 cm TL) also were landed The highest monthly catches were made in July (36.1% of total bass catch), folkmed by September (22.0% of total) and August (20.0%). The overall catch rate, i.e., catch per day, averaged 3.6+ bass, with monthly rates from July-September of 7.4+,4.1+, and 4.6+, respectively.
The seasonal catch of striped bass began slowly in May, but improved throughout June when bass were angled in all but one day. Catches peaked in July but then fell off the last week of this month and remained 53 t
o
relatively low throughout the first halfofAugust. In September, the catches continued to build again, which was l
followed by a decline in October, with no bass reported for November.
The total senped bass catdiin 1997 was only 35% of that recorded for 1996. The reason for the drastic catch decline is unclear, in that a much more plentiful supply of striped bass along the Atlantic coast in recent years casesianbly ahmM have resuhed in clovated candies of this species at the Shorefront again this year. PNPS's warm-water diadiarge has attracted bass over the years when the plant has been operating with both circulating seawater pumps in use. About a dozen " overwintering" striped bass were obsen ed in the discharge canal in December 1997, winch increases their susceptibility to mortality if PNPS were to expenence an outage during the winter.
Bluefish were first caught on 12 June 1997 and led monthly catches that month and from August through October. A total catch of 750+ bluefish in a variety of sizes was recorded 'at the Shorefront. Catches uneased throughout the summa months, pealang in Sw.M (25.6% ofseasonal total). Catch rates (i.e., catch per day and catch per angler trip) matched up well with the monthly totals: the former ranging from a mean of 4.5 fish per day in June to 6.4 in September, with the latter range at 0.15 fish per angler trip in June to 0.25 in September.
The 1997 recreational bluefish catch at the Shen ont was substantially down from that in 1996, with a reduction of 62.5% between years. Like the decline in striped bass catch, the reason for the bluefish decline is unclear, but may be partially e-p==4 by the incomplete reporting of the data collectors. We also cannot rule out a real decline in bluefish numbcss in the area. Nevertheless, it is readily evident that when PNPS is operating, the warm-water discharge current attracts and concentrates bluefish, which is advantageous to sportfishermen.
Recreational bluefish catches at PNPS Shorefront have been notable as to the number landed over the years when the station is operating. Conversely, power outages at the station result in markedly reduced sportfish catches at the Shorefront (Table 7).
54
Table 7. Recreational bhneneh cattlers reported by creel survey over three decades at the PSgrte Station Shoreheat be reassion to plant operation Year Nammber of Repeeted Period Plant Status illaeAsh 1973 Soo Septeenber-October On-line 1974 700 Septeanber-October On-line 1975 14 Septenebereber Ofr4ine 1983 IJ00 June-Novensber On -line 1985 2J00 Jame-Noveanber On-line 1984 less than 100 Ash Jame-Nevesnber Ofr-lbw
& for the two years 1986 cessbaned 1996 2A14 Jame-October On -line 1997 750+ June. October On-line Underwater finfish observations provided us with visual data on occurrence and general abundance of finfish in the immediate area of the thermal effluent. From late-July through December, monthly SCUBA dives were completed, investigating the mouth of the discharge canal and adjacent discharge area. Small aggregations of cunner were present throughout the summer and early fall dives. Striped bass were the most commonly observed species, with numbers peakmg in late July /early August at 150 to 200+ individuals per dive. Following a thermal backwash at PNPS in late November, striped bass were not observed within the discharge canal or thermal plume area. However, ca.12 bass were recorded inside the discharge canal underneath the pedestrian bridge on 19 December, while no bass were seen in the plume area. Tautog were noted on all dives up through early November, but following this they apparently had len the area as v/ater temperatures declined. Their numbers ranged from 24 to 75 individuals counted per dive. No tautog reportedly were caught in the recreational fishery at the PNPS Shorefront.
- 5. Imonet Perspective Cunner, winter flounder and rainbow smelt were selected for investigative work that involves assessing impact of PNPS on the respective local populations (Table 8). The response of these species to perturbation may 55
,o a , ,
n .
i Njl
!1 1lf{is! I I]
lIgl I l$I ll '
il Il t li !!! 1 811 ! il i
'{!j if l ,
fl I I Il
$ f ij!i i
p m i t. !! y. ,i ii h
- 1 JP 1 I
1 g! ;jsu it
!! j j' 1 i}i!
e!ji':1
!il!l P I
! I i l liI!!i' l4!l!l 1}1
![!
di' il l
1
! It j[II!II!
i jl!Il!
l !!l! !,l i i i i i i i 1 1 56
be illustrative of power plant-induced stresses on other manne finfish in the area.
Rambow smelt in 1993 and 1994, ranbow smelt annual impingements at PNPS were relatively high - about 9,500 and 10,600 fish, respectively, impingements of that magmtude undar a condition of low population abundance, which appears to be the case, were likely biologically inportant to the local rainbow smelt population. As a remedial measure to offset poww station impact, BECo funded our stockmg of over 1.8 million smelt eggs into the nearby Jones River over the years of 1994 and 1995. From 1995 through 1997, we also placed egg collecting trays into this stream to enhanos spawning habitat for the purpose of optamizing egg survival. This latter elTort is being continued in 1998 in a scaled-down version and also should be considered for implementation in the future if PNPS impacts large r.anbers of rainbow smelt.
Cunner Entramment ofcunner eggs and larvae in 1997 at PNPS equated by the Equivalent Adult Analysis model to the loss of si stamated 498,281 adults from the local population. Entrainment of this magnitude appears to be substantial, but the importance of this loss to the local cunner population is speculative. We had geographically bounded the local population which included all major recniitment sources. Absolute abundance of the local population is difficult to estimate because orlogistics and fmancial constraints. Instead, we conducted recruitment studies for three years to analyze for power plant efTects.
In 1995, our cunner recruitment census revealed that recruit success that year was regulated primarily during the post settlement period by cv. g==tary processes (density-dependent mortality), with the plant's impact oflarval entramment probably inconsequential. The recruitment survey in 1996 was inconclusive as to l plant impact. A number of storm events prematurely terminated the study and were the cause for altering .
recruitment patterns at the termination of our sampling. In 1997, recruitment success again waJ mediated by post-seedemmt density-depandet processes of predation and resource competition, with power plant impact of I
57
low importance. A dificnnee in habitat at one of the sites (Discharge) increased survival there, resulting in higher recruit densities at the Discharge reef by the end of the recruitment season.
Winter flounder Larval winter flounder entramment in 1997 was inordmately high (estimated total - 55.4 million larvae),
despite an outage penod at PNPS extendmg from mid-February through mid-April. In the past, reduced operation of the circulating water pumps during the winter flounder spawning period resulted in substantially lower larval entramment (Lawton et al.1996). It is not clear whether an increase of this magnitude in entrainment of winter flounder larvae in 1997 is representative of naturally high larval abundance because ofincreased egg sunival, or because of an artificially concentrated larval density resulting from the action of on-shore winds and water currents. An entrainment mortality oflarvae at this level would result in an estimated 47,087 equivalent adults (age 3) lost to the population.
We estimated adult winter flounder population size in 1997 at 321,832 individuals by an area swept approach (density extrapolation) using bottom trawl data. We extended the study area in 1997 to sample the recently estimat~i spatial range of the local population; however, much of the area is comprised of"untowable" bottom (i.e., rocky or otherwise unsuitable for trawling). As we were unable to sample within the entire bottom study area, we do not know what the mean density of winter flounder was on hard bottom. At this point in time, we assumed it was the same as on trawlable bottom. We calculated population abundance by expanding the average fish density over the entire study area. Entrainment in 1997 equated to a loss of 14.7% of this estimate of possible adults in the area.
Mark-recapture data were used to address the question of population discreteness and to generate independent estunates of population size to corroborate with the density extrapolation estimate. The low number of winter flounder tag retums obtained through the present time has hampered both these objectives. Despite the low number of tag returns, there is evidence of a fairly high fidelity of the local population. Mark-recapture 58
model estimates of population size (fish 2 250 mm TL) in 1997 ranged from 115,235 to 519,751 fish, with entramment loss as a percent of absolute abundance estimates for winter flounder ranging from 9% to 41%
Mantic silversides In 1994, there wac two acute incidents of high impingement of Atlantic silversides (Menid/a menidia) l at PNPS: 28-29 Novembcr- 5,800 fish and 26-28 December - 6,100 fish. In 1997, it was the dominant species impinged and typicall) has led all other species with an estimated several thousand impinged each year. No compensatory action was taken because the silverside is a prolific annual species with no commercial and only l
l limited recrearmal value as bait. However, the sdvcrside provides an important forage base for piscivorous fish.
Alewife A relatively high impingement of alewives (Alosapseudoharengus) occurred at PNPS on 8-9 September 1995 when an e= tim ='~i 13,100 individuals died. The alewife is important as bait for the lobster fishery and for sportfishing, while its roe and flesh are used for human consumption. Employing a special publication of the Amencan Fishenes Society (1992) we assessed the monetary value of this fish kill to be about $5,000.00. The DMF negneia'~i with BECo. for this sum ofmoney which was awarded to be used for habitat rehabilitation (i.e.,
the money will help rebudd or repair a river her.ing fish ladder in the local area). Large impingements of alewives have been uncommon in recent years at PNPS, and it appears that the number of river hemng are down in the nearby Jones River run. Nevertheless, impingement monitoring should be continued for species commonly impinged on the intake screens at PNPS, so appropriate mitigative measures can be undertaken if warranted.
1 59
V. CONCLUSIONS Cunner
- 1. Impingement of cunner at PNPS is not normally a major problem Annual impingement has been relatively low since 1980,when an estimated 1,683 cunner were affected. Impingement in 1997 totaled only about 39 fish.-
- 2. Large numbers ofcunner eggs and larvae are entrained routinely at PNPS cach year. In 1997 alone, the number entramed equated to the loss of ~498,000 adult fish from the local population.
- 3. The settlement phase of cunner recruitment in 1997 differed from that ic M6, as overall recruit ,
densities were considerably higher from the start at all IFree sites.
- 4. Analyses of the 1997 reauit data indicated that density-dependent compensatory processes (differential mortality via predation)in the post-settlement period likely drove recruitment success, with power plant impact (through larval entramment) probably of muumal ma==-
- 5. For the three tecnnt survey years (1995-1997), there were large differences in annual recruit abundance dunng the settkunent penod in the study area, with 1997> l995>l996. Annual larval entrainment values at PNPS matched up well with recruitment levels observed for those years and may have some predictive value for annual cunner recruitment success in the PNPS area..
- 6. A consistent hierarchy occurred annually in recruit densities spatially during settlement, with densities at Discharge > Rocky Point > White Horse Beach. 'Ihis was converse to our initial expectation oflowest recruitment at the Discharge site due to entrainment from nearby PNPS. A possible alteration in hydrodynamics, such as an eddy created by the thermal discharge current or other physical factors associated with PNPS may induce higher numbers of cunner larvae to settle at the Discharge site.
- 7. Any marked effect of PNPS entramment of pre-settlement cunner larvae on recruitment success is more likely to oaur during a recruit season where larval supply and subsequent settlement are relatively low overall a recruit hmitation scenario, which apparently occurred in 1996. In high recruit years such as 60
1995 and 1997, post settlement density-dependent effects on settled juveniles becone a major determmant of recruit success
- 8. Stnssful high water temperatures likely cause an avoidance response oflarger cunner to the discharge canal and near-thermal plume during late summer /carly fall.
- 9. The effluent at PNPS is of sufficient velocity to cause a small-scale shift in the distribution of cunner 4
by size, with only larger cunner seen residing in the path of the discharge current at flood tide.
Rainbow Smelt 1
- 1. To compensate for rainbow smelt impingement at PNPS, DMF, funded by BECo, stocked eggs and worked at improving spawning habitat in the Jones River for the last four years.
- 2. After two years ofegg stockings (ca.1.8 million smelt eggs) into the Jones River, this effort was ended because oflow source stream egg supplies.
- 3. For the last three years, specially-designed egg mil ~*iaa trays have been placed on the Jones River smelt spawrung grounds, which resulted in an lied number ofeggs being spawned on ideal habitat for egg survival.
- 4. Spawning tributaries should be cleared of any obstructions each year before anadromous fish species begin their spawmng runs. DMF helped with the removal of several tree snags from the Jones River in 1997 and would opt to lend assistance in the future when necessary.
- 5. In general, we would deem our initial smelt' efforts as being of moderate success. We did move numbers of smelt eggs into the Jones River, and smelt did spawn over our collecting trays, with generally higher egg densities obtained on the sphagnum moss. However, we did not measure hatching success and thus,
)
do not krum how many eggs that set on the trays survived to the larval stage. On a positive note, there appanntly were increased numbers of spawning smelt on the Jones River run in spring 1996, when we 61
l l
observed the highest smelt egg densities of the decade However, very little spawning activity and resultant egg deposition occurred on the traditional spawning grounds in the Jones River during spring 1997. Adverse conditions in the river (physical and/or chemical) may have been an impediment to l
spawning success. It is also possible that spawning numbers of smelt were low, and even under ideal enviraa- I conditions, few fish would have reached the spawning grounds.
- 6. A decline in smelt populations has taken place throughout Massachusetts Bay and in Quebec, as well.
Causality for the wide-spread declines is conjectural, although there are obvious environmental concerns ;
of storm-water runofr, toxicants, nutrient loading and sedimentation. These alterations can degrade water and habitat quality and are most likely linked to reduced smelt productiorr. Future remediation efforts in the Jones River and other smelt spawning streams should stress water quality issues in their respective watersheds. This last rw- -.a.dation should be a priority in future efforts to restore smelt popularwwis e.g., watershed runoff timie.t, purchasmg a " green belt" along a spawning river or stream to prevent development or detnmental land use.
Winter Flounder
- 1. The nearby location of winter flounder spawning (retention) grounds, the relatively limited movement pattems of flounder north of Cape Cod, and the geographic bounds of the local population make this species sensitive to impacts from entrainment and impingement at PNPS.
- 2. In late summer, water temperatures in the immediate vicinity of PNPS's thermal discharge can exceed
~
the avoidance temperature (24"C) for winter flounder and exclude them from this relatively small (-
4,047 m 2) area of stress.
- 3. The record 55.4 nsliian wmter flounder larvac entrained at PNPS in 1997 equated to an equivalent loss of 47,087 adult wmkr flounder from the local population. This level oflan al entrainment suggests that 62 i
the year was productive for flounder spawning.
- 4. In 1997, an estimated 770 winter flounder were impinged at PNPS, with the majority being juveniles.
Impingernent at PNPS as a soun:e ofnxxtality is relatively insignificant when compared to entrainment.
- 5. Our tag recovery rate (5.6%) is low, being hampered by the relative number of tagged fish for the size of the area, the sporadic reportmg of tagged fish, and the seasonal closure of the area to commercial fishing.
- 6. Our ppdian estunate ofwmter flounder in the PNPS study area (Figure 1) from density extrapolation is low in precision, while estimates from mark-recapture models are hampered by low recapture rates.
Thus, we cannot with confidence define the magnitude of entramment impact on winter flounder at PNPS, although population estimates by various methods lave been generated.
l 63
f VI. ACKNOWLEDGEMENTS h authors thank Wayne and Dana Bassett of Canal Marine Fisheries, Inc. for donating the bait used dunng cunner trappmg. hmas Hoopes of MDMF produced the GIS map of flounder recapture areas. Robert l
i Dylan and Harold Darnels of Boskm FAwwi Company collected sportfish data at Pilgrim Shorefront. Jay Burnett from the National Marine Fishenes Service (NMFS) in Woods Hole, MA, provided access to equipment used in the agemg work with cunner. Dave and Carole Amold provided their vessel Frances Elizabeth which was used for the tagging of winter flounder. Paul Rago of NMFS provided advice regarding the initial analysis of the wmacr flounder tagging data, and Gary White from Colorado State University assisted with technical expertisc regardmg population estunation using the MARK and CAPTURE tag recapture programs . We appreciate the guidance of Robert D. Anderson of BECo, W. Leigh Bridges of our Division, and members of the Pilgrim Adi.ie ive-Technical Committee. Their input on various studies and editorial comments on project reports and papers have been most helpful.
l l
t
Vll. LITERATURE CITED Amencan Fisheries Society.1992. Investigation and Valuation of Fish Kills. Special Publication 24 Raeha Aa, Maryland %pp.
Andsson, R.D.1990. *--- . =t of orgamsms at Pilgrim Nuclear Power Station. In Marine Ecology Studies Related to Operation of Pilgrim Station, Semi-annual Report No. 35. Boston Edison Company, Braintree, MA.
Andsson, R D.1993. Impingement of orgamsms at Pilgrim Nuclear Power Station. In Marine Ecology Studies Related to Operation of Pilgrim Station, Sani-annual Report No. 41. Boston Edison Company, Bramtree, MA. -
Auster, P.J.1987. The ofTects ofcurrent speed on the small scale spatial distribution of fishes. NOAA Symp.
Ser. for Undersea Res. 2(2):7-16.
Bqpelow, H.B., and W.C. Schroeder.1953. Fishes of the Gulf of Maine. U.S. Fish and Wildlife Service Fishery Bulletin. 53:577 pp.
Black, D.E., D.K. Phelps, and R.L. Lapan.1988. 'Ihe effect ofinherited contamination on egg and larval winter flounder, Pseudopleuronectes americanus. Marine Environmental Research 25:45-62.
Bradbury, C.1993. Ultrasome trackmg of activity and movement patterns of female cunner, Tautogolabrus adspersus, in Broad Cove, C==a*6 Bay, Newfourviland. M. Sc. Thesis. Memorial University of
' Newf-wiimad, St. John's, Newfoundland.
l Buckley, L.J.1982. ElTects of W.hne on growth and bWha =%! composition oflarval winter flounder, Pseudopleuronectes americanus. Mar. Ecol. Prog. Scr. 8:181. I 86.
l l Carr, M.H.1989. F.&ris of macroalgal assemblages on the recruitment of temperate zone reef fishes. Jrl. Exp.
l Mar. Biol. Ecol. 126:59-76. i Carr, M.H.1991. Habitat hun and ncruitmmt of an assembigp of temperate zone reef fishes. Jrl. Exp. Mar.
Biol. Ecol. 146:113 137.
l Dew, C.B.1976. A contribution to the life history of the cunner, Tautogolabrus adspersus, in Fishers Island I Sound, Caaaartimt cha==aa La Science 17:101-113.
Gibson, M.R.1994. Population dynamics of winter flounder in Mount Hope Bay in relation to operations at the l Brayton Point Electric Plant. R.I. Division of Fishenes and Wildlife. Kingston, R.I. )
l Green, J.M., and M. Farwell.1971. Winter habits of the cunner, Tautogolabrus adspersus (Walbaum 1792), !
in Newfoundland. Can. J. Zool. 49:1497-1499. I Green, J.M.1975. Restncted movements and homing of the cunner Tautogolabrus adspersus. Can. J. Zool.. .
53:1427-1431. l Holbrook, S.J., R.J. Sdunitt, and RF. Ambrose.1990. Biogmuc habitat structure and characteristics of temperate 65 1
f l
L _ _ _ ______ _ __________-_-_ _ __-_ __________ _ _ _ _ - _ - _ _ _ _ _ _ _ _ _ _ _ - _ _ _ - _ _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _
reef fish assemblages. Aust. J. Ecol. 15:489-503.
Howe, A., and P. Coates.1975. Winter flounder movements, growth, and martality off Massachusetts. Trans.
Amer. Fish. Soc. 104:13-29.
Howell, P., A. Howe, M. Cnboon, and S. Ayyazian.1992. Fishery Management Plan for Inshore Stocks of Winter Flounder (Pleuronectes americames). Fishones Management Report No. 21 of the Atlantic States Marine Fishones Commission.138 pp.
Jones, G.P.1984. Populaten ecology of the temperate reef fish (Pseudolobrus cel/dotus) Bloch & Schneider (Pisces:Labndae) 1. Factors influencing recruitment. Jrl. Exp. Mar. Biol. Ecol. 75:257 276.
Jones, G.P.1990. The importance of recruitment to the dynanucs of a coral reef fish population. Ecology 71(5):1691 1698.
Kmne, O., (Ed.) 1%9. Marine Ecology, "A Comprehensive Integrated Treatise on Life in Oceans and Coastal
. Waters". Wily-Interscience, landon. 681 pp.
Lawton, R.P., R.D. Anderson, P. Brady, C. Sheehan, W. Sides, E. Kouloheras, M. Borgatti, and V. Malkoski.
1984. Fishes ofwestern mshore Cape Cod Bay: studies in the vicinity of the Rocky Point shoreline, p.
191-230. In: J. D. Davis and D. Mernman (editors), Observations on the Ecology and Biology of Westem Cape Cod Bay, Massachusetts. Springer-Verlag, Berlin, F.R.G. 289 pp.
Lawton, R.P., P. Brady, C. Sheehan, S. Correia, and M. Borgatti.1990. Final Report on Spawning Sea-Run Rambow Smelt (Osmerus mordax)in the Jones River and Impact Assessment of Pilgrim Station on the Population, 1979-1981. Pilgrim Nuclear Power Station Marine Environmental Monitoring Program Series -Number 4: 33-43.
Lewton, R.P., B.C. Kelly, V.J. Malkoski, and J. Chisholm.1995. Final Report on Bottom Trawl Survey (1970-1982) and Impact Asacssment of the Thermal Discharge from Pilgrim Station on Groundfish. Pilgrim Nuclear Power Station Marine Enviro...s.:al Monitoring Program Report Series - Number 7. 56 pp.
levm, P.S.1991. Effcets of microhabitat on recruitment variation in a Gulf of Maine reef fish. Marine Ecology Progress Senes 75:183-189.
lawton, R.P., B.C. Kelly, V.J. Malkoski, J. Chisholm, P. Nitschke, and J. Boardman.1996. Annual Report on Assessment and Mitigation ofImpact of the Pilgrim Nuclear Power Station on Finfish Populations in l Western Cape Cod Bay. Project Report No. 60 (Jan.-Dec.1995). In: Marine Ecology Studies Related l~ to Operaton ofPilgrim Station, Semi-annual Report No. 47. Boston Edison Company, Braintree, MA.
Levm, P.S.1993. Habitat structure, nonspecific presence, and spatial variation in the recruitment of a temperate reef fish Occologia 94:176-185.
levin, P.S.1994. Small scale recruitmmt variation in a temperate fish: the roles of macroalgae and food supply.
Env. Biol. Fish. 89:77 88.
levin, P.S.1996. Recruitment in a temperate demersal fish: Does is.rval supply matter. Limnology and Oceanography. 41:672-679.
66
i dw4l, MJ.1939. A biologmal survey of the salt waters ofleng Island,1938. Report on certain fishes. Winter flounder,Pseudopleunnectes americanus, New York Conservation Department, Albany,28th Annual Report, Part 1, Supplement 14:63-%.
Im, F., A. Peterson, Jr.,' and R. Hutton.1970. Geographe variation in fin ray number in winter flounder, Pseudopleunnectes americanus (Walbaum), ofTMassachusetts. Trans. Amer. Fish. Soc. 99:483-512.
Marine Research, Inc.1986. Winter flounder early life history studies related to operation of Pilgrim Station -
A review 1975 1984. Pilgnm Nuclear Poww Station Manne Environmental Monitoring Program Repost Senes No. 2. Rnenn Edison Company, Bramtree, MA.
Marine Research, Inc.1988. Ichthyoplankton Entrainment Monitoring at Pilgrim Nuclear Power Station, Jan.-
Dec.1988 (Vol 2). In: Marine Ecology Studies Related to Operation of Pilgrim Station. Final Report.
Boston Edison Company.
Matthews, K.R., and R.H. Reavis.1990. Underwater tagging and visual recaptures as a technique for studying movement patterns ofrockfish. Amencan Fishenes Society Symposium 7:168-172.
McCracken, F.D.1%3. Seasonal movements of the wmter flounder, Pseudopleuronectes americanus (Walbaum) on the Atluitic coast. J. Fish. Res. Bd. Can. 20:551-586.
Meldnm, J.W. and J.J. GiA.1971. Temperature preference, avoidance, and shock expenments with estuarme fishes. Ichthyological Associates, Inc. Bulletin 7. 75 pp.
Normandeau Anancidas, Inc.1979. New Haven Harbor EmlayM Studies, Summary Report 1970-77 (prepared I for United Illuminating Co.), New Haven, CT. 720 pp.
NUSCo (Northeast Utilities Semce Company).1986. Winter flounder population studies, Section 7. In:
Monitoring the marme environment of long hiand Sound at Millstone Nuclear Power Station,
! Waterford, Connecticut. NUSCo, Annual Report,1985, Waterford, Connecticut.
Olla, B.L, R. Wicklund, and S. Wilk.1969. Behavior of winter flounder in a natural habitat. Trans. Amer. Fish.
Soc. 4:719 720..
Olla, B.L., A.J. Bejda, and A.D. Martin.1975. Activity, movements, and feeding behavior of the cunner, 1 Tautogolabrus adspersus, and cs,..yediou of food habits with young tautog, Tautoga onttis, ofTlang Island, New York. Fish. Bull. 73(4):895 900.
Pearcy, W.G.1%2. Ecology of an estuarine population of winter flounder. Bull. Bingham. Omanogr. Collect.,
Yale Univ.18(1):78 pp.
Perlmutter, A 1947. The blackback flounder and its fishery in New England and New York. Bulletin of the Bingham Occe..oy.phic Collection, Yale Univ. I 8(1):1-78.
Phelan, B.A.1992. Winter flounder movements in the Inner New York Bight. Trans. Amer. Fish.
Soc.121:777 784.
67
Piesce, D., and A. Howe.1977. A further study on winscr flounder group identification ofTMassachusetts. Trans.
Amer. Fish. Soc. 106(2):131-139.
Pottle, R.A., and J.M. Grem.1979 Temtonal behavior of the north temperate labrid, Tautogolabrus adspersus.
Can. Jrl. Zoology 57:2337-2347.
Rachards, W.J., and KC. Lmdeman.1987. Remutment dynanucs of reef fishes: planktonic processes, settlement, and fishery analysis. Bull of Marine Sci. 41(2):392-410.
Sails, S.B.1%1. A study of wmter flounder movements. Limnol. Oceanogr. 6:292-298.
Scarlett, P.G.1988. Life history investigations of marine fish: occurrence, movements, food habits and age structure of winter flounder from selected New Jersey estuaries. New Jersey Department of Environmental Protection, Technical Senes 88-20 Trenton, N.J..
Sm:huk, F.M., and C.F. Cole.1974. Age and growth of the cunner, Tautogolabrus adspersus (Walbaum) in the Weweantic River estuary, Massachusetts. Chesapeake Science 15:205-213.
Sinclair, M.S.1988. Marine Populations: An Essay on Population Regulation and Spectation. Washington Press, Seattle,252 pp..
Steele, M.A.1997. The relative importance of prme==ee affecting recruitment of two temperate reef fishes.
Ecology 78:129-145.
Stone and Webster Engmeering Corporation.1977. Supplemental Assessment in Support of the 316 Demonstration, Pilgrim Nuclear Power Station, Units I and 2. Boston, MA.
Sutter, F.C.1980. Reproductive biology of anadromous rainbow smelt, Osmerus mordar, in the Ipswich Bay area, Massachusetts. M.S. Thesis, Univ. Mass., Amherst. 49 pp.
Victor, B.C.1986. Larval settlement andjuvenile mortality in a recruitment-limited coral reef fish population.
Ecol. Monogr. 56:145 160.
Wilk, S.J., W.W. Morse, D.E. Ralph, and T.R. Azarovitz.1977. Fishes and associated enviromnental data collected in New York Bight, June 1974-June 1975. NOAA (National Oceanic and Atmospheric Admuustration) Techrucal Report NMFS (National Marine Fisheries Senice) SSRF (Special Scientific Report Fisheries) 716.
l 68
FINAL SEMI-ANNUAL REPORT Number 51 BENTIIIC ALGAL MONITORING AT TIIE PILGRIM NUCLEAR POWER STATION (QUALITATIVE TRANSECT SURVEYS)
January-December 1997 to BOSTON EDISON COMPANY Regulatory Affairs Department Pilgrim Nuclear Power Station Plymouth, Massachusetts 02360 From ENSR 89 Water Street l Woods IIole, MA 02543 (508)457-7900 l
1 April 1998
TABLE OF CONTENTS EXECUTIVE
SUMMARY
.............................................. 1 1.0 INTR ODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 2.0 FIELD ST UDIES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 2.1 METIIODS ................................................ 4 2.2 RESULTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 2.2.1 MARCH 1997 TRANSECT SURVEY . . . . . . . . . . . . . . . . . . . . . . . . 6 2.2.2 JUNE 1997 TRANSECT SURVEY . . . . . . . . . . . . . . . . . . . . . . . . . 13 2.2.3 SEPTEMBER 1997 TRANSECT SURVEY . . . . . . . . . . . . . . . . . . . 14 2.2.4 DECEMBER 1997 TRANSECT SURVEY . . . . . . . . . . . . . . . . . . . . . 15 2.3 DI SC US SION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . I 6 3.0 IMPACT ON ALGAL DISTRIBUTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 3.1 B AC KG RO UND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 3.2 QUALITATIVE TRANSECT SURVEYS: 1983 1997 . . . . . . . . . . . . . . . . . . . . 17 4.0 CONCLUSI ONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 5.0 LITERATURE CITED ............................................ 24 l
J l
i i
i 1
1 1
LIST OF FIGURES Figure 1. Location of Pilgrim Nuclear Power Station Qualitative Algal Survey Area ...3 Figure 2. Design of Qualitative Transect Survey . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Figure 3. Qualitative Algal Survey Data for 1997 Compared to Ilistorical Baseline Data
....................................................... 7 Figure 4. Denuded, and Sparse Chondrus Zones Observed in March 1997 . . . . . . . . . . 8 Figure 5. Denuded, Sparse, and Stunted Chondrus Zones and Dense Mussel Area Observed in June 1997 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 Figure 6. Denuded, Sparse & Stunted, Sparse, and Stunted Chondrus Zones and Dense Mussel Area Observed in September 1997 . . . . . . . . . . . . . . . . . . . . . . . . . 10 Figure 7. Denuded, Sparse, and Stunted Chondius Zones Observed in December 1997
....................................................... 11 Figure S. Results of the 1997 Qualitative Transect Surveys of the PNPS Acute Impact Zone off t' he Discharge Canal taken in March, June, September, and December 1997 ........................................... 12 Figure 9. Monthly PNPS Capacity Factor (solid lines) and Circulating Pump Activity (black bars at 100% = 2 pumps; at 50% = 1 pump; at 0% = 0 pumps)
Plotted for the Period 1983 Through December 1997 . . . . . . . . . . . . . . . . . . 18 I
Figure 10. Area'of the Denuded and Totally Affected Zones in the Vicinity of the PNPS Emuent Canal Plotted with the Monthly PNPS Capacity Factor (MDC) for the Period 1983 Through 1997 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 i
TABLE Table 1. Qualitative Algal Survey Data for 1997 Compared to IIistorical Baseline Data . 7 APPENDIX Appendix A. Quality Control (QC) Protocol for Qualitative Transect Surveys at PNPS O ut fall A rea . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 6 il l
l EXECUTIVE
SUMMARY
This report presents results of qualitative surveys of benthic algae performed in 1997 in the area of thermal effluent from the Pilgnim Nuclear Power Station (PNPS). The report summarizes the impact of the PNPS on algal distributions near the discharge canal. Field studies for 1997 were conducted in March, j
June, September, and December and included transect surveys designed to map algal cover in the area of water outflow. These investigations constitute the most recent phase of long-term monitoring of thermal effluent effects on the benthic algal community within and just offshore of the PNPS discharge canal. Field survey techniques were identical to those used in prior years. Starting in 1996, data from each quarterly survey were compared to the historical baseline (maximum measurements recorded prior to the 1996 survey year) for that season. Measurements greater than 15 % above the historical baseline trigger a report l to the Pilgrim Administrative-Technical Committee's (PATC) Benthic Subcommittee for review.
l The qualitative transect studies performed to evaluate the Chondrus crispus (Irish moss) community indicate that from October 1995 through December 1997 (with the exception of December 1996) the sizes of the denuded and totally affected areas in the thermal plume were larger for each season surveyed than in earlier surveys when the power plant was in full or nearly full operation (1983,1985,1989-1995). For the 1997 surveys, one (spring) to four (summer and fall) defined parameters exceeded the 15% trigger level. Chondrus denuded and totally affected zones were larger than historical maxima in all four 1997 surveys. A dense cohort of newly settled blue mussels (Mytilus edulis), first observed in June 1997, was much reduced in number by September. Damage to Chondrus plants from extensive mussel settlement appears to have caused the increase in area of denuded and total affected Chondrus zones between the spring and summer surveys. From June through December, the affected area extended northward from the central transect line (CTL) much farther than ever before (58 m in December). In September, the divers were unable to extend their smvey line to encompass the nomial Chondrus zone; this unprecedented increase in size of the Chondrus affected zones possibly was caused by some combination of siltation and/or light shading from turbidity resulting from the maintenance dredging operation that occurred in the intake embayment near the cooling water intake structure from mid-June until the end of August.
Although the annual capacity factor at PNPS for 1997 was 73.4%, less than it was in 1995 and 1996, this was not reflected in any size reduction of the Chondrus affected zones for 1997. The 1997 summer dredging operation may have exacerbated the usual effects of high power plant operational capacity, warm summer water temperatures, and extremely dense settlement of mussel larvae in late spring, on the algal community, resulting in the largest Chondrus denuded and totally affected zones ever seen.
1
l
(
l l
1.0 INTRODUCTION
The presence of hundreds of square meters of seafloor where the regionally abundant red alga, Chondrus crispus, is unnaturally absent, even in the presence of suitable substrata, provides evidence that the PNPS nearfield discharge area is affected by elevated temperature and high current velocity, causing bottom scouring, of the cooling water outflow. To study this acutely impacted area, a qualitative diver transect study was designed to map the effects of the thermal effluent on nearby algal distributions.
SCUBA divers perform quarterly transect surveys to measure the extent of denudation and other reductions in size or density of the algal flora, particularly Chondrus crispus, in the nearfield discharge area.
This report continues documentation of long-term (24 yr) benthic studies at PNPS designed to monitor the effects of the thermal effluent. The 1997 monitoring program was identical to those done since 1992 and involved qualitative underwater surveys of algal cover in the nearfield thermal plume of the effluent, within and beyond the discharge canal (Figure 1), performed in March, June, September, and December. Currently, no quantitative assessments of benthic flora or fauna are being made. Beginning in 1996, reports have been prepared after each quarterly survey to compare the collected data with an historical baseline that tabulates, for each parameter, the maximal sizes measured prior to the 1996 survey season (1983 through February 1996). This Semi-Annual Report includes seasonal qualitative observations, tabular and graphical comparison of these data with historical baselines, and a summary of the potential impact on algal distributions caused by PNPS. Work was performed under Boston Edison Co. (BECo) Purchase Order LSP007548, in accordance with requirements of PNPS NPDES Permit No.
MA 0003557.
1 PNPS is a base-load, nuclear-powered electrical generating unit designed to produce 670 megawatts i of electrical energy when operating at full capacity. The condenser is cooled by water withdrawn from Cape Cod Bay and subsequently returned to the Bay via an open discharge canal designed to dissipate heat through rapid mixing and dilution of the outflowing water. Two circulating water pumps produce a maximum water flow of approximately 20 m2s i. The PNPS cooling system may affect the benthic community in three ways: 1) by warming ambient waters (aT=18'C),2) through chemical discharge (mainly Cl 2), and 3) by seabed scouring from the rapid (- 2.1 mps at low tide) flow regime. Increasing temperature and chemical discharges may stress the algal community so that species composition and community structure change, with the extent of such change depending upon season and local I oceanographic conditions. A high current velocity directly affects the benthos by actually removing benthic organisms and inhibiting settlement and recolonization; where there is intense bottom scouring, rock surfaces may support fewer and smaller macroscopic organism:: than normally would be present.
i 2 '
l
Plymouth Cape Cod bay l . . , . .
Bay i ,P., \
.s
',: :r::. .
,'J[2 Outer Tautog Rock
-4,
.'.s. ,
-.....,,.sr'- . . ,
y.
~.
a',.'.'
... ... :: 1.. ', .
?, ,, s\ 's \., N y ...-
~' -
Survey Area
,[~' ',
9 '- pod '
k p., U x, .;.,,,,
PNPS]'" # 'i :. ! .i '
&gl 4 't.c /'), '
's 1 t,
'# '> White Horse Rocks 4Od, %. '
o . ',..3., .,
. K o .
- * *. ' f h y', '9 's,
- . s y ', 't, .
h O
t.,
%,., 'g,g h/, ] 3 o ~, =..,
' t* ' s ,
's
> e ., .~. ~.
04 ',;. ' . , O' ,,
--' s.
(ps - -
?n t r : .* -
..s
.~ . .
.. . N -
i .. ,',
w' ,: . ....*
- ,~ ', s'
- \ :.,;' , i tso@et90# ;;;
, 2000 Feet ..,...:i.,
- v. ..
, 1000 Meters ,A, j'
, g[
yj N:::
3 ,*l
,%.. r ps*
- W; ,.
- a. .e.' ,
{. '
- 4..s 1:
c.. = . . r
.. ,v:
l ./ . :."'-
Figure 1. Location of Pilgrim Nuclear Power Station Qualitative Algal Survey Area.
3
L 1.0 FIELD STUDIES 2.1 METIIODS The qualitative algal survey is performed by SCUBA divers in the same location and with the same techniques that have been used since the current monitoring program began, approximately 16 years ago.
The effluent area is surveyed by two or three SCUBA-equipped biologists operating from a small boat.
For all 1997 surveys, except March, the divers were able to launch their boat from the fishermen's launching site within the PNPS facility. For the qualitative transect survey, underwater visual observations are made along the axis of the discharge canal. A line is stretched across the mouth of the discharge canal (Figure 2). A weighted central transect line (CTL), marked at 10-m intervals, is then attached to the center of this line and deployed along the central axis of the canal to a distance of 100 m offshore, where it is anchored. Using a compass, divers extend a measuring line, marked at 1-m intervals, perpendicular to the CTL at each 10-m mark. For the first three 1997 surveys, a 45-m line (increased from the 30-m line used in previous years) was used; but because this proved to be too short for the September survey, a 90-m line was used in December. A diver swims along this third line, recording changes in algal cover from the CTL through the denuded, sparse, and stunted Chondrus areas, until the algal cover looks normal. A large boulder, nearly exposed at mean low water, is used as a landmark by dive teams and serves as a visual fix for proper alignment of the CTL. To ensure consistency among surveys, the divers make sure that the boulder is always located at 65 m along and just to the north of the CTL.
The terminology established by Taxon (1982) and followed in subsequent years uses the general abundance and growth morphology of Chondrus crispus to distinguish between " denuded" and " stunted" zones. The denuded zone is the area in which Chondrus occurs sparingly and only as stunted plants restricted to the sides and crevices of rocks. In this area, Chondrus is found on the upper surfaces of rocks only where the microtopography of the rock surfaces creates small protected habitats. In the stunted zone, Chondrus is found on the upper surfaces of rocks but is noticeably inferior in height, density, and frond development compared to plants growing in unaffected areas. In 1991 the divers began to discriminate between a stunted zone and a " sparse" zone. The sparse zone is an area with normal-lookin;; Chondrus plants occurring only at very low densities. The control zone begins at the point where Chondrus height and density are fully developed. The dive team must keep in mind while taking measurements that the shallow depths northwest of the discharge canal hamper normal Chondrus growth. In addition to 4
Discharge Canal V *- Barrier Net Sill J >
o
,o o
, %_ __ J
^
)
>N %
,o ,
oc o 3
O 00, OO Oo 60 0 Effluent Line W 10 gp W
to g Weight Marks at 10 - meter intervals ao q,
- qi t j Diver 1 50 g>
Diver 3 Diver 2 T
7o di Reference Boulder 30 Meter Measuring Tape I Central "qi l Transect Line "di (CTL) 1# qi Diver Safety Line ;
7 Anchor and Line Figure 2. Design of the Qualitative Transect Survey.
5
evaluating extent and condition of algal cover, the divers record any unusual recent events in the area, such as the occurrence of unusually strong storms, and note the location of any distinctive algal or faunal associations.
Since the April 1996 survey, Quarterly Progress reports have been submitted to BECo. These reports tabulate areal results of each SCUBA survey and compare them to previously measured maximal sizes of Chondrus denuded and toally affected zones, as well as other parameters, for that season.
Particular attention is pai ' a changes in the sizes of impacted regions that exceed earlier results (prior to 1996) by more than 15%, which requires a written report to be submitted to the PATC Benthic Subcommittee. Table 1 and Figure 3 summarize these comparisons for 1997. The quality control (QC) protocol for the 1997 benthic algal monitoring program is included as Appendix A.
2.2 RESULTS Qualitative transect surveys of acute nearfield impact zones began in January 1980 and have been conducted quarterly since 1983. Four surveys were performeo (March 28, June 22, September 17 and December 17) during the 1997 reporting period, bringhg the total number of surveys conducted to 58.
Results of surveys conducted from January 1980 to June 1983 were reviewed in Semi-Annual Report 22 to BECo (BECo,1983). A summary of surveys conducted between 1983 and 1996, including a review of the four performed in 1996, was presented in Semi-Annual Report No. 49 (BECo,1997). The present report summarizes the March and June 1997 survt vs, presents detailed results of the September and December 1997 surveys, and dismsses long-term trends.
Figures 4 - 8 show the results of the 1997 transect surveys. In the figures, the denuded zone is essentially devoid of Chondnes crispus, while sparse zones have normal looking Chandnis that is sparsely distributed and stunted zones contain smaller than normal Chondrus plants. Dislodged jetty boulders encountered by the divers along their transects are indicated. The landmark boulder (at 65-m) is plotted in all figures as are positions of other common algal and faunal species observed. In June, the divers del;neated a very large area completely covered by juvenile mussels that was nearly coincident with the totally affected Chondrus area. The dense mussel area was much reduced in size by September.
2.2.1 MARCII 1997 TRANSECT SURVEY The denuded and sparse Chondrus zones observed on March 28,1997, immediately offshore of PNPS, are shown in Figure 4. The denuded area was much larger than during most prior spring sutveys, 6
Dhble1. Qualitative Algal Survey Data for 1997 Compared to IIistorical Baseline Data.
Spring Summer Fall Winter Measure. March Kwoncal Percent June thstoical Percent Sept thstancal Percent Dec. thstorical Percent ment 1997 Dawlme Change 1997 Baschne Changs 1997 Baselme Change 1997 Dascime Change (Date) from (Date) from (Daic) from from (Date)
Baschne Baschne Baselme Baselme Total 1662 1321 +26% 2505 1835 + 37% >3587 2043 >+ 76% 2241 1961 + 14%
Denuded (3SI) (680) 2 (l0/95) (2Std Area (m )
! Total 2092 2029 + 3* 4 3972 2135 + 86% >4364 2348 >+ 86% 3512 2328 +51%
- Affected (4 '83) (6S0) (10/95) (2/96)
. Ares (m')
Maximal 100 94 +6% 130 los +24% 115 100 + 15% 101 100 +1%
Disiance of (3SI) (6S2) (990 (12S).
Affected 10/95 2S6)
Area from Diacharge Caal(m)
Maximal 37 40 8% 61 39 56*. >59 42 >+4/k O 42 +67%
Width of (4/83, (6/841 110/94) (12*83)
Affected 3'84)
Ares (m) 5000 140 4500 - ....
- 120 4000 - ... .. .. .b* **- -
3500 - g*.~."'.. ~~ T 1 -
- 100 .
3000 - *'
g :
- -- 80 e
- 2500 - -
__[ g
' 2000 - .i I '--
M-ar.( { ~ _' ' - 60 3
- '~
1500 - e G '
- h ; 40 $
$0I 0
]k I.
h -0
- 20 Spring Summer Fall Winter Season Total Denuded Area-1997 Total Denuded Area-Baseline
- Total Affected Area-1997 assa Total Affected Area-Baseline ior Maximal Distance-1997 - -A - Maximal Distance-Baseline
- C- Maximal Width-1997 --G-Maximal Width-Basel'ne i Uigure 3. Qualitative Algal Surv y Data for 1997 Compared to IIistorical Baseline Data.
7
I March 1997 N
-sm Normal
~ 110 Chondrus Littonna
-r l00
- u
$ i Sparse . .f l i,fS"88IlN):
~
l90l
.:',/
Chondrus N :.: ;.l;.};[e:i:'n'.'/,S.
..*)
,' llh!!!!!)!hhl,[
l Littonna
\\ '
f!30l!!!!!!!!!!f N .f.X@'.' .;;,;
' Corallina
, . i
- g
[' ' $'. s,' $ !
,0.$;i Littorina l:-l2
q
$ $ ' " Nh }.
'ie!!!9?4!$f9l;i i:: N Q f:: i +4e: i LittSbib} ($:
- ?.~' -lhh'iji' $:::L f Chondrus f:ke?i 5l?lfs!!i Isisnas,0. Denuded Zone
' $f!!I!}0!!
Corallena
..i :. i i ,,a ~i ll!!!!!i aisc hf w'd:llait!!!yjjyjgi uttonna i : !!!r wi];;l l. . iili!
~i littonna h;! .?$".b.
i.' ,!!!! hia,.i. :i 6P 'i e
j'iai:k!!$0Q
, . purpureume l l L
! #i l L;! ; W M Submerged Jetty ,
/Eftluent caNati :
I 3 1 4 I 5 5 5 40 30 20 10 0 to 20 30 NORTH METERS SOUTH i Area Legend Denuded Sparse Control Jetty Boulder II N i
? fffI:.$
Figure 4. Denuded and sparse chondrus Zones Observed in March 1997.
l 8
l Jun31997 N'" %
.'s N il fi:b.
, :J ik..lg[it.>.
,.s .
~.V:.:.i. :.s
,':.*h. I.. ..f*,' :@:.,
- . :: ~. Sparse
?
lij$0l! .'.::!b ;/ Chondrus s*
'u.: i 'v
?;,
v j,0 . ,'
- ili e !.
-! p o i. e
'!i6.
, Mussels anno ! Astenas j y' a
M.nm!!!!!!! !
i ,.. ussep . . . , . :h
- Stunted Amphipohn Chondrus\ Flounder i I'i./.ie
!! :<Vl/l!b'% ?:;.
h, N' '
!qi u%sels
.i ?.*. ' t i:; fl g Sparse Chondrus
.'N.p;n M.
lI.:..l;.,9;g, '
,, ;.:.q;.y2i's: :. R'* '. !! .r ..
,:::.t.:.Q:50,:'W Q: : ! - ?f
' :::K ~ : . . ' . . li$;i?, ~
,:.:<. f .n
.i cunner sses AO . Mussels;r:lti i
- ,3
. . . o ; :,. ..:,,
- ' :..).
.: i. k:..
9nli! , ! .t::;,
i v<;;. bob. ster C,rab:- : yJ34jif 3 Musjejs'Q
.1w . .
.m .
- u. c .#i u d :ly;b:p
.. ;.NOorattina . . !iWW "
. . il! ! l' r: :.
. 4. '. . . '.* . - . "!j
.'9;;;! L ussels: :
i j ll l I
- lN'8.
- '
PN
. .:.:s . .
-<e
- *
- Mussels Crab
- 3 L"im'ulus iGO' i* '~ -
.- L i:y u
Chondrus
- ii;ON..
' .Q N..' ' l :.;f9.:.
e.i.
- ;-,;E$
- * ' .\
i; [j.l .s
- Denuded Zone
.i;;,f.' .' .* 1
..y.,<..... .
i... -
< i;ia, , ,
. . ' . dus is, ,,g E,
f,: :: l /: :,-
' * ! Muss,e:s "
. ' . ' .^. .,
llW, l?' Mussels. ;Eriteromor'p ti5 ,
M:* ;* ^ ' : ;, !!]!!!!!#!!W@ $,
, ,j U!SCitartai !
. ., - .- ;; j
' ;. *.I ii;e r A,
3* :
< -- ;#dji' ) '
i i i I
g I i .
, ii .
l j !
1 4 ll !
Ai 'i . -!
l
!!Ni Submerged Jetty l
i j
i
- ;Efflud:MCA.: nal! I i a i e i i i . !
40 30 20 10 0 10 20 30 NORTH METERS SOUTH l Area Legend: Denuded Sparse Stunted Control Mussel Jetty Boulder Area )
T iyi. i ' ;.f.
. . 0;;0:
M
!!ji um!:;i!!i l 6/,. 'W::$.: ,.
Figure 5. Denuded, Sparse, and Stunted Chondrus Zones and Dense Mussel Area Observed in June 1997. i 9
S pt3mb r1997 $
N 120 Stunted Chondrus :
- .e -
\ f, * , h}';.,
ilMRfg...
L.gf,:,::.5 :,. .
a
.iD
.~
- f' Astenai ,
Sparse
- +,,'* . .=Astenas ' . , ,11l , ,' .E; :
ggg~ g 3' Chondrus s'
.. .- 4 i
W.d's s'<; 9 \* 'Xl .
t!.
CnW,*u, ' .. Y,%f?. hi':khkk$Y... :, ..
$llh ha. IUO!
fb h$hha p> Chondrus y
^
pgg; j g: 'i!n:
iAnemones f.
Ent morp a i t;
artna.
-]'
yne.q A -
.f>
- inti: M:j 3p :i L:n hracelarla ~
'gll Sparse &
?I$\l \
Stunted
! !!f*lilllll a8 .i Chondrus Sparse ak25,'
i!!!!!!!!%so. ;. p Chondrus %
Enteromorpha i;
$$iE0$' y
]$ : .
Chondrubl $ # $bE .t!!
'^!f,f #.purpureumii
,!3 Denuded Zone Sparse q! .
'll!
ipuvinnuill!!ha!!T. !
5D !!! !! Enteromorp $
Chondrus b ? rubilalidllj l ll @l rpureum [pu!liMih,' ',
ll,llj,iQ i
- i
- p
' !!!!!!' ) 8
' i !!!!. i
(.
N Il m jui 1'h Submerged Jetty i,
!! Eft:uerd danal i i i i i . . i 40 30 20 to 0 10 20 30 NORTH METERS SOUTH i
Area Denuded Sparse & Sparse Stunted Control Mussel Jetty Legend; Stunted a Boulder hk:k' -
,,M,.-
Figure 6. Denuded, Sparse & Stunted, and Stunted Chondrus Zones and Dense Mussel Area Observed in September 1997.
10 i
I l
L l
l December 1997
~ '38 1
N
-120 t
1 -110 l Mussels Astenas Cancer
~'
,.100
('!?.
9:: Stunted
. .:h..:g,.
l 4:G8:. k .m.e90.
Chondrus
.:i. ::.::::. ,
Sparse ~ .<
", W;_::li'%.taiciridsiis:
.-lq :;lLibanta:l, ".
Chondrus ..
1 i.?; Normal
'8 m!!n;
, , . :.7.;:. .' . . . ?. :W: N#.,Q;.ls
.. ' ' 't-':? . Chondrus
, ' I.);.;.,;
lh.,: {;f; Anemones, '*.,. '.
r ", *
..j !O. .f:.f ../<, .' f.y.fi.:<; <d " :'.. . ,
- A:i;!; ; : ;;
. 1 f,! f:nq.':?,
i
!. !!;h.. ;
.s . ucus :, 't.'
M!.::/ ':5..,I4?O'. .- .<!![j!I; ..
- encrustefred i!!!! 7d[ ii!!'I) algae,.I)
, z. e.
- !. s
- .m.mm :...
,.i
- . 9
' 4. ,, 4..i ., ~ijw! j
~
Nedeg..y q;p!;!:
- a.. .a.
w
.; if,. ,c..
3 .. v r " % y!
deramium
~! , . .m i i py : Y Chandrus racila ~ MM M
>,<..h.3.::.; f.l? *
.,. jj :.
. a,. i-a;i iat .>
s: :.;:::- .
- j. ::!:: ' -
9.i i
! c -.
iM
-. n, Enteromorphe }.;1
' . :0:::.:hl.!.\.:.< *.%
1
.. .. ..;. 5,e,.
! ,
- a c4. .= . j i- .4 je e,
- *N!,~.:,:[. :, . !a e ~ :.g a
v.
I, i efeli l I!.! . N
, 9i .
.:: i t\
g hil
.m
, l
.g;: . I i
ir2. d. ,'
, w .
Submerged Jetty l hffluen[dai[at I i i . . i a i i i i 60 50 40 30 20 10 0 10 20 30 NORTH METERS SOUTH Area Legend Denuded Sparse Stunted Control Jetty Boulder li
" ::!:/.Fi:
(T
- .::i:.:: ..
Figure 7. Denuded, Sparse, and Stunted Chondrus Zones Observed in December 1997.
I1
1 l March 1997 Ts Jure 1997 4" Ts N N
'm .
k'.'r !
cm sperne
... c,_ . . ii .
. . ,;i,;
'"hj , s!! he .f m.,, '
==
. :'5& !Y'N:. '
seweed -
i' i C"" ""-
ii u : m!! .:i l !
I'ii s n-!! :
- ii
- :'.!-l one. .
osaa 9 :n;y < e==== l/
. ., c/ . f.* . :#";. .at:!:!: :-
- ; s,,,,, 'lg s*** i.,.
- !"!'!:i.
?? ** t Cwow ;@'""=* -
r.*
i c-== . .
. ! !'jj'!
a!j
",fB.:p) h6'[ , ::.:M', ,'*
.j! '
- jh l f'k,y x' g:~ ;i ,
5/b,l+ !
- p.%ji'- l : :. '!
l l'
t:e..: w)i . w:!!
. !!;i ' C I
- . a.
y ,m l
, - .;w . -
, :n..
~ ;.:.
l . 'l.:
D p
'., u t m ===
GSl:
i:;::. v h..- - -'*==*c*.h
- l:l$.
- i
- gjgji:' -
Y*l :s,
-' w l .t.- ; ,; y c.. o.Ch.u.e. z,a,, i ,
y ~j- o.Cho,w. z.,e
^O, ,:i: !:
!$i.: ' . ' .
Ig itsn a l
!i.n;!;qi aatl
"$! jf !
nr y!laa- L a,l .
- liiij'i !jMy ? !;.;j:.;.
',,'- i !
- u. "'*- 1gm:::: wi $
! !. !!g)!i!!!; 'I . ; !!
' ij '
'i!I i
11'.!:!
llU'i!!LM i M.: .
N~~ !S .g!
j l
, g
!!Il
_ _ l _
.: .e . .:.n
- $ s.eme,,e w f :ai s.em vw a.n, i.!!!
!MM l Md 4 - A
,. . ,a h L j 4 s A .'. i ,a A k am wms soum ;
am w ens soum m.e t o ume s com % so , +=etev.= o=== sw= se-em come u m, e.,m., i l
pali;d;iii;;m n'h.**j ver l
g di gj:';;t Mid;;; *%I
.v A..'_
^'l*
, l September 1997 TN December 1997 Ts i N y seweed . , ,
Chomuue
- ., 59
.. :: !it m- .,,.
E cm,
'.:~i.+e. :%:] - n;.: L. l
- - .:/ s. .
.. . .. s m:!i![r /.W '/.,
saarne llE '$'b'" s'! Y' N *. ,
.l-l v:_m..h . c5 W
/ *t::~A...:. y- &' y;lo i:'de isa@:! ;! .*.: s-= -
C~~- .:
~, ; u C""'*"8 i' . , **
chormsw i
. t I
E.d , l i l l Ii!'.
a gd i i ,
.! f.f. e. i i- . !l*.' - *. ii a'
i -
! .h flji
!..- !!!;: /*.
- ,# tjf!!!!h.i
, . , M "- **.
,- A,al'/:< .. i :!!>w!p !.:i;:
l -
- r,4,,,.
4 .. i "'.: 6' : .
L. gig ,; yA Chorean 9 : ; t l'/- :9m .
{
'I ! : e son ne & Gera > Dmsos. Zone i:
(i.t4,.i; /,%.
t
.L : == sawsw a ::!: .
spa, . ." :i:. lap: } C#wous l' , j{ :i;i '
Causwee ' *
."***j '
.. ? 4ti
- /(s'/*.' '~#
i!! (!!l!f94
! ! ' " %'6'ei g- ti 34 3
- ,1,g ; ..,,
- g . ;
Demmim Zwie m,,phjjg 8,,ww
.,,,w j j j:j ,
. .,t,lp ,j ;;g, .,
16, .: .
r :jg:! .
- ;j[:e ! '!! [ ji;
"}!!!t?!;.
e i di Wi .
4.:
- ; in: _.
i q .;
, .n. I! '
i :
s -w
$l!
8- Jeny wd,;! ! w .; si
. l
~ =. n ,o. ,,. ,. ,. . . .
~~ .,m m,m = " ~ =. = . , ,. ,.
am wm, m,m is mee teseru, Denisised aper m ,,m.d carire Jessy EE M'"S S -
h em {3] gg g Figure 8. Results of the 1997 Qualitative Transect Surveys of the PNPS Acute Impact Zone off the Discharge Canal taken in March, June, September, and December 1997. l; I
12
l l
l while the totally affected area was only slightly larger than the historical maximum recorded in April 1983.
The denuded area (1662 m2) in March was slightly smaller (11 %) than in the spring 1996 survey but was i still larger (26%) than the previous spring high of 1321 m2 seen in March 1991 (Table 1). The total affected i 2
area (2092 m ) was one-third less than in the previous December survey (3111 m2 ),14% less than in spring 1996, and only 3% larger than the historical spring baseline obtained in April 1983. The denuded region extended 100 m offshore along the CTL and, as often seen before, was asymmetrically distributed with 62% of the denuded area north of the line. Other algae present included: the warm-water indicator, Gracilaria, near the discharge canal; a cold-water indicator, the kelp Laminaria, further offshore; as well
\ as Phyllophora spp.. Cystoclonium, Corallina, Ulva, and Fucus. Only a few blue mussels, Afytilus edulis, 1
l were seen. Other invertebrates included: a few starfish, Asteriasforbcsit periwinkle snails, Littorina littoreat and swarms of unidentified amphipods. No fish, crabs, or lobsters were seen.
2.2.2 JUNE 1997 TRANSECT SURVEY Results of the divers' survey for June 22,1997 are mapped in Figure 5. The Chondrus denuded and totally affected areas were much larger than in prior summer surveys. The denuded zone (2505 m 2) extended 130 m (30 m farther than in April) offshore along the CTL, was 51% :arger than three months earlier in March ,14% larger than in June 1996, and 37% larger than the June 1990 summer historical baseline of I 835 m2(Table 1). The sparse and stunted Chondrus zone (1467 m 2) had more than tripled in size since March 1997 and was even larger than it had been the previous June. The total affected area (3972 m 2) was 90%
larger than in March 1997 and 86% larger than the June 1990 historical summer baseline (2135 m2 ). The sparse and stunted Chondrus zones contributed heavily to the asymmetrical distribution of the affected area.
Gracilaria and Enteromorpha dominated the flora at the mouth of the effluent canal. Neither Laminaria nor Phyllophora spp. were noted, but Corallina and Ulva were seen. A dense array ofjuvenile blue mussels, 3-20 mm in length, similar to that seen in prior June surveys, was present from the 40-m mark seaward. The region with 100% mussel coverage (outlined by M's in Figure 5) was very nearly coincident with the totally affected Chondrus area. Dense aggregations of starfish, Asteriasforbesi, a mussel predator, were seen.
Other invertebrates present included: rock crabs (Cancer sp.), green crabs (Carcirms maenus), one horseshoe crab (Limulus polyphemus), and northern lobster (Homarus americanus). Striped bass ( Aforone saxatilis),
cunner (Tautogolabrus adspersus), bluefish (Pomatomus saltatrix), and winter f1ounder (Pleuronectes americanus) were seen.
13 l
i j
2.2.3 SEPTEMBER 1997 TRANSECT SURVEY Figure 6 shows the results of the transect survey conducted on September 17, 1997. Both the denuded and sparse Chondrus areas extended laterally to the north an unprecedented distance beyond the 30-m distance traditionally surveyed, and even beyond the 45-m distance more recently surveyed by the divers. Consequently, the reported sizes of the Chondrut denuded (>3587 m2) and the totally afTected zones
(>4364 m 2) are minimum estimates. The area of the denuded zone was much larger than in June 1997, a not unusual event, much larger than in last year's fall survey, and >76% larger than the historical baseline (2043 m2) of October 1995 (Table 1). The denuded zone extended seaward just beyond the 115-m mark on the CTL. The greatest lateral extent of the denuded zone was 45 m to the north of the CTL between the 60-m and 80-m marks. As has been seen often in the past, the denuded zone was asymmetrical around the CTL, with 74% of the area north of the line.
The areas of the sparse and stunted Chondrus zones totaled more than 777 m2 in September, covering an area larger than existed last fall (636 m2). The sparse Chondnes area extended far beyond the 45-m distance surveyed by the divers on all northern transect lines closer to the effluent canal than the 90-m mark along the CTL. The total affected area in September was >4364 m2 which was more than 10%
larger than measured in June 1997 and > 86% larger than the historical baseline (2348 m2) established in October 1995.
Most of the algal plants that had been covered with mussels in June were beginning to recover, but Chondrus appeared colorful and healthy only beyond the impact zone. As usual, the algal species seen within the discharge canal were Gracilaria spp., and Cystoclonium purpureum. The warm water alga, Gracilaria spp., was the dominant macroalga species occurring close to the CTL out to the 50-m mark.
Rockweed (Fucus sp.) and the encrusting red alga, Corallina, were seen occasionally between the 40-m l and 60-m marks,10 to 20 m north of the transect. Phyllophora spp. was not noted.
Mussels, averaging 1.5 cm in length, had declined in abundance since the June survey; they remained abundant only in a narrow (5 m wide) strip 5 to 10 m south of the CTL, extending from the 40-m to the 90-m mark on the CTL (outlined by M's in Figure 6). The divers also saw starfish (fewer than in June), periwinkle snails, and unidentified anemones. An unusual observation was the presence of several dead (not exuviae) crabs and lobsters; the divers did not recall ever seeing dead crabs or lobsters before in the survey area. Only one demersal fish, a winter flounder, was observed during the survey.
14
2.2.4 DECEMBER 1997 TRANSECT SURVEY The results of the 1997 winter dive, performed on December 17,1997, are mapped in Figure 7.
The area (2241 m2) of the denuded zone was > 38% smaller than it had been in September but was still larger (+14%) than the historical baseline established in February 1996 (Table 1). The denuded zone extended seaward just beyond the 90-m mark on the CTL. The shape of the denuded zone was highly asymmetrical around the CTL with 67% of the area north of the line. The greatest lateral extent of the denuded zone observed was at the 70-m mark, where it projected 42 m north of the CTL.
The area inhabited by sparse and stmited Chondrus (1271 m2) was 61 % larger than that measured in September 1997 (777 m2), surrounded the denuded zone, and was arranged asymmetrically around the CTL with most (89%) of the area north of the line, where it was up to 20-m thick from the 50-m to 70-m marks on the CTL. The total affected area (3512 m2) in December 1997 was larger (13 %) than measured in the previous winter survey, and much larger (51 %) than the historical baseline measured in February 1996 (Table 1). But, as usually seen in conjunction with declining water temperatures, the affected areas were smaller (> 20%) than those measured in September 1997. The maximal extent of the totally affected area out along the CTL was approximately equal to the historical baseline, that is, near the 100-m mark on the CTL. The maximal width of the affected zone (70 m) was greater than seen in any prior survey and was 67% over the historical winter baseline of 42 m.
During this survey, the divers observed that " normal" Chondrus plants looked different on the two i, ides of the CTL. Normal Chondrus on the south side was old growth; the plants had been there for months as evidenced by the epifaunal assemblage (Spirorbis sp. and barnacles) growing on the plants.
Normal plants north of the CTL were clearly very new and growing rapidly as shown by the total absence of epiphytes and the presence of bright pink meristems (growing margins). The effect of the discharge, felt by the Chondrus plants during the hot summer and fall months, was that the plants died back to their holdfasts; however, by December, lower water temperatures allowed new growth to appear from the old holdfasts. Other algae seen included: Enteromorpha, Gracilaria, a coralline red alga, Ceramium sp., and an occasional Fucus sp. The red alga, Phyllophora spp., was not noted by the divers. Snails, Littorina littorea, were extremely abundant throughout the entire area surveyed. The very few blue mussels present were seen only at the more distal portion of the survey area. Other invertebrates included a rock crab, a spider crab, Libinia sp., and an encrusting colonial anemone, Zoanthus sp. Fish seen included a juvenile sculpin, Myoxocephalus sp., and a winter flounder being eaten by a large green crab.
15
2.3 DISCUSSION i The configuration of the Chondrus crispus denuded zone, that can extend seaward even farther than 100 m beyond the discharge canal, is readily apparent to SCUBA divers and easily mapped from the qualitative transect survey. Stunted and sparse zones are sometimes less obvious, but the sparse zones observed in 1997 were delineated without difficulty. For the spring survey, the sizes of the denuded and totally affected zones observed in March 1997 were exceeded previously only in the year before, in April 1996. For the June, September, and December 1997 surveys, the areas of the denuded and total affected zones were larger than any of those seen previously (1983,1985,1989-1996), when the power plant was in full or nearly full operation. In June 1997, an extremely dense mussel mat, similar to those seen every June since 1990, with the exception of 1991, was present. The areas of the denuded and totally affected zones were again greater in June than in March, the usual trend when early summer growth of Chondrus is adversely affected by high mussel settlement, and larger than in any previous June survey. In September 1997, although the mussel mat was much reduced in size, the denuded and totally affected Chondrus areas were still larger than ever had been observed before, for any time of year. In fact, although the divers surveyed up to 45 m north of the CTL, this was not far enough to locate sparse or normal Chondrus on many of the transect lines surveyed. This unprecedented increase in size of the Chondrus affected zones possibly was caused by some combination of siltation and/or light shading from turbidity caused by the dredging operation that occurred near the cooling water intake from mid-June until the end of August. In December 1997, the area of the Chondrus denuded zone was smaller than in the fall but was the largest so far recorded for any winter survey since the quarterly surveys began in 1983'.
The lower annual capacity factor at PNPS for 1997 (73.4%), less than in 1995 or 1996, was not reflected in any size reduction of the Chondrus affected zones. For 1997, it appears that ?.he summer dredging operation exacerbated the usual effects of high plant capacity, warm summer water temperatures, ,
l and extremely dense settlement of mussel larvae in late spring, on the algal community, resulting in the j largest Chondrus denuded and totally affected zones ever seen.
l l
l 16 I 1
i
)
3.0 IMPACT OF EFFLUENT DISCIIARGE AT PNPS ON ALGAL DISTRIBUTION l
3.1 BACKGROUND
liistorically, operational conditions at the PNPS have provided opportunities to assess long-term
, trends associated with impacts on the benthic community. Plant operations have included consecutive years l of high operation as well as times when there were complete shutdowns, sometimes for prolonged periods.
The longest outage in the history of the plant began in April 1986 and continued until March 1989. During 1
l this period the benthic community associated with the effluent canal and nearby areas immediately offshore experienced reduced current velocity as the use of circulating pumps was restricted to one or none (Figure 1
l 9). In addition, the discharge water remained at ambient temperature. As a consequence, the benthic
]
community normally affected by these effluent parameters recovered, so that by 1988 there was essentially no difference between the control stations and the areas near the discharge canal. I Studies conducted after the power plant resumed electrical generation at full operating capacity, with the consequent thermal discharge and consistent use of one or both circulating pumps, assessed the l impact of plant operation on a benthic environment that had returned to near ambient conditions.
Quantitative faunal and algal monitoring studies, and qualitative transect surveys were conducted through 1991. In 1992, community studies of the benthic algae and fauna were discontinued. From 1992 through 1997, the monitoring program has consisted of seasonal qualitative surveys of the discharge area.
PNPS operated at a moderately high capacity in 1997. Figure 9 shows the monthly maximum dependable capacity (MDC) factor and circulating water pump operation of PNPS since 1983. The percent MDC is a measure of reactor output and approximates thermal loading to the marine environment. A maximum MDC value of 100% equates to the highest allowable change in ambient temperature for effluent water discharged to Cape Cod Bay (18* CAT). In 1997, the monthly maximum dependable capacity factor was greater than 92% for 7 months and between 68% and 70% for 2 months. For the six months of May through October, the monthly capacity factor at PNPS was above 95% (mean = 97.3%), although the j annual capacity factor for 1997 was 73.4%, less than in 1995 and 1996.
I 3.2 QUALITATIVE TRANSECT SURVEYS: 1983-1997-l Results of the qualitative transect surveys from 1983 through 1997 are summarized in Figure 10.
The total acute impacted area (denuded, sparse,' and stunted), the area of the denuded zone only, and the monthly PNPS capacity factor (MDC) are plotted. The difference between the denuded and total acute impact zones represents the area of the sparse and stunted zones.
17
i l\lilli 1\1<
7rSEu. xME" U ego. $ 8E t 1 0 =
0 0 0 0 0 0 0 0 0 0 1 9 8 7 6 5 4 3 2 1 0 %
0
, D h I S
J 5
t a
E s I
M p D i h S J
n u
p 2
M D =
h S J
0 0
"M 1 D t h E B -
S J
a s
r a
- I M b D k h E S
J l
b9
(
c a 7.
M 9 y1 D
h t
I i r S i v e t b J l c
M n
o Aecm D it pe h
I 1
1 a r mD S e ui B p Pl E J g M O gn D
p nor r
a e
h ;
S h J
t n 'P m
u l i
t ah u T Y o c3 MM I
I r8 E i D C19 h B l
E S
J D C dd no ai r B
5 M M ) e 5
D sP e
h 5 5
5 S
J l ine l
d h
t
- r M iof o l
h m D S
J t s
( d r e ot t o
cl E M aP I
D F) s h E B
M E
N S
J t
y p ic n i E M a u i
p p a
h D S
J C0 S=
P h1 M N%
g P 0 h n u
H D
S J
l ht ya t p
g g M nm o u h- E '
D Mp S
J M
D .
9 0 0 0 0 5 e 1 r u
g n
$ 8ilao OE2 cc_3s i
i o m$o.
i F
} llIl\1ll',ll 1:1l
4 ; 1lljil Iil1lla
- SM gkg U eh[8 S7 0 P8 9 0
1 0
8 0
6 0
4 0
2 0 N1 P
ye r
[gm ~
l r . 8 D h b em t
S n
~
8 .:-l
" J ot p
Me M
[q ' ?;i, D eS _
ht -
S m
? :'.':js J ht or if a- M w e D dd e , S t e a l
'8 d J t
o mtn .
l 8 M P er se
[ ' D S
l C
l an wa as e eb
_ / ."-
J D Ct ner t '.
M M t ew
[f g4 D nms e
u e e S r n a
-l.
. J muo E a z s
- n - .M s ed S
,S D u r P mten
~
_m
- , J d
n Na P et u m,, . 1 ,-
, M h o
e ar s p j' C h d
[ -
,\
D S
f d
o t
fNao on J
e y .de t
, M d 7d
- , D
- ht u
n e
in9 u i
i1 c9 n e r
a t
, S n D V h d e J o egf Y MM h u o
- C t os
[ nr n D
i S ih o
- J a sTi e t a
~
M e r
n3 o8 r c
A Z9 1 a
[ _
D S
J d
t c
e d t
ed e c od m
~
e ei re M f f f f e v D A api i t
~
~
S la t
ye n i J o l l
at e hf T
n ', o rd
- M -
t D
Tf o es
- a- S m d) u J
nCa c
'-m t l
aDe
~
's , D M dMb d(e 98 S ur9 no1
' J et c l *- M Da en D ht eF u S y J t h m,i, h f i J oc g m M a e a pou 8l D r a hr
[ m!:g_.
- m. -
S J
ACt
- M 8 D 0 1
0 0 0 0 0 0 0 0 0 0 0 e 0 0 0 0 0 r 5 4 3 2 1 u g
i F
'~E 2<
A lag in recov'ery time in the acute impact zone during and following the 1984 PNPS power outage was reported in Semi-Annual Report No. 27 (BECo,1986). Evidence of this slow recovery included a decrease in the area of the total acute impact zone that began in mid 1984 (5 months after the cessation of power plant operations) and continued through mid-1985. Between December 1984 and December 1985, the total affected area was the smallest recorded between 1983 and 1986, indicating a delay in recovery in response to the absence of thermal discharge and reduced circulating water pump operation in 1984.
This delay phenomenon also held true when the situation was reversed, so that the size of the acute impact zone began to increase only 6 to 9 months (September to December 1985) after the resumption of thermal effluent discharge and normal circulating water pump operation. These results confirmed a delay of 6-9 months between the causal factors (cessation or resumption of thermal effluent discharge and normal pump operation) and associated responses (decrease or increase in size of the acute impact zone). In 1987, in response to the 1986-1989 outage, increased recolonization of the denuded and stunted zones by Chondrus made zone boundaries difficult to distinguish (no areal differences could be discerned from September 1987 through June 1989). As in summer 1984, the large size reduction of the denuded zone between December 1986 and June 1987 was primarily the result of the shutdown of the circulating water pumps in late February 1987 that continued into the summer (BECo,1988). Apparently, water current scouring is a greater stress to algal colonization than elevated water temperature. Generally, scouring denudes the substratum, whereas elevated temperature results in stunted growth (Bridges and Anderson,1984).
In 1988, low circulating water pump activity caused few scouring effects. The 1988 transect surveys showed such an increase in recolonization of formerly denuded and stunted zones by Chondrus, because of the continuing outage, that divers could not detect zonal boundaries or make area measurements. In March and June 1989, divers were still unable to detect boundaries of denuded or stunted zones (BECo,1990). In September and December 1989, presumably in response to increased PNPS operations with resultant thermal effects and scouring of the acute impact zone, boundaries began to be redefined and area measurements were made of the total impact zone.
During 1990, boundaries between the stunted and denuded zones became even more clearly defined and areal measurements of bodi zones were made. The denuded and total impact zones in June 1990 were the largest measured since 1983 (BECo,1991). The dramatic increase in total affected area that occurred between April and June 1990 had not been seen before. The typical pattern seen prior to 1990 was that during spring, with warmer temperatures and increased sunlight, algal growth flourishes, and the impact area declines even in years when the power plant is operating at high capacity. The pattern in 1990 ;
I appeared anomalous until, more recently, a correlation was made between the appearance of enormous i I
20 1 I
l
numbers of juvenile mussels and the occurrence of large denuded and total affected zones. The divers l noted remarkable numbers ofjuvenile mussels during the June 1990 dive. Thus, it would appear that the l
l large affected zones result, at least partly, from damage suffered by the Chondrus plants due to the massive i
settlement of mussels, in 1991, the boundaries of the acute impact zone remained well-defined, except that in June there was no true stunted zone but only an area described by the divers as " sparse", that is, where the algal plants grew normally but were thinly distributed. From March to June, the total affected area and the Chondrus denuded zone decreased in area, a return to the typical pattern seen before 1990 (BECo,1992).
This decrease in area continued through the October survey, perhaps aided by the power plant outage from May into August. There was a slight increase in the affected area in December.
During 1992, the divers were unable to discern a Chondrus stunted region. Except for June, they noted zones containing normal but sparsely distributed Chondrus plants. An enormous set of mussels that had reached 0.5 cm in length by June, totally obliterated the boundary between the denuded and sparse l areas. Parallel to results seen in 1990, the areas of the denuded and total acute impact zones in June 1992 were larger than any seen (except for 1990) since 1983, and the dramatic increase in total affected area that occurred between April and June 1990 occurred once again in 1992. Thus, the pattern seen in 1990 can no longer be considered anomalous but may be related to oceanographic conditions that lead to a large settlement of mussel larvae and consequent damage to Chondrus plants (BECo,1993).
In 1993, the June mussel set that hampers Chondrus growth was not as dense as those that occurred in 1990 and 1992, so that the denuded zone was smaller in June than it had been in April, the opposite of the situation seen in 1990 and 1992 (BECo,1994). The area of the denuded zone in September was slightly larger than it had been in September of 1990 and 1992, but the denuded zone in December was much larger than in previous years. In addition, the total affected area in December was the largest seen since 1983, rivaling the areas measured in the summers of 1990 and 1992; this may be partly due to the very early winter date (Dec. 2) of the survey and partly to damage imposed by a heavy infestation of the encrusting bryozoan, Membranipora membranacea.
In 1994, the denuded and total affected Chondrus areas in all four seasons were similar in size to those found during prior surveys at times of full or nearly full power plant operation (BECo,1995). The dense mussel settlement seen in June obscured the boundary between the denuded and sparse / stunted regions and damage caused by the mussels to the Chondrus plants contributed to the enlargement of both Chondrus zones between the April and June surveys. The three-month fall power plant outage (September 21
through November) appeared to have had no effect on the size of either the denuded or total affected Chondrus zones.
In 1995, the sizes of the denuded and total affected Chondrus areas were within the ranges seen in earlier surveys only for the early May and late June surveys (BECo,1996). The impacted areas l
I measured in October 1995 and February 1996 were much larger than those measured during any earlier fall and winter surveys and most closely approximated the impacted areas seen in September and December l 1993. The two-month (April /May) spring power outage appeared to have no effect on the size of the l
Chondrus affected areas seen in May or June. However, the high plant operating capacity in effect from June 1995 through February 1996, in conjunction with a high mussel set in June, may have contributed to the largest fall and winter denuded and totally affected Chondrus zones seen since the current monitoring program began in 1983.
In 1996, the sizes of the denuded and totally affected Chondrus areas continued to increase over i
the historical baseline measurements (1983 through February 1996) for the first three surveys. In December, the denuded zone declined in size to less than the winter historical baseline measurement but was still the second largest denuded zone ever observed in winter (BECo,1997). The large Chondrus denuded and totally affected zones seen in each survey since October 1995 may be due to a combination of the high plant capacity that was in effect for the 18 months starting in July 1995 (mean = 92.6%), high summer water temperatures, and extremely dense settlement by mussel larvae in late spring that totally covered and may damage the algal plants.
In 1997, the sizes of the denuded and totally affected Chondrus zones were again larger than
- historical baseline measurements. In March 1997, the impacted areas were the second largest ever measured in spring. For the remaining three seasons, the areas of the denuded and totally affected zones were the largest ever seen for the corresponding season. The sizes of the denuded and totally affected zones in 1997 were extraordinarily large, larger than in 1996 for three surveys, and appeared not to track the reduction in the annual plant capacity factor from 90.5% in 1996 to 73.4 % in 1997 that resulted from a two-month spring power outage. Turbidity from the extensive dredging operation that took place from mid-June until the end of August, in conjunction with dense se
- tlement by juvenile mussels that occurred sometime between March 28 and June 22, high summer water temperatures, and a moderately high 1997 I
power plant capacity, caused many Chondrus plants to die back to their holdfasts, yielding the very large l affected Chondrus zones seen in 1997.
l l
22
4.0 CONCLUSION
S
- The denuded and totally affected Chondrus areas of the acutely impacted region off PNPS for all four quarterly 1997 surveys were larger than historical baseline values (for each season, the largest area measured between 1983 and the 1995 survey seasons).
e The areas of the denuded and total affected zones were greater in June than in March 1997, the usual trend observed when early summer Chondrus growth is adversely affected by high mussel settlement.
e The denuded and totally affected areas in September 1997 were the largest ever recorded for any season and extended far beyond the region of dense mussel settlement observed during the June survey. Turbidity from the extensive dredging operation that took place from mid-June until the end of August possibly contributed to this unprecedented increase in affected Chondrus area.
e Dredging, in combination with a moderately high 1997 power plant capacity, extremely dense mussel settlement in June, and high ambient summer water j temperatures, may have contributed to producing the largest denuded and totally affected Chondrus zones seen since the present monitoring program began in 1983.
23
5.0 LITERATURE CITED Boston Edison Co.1983. Marine ecology studies related to the operation of Pilgrim Station. Semi-Annual Report No. 22. Boston, M A.
Boston Edison Co.1986. Marine ecology studies related to the operation of Pilgrim Station. Semi-Annual Report No. 27. Boston, M A.
Boston Edison Co.1988. Marine ecology studies related to the operation of Pilgrim Station. Semi-Annual Report No. 31. Boston, MA.
Boston Edison Co.1990. Marine ecology studies related to the operation of Pilgrim Station. Semi-Annual Report No. 35. Boston, MA.
Boston Edison Co.1991. Marine ecology studies related to the operation of Pilgrim Station. Semi-Annual Report No. 37. Boston, MA.
Boston Edison Co.1992. Marine ecology studies related to the operation of Pilgrim Station. Semi.
Annual Report No. 39. Boston, MA.
Boston Edison Co.1993. Marine ecology studies related to the operation of Pilgrim Station. Semi-Annual Report No. 41. Boston, MA.
Boston Edison Co.1994. Marine ecology studies related to the operation of Pilgrim Station. Semi-Annual Report No. 43. Boston, MA.
Boston Edison Co.1995. Marine ecology studies related to the operation of Pilgrim Station. Semi-Annual Report No. 45. Boston, MA.
Boston Edison Co.1996. Marine ecology studies related to the operation of Pilgrim station. Semi-Annual Report No. 47. Boston, M A.
24 i
l l
r l
Boston Edison Co.1997. Marine ecology studies related to the operation of Pilgrim station. Semi-
- l Annual Report No. 49. Boston, MA.
f-Bridges, W.L. and R.D. Anderson.1984. A brief survey of Pilgrim Nuclear Power Plant ' effects upon the marine aquatic environment, p. 263-271. In: J. D. Davis and D. Merriman (eds.)
i Observations on the ecology and biology of western Cape Cod Bay, Massachusetts,289 i
pp. Springer-Verlag. (Lecture Notes on Coastal and Estuarine Studies, Vol. I1).
- Taxon.1982. Benthic studies in the vicinity of Pilgrim Station. In
- Marine Ecology Studies Related t
j to Operation of Pilgrim Station. Semi-Annual Report No.19.
l t
l l
i I
!' l l
i 25 !
L l'
APPENDIX A
' Quality Control (QC) Protocol for Qualitative Transect Surveys at PNPS Outfall Area l
1 Field Oncration Planning Field equipment is organized by the scientist in charge of dive operations; for 1998, the chief diver will be Mr. Erich liorgan of the Woods 11 ole Oceanographic Institution. Mr. liorgan has been a diver or chief diver on four quarterly surveys at the PNPS outfall site since April 1996. The survey equipment includes a boat and associated safety equipment; anchor and line; buoy and diver safety line; SCUB A gear, including a collecting bag; 100-ft kevlar line to be deployed across the mouth of the discharge canal; grapnel to aid in tying off the kevlar line tojetty boulders; the weighted 100-m central transect line (CTL), marked at 10-m intervals; two 30-m measuring tapes; compass; clipboard; data sheets on plasticized paper; two #1 pencils.
Every attempt will be made to perform the dives as scheduled for March, June, September, and
- December. Windows of opportunity, considering times of high tide (less current for the divers to contend with) and other commitments for both boats and personnel, will be blocked out in advance of each scheduled month. Enough leeway will be planned to allow some flexibility for bad weather days. A trainee diver will be included for two field days to decrease the possibilities for postponements caused by illness or conflicting schedules during the appropriate months.
2 Pre- and Post-dive Briefings The chief diver and ENSR data manager, Isabelle Williams, will hold pre- and post-dive briefings.
The pre-dive briefing (may be made by telephone) will be the opportunity for determining the dive schedule, for reviewing data collection, and for informing the dive team whether or not any additional observations are requested. At this time, emphasis will be placed on the importance of the divers exploring the limits, and defining them, of the entire affected area so that a comprehensive survey map can be produced. The post-dive briefing (in person) will give tl e chief diver the opportunity to tell the data manager his immediate impressions about the region surveyed and whether any problems were encountered that need to be corrected before the next dive.
26 l
I i
3 Data Collection A diver swimming perpendicularly away from the CTL, along the measuring line, records the distance away from the CTL line that changes in algal cover occur, from denuded to sparse and/or stunted l . Chondrus areas, and from sparse and/or stunted Chondrus to normal-looking Chondrus. Positions of other
, algal species, especially Gracilaria, a warm-water indicator, and kelp (Laminaria), a cold water indicator, t
are noted. Positions of animals, including mussels, starfish, crabs, and fish, and any unusual activities are also indicated.
For 1998 and beyond, detailed observations will be made of Chondrus, including notes on robustness, color, occurrence of epiphytes, and qualitative descriptions of density and height. The divers will look for the presence of Phyllophora, the second dominant algal species in this communit), throughout the survey area; if necessary, they wil! collect an algal sample from the normal Chondrus zone for examination in the laboratory. Particular attention will be paid to the boundaries of the high-density mussel set that appears to often occur in late spring or early summer.
A sample blank data sheet is shown. A separate sheet is used for the north and south sides of the CTL. As the diver swims away from the CTL, distances and notes are recorded on the data sheet from left to right. For ease in working in an underwater environment algal cover is coded as indicated on the data l
sheet: 1 - denuded; 2 - stunted; 3 - sparse; 4 - normal. Codes for mussel cover are M1 - very dense; M2 -
. separated clumps; M3 - absent.
4 Data validation l
The diver recording data during the field survey is responsible for reviewing his work at the end of l the survey to ensure that the data are complete and accurate. The chief diver will submit to the data manager l
l the original field notes and a survey report, previously reviewed for accuracy and completeness by other members of the dive team, that includes the data on the total extent of the denuded and stunted Chondrus l
- zones as well as a general description of the area surveyed, including notes on flora and fauna observed. The data manager is responsible for reconciling data in the submitted field report to those recorded on the original data sheets. The data manager will discuss any questions that may arise with the chief diver. The data manager is responsible for constructing maps based on the survey data and for calculating the total areal extent of the denuded and totally affected Chondrus regions. All calculations performed by hand are checked for accuracy. The data manager is responsible for proof-reading the final computer-generated maps against the original maps for accuracy. All reports generated by the data manager will be reviewed by the ENSR Project Manager, Dr. James Blake.
27 l
l l
I 5 Observation The data manager will plan to accompany the divers on one or two field trips in 1998. She will be be on hand to accept any samples collected during those dives and to hear immediately the impressions of all divers about the conditions of the outfall area, as well as ensure that the entire affected area has been surveyed.
6 Meetings The project and/or data manager will attend full Administrative-Technical Committee and Benthic Subcommittee meetings when appropriate. The chiefdiver, Erich Horgan, will attend the June 1998 Benthic subcommittee meeting. His will help ensure communication between ENSR, the field team, and the A-T Committee so that the quality of the benthic survey will be maintained as guided by the Committee.
l l
l l
l 28
[ _ _ _ _ - _ _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ .
Date: Divers Down @:
Dind: Divers Up @:
Visibility:
CTL(m) NORTli/S 3UTil l
CHONDRUS 30 1 DENUDED 2 STUNTED l 3 SPARSE
'40 4 NORMAL l MUSSELS i50 M1 V. DENSE l M2 CLUMPS M3 ABSENT l60
- 70 l
80 l90 NORMAL CHONDRUS 100 ROBUSTNESS l COLOR EPIPHYTES
>100 HEIGHT COLOR' Qualitative Transect Survey Field Data Sheet.' ,
l 29 I
I i
l I
t ICHTHYOPLANKTON ENTRAINMENT MONITORING AT PILGRIM NUCLEAR POWER STATION JANUARY - DECEMBER 1997 l Volume 1 of 2 l (Monitoring) l l
l l
! Submitted to Boston Edison Company l Boston, Massachusetts l
l by Marine Research, Inc.
Falmouth, Massachusetts i
1
! l l April 1,1998 l
1 i
l I
I TABLE OF CONTENTS SECTION PAGE I EXECUTIVE
SUMMARY
l II INTRODUCTION 3 III METHODS AND MATERIALS 4 l IV RESULTS AND DISCUSSION A. Ichthyoplankton Entrained - 1997 12 B. Multi-year Ichthyoplankton Comparisons 17 C. Mesh Extrusion 23 D. Lobster Larvae Entrained 25 V HIGHLIGHTS 27 VI LITERATURE CITED 28 i
l APPENDICES A and B (available upon request) l l LIST OF FIGURES
' FIGURE PAGE 1 Entrainment sampling station in PNPS discharge canal. 5 2 Dominant species of fish eggs and larvae found in PNPS ichthyoplankton samples by season. Percent of total and summed monthly means for all species are also shown. 14 3 Mean monthly densities per 100 m2 of water in the PNPS discharge canal for the eight numerically dominant egg species and total eggs,1997 (bold line). Solid lines encompassing shaded area show high and low values over the 1982-1996 period. 31 4 Mean monthly densities per 100 m' of water in the PNPS discharge canal for the thirteen numerically dominant larval species and total larvae,1997 (bold line). Solid lines encompassing shaded area show high and inw values over the 1982-1996 period. 37 i
j LIST OF TABLES TABLE PAGE I P1iPS ichthyoplankton entrainment notification levels for 1997 by species category and month. See text for details. , 9 2 Species of fish eggs (E) and larvae (L) obtained in ichthyoplw.kton collections from the Pilgrim Nuclear Power Station discharge canal, January. December 1997. 45 3 Species of fish eggs (E) and larvae (L) collected in the PNPS discharge canal,197S-1997. 47 4 Densities per 100 m 2_ of water for tautog/ cunner eggs taken with 0.333 ]
and 0.202-mm mesh netting, 1994-1997. 50 i l
5 Densities per 100 m' of water for stage I cunner larvae taken with 0.333 and 0.202-mm mesh netting, 1995-1997. 52 6 Densities per 100 m$ of water for stage 2 cunner larvae taken with 0.333 and 0.202-mm mesh netting, 1994-1997. 53 LIST OF APPENDICES APPENDIX A* Densities of fish eggs and larvae per 100 m* of water recorded in the PNPS discharge canal by species, date, and replicate, January-December 1997.
B* Geometric mean monthly densities and 95% confidence limits per 100 m$
of water for the dominant species of fish eggs and larvae entrained at PNPS, i January-December 1982-1997.
- Available upon request.
i l
11
SECTION I EXECUTIVE
SUMMARY
Sampling of entrained ichthyoplankton at PNPS followed the revised protocol initiated in -
April 1994 and utilized in 1995 and 1996. In January, February, and October through December three samples were taken every other week each month, weather permitting. From March through
. September single samples were taken three times every week in conjunction with the impingement
' monitoring study.
1 A total of 37 species were represented in the 1997 collections, just one below the 22-year mean of 38 species. Numerical dominants during winter, early spring included yellowtail flounder, fourbeard rockling, winter flounder, and American plaice eggs along with rock gunnel, sand lance, sculpin, and Atlantic herring larvae. During the late spring, early summer season numerical dominants included tautog/ cunner and mackerel eggs, and larvae of the cunner, winter flounder, mackerel, rockling, menhaden, and sand lance. During late summer, autumn, collections consisted primarily of .
1 windowpane, tautog/ cunner, rockling/ hake, and menhaden eggs as well as hake, cunner, menhaden, and Atlantic herring larvae.
Comparisons ofichthyoplankton densities over the 1975-1997 time series suggested that Atlantic mackerel eggs, windowpane eggs, and Atlantic herring larvae were entrained in relatively high numbers in 1997 consistent with overall trends in fish stocks. Larval winter flounder were also entrained in high numbers in 1997 although stock size estimates have varied without trend over the past eight years. Similar results were obtained for tautog and cunner eggs and larvae and, while stock size information is lacking for cunner, estimates for tautog indicate low stock abundance. Larval hake were also numerous in PNPS samples in 1997, nearly double the previous high. Like tautog, stock l size estimates suggest that abundance is relati"ly low. l Mesh comparison collections completed from 1994-1997 with 0.202 and 0.333-mm mesh nets suggested that cunner eggs are retained in higher numbers by the finer mesh although a statistically significant difference could not be demonstrated within each season. Ratios for 0.202/0.333 mesh )
averaged 1.10 in 1997,1.14 in 1996, and 1.24 for all four years pooled together. Mesh differences were also observed among yolk-sac cunner larvae (stage 1) and post yolk-sac, stage 2 larvae.
1
Pooling pairs of data from 1994 - 1997 showed a mean ratio of 1.16 for stage I and 1.28 for stage <
- 2. In both these cases differences could not be shown to be statistically significant.
No lobster larvae were collected in 1997 for the second straight year. The total dating back to 1974 remains at 13.
2
SECTION II INTRODUCTION This report summarizes results ofichthyoplankton entrainment sampling conducted at the Pilgrim Nuclear Power Station (PNPS) from January through December 1997 by Marine Research, Inc. (MRI) for Boston Edison Compar.y (BECo), under Purchase Order No. LSP007550, in compliance with environmental monitoring and reporting requirements of the PNPS NPDES Permit (U.S. Environmental Protection Agency and Massachusetts Department of Environmental Protection). As a result ofstudies completed in 1994, conversion from 0.333 to 0.202-mm mesh was initiated from late March through late May 1995 to improve retention of early-stage larval winter flounder; this enhancement continued in 1997. Extrusion ofyoung larval cunner was also a concern at PNPS based on data gathered from 1994 through 1996. Additional 0.333 and 0.202-mm mesh samples were therefore taken in June and July 1997 to improve that data base and results appear in this volume.
In an effort to condense the volume ofmuerial pre ented in this report, details ofinterest to some readers may have been omitted. Any questions or requests for additional information may be directed to Marine Research, Inc., Falmouth, Massachusetts, through BECO. .
3
1 l
l l
SECTION 111 i
METHODS AND MATERIAL _S Monitoring Entrainment sampling at PNPS begun in 1974 had historically been completed twice per month during January and February, October-December; weekly during March through September; in triplicate at low tide. Following a PNPS fisheries monitoring review workshop in early 1994, the sampling regime was modified beginning April 1994. In January, February, and October through December during two alternate weeks each month single samples were taken on three separate occasions. Beginning with March and continuing through September single samples were taken three times every week. During autumn and winter months when sampling frequency was reduced, sampling was postponed during onshore storms due to heavy detrital loads. The delayed sample was taken during the subsequent week; six samples were ultimately taken each month.
To minimize costs, sampling was linked to the impingement monitoring program so that collections were made Monday morning, Wednesday aflernoon, and Friday night regardless of tide (see impingement Section). All sampling was completed with a 60-cm diameter plankton net streamed from rigging mounted approximately 30 meters from the headwall of the discharge canal (Figure 1). Standard mesh was 0.333-mm except from late March through late May when 0.202-mm rnesh was employed to improve retention of early-stage larval winter flounder (Pleuronectes americanus). Sampling time in each case varied from 8 to 30 minutes depending on tide, higher tide requiring a longer interval due to lower discharge stream velocities. In most cases, a minimum quantity of 100 m' of water was sampled although at astronomical high tides it proved diflicult to collect this amount even with long sampling intervals since the net would not inflate in the low current velocity near high tide. Exact filtration volumes were calculated using a General Oceanics Model 2030R digital flowmeter mounted in the mouth of the net. Near times of high water a 2030 R2 rotor was employed to improve sensitivity at low velocities.
Entrainment sampling was not possible from mid-February through the first half of March 1997 since both circulating water pumps were out of service during a refueling outage. Sampling from mid-March through mid-April was completed with one of two circulating seawater pumps out of service as the outage continued. Sampling also was not possible on April 2, April 18, and June 2 4
' ij, 9h
- y
- s 3r, DISCHAROC CANAL
".. f
~ N bcto'.vgv
\
BRLOGC
,o.,,.
N yC IN T Agg A : d
' h* '
- A sin o
..$ 9h*
HCA0 WALL .o o.
. o-
.. \
UNIT 1 INTAKC !
D ..
PNPS 9 (CHTHYOPLANKTON STAT!ON vNn t '
i \
goo METERS !
Figure 1. Entrainment sampling station in PNPS discharge canal.
5
1 due to stormy seas. Sampling under such conditions results in such heavy detrital loads that processing the samples is all but impossible. (In the past when storm samples have been processed, ichthyoplankton has been unconunon.)
All samples were preserved in 10% Formalin-seawater solutions and returned to the laboratory for microscopic examination. A detailed description ofthe analytical procedures appears in MRI (1988). As in past years, larval winter flounder were enumerated in four developmental stages as follows:
Stage 1 - from hatching until the yolk sac is fully absorbed (2.3-2.8 mm TL).
Stage 2 - from the end of stage 1 until a loop or coil forms in the gut (2.6-4 mm TL).
Stage 3 - from the end of stage 2 until the left eye migrates past the midline of the head during transformation (3.5-8 mm TL).
Stage 4 - from the end of stage 3 onward (7.3-8.2 mm TL).
Similarly larval cunner (Tautogolabrus adpersus) were enumerated in three develop-mental stages:
Stage 1 - from hatching until the yolk sac is fully absorbed (1.6-2.6 mm TL).
Stage 2 - from the end of stage I until dorsal fin rays become visible (1.8-6.0 mm TL).
Stage 3 - from the end of stage 2 onward (6.5-14.0 mm TL).
Samples were examined in their entirety for larval American lobster (Homarus americanus).
When collected these were staged following Herrick (1911).
Notification Provisions When the Cape Cod Bay ichthyoplankton study was completed in 1976, provisions were added to the entrainment monitoring program to identify unusually high densities of fish eggs and larvae. Once identified and, if requested by regulatory personnel, additional sampling could be conducted to monitor the temporal and/or spatial extent ofthe unusual occurrence. An offshore array of stations was established which could be used to determine whether circumstances in the vicinity of Rocky Point, attributable to PNPS operation, were causing an abnormally large percentage of ichthyoplankton populations there to be entrained or, alternatively, whether high entrainment levels simply were a reflection ofunusually high population levels in Cape Cod Bay. The impact attributable to any large entrainment event would clearly be greater ifichthyoplankton densities were particularly high only close to the PNPS shoreline. In past years when high densities were identified, additional l
l 6
entrainment sampling was requested by regulatory personnel and the unusual density in most cases was found to be of short duration (<2 days). 'Vith the change in 1994 to Monday, Wednesday, Friday sampling the temporal extent of any unusual density can be more clearly discerned without additional sampling effort.
Until 1994 " unusually abundant" was defined as any mean density, calculated over three replicates, which was found to be 50% greater than the highest mean density observed during the l same month from 1975 through to the current year. Restricting comparisons to monthly periods damped the large seasonal variation so readily apparent with ichthyoplankton. Starting with 1994
" unusually abundant" was redefined. On a month-by-month basis for each of the numerically dominant species all previous mean densities over three replicates (1974-1993; to be updated each year) were examined and tested for normality following logarithmic transformation (Ryan and Joiner 1976). Single sample densities obtained from 1994-1996 were added to the pool within each month.
Where data sets (for example, mackerel eggs taken in June) fit the lognormal distribution, then i " unusually large" was defined by the overall log mean density plus 2 or 2.58 standard deviations.'
Log densities were back-transformed to make them easier to interpret, thus providing geometric means. In cases where data sets did not fit the lognormal distribution (generally months when a l species was frequently but not always absent, i.e., many zeros occurred), the mean and standard deviation was computed using the delta-distribution (see for example Pennington 1983). The same mean plus standard deviation guideline was applied.
l The decision to rely on 2 standard deviations or 2.58 standard deviations was based on the relative importance ofeach species. The more critical criterion was applied to species ofcommercial, recreational, or biological interest, the less critical to the remaining species (i.e., relatively greater densities were necessary to trigger notification). Species of commercial, recreational, or biological interest include Atlantic menhaden (Brevoortia tyrarmus), Atlantic herring (Clupca harengus),
Atlantic cod (Gadus morhua), tautog and cunner (the labrids; Tautoga onitis/Tautogolabrus
!- ' Normal distribution curve theory states that 2.5% of the measurements in a normally distributed population exceed the mean plus 1.% standard deviations (= s, we rounded to 2 for simplicity),2.5% tie below the l mean minus 1.% standard deviations. Stated another way 95% of the population lies within that range and 97.5%
l lies below the mean plus 1.%s. Likewise 0.5% of measurements exceed the mean plus 2.58s. 99% lie witidn the range of the mean i 2.58s,99.5% tie above the mean + 2.58s.
7 l
l l
\
adspersus), sand lance (Ammodytes sp.), Atlantic mackerel (Scomber scombrus), windowpane (Scophthalmus aquosus), American plaice (Hippoglossoides platessoides), and winter flounder.
Table 1 provides summary data for each species of egg and larva by month within these two categories showing the 1997 notification level.
A scan ofTable I will indicate that, in cases where the long-term mean amounts to 1 or 2 eggs or larvae per 100 m', the critical levelis also quite small. This situation occurred during months when a given species was obviously uncommon and many zeros were present in the data set with an inherent small standard deviation. The external reference distribution methodology of Box et al.
l (1975) was also employed. This procedure relies on a dotplot of all previous densities for a species within month to produce a reference distribution. Densities exceeding either 97.5 or 99.5% of the reference set values were considered unusually high with this procedure.
Mesh Extrusion l
To potentially improve enumeration ofcunner eggs and larvae in PNPS entrainment samples, i preliminary sampling was conducted in 1994 to see if eggs and young larvae are extruded through the standard 0.333-mm mesh netting. The smallest stage I larvae were not present in 1994 and slightly larger stage 2 larvae were uncommon. Additional paired sampling was therefore completed in June 1995 and June and July 1996. The sampling in 1994 and 1995 was completed by alternating nets on the sample rig while modification of the rig in 1996 allowed dual-mesh samples to be taken simultaneously. These mesh comparison studies indicated that cunner eggs were retained at significantly higher densities in 0.202-mm mesh samples compared with 0.333 mesh. No statistically significant difference was detected for stage 1 or 2 cunner larvae although respective mean ratios of 1.10:1 and 1.28:1 were recorded. Dual-mesh samples were taken on four occasions in June and July 1997 to provide a second year of simultaneously streamed nets. On each occasion three pairs were taken for a total of 24 samples. As in previous years, dates were selected based on previous samples 1
. and historical data to correspond to the likely period ofoccurrence ofsmall, early-stage larval cunner.
All samples were taken near low water when velocity and potential extrusion would be greatest with each collection eight to twelve minutes in duration. Since flow rates in the canal visibly vary across its width, the position of the nets was reversed between replicates to compensate for sampling position. Methodology followed that described for the routine sampling.
8 l
l
Table 1. PNPS ichthyoplankton entrainment notification levels for 1997 by species category and month. See text for details.
Densities per Long-term Mean + Mean +
100 m$ of water Mean' 2 std dev. 2.58 std.dev.
January LARVAE Atlantic herring' O.2 1 Sculpin Rock gunnel 0.8 _ l.4 Sandlana 5 11 February LARVAE Atlantic herring2 0.1 0.8 Sculpin 2 65 Rock gunnel 3 99 Sand lance 9 16 MilfEll EGGS American plaice2 2 3 LARVAE Atlantic herring8 0.9 1.3 Sculpin 17 608 Seasnails 0.6 i Rock gunnel 10.7 723 Sandlance8 7 164 Winter flounder 2 0.4 0.7 Antil EGGS American plaice2 3 32 LARVAE Atlantic herring' 1 2 Sculpin 15 391 ,
Seasnails 6 10 i Radiated shanny 3 6 Rock gimnci 4 142 Sandlance2 21 998 Winter flounder' 7 12 MaY EGGS . ,
' Labrids' 36 3514 Mackerci' 18 4031 Windowpane2 9 147 American plaice' 2 15 9
i I
Table 1 (condnued).
Densides per long-term Mean + . Mean +
100 m$ of water- Mean' 2 std.dev. 2.58 std dev.
Max LARVAE Adande herring: 0.7 1,1 Fourbeard rockling 2 5 Sculpin 3 4 P MinsM shanny 7 236 Sandlance2 22 32 Winter flounder2 9 123 hillt EGGS 2
. Adantic menhaden 4 6 Scarabins 3 4 Labrids2 958 21599 Mackerel 2 63 3515 Windowpane2 27 261 American plaice' I 2 LARVAE Atlantic menhaden 2 6 10 Fourbeard rockling 9 634 Cunner2 6 265 Radiated shanny 1 15 Mackerel 2 91 155 2
Winter flounder 2 20 hl!E EGGS Atlantic menhaden' 2 4 Labrids2 615 13349 Mackerel 2 9 16 Windowpanc2 12 156 LARVAE Atlantic menhaden 2 2 3 Tautog 2 2 2 Cunner 2 7 318 Mackerel 2 2 3 August l
EGGS Scarobins 4 6 i Arids: 23 936 Windowpane2 15 136 LARVE Atlantic menhaden 2 0.4 I Fourbeard rockling 6 10 Hake 2 4 Tauiog2 1.6 2.2 Cunncr8 10 15 10
i Table 1 (continued).
Densities per Long-term Mean + Mean +
100 m' of water: Mean' 2 std dev. 2.58 std dev.
Scotember EGGS Labrids' 2 3 Windowpane 2 11 159 LARVAE Fourbeard rockling 4 6
}{ake 5 9 Tautog' 1 2 Cunner' I 2 October EGGS Atlantic menhaden' 2 6
.Windowpanc' 1 2 LARVAE Atlantic menhaden' 2.3 4 Fourbeard rockling i 16 Ilake 1 2 November LARVAE Atlantic menhaden' O.4 1 Atlantic herring2 4 8 December LARVAE Atlantic herring2 2 3
' Geometric or Delta Mean.
' Species of commercial, recreational, or biological interest for which more critical notification level will be used.
I1
SECTION IV RESULTS A. Ichthvoolankton Entrained - 1997 Population densities per 100 m$ of water for each species listed by date, station, and replicate are presented for January-December 1997 in Appendix A (available upon request). The occurrence of eggs and larvae of each species by month' appears in Table 2. Ichthyoplankton collections are summarized below within the three primary spawrdng seasons observed in Cape Cod Bay waters:
winter-carly spring, late spring-carly summer, and late summer-autumn.
Winter-carly sorina spawners (December-Aoril)
When discussions are calendar-year based, this spawning season is split between the beginning and end. Many species spawning during this season employ a reproductive strategy relying on demersal, adhesive eggs which are not normally entrained. As a result, more species are typically represented by larvae than by eggs. Over both life stages number of species represented in the catch increased from 6 in January to 20 in April; eggs were absent in December. Considering the season as a whole,9 species were represented by eggs, yellowtail flounder (Pleuronectesferrugineus),.
fourbeard rockling (Enchelyopus cimbrius), winter flounder, and American plaice being numerically dominant (Figure 2). Yellowtail eggs did not appear in the @llections until April at which time they represented 28% of the month's total with a geometric mean density of 5 per 100 m' of water.
Rockling eggs were taken in March and April, accounting for 6% and 21% of those respective totals with monthly geometric mean densities of 0.2 and 4 per 100 m' of water. Wm' ter flounder eggs were found in Febmary, March, and April at respective geometric mean monthly densities of 0.2,0.9, and 0.7 per 100 m' of water, densities which accounted for 23,94, and 6% of those monthly egg totals.
American plaice eggs appeared in April when a geometric mean density of 5 per 100 m' accounted for 18% of the month's total.
Winter flounder eggs, although among the winter-early spring dominants in 1997, are demersal and adhesive and not often entrained by water intake systems. Their densities in PNPS samples are therefore not considered representative of numbers present in the surrounding area.
Those entrained were probably dislodged from the bottom by currents and perhaps the activities of other fish and benthic invertebrates.
12
Winter - Early Spring December- April Eggs Larvae Ye 11 flounder r7
'* $ nd knee '
252% ^ fE Sculpine 29.7 % ppWh.: "28.1%
Fourbeard rockling -
9 tlanbc hening Amehn phice .
15.7 % Winter flounder Rock gunnel 16.5 % 31.4 %
Sum of monthly means = 43.60 Sum of monthly means =582.99 c MWtcorrp07 p4 Late Spring - Early Summer Season May-June Eggs Larvae ,,_,,,
TauwCunrer Wirter flounder F 6.6%
74 8 % 7 19 9 % ] Sandlance F 64%
Atlantic mackerel 12.9 %
Cunr.er
- g< 25.o%
Rockling s fjl'; ..
A 10M x_ , 74p y llothers Atlartic mackerel 10.7% All others 14.6 % 19.2 %
Sum of monthly means =3549.33 Sum of monthly means =440.83 l
Figure 2. Dominant species of fish eggs and larvae found in PNPS ichthyoplankton samples by season for 1997. Percent of total and summed monthly means for all species are also shown.
13 I
Late Summer- Autumn Season t
August - November Eggs Larvae Rockkn0/ Hake 12.6% Tautogcunner #.nartic menhaden
%,, 15.5 % 10%
Ausnuernenha e _ h-Windowpene Cunner O
$2.1 % 23 8 %
Sum of monthly means =103.93 Sum of monthly means =294.73 Figure 2 (continued)
Larval collections during the winter-early spring season contained 20 species of fish.
Numerical dominants included rock gunnel (Pholisgunnellus), sand lance, sculpin (Myorocephalus spp.) and Atlantic herring (Clupea harengus; Figure 2). Rock gunnel accounted for 27% of the January total with a geometric mean density of I per 100 m',20% of the February total with a geometric mean of 5 per 100 m',44% of the March total with a geometric mean of 52 per 100 m',
and 7% of the April total with a geometric mean of 5 per 100 m'. Larval sand lance accounted for 36,25,24, and 51% of the four respective monthly totals with geometric mean densities of 1,7,43, and 52 per 100 m'. Sculpin, a group actually consisting of three species, represented 36% of the January catch,52% of the Febmary catch,30% of the March catch, dropping to 18% in April.
Respective monthly geometric mean densities amounted to 1,17,14, and 17 per 100 m' of water.
Among the three species oflarval sculpin the grubby (M. acnaeus) was most abundant over the season as a whole accounting for 90% of the group total. The shorthorn sculpin (M. scorpius) 14
followed at 7% and the longhorn sculpin (M. octodecem3pinosus) at 3%. Lastly, Atlantic herring larvae were taken only in December when they accounted for 94% of the month's total with a l geometric mean of 10 per 100 m' Late Spring-Early Summer (May - July)
Egg and larval densities, particularly among species with pelagic eggs, typically increase during this season along with expanding day length and rising water temperature. Considering both eggs and larvae together,14 species were represented in May, increasing to 19 species in June and 20 species in July. Numerical dominants included tautog/ cunner and mackerel among eggs and cunner, winter flounder, mackerel, fourbeard rockling, menhaden, and sand lance among the larvae.
Based on a study completed at PNPS in 1975 and 1976 (MRI 1978), over 90% of tautog/ cunner eggs taken in the PNPS area are believed to have been spawned by cunner.
Tautog/ cunner eggs accounted for 26% of the May total and 82% of both the June and July totals.
Respective monthly geometric means were 21,1105,128 per 100 m'. Mackerel eggs contributed an additional 63% to the May total and 10% to the June total with monthly geometric means of 21 and 14 per 100 m'. Mackerel eggs were not present in the July collections.
Cunner larvae first appeared during the second week of June and ultimately accounted for 14% of the month's total with a geometric mean of 10 per 100 m'. They increased in number in July accounting for 63% of that month's total with a mear. of 56 per 100 m'. Larval winter flounder overlap the winter-early spring and late spring-early summer seasons, first appearing in March. They accounted for 44% of the May larval catch,14% of the June catch, and less than 1% of the July catch; monthly geometric mean densities were 45,5 and 0.1 per 100 m', respectively. Although they appear to prefer the cooler months, rockling is one of the species which has been taken year-round at PNPS when the entire time series is considered. Their larvae added 11 % to the May catch with a geometric mean of 5 per 100 m',16% to the June total with a geometric mean of 12 per 100 m',
and 3% to the July total with a geometric mean of 3 per 100 m'. Atlantic menhaden larvae first appeared early in June, accounted for 5% of that month's total with a geometric mean density of 3 per 100 m', and increased in number in July when they accounted for 16% of the month's total with a geometric mean of 11 per 100 m'. Lastly sand lance, a species which like winter flounder often spans two seasons, occurred in May and the early part ofJune. Overall they contributed 18% of the 15
May larval total with a geometric mean of.16 per 100 m' and 0.3 % of the June catch with a geometric mean of 0.2 per 100 m'.
Late Summer - Autumn Soawners (Aunust - November)
This season. is typically described as one where a marked decline in both overall ichthyoplankton density and number of species occurs. Considering eggs and larvae combined,16 species were taken in August,17 were taken in September, followed by 12 in October dropping sharply to 4 in November. Numsal dominants included windowpane, tautog/ cunner, rockling/ hake, and menhaden among the eggs and hake, cunner, menhaden, and Atlantic herring among the larvae.
l Windowpane eggs are reported under the Paralichthys-Scophthalmus group because they cannot be distinguished from fourspot flounder eggs, Because windowpane larvae are far more abundant then fourspot larvae in PNPS collections, the majority of the grouped eggs are assumed to be in fact windowpane. These eggs occurred from August through September accounting for 47%,
55%, and 60% of those three monthly totals; geometric monthly means were 6,21, and 3 per 100 m'.
Tautog/ cunner eggs (presumably cunner) were taken throughout the season although their numbers declined steadily. They accounted for 34,10,2, and 7% of the respective monthly totals with mean densities of 5,3,0.2, and 0.1 per 100 m'. Rockling and hake eggs were entrained from August through October with monthly geometric means ranging from 1 to 4 per 100 m' and percent contribution ranging from 13 to 23%. Based on collection ofspecifically identifiable late-stage eggs, hake and rockling probably occurred in approximately equal numbers within the egg group. Lastly Atlantic menhaden eggs were found in September and. October with monthly geometric mean densities of 1 per 100 m' in both cases, values which accounted for 10 and 15% of each month's egg catch.
Among the larvae, hake were collected in August at a mean density of 4 per 100 m', in Septemb.s at a mean density of 24 per 100 m', and in October at a mean density of 3 per 100 m'.
These densities accounted for from 8 to 84% of those three r.ionthly larval totals. Cunner, ranking second among seasonal dominants, were present in samples from August and September at monthly geometric mean densities of 34 and 0.3 per 100 m'; these densities accounted for 53 and I % of those monthly totals. Larval menhaden occurred in August, September, and October, with mean densities l which ranged from 2 to 13 per 100 m' and accounted for 4 to 65% of the totallarval catch. Lastly, l
l l
o Atlantic herring appeared only in November, the typical beginning of their larval season, with a geometric mean density of 12 per 100 m' accounting for 99% of the month's catch.
B. Multi-vear Ichthyoolankton Comparisons A master species list for ichthyoplankton collected from the discharge canal at PNPS appears in Table 3; the years during which each species was represented is indicated for 1975 through 1997.
A total of 37 species was represented in the 1997 collections,just below the 22-year mean of 38.
Appendix B (available upon request) lists geometric mean monthly densities along with 95%
confidence limits for each of the numerical dominants collected during each year dating back to 1982.
Geometric means are reported because they more accurately reflect the true population mean when the distribution ofsample values are skewed to the right as is commonly the case with plankton data.
Generally low values obtained for both eggs and larvae during April-June 1984 and 1987 were shaded
^
because low through-plant water volumes during those months probably affected densities of ichthyoplankton (MRI 1994); shaded values were omitted from the discussion below. Entrainment i data collected from 1975-1981 remain in an outdated computer format requiring conversion before geometric mean densities can be generated. These years were therefore excluded from Appendix B but are discussed in the multi-year comparisons if noteworthy.
To help compare values over the 16-year period, egg data were plotted in Figure 3 for those species whose combined total represented 90% of the 1997 egg catch. For this figure cod and pollock eggs were combined with the Enchelyopus-Urophycis-Peprilus group, and labrids and yellowtail flounder were combined with the labrid-Pleuronectes group. For each category shown, the highest monthly geometric means obtained from 1982 through 1996 werejoined by solid lines as were the lowest monthly geometric means, and the area between was shaded, indicating the range of these values. Monthly geometric mean values for 1997 werejoined by a solid line. Alongside each plot is a bar graph showing annual abundance indices for each year. These were generated by integrating the area under each annual curve using trapezo dal integration. One set ofbars was based on geometric monthly means (1982-1997) and the other, longer time series, on arithmetic monthly means (1975-1997). Appendix B and Figure 4 contain corresponding data for the 13 numerically dominant species of fish larvae, those accounting for 99% of the 1997 catch, as well as total larvae 17 i
l (all species combined). As mentioned for eggs, low values obtained for both eggs and larvae during April through August 1984 and 1987 were flagged in these figures and omitted from the following discussion.
In many cases densities of fish eggs and larvae vary considerably from year to year. For example, over the 16-year geometric mean time series the highest annual abundance index divided by the lowest for Atlantic menhaden eggs amounted to 292. In spite of such pronounced variation, no consistent upward or downward trend was apparent over the time series for many species such as menhaden and windowpane eggs, menhaden, sculpin, seasnail (Liparis atlanticus), tautog, and rock gunnel larvae. Following are noteworthy observations concerning the multi-year time series.
Since densities of each ichthyoplankton species rise from and fall to zero over the course of each respective occurrence season, inter-year comparisons are often conveniently made within monthly periods.
Atlantic menhaden eggs were very abundant at PNPS in June 1997. This was clearly reflected in the notification program where " unusually high" densities were recc;ded on 10 of12 sampling occasions during that month. On June 11 (228 per 100 m'), June 20 (425 per 100 m'), and June 27,1997 (228 per 100 m') densities exceeded all previous June v'alues (Table 1, Volume 2). A monthly geometric mean of 20.3 per 100 m' was recorded which exceeded all previous values except that noted in 1989 (21.1 per 100 m'). Primarily because of the high June densities menhaden eggs proved relatively abundant for the year in general, 1997 ranking fifth among the geometric indices dating back to 1982 and fourth among the arithmetic indices dating back to 1975. Considering the protracted period of high egg densities it is perhaps not surprising that menhaden larval densities were also relatively high in 1997. Larvae were abundant from June through August with a number of samples, particularly in August, exceeding all previous densities for the respective month (Figure 4).
Overall the 1997 larval abundance indices exceeded all previous years. The geometric index (1145) exceeded the previous high (382,1995) by a factor of three and the arithmetic index (2801)just exceeded the previous high for that time series (2708,1981).
Menhaden are coastal migrants which travel in schools that can often be quite dense. The great variability in numbers of eggs taken at PNPS probably reflects not only numbers of I8
adults in the surrounding waters but variability in the distance from PNPS at which
! spawning iakes place. Spawning stock biomass increased from 1993 through 1995 (Cadrin and Vaughan 1997) which is consistent with the observed increase in egg densities.
Atlantic cod eggs were typ% ally collected in low numbers (5 per 100 m' of water for example) at PNPS during winter months from 1975-1987. Following 1987 they became uncommon particular2y during January and February. None were taken either month in 1993 or 1994 and only one was taken in 1995. In 1996 collections rose to three eggs, all taken in February. The gadidae-Glyptocephalus group in general showed a significant decline from 1975 to 1993 (p<0.001), based on a nonparametric sign test, which is consistent with the I
downward trend reported for Atlantic cod and witch flounder stocks apparently resulting, at least in part, to overexploitation (NOAA 1995, NFSC 1996). - Annual geometric mean indices suggest the decline has ended if not reversed, at least locally, since values for 1994 through 1997 appear stable at about three times the low value recorded in 1993.
Eggs of the fourbeard rockling and closely related hake (grouped in the early developmental stages with far less common butterfish as EnchelyopuM/rophycis-Peprilus) have been uncommon in recent years. Trend analysis using the longer-term arithmetic time series indicated that a significant downward trend occurred from 1978 through 1996 (p = 0.05) in spite of a moderate catch in 1995. Any suggestion of a reversal in 1995 was erased by the 1996 value which was similar to values observed from 1992 to 1994. In spite of relatively high densities in April 1997 (Figure 3), the 1997 indices represented but a slight improvement over 1996. Fourbeard rockling dominate within this groupmg based on late- I stage egg as well as larval collections. Since this a small bottom fish with little or no commercial value, stock size data are not available with which to compare trends. Hake on the other hand contribute to the commercial bottom fishery, and stocks in the GulfofMaine and northern Georges Bank are considered to be underexploited. In contrast, stock abundance ofred hake on southern Georges Bank and in Massachusetts waters are relatively i low according to the Northeast Fisheries Center survey index (NOAA 1995) which might explain, at least in part, the apparent decline in grouped egg abundance.
l 19
i
. Tautog/ cunner eggs, composed primarily of cunner (Scherer 1984) appeared to be in a downward trend from the late 1970's through 1994 although a sign test failed to confirm it using the conventional 95% significance level (p = 0.055). In contrast the arithmetic and geometric indices both showed an increase in density in 1995, the geometric index continuing to rise in 1996. The 1995 arithmetic index appeared exceptionally high and disproportionate to the geometric value due to a single high density in June (37,282 per 100 m' of water) which greatly skewed the arithmetic mean for that month. The 1907 indices declined from 1996 but remained well above the low values observed in 1990,1991, and 1994. The downward trend noted through 1994 is consistent with finfish observations in the PNPS area as well as impingement collections at the Station (Lawton et al.1995). Changes in sampling protocols at PNPS have negated the ability to monitor general cunner population trends beyond 1994 which in the past were sampled by gill net, trawl, and diver surveys. Numbers impinged appeared to systematically decline from 1980 through 1992 (annual totals dropped from 116 to as law as 2 in 1988) then increase from 1993 (104) through 1995 (288). They remained high in 1996 (211) which appeared to roughly parallel the egg abundance data.
The impingement total for 1997 (39, see impingement Section) represented a sharp drop relative to the preceding four years and appeared out of step with the ichthyoplankton collections.
Eggs of the yellowtail flounder were also relatively abundant in April 1997. While early staged eggs of this species are similar to and grouped with the labrids they are believed to account for all eggs of that type collected in April since the labrids are not likely to spawn until May. The geometric mean density for that month was 4.6 per 100 m', exceeding the ;
previous high of 1.8 per 100 m' noted in 1983 (Figure 3).
. While mackerel egg densities declined in 1997 they have clearly been more abundant since i
1988 when compared to the 1975 through 1987 period. A sign test using the arithmetic '
index time series supported this upward trend (p<0.006). Mackerel eggs typically display a sharp peak in their abundance curve often with one or two very high densities. For example in May 1995 a single density of 19,203 eggs per 100 m' was recorded on May 26, dropping to 557 eggs per 100 m' on the 29* The second highest density occurred on June 20 l
9 at 4,754 per 100 m'. Due to these briefsharp peaks, arithmetic and geometric indices are often quite far apart (Figure 3). Entrainment ofhigh densities ofmackerel eggs over the past decade is consistent with a dramatic rise in stock biomass attributable to reductions in foreign fishing and under exploitation by U.S. fishermen (Overholtz 1993, NOAA 1995, NFSC 1996).
- Windowpane eggs, assuming, based on larval collections, that they predominate within this egg group, have increased from 1994 through 1996. The annual geometric index for 1997 (3144) was essentially equal to 1996 (3147). Over the entire 23-year time series the arithmetic index for 1997 ranked sixth just below 1996. In general these eggs have not shown wide variations in number, at least not compared with other species regularly entrained. Consistent with the recent egg collections, current abundance indices for windowpane, based on Massachusetts Division Of Marine Fisheries spring and fall surveys, suggest that stocks are relatively high at least through the 1996 assessment, values having increased steadily since 1991 (Steve Correia, MDMF, personal communication).
- For American plaice eggs, the arithmetic time series shows a more or less steady decline from 1978 to 1986 followed by a steady increase through 1995. While 1996 represented a drop in overall abundance,1997 turned upward again. Patterns of egg abundance appear to follow, in a general way estimates of finfish abundance According to the Northeast Fisheries Center survey results for Massachusetts (NFSC 1996) plaice were at relatively low abundance levels in the mid 1980's, when egg entrainment was also at its lowest, then rebounded following the production of a strong year class in 1987. They again declined from 1989 through 1993 and while the finfish series does not yet run to 1997 they appear to have begun to increase again.
- Atlantic herring larval abundance indices have proven valuable in management of herring stocks on Georges Bank, Nantucket Shoals, and in the Northeast Atlantic in general (see for example, Smith and Morse 1993). The stock was seriously depleted during the 1970's and collapsed on Georges Bank in 1976 (Anthony and Waring 1980, Smith and Morse 1993).
The stock has increased more or less steadily since 1986 following reductions in fishing pressure. Presently the Atlantic coast stock is increasing in size and considered to be 21 l
i I
extremely underutilized (NFSC 19%). Iarval collections at PNPS from 1994 through 1997 reflect the general increase in stock size, the geometric index for those four years ranking among the top five (Figure 4).
. Larval hake were abundant in 1997. The geometric index for that year (994) exceeded the previous high (514,1985) by a factor of 1.9. Data available through 1995 suggest that hake stocks in Southern New England have declined by about 50% since the late 1960's and surveys in Massachusetts waters confirm that abundance is relatively low (NFSC 1996).
High larval abundance at PNPS in 1997 may indicate production of a strong year class or simply reflect a localized spawning aggregation.
. Larval sculpin abundance was high at PNPS in 1997, the geometric index (2249) ranking .
third over the 16-year time series and the arithmetic index second over the 23-year series.
Since these fishes have no commercial or recreational significance no stock size data are available with which to compare the larval abundance.
. in spite of unremarkable egg abundance in 1997, larval tautog and cunner were relatively abundant in PNPS entrainment samples. Tautog were panicu-larly abundant in July, cun ner were particularly abundant in August (Figure 4). The 1997 geometric index for tautog (219) ranked second behind 1989 (324) and represented the third straight year ofincreasir.g abundance The geometric index for cunner (3096) also ranked second behind 1990 (3680).
Current stock size data for cunner are not available but tautog are believed to be overfished and at very low levels (NFSC 1996).
- Larval winter flounder densities were relatively high throughout their occurrence period in 1997panicularly in April and May. April 1997's geometric mean of 8.2 per 100 m' and May's geometric mean of 45.3 per 100 m' clearly exceeded all previous April and May i values which ranged from 0.2 to 5.2 per 100 m' and 5.1 to 17.3 per 100 m', respectively.
June's geometric mean of 4.9 per 100 m' also exceeded all previous June values with the exception of 19% (6.5 per 100 m'). These repeatedly high mean densities translated into the highest overalllarval abundance index yet observe (' .t PNPS. This was true for both the arithmetic and geometric indices. The geometric index for 1997 (1800) exceeded the 22
previous high of 621 noted in 1995 by a factor of 2.9 and the arithmetic index (3196) exceeded the previous high of 2735 noted in 1978 by a factor of 1.2.
Larval winter flounder abundance in the Mount Hope Bay section ofNarragansett Bay was average to somewhat below average in 1997 (40* and 50* percentile by two abundance d
indices) while in Niantic Bay, Connecticut, abundance was relatively high (2 out of 15 years; Dale Miller, Northeast Utilities Service Co., personal communication). In spite of the differences in 1997 between these areas, in general larval flounder abundance has been found to be correlated between these two locations (NUSCO 1997), suggesting that widespread regional signals exist. For the 1975 through 1997 period no comparable correlation was detected between Mount Hope Bay and PNPS larval flounder abundance (Pearson correlation coefficient, r = 0.036, p > 0.05). Since Cape Cod appears to serve as a faunal barrier (see for example Anraku 1964, Davis 1984, Scherer 1984), regional signals may be expected to dissolve or weaken across that barrier. Stock abundance based on the Massachusetts Division Of Marine Fisheries spring, northern stock assessment appears to have been relatively stable since 1988. From that time through 1996 trawl estimates have varied without trend from 10 to 15 kg per tow, down from a high of 29 per tow in 1983.
There is no evidence in that time series to suggest why particularly large numbers oflarvae would be present in PNPS waters in 1997. Hopefully the high larval abundance will result in production of a strong year class although this is unlikely to be observed until they recruit to trawl surveys C. Mesh Extmsipfl Densities per 100 m' of water for tautog/ cunner eggs and cunner larvae by stage for both 0.333 and 0.202-mm mesh collections completed from 1994 through1997 appear in Table 4.
Eggs: Paired sample t-tests on log-transformed data for the 12,1997 samples failed to detect a significant difference between mesh samples (mean 0.202/0.333 ratio = 1.10, p = 0.270). Similar results were obtained in 1996 when simultaneous, paired samples were also collected (mean ratio =
1.14, p = 0.093, n = 12). Although the 0.202-mm mesh netting retained somewhat more eggs both seasons, the difference was not large. Power analysis indicated that 93 pairs of samples would have 23 l
been necessary to detect the observed difference in 1997 with a probability ofType 1 error of 5% and probability of Type 2 error of 10% (power = 90%, Cohen 1988, Hintze 1996). A corresponding sample size of 38 would have been necessary in 1996.
Pooling samples from all four seasons, 1994-1997, resulted in detection of a significant difference ( mean ratio = 1.24, p = 0.0006, n = 45). The greater ratio for all samples combined resulted from relatively high ratios of 1.58 in 1994 and 1.20 in 1995. The similarity in mean ratios from 1995 and 1996 at least suggests that great sampling error was not introduced by alternately collecting samples (1995) rather than streaming both nets simultaneously (1996).
Mesh collection ratios appear to vary from one year to another. This is probably attributable to differences in mean egg size which can vary with water temperature (see for exampic Bagenal 1971), female size and condition (Kjesbu et al 1992, Kjesbu 1994), and other plankton in the water column at time of sampling. Population densities and types of other plankton can have a profound effect on net performance by the purely physical process ofcoating the mesh (Tranter and Smith 1968 among others).
In spite of failing to detect a difference between nets in 1996 and 1997 when differences were subtle, the respective mean ratios observed (1.14 and 1.10 respectively) were applied to egg densities to be conservative when calculating numbers entrained for impact assessment (see Volume II). Since nets were streamed simultaneously during those years, results are likely to be more accurate. For years preceding 1996, the pooled-data ratio of 1.24 was used.
LarYae Recently hatched larval cunner were present in both nets of seven sample pairs in 1997. Pooled with the 1995 and 1996 data,22 pairs were available for comparison. A paired sample t-test was unable to detect a statistically significant difference between nets as the difference was small and variable (mean ratio = 1.16, p = 0.514, Table 5). Power analysis indicated that over 500 pairs of samples would have been required to detect a difference of the observed magnitude with Typel error equal to 5% and Type 2 error equal to 10% (power = 90%).
Stage 2 larvae were collected more often each season than stage I larvae resulting in 32 pairs of samples for analysis. Again no mesh difference was detected although the ratio between nets was greater than for the younger larvae (mean ratio = 1.28, p =0.096, Table 6). As the probability level indicates, the 0.202-mm mesh netting collected significantly more stage 2 larvae then the 0.333-mm 24
mesh at the 90% probability level (10% chance of accepting a difference which was not real).
Examined separately, the 1997 samples showed a mean ratio nearly equal to the pooled data (mean ratio = 1.26, p = 0.208, n = 1I). The 1996 samples showed a somewhat smaller difference (mean ratio = 1.14, p = 0.559, n = 12) although overall larval mesh collections were more similar over time than the egg collections. Power analysis indicated that 115 samples would have been necessary to detect a statistically significant difference with the pooled data at a Type 1 error rate of 5% and Type 2 error of 10%.
Stage 3 larval cunner were not collected as often as the younger stages because there are fewer of them to be caught and also because sampling dates were selected to collect younger larvae.
Consistent with earlier results, larger stage 3 larvae were taken in similar densities in both nets when they were available (mean ratio = 1.09, p = 0.646, n = 14).
Consistent with the egg densities, mesh conversion constants were applied to stage I and stage 2 larval cunner densities as a conservative measure when assessing impacts by the equivalent adult approach. Stage i densities were scaled upward by 1.16 and stage 2 by 1.28. These values we.e used for all years since larval differences were fairly consistent between years and fewer pairs were ava.hble to study years individually.
l D. Lobster Latvae Entrained l l
No lobster larvae were found in the 1997 entrainment samples, the total, dating back to 1974, l remaining at 13. Following is a tabulation of previous collections:
1996: none found j 1995: 1 larva - stage 4-5, July 28. j 1994: none found. j 1993: 1 larva - stage 4-5, July 21.
1991-1992: none found.
1990: 2 larvae - I stage 1, June 26; I stage 4 August 23.
l l 1983-1989: none found.
1982: 1 larva - stage 1 on June 14.
1981: 1 larva - stage 4 on June 29, 25
i 1980: none four.d.
1979: 1 larva - stage 1 on July 14.
1978: none found.
1977: 3 larvae - 1 stage 1, June 10; 2 stage 1, June 17, 1976: 2 larvae - I stage 1, July 22; I stage 4-5, August 5.
1975: 1 larva - stage 1, date unknown.
1974: none found.
The lobster larvae collected in 1976 were obtained during a more intensive lobster larvae program which employed a 1-meter net, collecting relatively large sample volumes, in addition to the standard 60-cm plankton net (MRI 1977). Both larvae taken in 1976 were collected in the meter net; none were found in the routine ichthyoplankton samples.
During the three-season Cape Cod Bay neuston study for larval lobster begun in 1974, larvae were found from May through September at monthly mean densities ranging from 0.2 (September) to 3.8 per 100 m' (July; Matthiessen and Scherer 1983). Considering that a minimum of roughly 10,500 m' of water were sampled during these months each year, larval lobster must indeed be rare in the PNPS circulating water system.
i t
26
SECTION V HIGHLIGHTS
- 1) Numerical dominants among the ichthyoplanktoq entrained in 1997 included yellowtail flounder, fourbeard rockling, winter flounder, American plaice, rock gunnel, sand lance, sculpin, Atlantic herring, cunner, Atlantic mackerel, Atlantic menhaden, and windowpane.
- 2) Atlantic menhaden eggs were very abundant in PNPS entrainment' collection in 1997 particularly in June. This was reflected in the notification program and also in the 1997 geometric mean abundance index in general which ranked fifth dating back to 1982.
- 3) Fourbeard rockling and hake eggs have shown a significant downward trend in abundance j from 1978 through 1996. Abundance in 1997 showed only a slight improvement over 1996.
- 4) While mackerel egg densities declined in 1997 these eggs have clearly been more abundant since 1988 when compared to the 1975 through 1987 period. This is consistent with a dramatic rise in stock biomass attributable to reductions in foreign fishing and under-exploitation by U.S. fishermen.
- 5) Atlantic laring larval abundance indices at PNPS have generally reflected an overall increase in the Georges Bank, Nantucket Shoals, and Northeast Atlantic stocks. Larval collections from 1994 through 1997 have ranked in the top five years dating back to 1982.
- 6) Larval hake collections at PNPS in 1997 exceeded the previous high year by a factor of 1.9 in spite of relatively low egg numbers.
- 7) Larval winter flounder abundance in 1997 was the highest yet observed at PNPS, densities being high throughout their period ofoccurrence. Stock abundance data suggest the adult population has been relatively stable since 1988.
- 8) Mesh comparison studies of cunner eggs and larvae were completed for the fourth season in 1997. These data suggest that cunner eggs are extruded through 0.333-mm mesh at a ratio of 1.24:1 when compared to 0.202-mm mesh. Recently hatched cunner larvae also appeared to be extmded although the observed mean ratio of 1.16:1 was not found to be statistically
, significant using paired t-tests. For older stage 2 larvae 0.202-mm mesh retained significantly more individuals when tested at the 90% probability level with a mean ratio of 1.28:1.
- 9) No lobster larvae were found in the 1997 entrainment samples, the total remaining at 13 dating back to 1974.
27
SECTION VI LITERATURE CITED Anraku, M. 1964. Influence of the Cape Cod Canal on the hydrography and on the copepods in Buzzards Bay and Cape Cod Bay, Massachusetts. I. Hydrography and distribution of copepods. Limnology and Oceanography 9:46-60.
Anthony, V. and G. Waring. 1980. The assessment and management of the Georges Bank herring fishery. Rapp. P,-V. Reun. Cons. Int. Explor. Mer. 177:72-111.
Bagenal, T.B. 1971. The interrelation of the size of fish eggs, the date of spawning and the production cycle. Journal of Fish Biology 3:207-219.
Box, G.E.P., W.G. Hunter, and J.. Hunter.1975. Statistics for Experimenters. John Wiley & Sons, New York.
Cadrin, S.X. and D.S. Vaughan. 1997. Retrospective analysis of virtual population estimates for Atlantic menhaden stock assessment. Fishery Bulletin U.S. 95:445-455.
Cohen, J. 1988. Statistical Power Analysis for the Behavioral Sciences. Lawrence Erlbaum Associates, Publishers, Hillsdale, New Jersey. 567p.
Davis, J.D. Western Cape Cod Bay: hydrographic, geological, ecological, and meteorological backgrounds for environmental studies. pl-18 In: J.D. Davis and D. Merriman (editors).
Observations on the Ecology and Biology ofWestern Cape Cod Bay, Massachusetts. Lecture Notes on Coastal and Estuarine Studies. Volume II. Springer-Verlag, New York.
Herrick, F.H.1911. Natural history of the American lobster Bulletin U.S. Bureau of Fisheries 29:149-408.
Hintze, J.L.1996. Pass 6.0 User's Guide. NCSS, Kaysville, Utah. 245p.
Kjesbu, O.S. 1994. Time of start of spawning in Atlantic cod (Gadus morhua) females in relation to vitellogenin oocyte diameter, temperature, fish length and condition. Journal of Fish Biology 45:719-735.
, H. Kryvi, S. Sundby, and P. Solemdal. 1992. Buoyancy variations in eggs of Atlantic cod (Gadus morhua L.) In relation to chorion thickness and egg size: theory and observations.
Journal of Fish Biology 41:581-599.
l l
28
Lawton, R.P., B.C. Kelly, V.J. Malkoski, and J. Chisholm. 1995. Annual report on monitoring to assess impact ofthe Pilgrim Nuclear Power Station on selected finfish populations in western Cape Cod Bay. Project Report No. 58 (January-December 1994). IIIA.i-77. h: Marine Ecology Studies Related to Operation of Pilgrim Station, Semi-annual report No.45. Boston Edison Company.
MRI (Madne Research, Inc.). 1977. Entramment investigations and Cape Cod Bay Ichthyoplankton Studies, July-September 1976. III.C 1-1-71. b: Marine Ecology Studies Related to Operation of Pilgrim Station, Semi-annual Report No. 9. Boston Edison Company.
1978. Entrainment investigations and Cape Cod Bay Ichthyoplankton Studies, March-December 1977. III.C.2-34-38. b: Marine Ecology Studies Related to Operation of Pilgrim Station, Semi-annual Report No. I 1. Boston Edison Company.
1988. Entrainment investigations and Cape Cod Bay Ichthyoplankton Studies, March-December 1987. III.C.1-6-10. h: Marine Ecology Studies Related to Operation ofPilgdm Station, Semi-annual Report No. 31. Boston Edison Company.
1994. Ichthyoplankton entrainment monitoring at Pilgrim Nuclear Power Station January-December 1993. Volume 2 (Impact Perspective). IllC.2.i-33. h: Marine Ecology Studies Related to Operation of Pilgrim Station, Semi-annual Report No. 41. Boston Edison Com-pany Matthiessen, G.C. and M.D. Scherer. 1983. Observations on the seasonal occurrence, abundance, and distribution oflarval lobsters Glomarus americanus)in Cape Cod Bay. p41-46 M: M.J.
Fogarty(ed.). Distribution and relative abundance ofAmerican lobster, Homarus americanus.
larvae: New England investigations during 1974-79. NOAA TechnicalReport NMFS SSRF-775.
NFSC (Northeast Fisheries Science Center). 1996. Report of the 21st Northeast Regional Stock Assessment Workshop (21st SAW). Stock Assessment Review Committee (SARC) consensus summary ofassessments. Northeast Fisheries Science Center Reference Document 96-05d. 200p.
NOAA (National Oceanic and Atmospheric Administration). 1995. Status of the Fishery Resources off the Northeastern United States for 1994. NOAA Technical Memorandum NMFS-NE-108. 140p. i NUSCO (Northeast Utilities Service Company).1997. Monitoring the marine environment of Long !
Island Sound at Millstone Nuclear Power Station.1996 Annual Report. NU Environmental Laboratory, Waterford Ct. 248p.
29
Overholtz, W.J.1993. Harvesting strategies and fishing mortality reference point comparisons for the Northwest Atlantic stock of Atlantic mackere!(Scomber scombrus). Canadian Journal of Fisheries and Aquatic Science 50:1749-1756.
Pennington, M.1983. Eflicient estimators of abundance for fish and plankton surveys. Biometrics 39.281-236.
Ryan, T.A., Jr. and B.L. Joiner. 1976. Normal probability plots and tests for normality. Minitab, Inc., State College, PA. 19p.
Scherer, M.D. 1984. The ichthyoplankton of Cape Cod Bay, p151-190 In: J.D. Davis and D.
Merriman (editors). Observations on the Ecology and Biology of Western Cape Cod Bay, Massachusetts. Lecture Notes on Coastal and Estuarine Studies. Volume 11. Springer-Verlag, New York. 289p.
Smith, W.G. and W.W. Morse. 1993. Larval distribution patterns: Early signals for the collapse / recovery of Atlantic hering Clupea harengus in the Georges Bank area. Fishery Bulletin, U.S. 91:338-347.
Tranter, D.J. and P.E. Smith. 1968. Fi;tration performance. p27-55 in Zooplankton Sampling.
UNESCO (United Nations Educational, Scientific and Cultural Organization), Paris, France.
! 1
! 1 30
)
Figure 3. Geometric mean monthly densities per 100 m' of water in the PNPS dischwge canal for the eight numencally dominant egg species and total eggs,1997 (bold line). Solid lines encompassing i shaded area show Ngh and low values over the 1982 1996 period.
Brevoortia tyrannus labrsgae-Pleuronectes Gadadae-Glyptocephalus Scomber scombrws l
Enchel>vpus-Urophycis Paralichthys-Scophthalmus Peprilus Hippoglossoidesplatessoides Prionotus spp.
Total eggs To the right are pioned integrated areas under the annual entrainment abundance curves for 1975-1997. An asterisk above 1984 and 1987 marks the two years when values may have been low due to low through-plant water volumes from April August. An asterisk above 1976 indicates abundance value may be low due to absence of sampling during January-late April; see text for clanfication. Light bars represent indices based on monthly arithmetic means, solid bars (1982-1997) indices based on monthly geometric means.
I 31
Brevoortia tyrannus
,00 Eggs ,
21.233 to ' ' '
isao I. .. .
i 8
q ..
u 0, .. ... .
} ,
O9 9 ".
" " " ' l'
, . , , . , ",P,,",:'n"~"."
0 001 -- -- - - --
Yast c;~ =a~o
. . , . . , , , e - - -'~i Gadidae - Glyptocephalus 10 2000
- .' y - >i.- ,. 's D00 t 'lif .g .. ., 4 c-gn -
g 0% , (; "
-3. .' !
, ,, 3 Li,,
wp g; _
",." ".e'.,. z'.e>, -
'~
0 005 -
Year
^' Ahurwaanos huica m (C3H@ew **1997]
Indwins O mwees P weena endo c.wingkasme l 32
Enchelyopus - Urophycis - Pepriha Eggs i.0 30 21 - - . - . -
I" N!!hf i'Df!P MhM!"t !py.ppe;gy;pyy -
0' 1M"!!!!"!!!!!!!!P in;p;7pi
- ' I$
I !q #
- 0!!PH!PH !!DHi!ppit!lfah . . . .
OI 3.pn. un,g!n ,,g, ntp 0 04 ,
ngp!,g, n f
,n !
p l l l l l
,, 7, ,9 .I . ,, ,, 9 9, 9, 9, 9,
' * " - - 76 75 30 32 34 e6 as 90 92 94 96 JFMAMJJA8ONb Abundenne inden based am
% - - -e --
t..._ ,99,3 3d*I'* I W MJem w.and P %
Prionotus spp.
Eggs
- i. 1400 2391
~~
iam .. ? ..
10 r 3 , ,
1000 I, .
1-22' 600 P
, 400 , , .
0I i i ,
Q N 2#
) \
,1 nllll lib 75 7, 79 31 33 g3 31 gg gg 93 76 75 30 82 84 to 58 90 92 94 96 l,t llfh m.
- n. l 97 3;.
j hd hs;- [ y,,,
JFMAMJJA5ONh fP Ardenene Abundanos hussa bened an sneene .Geommenc y.g
$3thgh/1m es1997}
33
Labridae - Pleuronectes l-Eggs ,
1000 mp. 2 90 100 mm;pmpppqn. !m; spn;pigmypprp 200 - -
1 10 ISO 99pippyfpp fpunn.e mv.uimmep.
8 . . . . . .
I mannier 1"
!"!T!H"! !!"'M!""" p 08 !ip!p i pq7,o 30 0 01 tpmpgp m,ng !p!. 0 75 ?? 79 Si 23 R1 87 39 91 93 91 97 76 73 30 82 34 B6 E8 90 92 94 96 0 008 - - - Year J F M A M j jA 8 O N O M pwmut mo s.w (um> w =nm) includes Labridae and P.ferrvgtness Scomber scombrus 10000 250 low op7 gppoi ; i i ,
200 - - - - -
100 5 i i,. . :. pi . .. ,
an _
10 pipipmuqppi qpipmp nip 8 . . . . . .
f 'f " , ,
., ..: ~ ~;$.;j
, ,, n g ., ,
M. 30 01 ogpqpygg n, pgp! , ppq g;q
$ ,m , n, _ fi ri n rw $ M l i . fl_ _ .Il 0 01 puup:9p> spoupy pupunua 75 77 79 Bl 83 39 B7 89 91 93 95 97
.4
- 'O 76 78 B0 82 84 B6 as 90 92 94 96 x
']" sk)C.s '
Year J F M A M J J A 8 O N D ^ "" "
(_
p Arnimmtz nuers acewnstru a (C3HgMew as1997}
34
Paralichthys - Scophthalmus
,0. Eggs,,
iOO m,q.nmncemo ,n. m.mmn _
S -
10 n .,ui p.p;.. . ..
1 y
==
j 6
, i mnen , w 4 I OI ""' " '"
2
- 1 1 Oa e ug, - .. .. .
76 78 30 52 84 36 88 90 92 94 96 J F M A M J J A 5 O N D ^ "
[EllH@ 891997)
Hippoglossoidesplatessoides 10 1600 Q' ,',.
1400 f ij:
1200
[.h Y. %*
i .yl: f4)
. .d - ?+ - 1000 I b. .
I 't . *. suo g 01 ^7 < 600 400 g$:L1g, 2.
.($ #'I' y
/I
, [1 . &
73 77 79 st 33 ES 37 39 91 93 95 97 l :. ; I:i 76 18 30 52 84 86 85 to 92 94 96
) fly #, 2. :.. m&#.k,,, veer J F M A M J J A 3 o le D ^ "
Anthmens means MGenestne meer (DHgWim est991) 35
1 l
Total Eggs 10000 Son 1000 i i ,o g i g
100 ig;p!!yn i ni o 1
g g
10 i
! iti yp i 200 i 8 . . _ . . .
I e gypipapount: pong , ,
100 .*
0I 1 .munim:am ia o ei l
l 0 01 73 77 79 81 83 85 37 39 91 93 91 97 iny;put i i n; p;!!'"' i 76 78 80 32 84 36 as 90 92 94 96 Year 0 001 J P' M A M J J A 3 O N D
( Abwannae W based an D Anthmete means WGeonstne Month (C2HW es1997]
f 36
l l
Figure 4. Geometric mean monthly densitics per 100 m' of water in the PNPS discharge canal for the thirteen numerically dominant larval species and total larvac,1997 (bold line). Solid lines encompassing shaded area show high and low values over the 1982-19% period.
1 Brevoortia tyrannus Tautogolabrus adspersus Clupea harengus Ulvaria subbifurcata Enchelppus cimbrius Pholisgunnellus Urop%ycis spp. Ammo &tes sp.
Mprocephalus spp. Scomber scombrws Liparts upp. Pleuronectes americanus Tautoga onitis Totallanae To the right are plotted integrated areas under the annual entminment abundance cun'es for 1975-1997. An asterisk above 1984 and 1987 marks the two years when values may have been low due to low through-plant waler volumes from April August. An asterisk aboyc 1976 indicates ahnadae value may be low due to absena of sampling during January-late April; see text for clarification.
Light bars represent indices based on monthly arithmetic means, solid bars (1982-1997) indices based on monthly geometric means i
i 37
I I
i Brevoortia tyrannus Larvae 100 le gg - .1Y 10 gg ,
1000 -
OI 400 .
~
200 - -
0 00 e i r ni 4
0 71 77 79 si 33 35 87 89 91 93 93 97 76 75 30 52 84 86 88 90 92 94 96
^ - '
0 001 - YEAR J F M A M J J A 8 O N D 3 ,
M'** p Ard2mnenc mesens WGesunetnc (IDHW selv97)
Clupea harengus Larvae 100 1600 1,4 j gg _
10 '
1200 Shp ; -
l 4 6Mes/
1 1000 3 .jij 300 ~
_..y $N?h
-8Ei 8
- 'm_ y;;:g: -< eon . . _ . . .
1 ( j Q ;I s[F[lf:'
M ~
i 1k '*9 l[' lik . l
$$ $. 200 l. -
0 01 () f' @f:
l '- f; y GnD0%
i k % 57 75 77 19 st 23 35 57 39 91 93 95 97 76 78 80 R2 84 36 R8 90 92 94 96 l
0 001 j
{. 39
YEAR J F M A M J J A S O N D
<-n M-.
6._ ._ . ]
38
Enchelyopus cimbrius
- i. Larvae.
S l0 1 1 4 ..._...
, n.
8
\ on ,, ,
. A nm
,e 2. 80 .2 .4 .6 11 8 ,o 92 .4 .6 0 001 .. . Year J F M A M J J A 8 O N D
-- 6g .-. c J
c., n ,,,,,
Urophycis spp.
Larvae
- i. i 9
4 . . . . .
I i ~. e 2 :
- ug i
?Y(
8 s f . .
2 el , ,
I k$
j;f i c
i^ %:!
-, ,1, , ,,
,110 .il
,, . i o ., ,,
p kal[u in ,, ,,
y g. ,. ,, . n ua a .0 nu u
, , . . . ; , .'. o o n, _-
..,3 6_--- ---- J _
39
Myoxocephalus spp.
Larvae
- i. 7 6
10 i ^
5 ' ~
=:.
l , . , . . . . . ,
c, 4 1 5 2 s 1
] ol j2 ,
d
' l Om .
. n m r i 75 77 79 31 83 85 87 89 91 93 95 97 76 78 80 82 84 86 88 90 92 94 96 0 00) - - - - - - I'"
J FMAM J J A 3 OND M- s (GEgh/Lew **1997) 6~A- .-
- ind.a ni d e
-c -)
I Liparis spp. 1 1
Larvae 100 2$00 to
~"
3; }
.) 33an
[kp , ,.0 01 :$
< lI ===
- 0 01 ; .
p l l' h .
75 77 79 81 21 81 57 89 91 93 95 97 Bd 76 73 to 82 84 56 88 90 92 94 96
, ,E %V # .
y,,
J F M A M J J A 8 0 N D 3 g, y , 1 Manih p Anthmanc mouns WGemsneev j (DHgh/tmr #1997) '
40 i
l I
l Tautoga onitis Larvae 30 iam 300 i
1 Oi
]
200 , ~~ b 0 01 n\ . . l l . .
n ,, ,, ,9 .1 ., ., ,, .9 9, 9, 9, ,,
16 73 80 82 84 86 3e 90 92 94 96 0 00 t ._ . . . -
Year 1 F M A M J J A 3 O N D _
M*4 pArdmene snown .c - sneer (IEW e.1997)
Tautogolabrus adspersus Larvae 1000
'::L;::
25.901 100 ; i ri i ii i p!g;' "
10 10 iit i 8 ~
qqg:t nue n -
i i . . . . . . . .g.,,
6 - - - -
g i *'
i*"qgyp "qpp! '"g;
- t A - 4
' J '-
- l 0l e ni i t ii 2
00i t i >
l <; Ik N '
O]
t l
l Ah .; bw 7, 77 79 85 B, 85 87 89 91 93 9, 9, N
0j_$
s 16 78 50 82 84 86 B5 90 92 94 %
0 001 . Year J F M A M J J A 5 O N D
,_ . 73 M-. 6 ._..c __._J 41 l
Ulvaria subbifurcata Larvae 100 IM 2000 10 n 1500 , ~ ~
i l 8 '"
g r8 l i
,1 ,, ,9 Bl t, 81 87 89 91 9, 93 9,
>6 ,s 80 32 54 e6 all 90 92 94 96 0 001 . Y
J F M A M J J A S 0 N D s
M 6Abumience inden tened on- .-. u- -eeJ
<-n .o ,99,3 l
Pholis gunnellus Larvae 100 6 s'
- - s 10 b
$ "$~
djf '
W.%it l
i I
gp ,
, = .
N.- 2 l 0I ,
'pg
& I m3 , ,
j 0,,'",,
>9 .1 ., ., ,, .9 91 9, 9, 9, ;
z 6 ,s so s2 u a s 90 92 a m i i M
, y, l J F M A M J J A 3 0 N D ggg y, Month p Arnhnwee means MC _n (C3HyWLow se190,)
42 1
Ammodytes spp.
Larvae i=0 i0
- 2::: . .:.Z:: 262,6,,S 1 l i
- i. unm m en:m, n 4 ,.nno.e e
.. uy Im <
( l .9i to n, . , ,in j -. j g I ' ' ' ' " '
4 . . . . . .
.seae.
l l l s _ :.[55!
0 01 : noi _$ sh
, , , , > . . . o ., ,, , ,, ,, ,,
. . . . 26 28 .0 82 84 .6 .5 .0 ,i,2 96 J F M A M J IA S O N D t.
_ . ,3 6 ._..<____J Scomber scombrus Larvac 100 to
- ::. Ilon6 10.( 10 10 _
- , .;;J.:.
l I f6 1
=
- .s I
8
$#i($ i .
3 j<
b 0 01 i[j
~.... .
~
, , on n .
rL !
- I I L n. hP-. l l
$ ,s ,, ,, si is e, , ,3 ,, ,,
a 26 2 . .3 .6 . . ,i,2 .
"' ~
J F M A M J J A 3 O N D c._
i 6 ._.. _ _ _.]
43
l l
1 i l l
l 1 1
i Pleuronectes americanus Larvae iOO >=
10 >
2500 - - - - -
~
II i J:P:! i 1100 - --
= .
sao -
0 01 $7 75 77 79 sl 83 85 57 39 91 93 93 97 e 76 78 30 52 84 96 38 90 92 94 96 l 0 001 -- - --
Ye.r -
3 F M A M J J A 8 O N D
--- 6- .- --- J <
[GH@ se1997)
Total Larvae 1000 60 100 n i e, y))
l SF um,_VMN$j
~' Mg~ .,M.$ ind% E 1
N '
y
~
70$P Mj$F@33y g
~
I .' ' ' '
M y{$#
2 n@ 20
- S 01 ; i s, ,, i mgt, @M*
M 30 e o
>[ $ i i g
l 0 01 2 m rppn n , ,i l g l
'l= 75 77 79 BI 83 55 37 39 91 93 95 97 fh^ 16 78 to 82 84 36 83 90 92 94 96 0 001 E yw J F M A M J J A 8 O N D tm,._
M--
,i 6- ---- J
/4
v o L L E N
l a t c L L
- n a
c O
/
E L L /
E L L L e
g r
5 a t o L L L L h
c c /
L / / /
L L /L L L s S E E E E E id n
o g L L L L it a u L L / /
/ L L L / L t A E E E E S
r e
p P w
o r
a ly J
u L
/
E L
L
/
E L E L
/
E L L E L l
e c e u n L L L L L N u /
E
/
E L /
E
/
E
/
E L E L J
i m
r lg y i
P a L L L L p
/
ht e M E m
o r
r L L Ae L f L L / / L L L L L L s E E n
iot c r e a l
l L L E L L L L L L o
c M n
o t
k b n e L E L E L L L l
a F p
o y
ht h n c J a L E L L i
i n
d e
i n
a t
b s o u
)
s u s m e a su L s u
i r
i r d u s u n osn : a ps
( a e co s t
a 's e a os n
n n b s e a n s r m a u a i u i r y a
v t
a r ra gu a i dr c a l i
n om rne p .
ig c c s . us sp c t s h r
r a t y n u b i t r i
r u p l m
i t
s c l s t o a a e
r p . o s u h r s u s
i v
s sp s a e f p m m u ss hp e c u s a e s a n i t s d
r p m p o u d u n as h u e d l s u n a ri h s t a 7.
t o a a us oy mic c gah i iy io r t c o rip a n i ir mo p
) 9 l
l iu ov e po h r l e he s u o c h d u a ot o t o is r h e o t i h i n gn o x v c c u a p r o E9 l r o s a o t g e u c m c ds r n a e fic o r p o y r f> L I. p c ne en l
(
s 1
n r l n l o h p n i gr e gb A B C A O E G A fA P U O L S P A A A L L C S i t e s hs ecm i
e ice f e p fo D - S s yr n g ea d e
g i n h n n i i
c u a l d l c is ip lp e h nr lt k o f pa n c e- s e i e c e lu u h s en l
o p c cs a SJ e r m r d m k is s dr c o k ea e s ip s b n mh o u hs n nr f
ns l
gw n ian e as
. 1 2
ia c c hr i ck c ciebbhh y or o l e c i c r na na h bo t it b o ea i t
n e n o e t d f e e h r i o g t ia n s k e n r ip s b r p n a ou lta v a o ak t r f c l&
T ml t lt c i l
i l
t l
l t
r o
o o r a e
ur no oh a se lu l a uc A A A A R F A S A P H S G N S G L S S G B S I
1
!: 1Ill c
e L D L 4 v
o N L 4
t c L 2 O E L L L /
E 1 t
o L L L 7 S
c E L L L /
E L /E /
E 1 r L L u E L L / L / L 6 A E E 1 y L L l
u E L L L L E L / E L / 0 J E E 2 e L L L /L L n
u E L L L L / / / L/ 9 J E E E E E 1 y
a L L L /L L L L 4 E /
E E
/ E / E M E E 1 r
p L L L L 0 A E L L L E / / / /
E E E E 2 r
a L 3 L L L /
M E 1 b 0 e L L E 1 F
n J
a L L E 7 s
s s e s u u d s i u t s
s r l a o o s su u s us su go es an l
e c p a a s u g s t a cr u m on o cnp s t d a m s u l i a cr s s u h o l q y e is us fiu ud e . b r t n t s b s a su ed a ss u sm it r i n b b l
e o sp m ca o ro s l u sa o e en i h s co o al b a o su unn a nt et s ir a cmh t
i y mh t
l a ep os ec g t
i e
a gogo a g c d e u y r s s h c g n u l
u c h u u ar s ap mo bmi pr o ar poh p po o o r t
d i t i u ee r r i
t t i t l t r l h r y m o p r b a a v o y p s U P C A Sc Pe E P aSc G lf P Pf.
a T T l s l i l e L ic e
s p i
e s c f e o p r r
)
S e r e d d e b e l e n r d m u y u e e c er un u r
n n e o d r n k l n e i N o n n p u dn o u e d a i f t
n a l c ht l l o hs en h e a f c n c mh su l
f a cu n p o l i
( t o l a
s g e red u gu on a m o n a t
e 1 f t
2 e s o n t i
1 r p o 1 c r w l
e s a a t
u n i u a n
kc ymd n i l t n e a t l
l a s d h c e ir r t e
n o
W T C n B R WS a A B uSmFo WWA WY b
T a r o r l t
t u i n t i mie l l
1 l
j I 'i & 8i .
j a i n 2 8. ). !. !. 4 2 182 u. 5. g ig 8 1 8. z a 4 , 5. .
- \
I ' dd d d dd dd "dd d
,-, . ,,.,. , e -
J ,
$'" " 6$ $d 6 a f $
i E
- s. 3e ss a s s . -
A
{ j da s - s dd da dd d -s i -
t 2
~
$""- tt t "t d a ~
I l ,,ga a - gs-ga a g g a
, , - - - , , , , --, c -
1 i g, s. . 33. s . .
I e-- ee s-ss 3 ! - - -
j gs s - a gg gs s s . .
i - a a gg ga a - g a l : e-- - -ee-e- e e e-e
$ E da s ds =dd da $ $ -
! $s s - s-d$ da d d d d j n,
~
- s. ese e se . s ,
h ,,s l
, g ,
s . , , _ , .,.
j j ,sa s - s ssssa $$ $ -
g j i
,a da s - a ddddds - d ss - s -
a s, s. . sseee ee - - - e !
l i :-,
~
- s. . ees s se-e e sg fta s <i
' I
}! i , 'I 11{ I I
, n4ja,i;Jj}{}'(jIc1m,e1!fjij1aj 4
3-
- i . a -
47
i l Ii)iI } ji]
p p y r e t t e S e t n 2_
S lg M
t r
M t
e Mt e
dJ M 4 e A t
p
!g u
o o is c
lg c
e e
J l
u u p r p e r p e r a
.tc a p r ru ru s u u n nu p r e O n u
A(
- M A
(
A t A
(
p A A p
Mt J J t
M( M e
s.
M-A u A(
A
( A(
J u J
( At A p
C r -
d e
y n - ty
_ t e p J r u
b t
r e
t r
e t
r b r a p t
e p
l e
p t
b r u e in e n a
t r
l b be d
A J (y
F e
m m
MM @ A e F e
M A A(
At e
F(
M e A J M Mr t u
J ru J A p e Ft F
o e e n r a n a Jtv ty <
n r e re r e a e M o e e e e' M D J J J M J MM F J P -
7 9 L l E L L L L L L L L L E
L L L 9 L L 1
8 E 9 L E L L L L L L L 9 L L L L L L 1
5 9 L E E 9 L L L L L L L L L L L L L L L 1
4 9 L E E 9 L L L L L L L L L L L L L L L 1
3 9 A E 9 L E L L L L L L L L L L L L L L L 1
2 9
9 L E L L L L A E
L L L L E
L L L L L 1
1 9
9 L E A 1E E
L L L L L L E L L L L L 1
0 9 L 9
E L L L L A L L L L L E L L L L L L 1
E 9
8 L L E l E L t E L L L L L L L E E L L L L L L 19 0
3 L 9
t E L L L L L L L L L L E L L L L L 1
7 8 L 9
t E L L L L L L L L L E L L L L 1
0 8 L A E L L L L L L L L L E L L L L 9 L L 1
5 8 t l E 9 L E E L L L L L L L L L L L L L L L 1
4 8
9 L t L L L L L L L L E E L L L L L L 1
3 8 L 9
A A L L E L L L L E 1
E E L L L L L 2
8 L E 9 L L L L L L L L E L L L L L 1
1 8 L 9
l E L L L L L L L L L E L L L L L L 1
0 8 L A L L L L L L L L E E L L L L 19 9
7 9 L E L L L L L E L L L L L 1
8 7
9 L E L L E L L L L L 1
7 7
9 L L L L L E L L L L 1
4 7
9 L E L L L E L L 1
L L 5
7 9 L A L L L L L E 1
E L L L L L s
s s '. u s .
o se s u m s r
re s c s s e v e
p p n s u
u e
n o
p p J-_
s u fe o i.. d s "- e e e c p n a - p e s s e ;e re m u s r _ = h y
- r e s n u e . es e e r - s s y n e vr m s h
- u e p _ g 1 d
_. su f A s c ir h o p u s u s e t r m d o re -
n o r u __ s -
es r s e n 6 A e _
'.. ip c a e h
(
c ^
s e
s u
o rp c s
s a t r
o " o u
r p
p o
e sn t
n e nm a n r
u hr e
p
- p s ue s b
a s
n e
e a s.c p 3 ,
rs f. r e s rp ar h e o w o o o . c e . e f
n s o o s re c e e o dt
. w A n e c r o 'e d s u fo y ne s r e
- 6 e . a e oy M M. M
, c y
a e L L t
n e
n n se b a e e e p
=ra ev y
S, y re T T T S . H M A C L C C S s A L e % r C
e* L t 7 i 1llll ,
Il 11)l1IillI jl l Ilji1 I )j I l ll l )lll4llj]
e, n p t v , g o u e e o u
. J p s o tp mi %e A %
)y i
t u t w t e
n d n p g s p t u e c e c
u e u O. -Oc - wO c S p Jru ru Jt e
m u
J u C t.
S t
t
- s. A y
f o re t l u ev o
y e lu p e o v tv o
ee t r
a l
e p
t e
r a e lu J M J S M A p M M J d
e 4 M Mt r M( Mt M 4 4 A
t n n p r p
r a n n b e e e e e P J A A M J J F 7
9 L A L A E
A AE A EA AE A 7 9 E E E E 3 1
6 9 L A L t L A AE A EA AE A E 8 9 E E E E 3 1
5 9 L 9 L A E
AE A E A EA A E AE A E A EA E E L 2 4
1 4
9 L 9 L t E
L E A LE A E A A E E AE E 0
4 1
3 9 L A L E A L i A AE A EA AE 8 3
9 E E E E E 1
2 9 L 9
A E
E L i
E A AE A EA AE E E 6
3 1
1 9 L 9
A E
AE E A F
AE A E A EA AE E E 4 3
1 0
9 L A L L A AE A E A EA AE A E
2 9 E E E 4 1
9 8 L A AE E A AE t. A AE A A E
0 9 E E E E E 4 1
0 8 L 9 L A E
AE E A E
AE A EA A E AE E E 1 4
1 s
7 8 A A AE A EA LE A E A 6 e .
v 9 L E E 3 r 0 E E E E Be 99 1
f o 1 6 ; 5 8
9 L A A A AE A EA AE A E E E 7 3
7 E E E _
9 1 1
. r 5
8 L 9 E A L E L A E
AE A EA AE A E E L 2 4 - fosn _
1 e a _
h t
L o _
4 n f 8 L 9
A AE A AE E A AE E 3 4 o o E E E d s 1 e n e e ic 3 b ep 8 L A AE A AE A E A EA AE A E 7 n s 9 E L E E 3 wo 1
how t s
2 s d n
8 L 9 L A E
L A E
AE A EA AE A E A E E E 8 3
e a d 8 1
e r 7 r
u 91 1
c 8 L 9
A AE E E A AE t EA AE A E E L E 0 4 co 75 1
e 9
- r. w 0 b e oh r o 1 8 L 9
A AE E E A AE A E A EA AE E 5
3 mto f e d 1
c f e e
9 D : mu 7
9 E A L
/
E E A E
AE A Et AE A E E 7
3 b r
e s e
-- es
_ 1 m
e e e er v
r e
_ 0 o e w 7
9 L A E A AE A A E E AE E E 5 3 N w e E d y ev 1
_ n lo
. a h h yt. e c 7 y p 7 A AE A AE A t A 3 e 9 L L L E L E E E E E 4 M Re n 1
4c c 6 ouy e
r c M pef M e
6 7
9 L A AE E
A AE A EA A E E AE E 6
3 s o 1
= d es e ie e
_ M 9 L L A AE L A E
A E
AE AE A E A 1 4
k e
p od s f
a ne pec E E 1
e ito er s *_. n ht s e u f m _. e se s t r
e s o p s
s a
eg s s
e
- e. e v "s S e w osp s s u s o - s re o
e e r
. ir u re c d g u uhs mo
) s r e n .
u n g p w u re u g
y e
ap -
r tu c
e p e o 6 r e we
- s e t n e s te s S t n e s e u s a
p em u es ne p ts _
_- o e o d s u e c ni f o
e r u d e w s u s g h
_ s s e e e s
e s
e w
ey 'su e n h a r e e er n a
r e u t t
e
. e r p r e p s e n ro A h p c
i m
o e *r p n hr e 6e e y re s
e 6r . gm s c t 3 e f_
n s
kn s
up s c h
so ap o o n o u
r r
n r c
- ru p ne uo oc e e le S o o b so p r u e fu p N - bs h eo b
a
- 6 o
_ p a r r e
o oc h y o f r e t t
T A G O P
_ f E P P. S G N M P. P. T i
G 'J 'A ' A* F pC
Table 4. Densities per 100 m' of water for tautog/ cunner eggs taken with 0.333 and 0.202-mm mesh netting, 1994 1997.
Mesh Date Rent!~, 0333 0202 Ratio o' 1994 May 4 1 2.9 16.1 5.55 2 3.2 9.0 2.81 3 5.3 4.4 0.83 May 9 1 1.1 3.9 3.55 2 4.7 4.9 1.04 3 1.8 2.9 1.61 July 21 1 1194 1330 1.11 2 1028 1462 1.42 3 1377 2259 1.64 August 8 1 134 110 0.82 2 134 172 1.28 3 134 152 1.13 0.02 Geometric mean 33.3 52.6 1.58 95% confidence limits 1.08 - 2.30 1995 June 16 1 1364 1959 1.44 2 1405 1514 1.08 3 1609 1299 0.81 June 26 1 386 675 1,75 2 631 675 1.07 3 515 570 1.1I June 28 1 17447 17658 1.01 2 16432 24925 1.52 3 21671 26357 1.22 0.05 Geometric mean 2375 2857 1.20 95% confidence limits 1.00 - 1.44 19 %
June 19 1 1959 2150 1.22 2 1739 2128 1.22 3 1382 1351 0.98 June 24 1 3637 4123 1.13 2 2572 3413 1.33 3 3865 2782 0.72 4 2893 3637 1.26 July 1 1 871 1092 1.25 2 495 850 1.72 3 959 794 0.83 July 5 1 4168 4388 1.05 2 3118 3%3 1.27 0.09 Geometric mean' 1928 2176 1.13 95% confidence limits 0.98 - 1.30 50
Table 4 (continued).
Mesh Dale Replicate 0.333 0.202 Rctio o' 1997 June 13 1 775 987 1.27 2 762 733 0.%
3 506 299 0.59 June 18 1 1081 1120 1.04 2 2087 2636 1.26 3 1684 1843 1.09 June 27 1 2106 2718 1.29 -
2 1565 1872 1.20 3 1644 1660 1.01 July 2 1 1371 1645 1.20 1 2 1145 2252 1.97 l 3 3244 2709 0.84 0.27 Geometric mean 1332 1467 1.10 95% confidence limits ' 2.54 - 3.87 Pooled 0.0006 Geometric mean 617 766 1.24 95% confidence limits 1.09 - 1.41 l
l 51 a
l 1
Table 5. Densities per 100 m' of water for stage I cunner larvae taken with 0.333 and 0.202-mm mesh netting, 1995-1997.
Mesh Date Replicate 0 133 0.202 Ratio o' 1995 June 16 1 59.7 25.0 0.42 2 30.7 18.4 0.60 3 69.3 39.7 0.57 June 26 1 0.6 5.4 9.82 2 0.8 7.3 8.80 3 0 0 -
19 %
June 19 1 8.4 2.2 0.26 2 10.2 26.8 2.63 3 5.4 12.7 2.35 June 24 1 3.0 5.8 1.93 2 3.0 2.1 0.70 3 4.I 3.1 0.76 4 5.0 9.7 1.94 July i I 20.8 5.0 0.24 2 13.0 15.9 1.22 3 14.6 5.1 0.35 1997 June 13 1 2.0 4.3 2.15 2 5.5 5.1 0.93 June 18 1 36.6 13.1 0.36 2 27 114.3 4.23 3 41.2 102.6 2.49 June 27 1 6.4 6.5 1.02 2 5.3 4.6 0.87 Geometric mean 8.54 9.90 1.16 95% confidence limits 0.73 - 1.84 1
n 1
)
i
1 Table 6. Densities per 100 m' of water for stage 2 cunner larvae taken with 0.333 and 0.202-mm mesh netting, 1994-1997.
W Date Phaa 0333 0,202 Ratio o' 1221-July 21 1 1.1 7.8 7.09 L922 June 16 1 56.8 60.0 1.06 2 36.3 12.3 0.34 l 3 72.2 34.0 0.47 June 26 1 16.6 43.5 2.62 2 56.2 90.7 1.61 3 85.9 36.3 0.42 l . June 28 1 1.5 10.4 6.75 2 4.5 14.0 3.13 3 14.4 0 -
12 5 June 19 1 1.8 2.2 1.22 2 1.7 6.7 3.94 3 2.7 33 1.22 June 24 1 3.0 2.9 0.97 2 6.7 2.1 0.31 3 *!
.. 1.8 0.44 4 2.8 6.8 2.43 Julyi I 35.4 24.9 0.70 2 39.8 39.3 0.99 3 40.2 41.I 1.02 July 5 1 1.3 4.3 3.31 2 10.9 11.6 1.27 1997 June 13 1 6.0 10.5 1.75 2 22.7 14.2 0.63 June 18 1 62.9 67.2 1.07 2 54.1 95.9 1,77 3 77.6 100.6 1.30
' June 27 1 2.6 1.6 0.62 2 3.7 3.2 0.86 3 0.8 3.8 4.75 July 21 1 10.2 14.4 1.41 2 9.3 8.8 0.95 3 9.6 13.6 1.42 Geometric mean 9.67 12.39 1.28 95% confidence limits 0.95 - 1.72 53
APPENDIX A*. Densities of fish eggs and larvae per 100 m' of water recorded in the PNPS discharge canal by species, date, and replicate, January-December 1997.
- Available upon request.
i
APPENDIX B*. Geometric mean monthly densities and 95% confidence limits per 100 m' of water for the dominant species of fish eggs and larvae entrained at PNPS, January-December 1982-1997.
Note the following:
When extra sampling series were required under the contingency sampling regime, results were included in calculating monthly mean densities.
Shaded columns for certain months in 1984 and 1987 delineate periods when sampling was conducted with only salt service water pumps in operation.
Densities recorded at those times were probably biased low due to low through-plant water flow (MRI 1994).
- Available upon request.
I ICHTHYOPLANKTON ENTRAINMENT MONITORING AT PILGRIM NUCLEAR POWER STATION JANUARY-DECEMBER 1997 4
Volume 2 of 2 (Impact Perspective) i Submitted to Boston Edison Company Boston, Massachusetts I
by Marine Research, Inc.
Falmouth, Massachusetts I
April 1,1998 l
i I
i
TABLE OF CONTENTS SECTION PAGE I EXECUTIVE
SUMMARY
l II INTRODUCTION 2 III IMPACT PERSPECTIVE A. Notification Plan 3 B. Ichthyoplankton Entrainment - General 3 C. Ichthyoplankton Entrainment - Specific 7 IV LITERATURE CITED 19 LIST OF PLATES PLATE I Plankton net streaming in the discharge canal at Pilgrim Station for the' collection of fish eggs and larvae (lobster larvae are also recorded).
A single, six-minute collection can contain several thousand eggs and larvae representing 20 or more species. i LIST OF FIGURES
)
FIGURE PAGE I Estimated numbers of fish eggs entrained at PNPS by species or group,1997. 8 2 Estimated numbers of fish larvae entrained at PNPS by species or group,1997. 9 3 Numbers of equivalent adult winter flounder estimated to have been lost to entrainment at PNPS, 1980-1997. 22 4 Numbers of equivalent adult cunner estimated to have been lost to entrainment at PNPS, 1980-1997, 23 5 Numbers ofequivalent adult Atlantic msckerel estimated to have been lost to entrainment at PNPS, 1980-1997, 23 i
i
l LIST OF TABLES TABLE PAGE I Ichthyoplankton densities (number per 100 m' of water) for each sampling occasion during months when notably high densities were recorded,1997. 4 2 Numbers oflarval winter flounder entrained at PNPS annually, by stage,1980-1997. Number and weight of equivalent age 3 adults calculated by two methods is also shown. 24 3 Area 514 commercial landings and Massachusetts recreational landings from ' Sand waters (pounds), 1982-1996. 25 4 PNPS area fish trap collections by age class determined by age length key, cunner, 1992-1997. 26 5 PNPS area fish trap collections weighted by overall effort, cunner, 1992-1997. 26 6 PNPS area fish trap collections weighted by mean annual catch, cunner, 1992-1997. 26 7 Instantaneous mortality rates for cunnu collected in the PNPS area. 27 8 Calculation of average life-time fecundity for PNPS area cunner. 28 9 Numbers of cunner eggs and larvae entrained at PNPS annually, 1980-1997. Numbers of equivalent adults are also shown. 29 10 Calculation of average annual fecundity for PNPS area cunner. 30 11 Densities per 100 m' of water for mackerel eggs taken with 0.333 and 0.202-mm mesh netting at PNPS, 1994-1997. 31 12 Densities per 100 m' of water for mackerel larvae taken with 0.333 and 0.202-mm mesh netting at PNPS, 1994-1997. 31 11 Numbers of Atlantic mackerel eggs and larvae entrained at PNPS annually, 1980-1997. Numbers of equivalent age I and 3 fish are also shown. 32 l
l ii 1
SECTION I EXECUTIVE
SUMMARY
Unusually high entrainment densities, as defined under PNPS's notification plan, were identified on a number occasions in 1997. These involved Atlantic menhaden eggs and larvae as well as the larvae of sand lance, hake, tautog, cunner, winter flounder, and Atlantic herring. Menhaden displayed the most protracted period of.%,h densities with unusually high numbers being recorded each month from June through August along with October and November.
Estimated numbers of eggs entrained by PNPS during 1997 ranged from 2,800,000 for searobins to 1,718,290,000 for tautog/ cunner and totaled 2,390,190,000 for all eggs combined.
Corresponding values for larvae ranged from 8,1%,000 for seasnail to 106,912,000 for sand lance, totaling 694,854,000 for all larvae.
Entrainment of winter flounder, cunner, and Atlantic mackerel, was examined in more detail
, dating back to 1980 using the equivalent adult (EA) approach. Winter flounder estimates for 1997 were 3,414 and 47,087 age 3 adults bed on two suites of survival values, the highest yet observed because larvae were very abundant at PNPS throughout the 1997 season. These values were compared with estimates ofcommercial and recreational landing as well as local population estimates determined by trawl and mark-recapture. Recent, dramatic declines in commercial flounder landings i
reduce the value of that variable as a measure of EA impacts. The 1997 EA estimate from the larger
. staged approach amounted to 14.6% of the local area swept population estimate and 9-41% of j several mark-recapture estimates. A respective EA estimate for numbers of cunner eggs and larvae entrained in 1997 amounted to 498,281 fish. Comparable values for 1980-1996 ranged from 119,342 :
to 2,571,973 adult fish. The mean cunner EA value for the time series (500,592) represented less than one percent of an estimate of the number ofcunner in the PNPS area. For Atlantic mackerel EA j.
estimates of 1,712 age i fish or 1,103 age 3 fish were obtained for 1997. Average values of 5,489 and 3,535 age I and 3 fish, respectively, were obtained over the 1980-1996 time series. Each of these values represented less than one percent of the commercial mackerel landings for area 514 which ;
I encompasses Cape Cod Bay and Massachusetts Bay. )
I
l 1
SECTION 11 INTRODUCTION This report focuses on the potential impact of ichthyoplankton entrainment at PNPS.
Discussions are based on results presented in "!chthyoplankton Entrainment Monitoring at Pilgrim Nuclear Power Station January-December 1997", Volume 1 - Monitoring. Work was conducted by Marine Research, Inc. (MRI) for Boston Edison Company (BECo) under Purchase Order No.
LSP007550 in compliance with environmental monitoring and reporting requirements of the PNPS NPDES Permit (U.S. Environmental Protection Agency and Massachusetts Department of Environmental Protection). In a continuing effort to condense the volume of material presented in this and related reports, details ofinterest to some readers may have been omitted. Any questions or requests for additional information may be directed to Marine Research, Inc., Falmouth, I Massachusetts, through BECO.
Plate I shows the ichthyoplankton sampling net being deployed on station in the PNPS discharge canal approximately 30 meters from the headwall.
l 1
2 l
., t
- t.-
vw :- ,
Yhti - -? *
. -P
. R . J. ,
~ .. & d,
~
~~
,, . 4M ., J-1 o[.' ,.
w g.-
.4 ,-
/
s
~ .
+ m'
- y ,
s ,
mer%,g Plate 1. Plankton net streaming in the discharge canal at Pilgrim Station for the collection of fish eggs and larvae (lobster larvae are also recorded).
A single, six-minute collection can contain several thousand eggs and larvae representing 20 or more species.
SECTION 111 IMPACT PERSPECTIVE A. Notification Plan Ichthyoplankton densities reaching the unusually high level during the 1997 sampling season occurred on a number of occasions. These involved Atlantic menhaden eggs and larvae as well as the larvae ofsand lance, hake, tautog, cunner, winter flounder, and Atlantic herring (Table 1). Since winter flounder densities recorded in March and April were based on 0.202-mm mesh samples they were scaled downward using mesh conversion factors for stage 1 and 2 larvae to make them comparable to past 0.333-mm mesh based densities (see Volume 1,Section IV). Among the above species, Atlantic menhaden displayed the most protracted period ofhigh numbers with unusually high densities of eggs and/or larvae being recorded each month from June through August as well as October and November. Menhaden egg densities exceeded all previously recorded densities on three occasions in June. Menhaden larvae exceeded all previously recorded densities on five occasions in August, two occasions in October, and once in November. Since menhaden eggs hatch in as little as two days during the summer (Kuntz and Radcliff(1917), the decline in high egg densities after June while larval densities remained high suggests spawning continued through the summer but at some distance away from PNPS. Although we did not measure larval menhaden, qualitative observations suggest relatively young larvae continued to enter the area presumably from protracted spawning. Hake larvae were also abundant for an extended period of time, beginning in August and lasting into October. In September densities were particularly high, exceeding the previous high density for the month (122 per 100 m') on two occasions. For both menhaden and hake larvae 1997 was in general a productive year (see Volume 1).
B. Ichthvoolankton Entrainment - General ichthyoplankton entrainment at PNPS represents a direct negative environmentalimpact since fish eggs and larvae passing through the station are subjected to elevated water temperatures, shear forces, and periodic chlorination. In effect PNPS operates as a mechanical predator increasing overall mortality rates in western Cape Cod Bay. When PNPS is not on line, elevated temperature is not a factor but fish eggs and larvae may still be wbjected to mechanical forces and periodic chlorination 3
Table 1. Ichthyoplankton densities (number per 100 m' of water) for each sampling occasion during months when notab;fhigh densities were recorded, January-December 1997. Dt.nsities marked by + were unusually high based on values In Table 1. Number in parentheses indicates percent of all prnious values during that month which were lower.
Winter flounder larvae' Sand lance larvae March 14 0 May 2 124.6 + (93) 17 0 5 40.1 19 0 7 53.2 21 1.6/1.0 9 80.3 + (92) 24 0 12 18.8 26 0 14 16.7 28 0 16 57.9 31 16.2/10.0 + (100) 19 26.0 21 10.6 l Previous high: 8 (1982) 23 24.3
- Notice level
- 1 26 1.9 l Note: no sampling prior to 14th 28 0.6 due to plant outage 30 2.0 April 2 storm Prnious high: 639 (1996) 4 0 Notice level: 59 7 17.0/10.5 9 21.8/13.5 + (91) 11 0 14 2.9/1.8 16 0 18 storm l 21 8.2/5.1 I 23 13.3/8.2 l 25 14.9/9.2
! 28 8.5/5.2 30 17.3/10.7 Prnious high: 198 (1974)
Notice level: 12 June- 2 storm j 4 3.9 l
6 84.3 + (98) 9 46. ! + (98) 11 9.6 13 103.6 + (98) 16 0.7 18 2.8 20 8.9 23 0 25 1.7 27 0 29 0.7 Prnious high: 154 (1996)
Notice level: 29
'0.202 mesh densities adjusted to 0.333 mesh. Both are shown as follows: 0.202/0.333.
l l
1 l
l
l l
Table 1 (mntinued).
A*'--*b sienbades EGGS 1.,ARVAE June 2 stonn 4 24.9 + (94) 0 6 27.3 + (94) 0 9 17.3 + (91) 0.8 11 227.8 + (100) 3.8 13 33.0 + (%) 1.6 16 13.9 + (90) 12.5 + (88) 18 2.8 26.3 + (92) 20 425.2 + (100) 4.4 23 19.2 + (91) 1.7 25 8.4 + (88) 0 27 227.6 + (100) 19.1 + (90) 30 0.7 6.4 Previous high: 116 (1994) 496 (1981)
Notice level: 8 10 July 2 0 13.2 + (98) 4 high water 7 0 7.6 + (%)
9 0.9 0.9 12 0 48.7 + (99) 14 0.8 6.6 + (95) 16 28.9 + (98) 66.I + (99) 18 0 2.5 21 dredging 23 dredging 25 storm 28 0 17.7 + (98) 30 1.6 3.2 Previous high: 59 (1978) 124 Notice level: 4 3 Atlantic menhades lanse Aug i 47.0 + (100) Oct 6 7.8 + (92) 4 dredging 8 0 6 11.6 + (100) 10 70.3 + (100) 8 19.3 + (100) 22 2.9 11 1.8 + (97) 29 56.1 + (100) 13 3.6 Previous high: 41 (1995) 15 183.4 + (100) Notice level: 4 18 27.0 + (100) 20 1.8 Nov i 57.1 + (100) 22 15.6 + (100) 10 0 25 dredging 12 0 27 0 21 0 29 1.6 + (97) 24 0 Previous high: 15 (1974) 26 0 Notice level: 1 29 0 Previous high: 24.3 (1980)
Notice level: 1 5
i Table 1 (continued).
Hake larvae Tautor larvae Aug 1 17.2 + (93) Aug 1 17.2 + (97) 4 dredging 4 dredging 6 1.9 6 0 8 81.9 + (98) 8 13.8 + (95) 11 0 11 5.4 + (91) 13 10.8 + (89) 13 12 15 1.8 15 7.1 + (93) 18 0.9 18 0 20 9.2 20 1.8 22 3.5 22 41.5 + (100) 25 dredging 25 dredging 27 0.9 27 2.7 + (80) 29 6.2 29 0.8 Previous high: 196 (1995) Previous high: 28 (1995)
Notice level: 10 Notice level: 2 l
Sept i 1.4 i 3 0 Cunner larvae 1 5 327.2 + (100) Aug i 254.0 + (100) 8 144.8 + (100) 4 dredging 10 116.7 + (99) 6 13.5 l
12 104.6 + (99) 8 93.0 + (98) 15 33.2 + (%) 11 62.6 + (97) 17 11.8 + (84) 13 84.3 + (97) 19 2.6 15 123.2 + (98) 22 108.2 + (99) 18 9.3 24 34.8 + (97) 20 72.4 + (97) 26 20.0 + (91) 22 228.2 + (100) 29 8.4 25 dredging Previous high: 122 (1985) 27 6.3 Notice level: 9 29 1.6 Oct 6 5.5 + (95) Previous high: 165 (1974) 8 4.6 + (95) Notice level: 57 l 10 10.4 + (97) 3 22 0 29 0.8 Previous high: 14 (1995)
Notice level: 2 Atlantic berrier larvae l Nov i 9.9 + (80) Dec 8 28.7 + (97) 10 13.7 + (77) 10 44.3 + (97) 12 107.4 + (98) 13 10.0 + (88) 21 20.1 + (87) 22 2.6 24 5.6 24 0 26 26.I + (90) 26 29.7 + (97) 29 0 Previous high: 217 (1995)
Previous high: 125 (1995) Notice level: 3 Notice level: 8 i )
6
4 when circulating seawater or salt service water pumps operate. Although survival has been demonstrat.d for some species of fish eggs at PNPS such as the labrids (45%; MRI 1978) and winter flounder (MRI 1982) and among larvae at other power plants (0-100% initial survival depending on species and size; Ecological Analysts 1981), entrainment mortality is conservatively assumed to be 100% in all PNPS assessments.
To place fish egg and larval densities recorded in the PNPS discharge canal, expressed as numbers per 100 m' of water, in some perspective in relation to amounts ofwater utilized by PNPS, they were multiplied by maximum plant flow rates over each respective period ofoccurrence. This was completed for each of the numerically dominant species as well as total eggs and total larvae.
Mean monthly densities were multiplied by 17,461.44, the full load flow capacity of PNPS in 100 m' units per 24-hour day, then by the number of days in each respective month they were collected.
Values for each month in which a species or species group occurred were then summed to arrive at a seasonal entrairment value in each case (Figures 1 and 2). For cunner, mackerel, and winter flounder, egg and larval totals were calculated using individual densities and mesh adjustment where appropriate as part of an adult equivalent analysis (see next section). Among the eight numerically dominant groups, numbers of eggs entrained ranged frcm 2,824,000 for searobins (Prionotus spp) to 1,718,290,000 for the labrids. Corresponding values among the thirteen dominant larval species varied from a low of 8,196,000 for seasnail (Liparis spp.) to a high of 106,912,000 for sand lance (Ammodytes spp.). For all eggs and larvae combined, values amounted to 2,390,190,000 and 702,147,000 respectively. These totals state the extent to which large quantities of eggs and larvae can be entrained by the circulating seawater system at PNPS during a single year; based on the assumption of 100% mortality all are lost to the local population.
C. Ichthvoolankton Entrainment - Specific Estimated numbers of eggs and larvae entrained annually at PNPS were examined in greater detail for three species of fish using the equivalent adult procedure (EA, see Horst 1976, Goodyear 1978, for example). Somewhat arbitrarily this review dates back to 1980 so that with the addition of 199718 years are included. The adult equivalent methodology applies estimated survival rates to numbers of eggs and larvae lost to entrainment to obtain a number of adult fish which might have 7
0 0
0 0
0 0
0 0
0 0
0 6 1 0 5 0 2 5, 0 0
7, 9 0 9 8 0 1, 0 8 0 2, 0 0 8 9 0 7 1 9 3, 0 9 7 1 7, 0, 2 9 1 3 !
0 0
1 9
4, 1
@ 0 0
0 0
d 4 7 0 e
n 9
3 g:- +
4 y
- 0 0
1 i
a r
7*
9 7
]*
N
g f 0 t
6, /
$; 0 n g' .
8 0
g- , 0 E 6 0 6 n^
9 0 s 9 0 g ~ ' M ' 0
^
4 ;
g t
g 2 E 5 1
7* M + ^
M 4 g 9 3 1, 0 0
f 4 1 9 0 '
o 4 6* 2* M 0 0
r 8* s"- 5 4 2 !w-8, f" 0 0
0 gN e 4 M* 6 -
$ 4 " o 5* 8 g' $ 1 b 2' m
r l
$g ( '
m u b' J *'2 + h ea s 0 N i-J' d
J @n#"' -
4 g 4 g J
i "a
0 0
0
)
0 e 0 c )
r r e
e i .)
pr e a )e r
" )er s u
- l)y ,
=be 1
n k" b) l
- u. pb o tc b b" b m d u ,
e s e ir o habn s m em nme d
'J er = me pt e c p. e ne "hc -
4 ir e ut ot ht a, r cu ec vt t p a e PO ol cc o.
t ope o v o cp e e t "a r o y 7 p
p o IwN aM(
pc SO it S-e ic i y p r t lyh
- imS r y P y nS ey
- 'p 9 A +a n a m
f = n u
h(
p o G. -
c Pl u J eM r oM s d
" u n
o " J ( r o eM
( d(
n c( ht y
io t
n d 8' U -
d( b S
- e io e i e h * \
e u d L c o p r e
'o i
e p le n o G o p P
( ly r
e p i o P p b f
h W j
\
c :
n c E
o c
llllii11lIii!liil l1J 0
0 0
0 0
0 0
0 l 0 i
3 1
9, 4
5 .
8, 7 4 9 9 9 6 o 1
- o ,
- o. p 7 l o o u 9 l l
l l
0 o. o r
5'- 7 7 g 9 3 0
0: 7, m r 1
]- A6 4
y 0 1
1 l
o
- t- g 1 9, 9 s 1
3, , 5, 6 p 3 e d ] M 8 9
9 7
f-9 0 3 i
c e
e 1 8 ;1 l 7
g 40 p l
n
@%g 2 s i- 6 i y t
a r g 7
g 5
~
+
g l
7 '*
0, 8
7 4
y
^ 0 0
0 b
S E
n e h s
+
@ n 4
g- ic g"-
[
j p
~
~
]
g g
p<
5 9
3, 8
3 J
E n E-7, 5
@2 s r
~
N 0
0 0
o 1
P N
P g
i
- 1 g < -
t a
a s 5-2 8- -
+
g v
r s
y $" g 9 \
- 1 1 d e
a ' E s
i n
L $
E - a r
f / !
j 3'
o t
n e
o ..
${,f;$ >
) ~ ^
s o
- c. e r = ($ ^s s.
p (p
~
^
o o a e ;g o. vr b
sj i
a g '-
l m , h
< s u -
~
i f
N ^
^
i e g
+ . '
f o
s r
e o b 2
f Q 9 g { g H o z_ 4
$ $. ' ,~-
- o. m u
)
e ~ - _ - _ _' _ \_ - 4_ _ - - o o n s c e n t
) -) ) )
= )y
- P' )r P )e ie .
. Py n a y a = d c r ...iTh i a e
' k T a Tunbem r ~ n e
pcu M J a t tmM**M$3 gi a
- o m y- l y - -
y L _
S c m7Oi r r gt p r r O "P "N -
8' -
n a
n %i Apes
( aa-e i-na u4u a a n
, a m
it m
TeW'u a T f
o " PyWJ
' m u J
( T nu o *,t= J ' (
a J d= s d
7i h
(
(
"7 J n( :
E o
i r
e
~ M
(
W "" ?- -
- 2 P w e
( T r- r u
ig F
c _
' ! lll, 1'
entered the local population had entrainment not occurred. The consequences, ifany, of the loss can then be considered if the size of the extant population is known or numbers can be compared with commercial or recreational landings.
Many assumptions are associated with the EA procedure. The fish population is assumed to be in equilibrium, therefore in her lifetime each female will replace herself plus one male. It is also assumed that no eggs or larvae survive entrainment and that no density-dependent compensation occurs among non-entrained individuals. The later two assumptions lend conservatism to the approach. As pointed out earlier, numbers of eggs and larvae entrained were determined using the 1
full-load-flow capacity of the plant. This value was used even if the station was out of service and less than full capacity was being circulated. In those cases the adult equivalents are conservatively high.
Since plankton densities are notorious for deviating from a normal distribution but do generally follow the lognormal, geometric mean densities more accurately reflect the true population mean. For data which are skewed to the right such as plankton densities, the geometric mean is always less than the arithmetic mean (See Volume 1, Figures 3 and 4). In calculating total entrainment values for the adult equivalent methodology we chose to use the larger arithmetic mean for all sampling dates preceding April 1994 when three replicate samples were taken per sampling occasion to lend additional conservatism to the assessments. Beginning with April 1994 each individual sample density was utilized so that no averaging was necessary.
In summary, four opportunities were chosen to overestimate the impact of PNPS
- All eggs and larvae were assumed killed by plant passage regardless of thermal load.
- No density-dependent survival compensation was assumed to occur.
- PNPS was assumed to operate at full-flow capacity year round.
1
- Mean entrainment densities were overestimated by the arithmetic mean for sampling dates l when three replicates were taken.
The three species selected were winter flounder, cunner and Atlantic mackerel. Flounder were chosen because of their commercial and recreational value as well as their importance in PNPS ecology studies. Cunner were selected because they are abundant in entrainment samples and in the local area and PNPS finfish studies have been focusing on that species which appeared to be in a lo
1 declining trend from 1980 to 1994 (Lawton et al.1995). Mackerel were included because they are abundant among the ichthyoplankton entrained, both eggs and larvae being removed from the local population, and they are commercially and recreationally valuable.
! Winter Flounder The annual larval entrainment estimates were converted'to equivalent numbers ofage 3 adults, the age at which flounder become sexually mature (Witherell and Burnett 1993, NOAA 1995).
Numbers of stage I and 2 larvae collected prior to 1995 were scaled upward by 1.62 to correct for mesh extrusion (MRI 1995). Two sets of survival values were used. The first set followed NEP (1978) using data from Pearcy (1962) and Saila (1976). Briefly, this consisted of dividing the total number of entrained larvae by 0.09 to estimate the number of eggs which hatched to produce that number oflarvae. The number of eggs was then multiplied in succession by 0.004536, an estimate of survival from a newly hatched egg to day 26; 0.2995, survival from day 27 to metamorphps; 0.03546, survival ofjuveniles from 3 to 12 months; 0.3491, survival from 13 to 24 months; and finally 0.33, survival from 24 to 36 months. The second approach followed larval stage-specific survival rates (s) derived by NUSCO (1993) as modified by Gibson (1993a). These are as follows:
S (stage 1) = 2.36E-01 S (stage 2) = 1.08E-01 S (stage 3) = 1.54E-01 S (stage 4) = 6.23E-01 S (age 0) = 7.30E-02 S (age 1) = 2.50E-01 S (age 2) = 4.77E-01 In using the stage-specific rates it is recognized that NUSCO employs different morphological stage criteria than those used at PNPS. However a comparison ofsamples from both studies showed stages to be quite comparable until larvae approach metamorphosis, a size not often collected because these individuals begin to take up a benthic life style. Although small numbers are entrained each year, flounder eggs were ignored because they are demersal and adhesive and not generally impacted by entrainment.
Recently Rose et al. (1996) presented information on a population dynamics model for winter flounder consisting of separate young-of-the-year and adult components. The young-of-the-year model includes survival rates for eggs, larvae, early and late juveniles stages. Since the model is 11
1 designed to mathematically represent numbers ofindividuals as they develop from one stage to another, it is difficult to apply their survival rates to the mixed age pool oflarvae entrained at PNPS.
All individuals would need to be converted to a common starting point such as newly hatched eggs I
as is done with the unstaged approach. By using a value of 0.09 to step back from mixed-age larvae to hatched eggs, the rates utilized by Rose et al. produce approximately twice as many fish as the staged survival values provided above. Since the staged survival values were adjusted by Gibson (1993a) to provide an equilibrium population the Rose et al. values likely overestimate EA values in this instance.
The general, unstaged larval survival values produced an adult equivalent value of 3,414 age 3 fish for 1997 (Figure 3, Table 2). The stage-specific values produced an EA total over fourteen times higher at 47,087 age 3 individuals. Based on a weight of 0.6 pounds per fish (Gibson 1993b),
these values convert to 2,048 and 28,252 pounds, respectively. Comparable values for 1980 -1996 ranged from 535 to 1,841 fish (mean = 990 fish, 594 pounds) for the general approach and 2,624 to 15,558 (mean = 8,169,4,901 pounds) for the staged approach. EA totals for 1984 and 1987 were omitted here because both circulating seawater pumps were offfor most ofthe larval winter flounder seasons during protracted maintenance outages. There is some indication that ichthyoplankton entrainment is disproportionately low when only the salt service water pumps are in operation (Marine Research 1994). Values for 1997 using the unstaged general approach exceeded the previous high recorded in 1981 (1,841) by 85%, values based on the staged approach exceeded that previous high recorded in 1988 (15,558) by over 200%. The relatively high EA values noted in 1997 i l
are directly attributable to the relatively high number oflarvae entrained (see Volume ~ 1). The large differences between the two sets of survival estimates clearly show how relatively smt variations in survival values when applied to large numbers oflarvae can result in relatively large vanations in adult numbers (see Vaughan and Saila 1976 for example).
Over the 1982 through 1996 period an annual average of1,661,069 pounds (s.e. = 272,426 pounds) of flounder were landed from NOAA statistical area 514 which covers Cape Cod Bay and Massachusetts Bay (Table 3). Based on a weight of 0.6 pounds per fish, the average estimated loss of 685 or 6,361 pounds of equivalent adults represents 0.04 or 0.4% of those landings. Area 514 landings, declined sharply after 1993 froml,057,211 pounds that year to 16,788 pounds in 1995 and 1
12
only 2,961 pounds in 1996 (Table 3). The precipitous drop is attributable to increased fishing restrictions and stock declines. EA values for 1994 through 1996 alone appear quite high compared to the reduced landings and in fact the unstaged value exceeds the landings for 1996.
Winter flounder also have considerable value as a recreational species. Based on NOAA records' an annual average of 1,029,652 fish (s.e. = 333,404) weighing an average of about one pound each were landed from Massachusetts inland waters over the 1981-1996 period (Table 3).
More recently (1990 -1996) recreational landings have been well below earlier years consistent with commercial landings because of stock declines and area closures; an annual average of 115,158 fish (s.e. = 12,190) were reported landed in the state from inland waters during that more recent period.
These fish were also apparently smaller, weighing an average of 0.75 pounds each. Unfortunately these landings are compiled by state within distance from shore areas (inland, <3 miles from shore,
> 3 miles from shore) and the number of fish taken from a more appropriate area such as Cape Cod Bay are not available. Arbitrarily adding 20,000 pounds of recreationally-caught flounder to the 1994-1996 Area 514 commercial landings would bring the respective totals for those three years to 348,706,36,788, and 22,961 pounds. The EA values from the unstaged approach for those years then amount to 0.I,2.6, and 8.9%, respectively. For the staged larvae approach the three values amount to 1.7,25.1, and 123%, respectively. Clearly the decline in commercial tradings in 1995 and 1996 suggest that those values are no longer a realistic measure of EA losses.
Massachusetts Division ofMarine Fisheries (DMF) personnel made estimates of the number of adult w;nter flounder (>280 mm TL - age 3+) in a 1% square mile area in the vicinity of PNPS using area swept by a commercial trawl and using several mark and recapture models in 1997 (see Section IIIA, this report). While reliable estimates oflocal population size are difficult to make, they can provide realistic numbers with which to compare EA values. Landings data typically represent numbers caught over a very large area or as displayed by the most recent commercial landings can be subject to catch restrictions or changes in fishing effort which make them less useful. The DMF area swept estimate equaled 321,832 adults based on gear efficiency of 50% with broad confidence limits ranging from near zero to over 800,000 fish. DMF's mark and recapture study was limited by 5
Recreational landings data were obtained via the internet at http:// remora. ssp.nmfs. gov /mrfss.
13 i
disappointing tag returns. Estimates ranged from 115,000 to 520,000 adults depending on the model employed. EA estimates for 1997 using the unstaged survival values amount to 1.1% of the area swept estimate and 3.0% of the low mark and recapture estimate. The 1997 EA estimate from the staged approach amounted to 14.6% of the area swept estimate and 40.9% of the mark-recapture estimate.
Cunner Goodyear's (1978) basic procedures were used to estimate equivalent adult values. This method converts numbers of eggs and larvae to numbers of fish at age of sexual maturity which occurs for approximately half the population at age 1 (P. Nitschke, University of Massachusetts, Amherst, personal communication).
Assuming all labrid eggs were cunner eggs in PNPS entrainment samples (Scherer 1984),
cunner larva: egg ratios were determined from PNPS samples to provide an estimate of survival from egg to larva. Mesh correction values were first applied to both eggs and larvae. Presented earlier these were 1.24 for eggs taken from 1980-1995,1.14 for eggs taken in 1995, and 1.10 for eggs taken in 1997. Cunner larval mesh values applied were 1.16 for stage I and 1.28 for stage 2, irrespective of year. From 1980 to 1997 the larva / egg ratio ranged from 0.001284 to 0.128812 and averaged 0.027055; 1984 and 1987 were excluded because ofextended circulating seawater pump shutdown.
Average lifetime fecundity was calculated from fish in the PNPS area provided by Nitschke (1997).
He provided numbers of eggs produced at age in the second order form:
Log F = [2.891 log A) - [1.355 log A 2] + 3.149 where I i
F = fecundity at age A ]
Age-specific instantaneous mortality necessary for calculation of average lifetime fecundity was calculated from fish trap collections made from 1992 - 1997 (Brian Kelly, Massachusetts Division Of Marine Fisheries, personal communication). Fish trap collections were converted from length to age using an age-length key developed by Massachusetts Division of Marine Fisheries (MDMF) personnel working around PNPS (Lawton et al.1997, Brian Kelly personal communication, Table 4). Overall effort and distribution of effort by station varied by year in the MDMF study which resulted in changes in apparent numbers offish within year class attributable to sampling alone. This
!- made it impossible to determine estimates of mortality by tracking individual age classes. Two 14
approaches were used to weight the trap catch by sampling effort to overcome this problem. The first adjusted catch by overall annual sampling effort (Table 5), the other adjusted catch to the mean number offish taken each year (Table 6). Mortality rates were then calculated for each age class over the six year study period since that is preferable to assuming a constant mortality rate for all age groups (Rivard 1989). Since the mean number collected approach provided a better distribution of fish within age class and therefore a greater number of mortality estimates, that approach was ultimately selected (Table 7). Average instantaneous mortality rates for the PNPS area collections from 1992 through 1997 using this approach were as follows:
Age 3 = 0.286 Age 4 = 0.342 Age 5 = 0.645 Age 6 = 1.260 Age 7 = 0.653 Age 8 = 1.463 Age 9 = 0.728 Utilizing data from Serchuk and Cole (1974) for age 1 through 5 cunner collected with assorted gear, a survival rate of S = 0.605 was obtained (Z = 0.5025) which appears comparable to the PNPS values. Age 1 and 2 fish appeared less abundant in the PNPS collections than age 3 fish (Tables 4, 5), suggesting they were not fully recruited to the trap collections, perhaps due to their small size or behavior. Fish older than age 10 were rarely taken both because they are uncommon and because they can exceed the maximum size susceptible to the fish traps. In the absence of additional information an overall mean value of Z = 0.831 was substituted for age 2 and age 10.
Based on the PNPS area fecundity study (Nitschke 1997) 50% of age I females were assumed -
to be mature; complete recruitment was assumed by age 2. Following Goodyear (1978), an average lifetime fecundity of 21,656 eggs per female was calculated (Table 8). Utilizing the survival estimate for eggs to larvae and average lifetime fecundity, a survival estimate for larvae to adult of 3.41E-3 was calculated. Converting numbers of eggs to larvae utilizing the larvae / egg ratio and then converting numbers oflarvae to adult produced an estimate of 498,281 cunner potentially lost to entrainment effects in 1997. Comparable values for 1980-1996 ranged from 119,342 to 2,571,973 adults averaging 500,592 (s.e. = 130,709) over the 17-year period (Figure 4, Table 9) The high value I
of 2,571,973 recorded in 1981, attributable to high egg and exception-ally high larval densities, 15 l
skewed the mean EA value; without that high value a mean of 378,746 (s.e. = 50,182) was obtained.
Cunner have no commercial value and little recreational importance (although many may be taken unintentionally by shore fishermen) so that current landing records are not available. To shed some light on their abundance in the PNPS area, calculations were performed to estimate the number of adult cunner which would be necessary to produce the number of eggs found there. The PNPS area was defined by Cape Cod Bay sampling stations 2,3,4,7,8 (MRI 1978), the half-tide volume of which was estimated by planimetry from NOAA chart 1208 at 22,541,000 100 m' units. Labrid egg densities were obtained at those stations on a weekly basis in 1975; they were integrated over time (April-December) using the mean density of the five stations. The integrated values were multiplied by 1.40 to account for extmsion through the 0.505-mm mesh used in that survey (MRI unpublished data), then by the sector volume. Based on the 0.333/0.202-mm mesh data collected from the PNPS discharge stream from 1994 through 1997, additional upward scaling might be appropriate. No specific data for towed samples are available and an estimated value was not applied to provide additional conservatism to the comparison. The resulting value was divided by 2.2, the estimated incubation time in days for cunner eggs (Johansen 1925), then divided by 30,230, an estimate ofmean annual fecundity per female derived from Nitschke (1997). To derive average annual fecundity per female (not to be confused with average life-time fecundity) each age class was weighted by its relative catch in the DMF trap collections (Table 10). Since age 2 fish appeared to be under-sampled their numbers were estimated using linear regression over ages 3 through 10. Lastly the resulting value was multiplied by 2 assuming an even sex ratio. These calculations resulted in an estimated production of 6.899E12 eggs by an estimated 207,473,000 adult fish. The annual mean loss of
- 500,592 fish due to PNPS operation represents 0.24% of that value.
MDMF personnel have chosen cunner as an indicator species for PNPS impact investigations.
Tagging studies were conducted during the 1994-1996 seasons to estimate the size of the cunner population in the immediate PNPS area. Minimum tagging size and therefore the minimum size fish enumerated was 90 mm TL. Estimates were highly localized since individual cunner have a very small i
home range measured on the order of 100 m2 orless (Pottle and Green 1979). Estimated population !
size for the outer breakwater and intake areas combined were 7,408 and 9,300 for the two respective years. Combining upper 95% confidence limits produced totals of 10,037 and 11,696 fish, 16
i respectively. Since the upper confidence limit total is only 0.003% of the egg based population estimate, it is clear that eggs must arrive at PNPS from areas removed from the immediate vicinity of the Station. A hydrodynamic modeling study completed by Eric Adams of MIT (see section III. A) predicted that 90% of the cunner eggs and larvae entrained at PNPS come from within about 5,5 miles of PNPS to the north to White Horse Beach, about one mile to the south. This area extends further to the north than the area 2,3,4,7,8 used in the above egg estimates. The number of eggs entrained indicate that cunner must be abundant in these waters.
Atlantic Mackerel Procedures outlined by Vaughan and Saila (1976) were used to derive a survival rate for mackerel eggs to age 1 fish. This procedure utilizes the Leslie matrix algorithm to estimate early survival from proportion mature, fecundity, and survival within each age class assuming a stable population. Fecundity for Atlantic mackerel was obtained from Griswold and Silverman (1992) and Neja (1992). Age-specific instantaneous mortality was obtained from Overholtz et al. (1988) and NOAA (1995). Since two fecundity profiles provide two egg to age I survival values: 2.2231E-6 for Griswold and Silverman,2.3162E-6 for Neja, values were averaged (2.2697E-6). The observed average ratio ofeggs to larvae for PNPS of 0.08236 (1980-1997) provided a larva-to-age I survival rate of 2.75583E-5. in calculating larvae /cgg ratios 1981,1984, and 1987 were omitted,1981' because larvae were more abundant then eggs and 1984 and 1987 because both circulating seawater pumps were off for the mackerel egg and larval seasons during protracted maintenance outages. A mesh adjustment factor of 1.12 was applied to the egg data based on mesh comparison collections completed from 1994 through 1997 (Table 11). No mesh adjustment was justified for larvae (Table 12). According to NOAA (1995,1996) stock biomass consists of fish age I and older while fish completely recruit to the spawning stock by age 3. Therefore, adult equivalent values are shown for both age groups (Figure 5, Table 13). Age 3 individuals were estimated using an instantaneous fishing mortality rate of F = 0.02 (NOAA 1995,1996). These values provide an annual survival rate of S = 0.8025.
Equivalent adult estimates for 1997 amounted to 1,712 age 1 fish or 1,103 age 3 fish.
Corresponding age 1 values over the 1980 through 1996 time series ranged from 504 to 16,583 fish with an average of 5,489 (s.e. = 1,180). Age 3 values ranged from 325 to 10,679 with an annual l
17 l
average of 3,535 (s.e. = 760) individuals. Data from 1984 and 1987 were omitted here because values were unusually low as described above for the larvae / egg ratio calculations. Converting numbers of fish to weight using 0.2 and 0.7 pounds per individual (Clayton et al.1978) resulted in an estimated average annual loss of 1,098 pounds (s.e. = 236 pounds) or 2,474 pounds (s.e. = 532), respectively (1984 and 1987 excluded). Weight values for 1997 alone were 342 pounds of age I fish,772 pounds of age 3 fish.
According to NOAA statistical records, an annual average of 384,140 pounds (s.e..= 94,831) of mackerel were taken commercially from statistical area 514 over the years 1982-996. The loss of an average of1,050 pounds of age I fish (1980-1997,1984 and 1987 omitted) amounts to 0.3% of those landings and the loss of 2,368 pounds of age 3 fish,0.6%. In addition to commercial landings, mackerel have considerable recreational value. For example, over the years 1981-1996 an average of 658,099 fish (s.e. = 132,192) were landed in Massachusetts by fishermen working within three miles of shore. These fish had an average weight of about one pound. Unfortunately these landings are available only by state and therefore the portion attributable to Cape Cod Bay is not known.
Arbitrarily adding 200,000 one-pound fish to the commercial landings brings the harvest total to 584,140 pounds and the mean EA total to 0.2 and 0.4%, respectively.
18
l SECTION IV LITERATURE CITED Clayton, G., C. Cole, S. Murawski and J. Parrish. 1978. Common. marine fishes of coastal Massachusetts. Massachusetts Cooperative Extension Service, Amherst, Massachusetts.
231p.
Ecological Analysts, Inc.1981. Entrainment survival studies. Research Repon EP 9-11. Submitted to Empire State Electric Energy Research Corporation, New York.
Gibson, M.R. 1993a. Population dynamics of winter flounder in Mt. Hope Bay in relation to operations at the Brayton Point electric plant. Rhode Island Division Fish and Wildlife, West Kingston, R.I.
1993b. Stock assessment of winter flounder in Rhode Island,1992: A report to the RI Marine Fisheries Council. Rhode Island Division Fish and Wildlife. Res. Ref. Doc. 93/1.
Goodyear, C.P.1978. Entramment impact estimates using the equivalent adult approach. U.S. Fish and Wildlife Service, Biological Service Project. FWS/OBS-78/65. 14p.
Griswold, C. A. and M.J. Silverman. 1992. Fecundity of the Atlantic mackerel (Scomber scombrus) in the Northwest Atlantic in 1987. Journal ofNonhwest Atlantic Fisheries Science 12:35-40.
Horst, T.J.1975. The assessment ofimpact due to entrainment ofichthyoplankton. In: Fisheries and Energy Production: A Symposium. S.B. Saila, ed. D.C. Heath and Company, Lexing-ton, Mass. p107-Il8.
Johansen, F. 1925. Natural history ofthe cunner (Tautonolabrus adsocrsus Walbaum). Contribution to Canadian Biology new series 2:423-467. l Kuntz, A. and L. Radcliff. 1917. Notes on the embryology and larval development of twelve teleostean fishes. Fishery Bulletin U.S. 35:87-134.
Lawton, R.P., B.C. Kelly, V.J. Malkoski, and J. Chisholm. 1995. Annual report on monitoring to assess impact of the Pilgrim Nuclear Power Station on selected finfish populations in western Cape Cod Bay. Project Report No. 58 (January-December 1994). IIIA.i-77. In: Marine Ecology Studies Related to Operation ofPilgrim Station, Semi-annual repon No.45. Boston Edison Company.
19
7
, B. Kelly, V. Malkoski, J. Boardman, and G. Pintarelli. 1997. Annual report on assessment and mitigation ofimpact of the Pilgrim Nuclear Power Station on finfish populations in western Cape Cod Bay. Illa.i-55. h: Marine Ecology Studies Related to Operation of l
Pilgrim Station, Semi-annual report No.49. Boston Edison Company Marine Research, Inc. 1978. Entrainment investigations and Cape Cod Bay ichthyoplankton studies, March 1970-June 1972 and March 1974-July 1977. Volume 2, V.1-44. M: Marine Ecology Studies Related to Operation of Pilgrim Station. Final Report. July 1%9-December 1977.
Boston Edison Company.
l 1982. Supplementary winter flounder egg studies conducted at Pilgrim Nuclear Power l Station, March-May 1982. Submitted to Boston Edison Company. 4p.
l 1994. Ichthyoplankton entrainment monitoring at Pilgrim Nuclear Power Station January-l December 1993. Volume 2 (Impact Perspective).III.C.2i-27. M: Marine Ecology Studies l Related to Operation ofPilgrim Station, Semi-annual report No.43. Boston Edison Company 1995. Ichthyoplankton entrainment monitoring at Pilgrim Nuclear Power Station January-l December 1994. Volume 1 (Monitoring) IllC.i-67 h: Marine Ecology Studies Related to l Operation of Pilgrim Station, Semi-annual Report No. 45. Boston Edison Company.
l Neja, Z. 1992. Maturation and fecundity of mackerel, (Scomber scombrus L.) in Northwest Atlantic. Acta Ichthyol. Piscatoria 22(1):125-140.
Nitschke, P.C. 1997. Assessing factors that influence cunner (Tautogolabrus adspersus) l reproduction and recruitment in Cape Cod Bay. Masters thesis, University ofMassachusetts L Amherst.
5 i
NFSC (Northeast Fisheries Science Center). 19%. Report of the 21st Northeast Regional Stock l Assessment Workshop (21st SAW). Stock Assessment Review Committee (SARC) l consensus summary ofassessments. Northeast Fisheries Science Center Reference Document l %-05d. 200p.
l l NOAA (National Oceanic and Atmospheric Administration).1995. Status of Fishery Resources off the Northeastern United States for 1993. NOAA Technical Memorandum NMFS-NE-108.
140p.
' NUSCO (Northeast Utilities Service Company.1993. Monitoring the marine environment ofLong Island Sound at Millstone Nuclear Power Station, Waterford CT. Annual Report.
Overholtz, W.J., S.A. Muraski, W.L. Michaels, and L.M. Dery. 1988. The effects of density dependent population mechanisms on assessment advice for the northwest Atlantic mackerel !
stock WoodsHole,MA:NMFS,NEFC NOAATechnicalMemorandumNMFS-F/NED-62.
~
49p. j 20
Pearcy, W.G. 1962. Ecology of an estuarine population of winter flounder Pseudooleuronectes americanus Bulletin of Bingham Oceanographic Collection 18:1-78.
Pottle, R.A. and J.M. Green. 1979. Territorial behaviour of the north temperate labrid, Tautomolabrus adsocrsus. Canadian Journal ofZoology 57(12):2337-2347.
Rivard, D. 1989. Overview of the systematic, structural, and sampling errors in cohort analysis.
American Fisheries Society Symposium 6:49-65.
Rose, K.A., J.A. Tyler, R.C. Chambers, G. Klein-MacPhee, and D.J. Danila.1996. Simulating winter flounder population dynamics using coupled individual-based young-of-the-year and age-stmetured adult models. Canadian Journal of Fisheries and Aquatic Sciences 53(5):1071-1091.
Saila, S.B.1976. Effects ofpower plant entrainment on winter flounder populations near Millstone Point. URI-NUSCO Report No. 5.
Scherer, M.D.1984. The ichthyoplankton of Cape Cod Bay. In: J.D. Davis and D. Merriman (eds.). Observations on the Ecology and Biology of Western Cape Cod Bay, Massachusetts.
Lecture Notes on Coastal and Estuarine Studies. Volume II. Springer-Verlag, New York.
289p.
Serchuk, F.M. and C.F. Cole. 1974. Age and growth of the cunner, Tautonolabms adsoersus. in the Weweantic River estuary, Mass. Chesapeake Science 15(4):205-213.
Vaughan,D.S. and S.B. Saila. 1976. A method for determining mortality rates using the Leslie matrix. Transactions of the American Fisheries Society 3:380-383.
Witherell, D.B. and J. Burnett. 1993. Growth and maturation of winter flounder, Pleuronectes americanus. in Massachusetts. Fishery Bulletin U.S. 91(4):816-820.
i l
21 I
Winter Flounder I Equivalent Adult Summary l
1980 l j
I",3 9
UMed Approach
~
1983 Age 3 Fish 1984 1985 l 1986 1987 i 1988 ,
1989 1990 1991 1992 1993 1994 1995 i 1996 1997 .
0 $00 1000 1500 2000 2500 3000 3500 4000 Winter Flounder Equivalent Adult Summary 1980 j f'9 Staged Approach 1983 Age 3 Fish 1984 1985 1986 1987 1988 1989 1990 1991 1992 ,
1993 1994 1995 !
1996 1997 0 10 20 30 40 50 Thousands Figure 3. Numbers of equivalent adult winter flounder estimated to have been lost to entrainment at PNPS, 1980-1997.
22
I Cunner I Equivalent Adult Summary 1980 i i j 1981 l 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 :
1997 0 500 1000 1500 2000 2500 3000 Thousands Figure 4. Numbers of equivalent adult cunner estimated to have been lost to entrainment at PNPS, 1980-1997.
Atlantic Mackerel Equivalent Adult Summary 1980 l9$ Age i Fish 1983 1984 1985 1986 1987 1988 1989 1990 1991 .
i 1992 1993 1994 1995 1996 ;
1997 '
O 5 10 15 20 Thousands cNnikenengraphsWaea97.pr4 j 1
1 Figure 5. Numbers of equivalent adult Atlantic mackerel estimated to have been lost to entrainment at PNPS, 1980-1997.
23 .
Table 2. Numbers oflarval winter flounder entrained at PNPS annually by stage, 1980 - 1997.
Number and weight of equivalent age 3 adults calculated by two methods is also shown.
Number Of Larvae Entrained Equivalent Age 3 Adults Stage Year 18 28 3 4 General Staged Number Pounds Number Pounds 1980 8,694,456 12,714,822 7,317,129 0 1,771 1,063 7,443 4,466 1981 7,606.942 19,133,121 3,073,126 43,304 1,841 1,105 4,689 2,813 1982 2,706,834 6,724,795 11,583,134 425,011 1,322 793 12,643 7,586 1983 1,933,453 2,246,172 7,558,534 260,350 740 444 7,%9 4,781 1984 248,082 0 7.570,145 516,247 514 308 9,128 5,477 1985 1,039,001 2,312,789 8,025,452 130,786 710 426 7,643 4,586 1986 5,397,403 5,783,669 3, % 3,747 77,005 939 563 4,365 2,619 1987 0 437,608 3,088,405 0 217 130 2,619 1,571 1988 1,995, % 8 1,656,376 15,079,960 511,009 1.187 712 15,558 9,335 1989 1,668,823 5,755,240 2,224,675 39,114 597 358 2,624 1,574 1990 643,683 1,155,404 6,846,718 33,002 535 321 6,016 3,610 1991 3,471,022 3,908,488 5,188,056 37,717 777 466 4,966 2,980 1992 873,660 876,914 7,034,690 26,192 543 326 6,114 3,668 1993 1,595,700 3,540,750 4,934,952 88,617 626 376 4,958 2,975 1994 1,034,617 6,433,716 13,060,373 172,606 1,276 766 12,446 7,468 1995 1,632,907 2,820,023 8,826,4 % 375,857 842 505 9,699 5,819 19 % $04,810 5,818,499 11.329,855 995,127 1,150 690 15,395 9,237 1997 2,225,634 9,537,788 41,484,016 2,126,280 3,414 2,048 47,087 28,252 Mean 2,404,055 5,047,565 9,343,859 325,457 1,056 633 10,076 6,045 s.c. 578,064 1,140, % 8 2,067,069 122,466 173 104 2,368 1,421 1984,1987 Omitted 2 Mean 2,689,057 5,651,160 9,845,682 333,874 1,142 685 10,601 6,361 s.e. 614,157 1,200,337 2,292,345 136,105 !83 110 2,626 1,576 8 Mesh factor = 1.62 applied to Stages I and 2 prior to 1995.
2 See text for details.
24 1
l Table 3. Area 514 commerciallandings and Massachusetts j recreational landings from inland waters (pounds). I 1982 - 1996.
Cosassential Recreational Total (ponads) (pounds) (pounds) l 1982 3,830,162 4,146,553 7,976,715 1983 2,936,176 874,245 3,810,421 1984 2,558,483 839,561 3,398,044 1985 2,450,319 1,858,645 4,308,964 1986 1,667,938 708,677 2,376,615 1987 1,739,664 568,822 2,308,486 1988 1,846,171 729,200 2,575,371 1989 1,896,609 1,163,315 3,059,924 1990 1,737,733 139,641 1,877,374 1991 1,520,470 67,659 1,588,129 1992 1,326,646 85,256 1,411,902 1993 1,057,211 147,287 1,204,498 1994 328,706 71,403 400,109 1995 16,788 43,362 60,150 1996 2,%1 69,871 72,832 i
1 25 i
I j!jiI lll l1il 9
y h a 94 tcD a r 959.421 5 0 57 28 5
CeP 8 1 48 1 1 tr )s oy 00.000 f a 079679 8 1 221 2 E(D
_ f l 892733 333333 t
a 9058253 6
o 8 01 6775 l
t a
o 959.4 5 57 2.9 1 54 28 l
t a
o 333 333 555555 T 6 1832032 1 2252 T 8 0 48 1 1 1 T 222222 222222 0 3561 22 0 041 01 2 0 91 71 32 1 1 1 061 1 1 5 000000 0821 1 2
_ 9 978786 2 9 1 61 1 2. 4 9 92 1 571 3
_ 09332 1 000000 2 1 6335 44594 281 452
_ 7 8 896575 8 87.661 9 8 268593 9 6 21 1 1 5 01 001 1 2321 1 2
_ 9
_ 1 2
9 9
1 24431 3 9 6 7.0.1 98 7 1 401 98 7 33083 7 9 7 3 92855
_ r, 1 6323 0 1 31 1 1 3 9 475234 1
e 1
_ n -
n 7 2
u c
9 9
_ 9 9 y, 00000 1 1 85421 2
- k e
s 6 830765 7
4 91 1 1 7 2
s 6 0.4.83465 8 29 6
1 1 1 r,
e s 61 8667 6 5481 25 ht s 1 1 1 1 3 9 s n s 1 21 1 1 1 g
a 9 a n l a
n C l 1 l C u C l
e e r,
e e c, e g n g h g e c 1 58423 00000 n A A g
a y
A 5 3342169422 9 31 227 06 u c
5 74.6325 1 1 1 2 69 90 1 2 l t
a c 5 658994 0523 3 0 343223 b t r a d o u e f n
_ n f e n i
0 l a
m n 1 48530 r
e 000003 4 35402 l
ar 4 99.4475 686 a 4 947783
_ t e 47480 1 4233 0 3
e 77 1 2 8 1 1 3 e m
691 723 454334 d 1 m
s y y s b l
a b d
c d e e
8971 8203 444232 t
eg h t
735808 h a
y 3 1 51 35 89697 1 3354 5 ig e
3 232282 221 225 i
g e
3
%59972 91 81 4 546556 b
1 w w s s s n n n a o o 6 0 i it c
000006 t c 0 0.7 3 2 4 l
e 2 251 3 4 9 c
e 2 5 61 5.3.7 8 1 37 28 l l
e 2 3 42939 3 41 51 5 l
o c
39466 1 1 255 0 1 l
l o
1 1 233 o c 424 564 p
c p p a a
r a r t r t
t h
hs i
f 1 995590 78 1 598 1 233 h
i f
s 1 51 844 0.5 2040 9 1 21 i
f s
a 1 7.27390 2
2 6 9268 8
1 456 a 66 1 334 a e 2 1 331 e
r e r r a c a S S S P P P N N N P P P 234567 s 234567 234567 4.
999999 999999 es 5 999999 999999 6
. 999999 999999 1 1 1 1 1 1 l
e 1 1 1 1 1 1 M l e 1 1 1 1 1 1 l
b e
b b a a a T T T w*
l1 il ll i ! llll l 32280 3 1 2940 1 07378 3 69595 3 Z53 484 3 4 5 6 Z01 68963 403 6 4
1 0.2 0 01 1 01 1 00 1 1 1 8 9 e78593 e2 g321 1 2 g1 6435 A A 7 8 e392866 e278593 g2321 1 2 g475234 A A 234567 234567 999999 999999 999999 999999 1 1 1 1 1 1 1 1 1 1 1 1 n n a
a e e M M 89522 1 8621 o 5 83 200 5 63407 0 Z38442 4 6 Z8307 65.91 0 0 6 2
2 8.0 6.4 001 00 0 01 1 1 1 1 6 7 e 1 8667 e92866 g481 25 g75234 21 1 1 1 A A 5 6 e 659994 e 71 8667
- a. g052330 g5481 25 e
r A 343223 A 1 21 1 1 1 a
S P 234567 234567 N 999999 999999 P 999999 1 1 1 1 1 1 999999 1 1 1 1 1 1 e n n ht a a e e i
n M M d
t e
c l
e l
o c
r e
n n
u c 41 943 2 38080 4 2 0509 6 r - 39557 6 00955 2 7 0744 0662 7 fo Z06987 01 451 8
2 39557 05540 4
3 Z 20.5703 2 7
1 s 00000 0 00000 0 - 1 1 1 0 0 t
e a
r y 0 t 4 5 i 1 i 48783 e 59994 r e91 g 723 g52330 e 1 71 32 g821 1 2 i
v A 54334 A'43223 A e
s u
o e 3 4 9 n e%59972 948783 e691 723 e91 571 3 a g 91 81 4 g g2 2 6335 t
n 546556 454334 a A A A 1 t
s n 234567 234567 I 234567 999999 999999 999999
- 7. 999999 999999 999999 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 e n a
n n l
b e a e a a e T M M M U
5 208608 81 7 6 03 3 1 2 1 5 8 62 5
_ j 7 4 48 9371 6 3553 1 1
_ E j
2 P
j S
)
j E(
_ y 99 27 0868 3 7 t
07 9 2 95 1 5 91 )
i d 4 8 5 01 81 1 1 4 a e 1 7 65 2 72 57 8 F
_ e 1 23 34 4 4 4 =
(
c m e
F u S
n j) oP i(
t r
_ er pu0 e 5. 1 1 1 1 1 1 1 1 1
_ ot a r
PM r
laJ vS 635 0602 3 6 e 5 72 24 8 4 1 0 n i (e I 3 23 23 1 000 n nta 4321 00000 SeR u 000000000 c
a e
r a
S P
N P
r f
o y I 1 79 4 47067 t s 3 1 5 061 8 4 7 i w 81 41 30464 d
n u 01 1 23 4567 u C c
f e
e i
m .
t
- 0 f
e 1 l
i 2,
e g eg a 1 62 5 03 3 61 r Z 3 8 4 4 65 66 3 A e 82 3 62 641 8 r v 00001 01 1 0 o a f f d c e n s o u i
t
)
1 l
a 3 u 8 l
c 0 a (
C e g 0 Z 1 23456789 1 e
- 8. A g a
le r e
b v a
T A l
t 63560371 69729531 71 n 21 990 64 l
est 47771 376491 4481 978 1, 9,0,7,1,5,5,9,1, 2, 1, 3, 1, 4,8,4,3,2, 95 008 40 2 8
al v u31 0872947009007088 5,5,8 7,5,2, 7,9,7 1 id1 7286221 691 1 1 5741 9 041 065 861 0 uA75,261 q 3241 571 231 534 05 55 336 1 33 70 32 5 E 2 061 841 842077592834 85 251 51 3 46673773039270771 9 45 820 5,5,5,9,0,2,7,4,5,1, 7,6,8 1,0,1, 8, 9, 8, 3, 2, 0, 3, 00 4 la1 268623068851 06084 61 786 8, 9, 1 to 9238081 1 789395831 3 80 8 397 1 37338 331 81 1 T 9, 3,1, 4,2,8,3,8,3 1, 0,7,7,2,9,1 4, 6, 1, 8, 1 5, 3, 0, 5, 9, 3 06021 9607226259779 23 1 37 26 1 271 4 324 571 1 41 9 73 332 1 5 1 1 44 1
- 7. 866227882886535299 91 592 9 475051 635326509945 38 1 37 5 4 0 9 3 2,8,3,8,3,6,3,7,4,9,5,9,0,3 0,7,8,5, 4, 7, 2, 3, 7, 2 4 5 1
e987597781 343530784 52 81 435 22 8, 0, 8 n20952094661 4024559 30 5 1 7 3 0 to 2,5,5 9,0,3,2 2 4,8,0,2,6,9,4,2,5,2, 9,9,1 0, 7, 2, 9 51 1 8 2, 2 9 S 43 31 2 22471 74951 72 333 1, 8 13 1
4 4 1 3 ly e la a 2000258471 80937891 22 853 u v r 356381 1 774991 37401 92 720 a 8, 3, 3, 3, 8, 22 n 0,9,7,5,6,6,3,4,0,1,2,4,8,1,9,7,5,5, 41 1 n L2 0771 62744481 685045 44 691 2, 3, 8 a e86877991 22708741 99 35 651 0 S g 4, 5,1, 5, 1 3,1,3 6,2,3,7,1, 1, 5,9,0,8, 0,2,8 9, 6, 9, 807 0 ta 0637 P 1 691 3 02 25231 75981 1 9 31 8 7, 0, N
P S45 2 22 1 6 2 5 35 1 1 79 6
2 4 t 1 1 a .
d n
e w no h 095609222202040853 72 74 ia r s 6344 3927577 7 352 1 9 48 2,7,4,6, 0,0,2,9,0,9 1, 6, 6, 4, 9, 5, 5, 1, 1, 98 n
t els o 1 251 1 2970050 8 1 54 62 98 33 4 0
e 8456 81 49009 4 664 1 4 99 7, 0, 6 e a g 2,2,3,9, 1, 4,2,2,1, 7, 7, 1 9, 7, 4, 2,2,81 4,3, 767 ae vr a6660 7402455 736 34 76 661 5 1
r a t S 713 1 1 4 36 37 1 5, 5, 7 la s 63 dl t
1 2 5 nsd u
s a
t gn ge 650741 908420877490 20 959 89 el 71 5467881 01 9706702 37 775 66 1 r a 7,9,1,3,8,0,2,0,3,0,1, 3,0,2,6,8,1,7, 8, 3, 4, 5, 5, 7, 9, 4 ei v s 1 499463496801 731 49 00 939 1 5 3 nu nq g 9972685681 4061 5928 878272 94 7 g 8,2,7,3,7,8,6,5,0,4,0,0,6,3,2,4,1, 2, 1 ue c i E
760561 359965458678 8,2,1 3, 2,1, 90 1 81 0,9,7 05 1
4, f o 571 9629636377351 01 36 992 21 3 o s 2,5,0,8,7,0,4 4,5,4,3,6 1,2,5,1, 8, 7, 8, 6, 3 0,4, 6, 4, 4 s
r r e 36251 21 41 41 231 421 21 22 21 3
eb bmu m 1 uN 8 r 9 N e 01 2345678901 234567 1
. n 888888883899999999 v n v.
n e e t e
9 le C u 99999999%999999999 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 n Dt a .n ne iam h u
o an Dt .n b edt oue..do it ed c a e e T MScsMG WMScs t o 1
I Table 10. Calculation of average annual fecundity for PNPS area cunner.
l l
Age Fecundity Relative Fecundity x Catch Catch i 2 7,879 162.3 1278814.1 3 16,592 162.11 2689730.7 4 25,027 122.23 4057046.1 5 32,190 86.94 3934617.9 6 37,858 45.38 3291365.8 7 42,116 13.23 1911236.9 i 8 45,158 6.7 597439.0 l
9 47,193 1.55 316195.5 10 48,417 0.76 75046.7 Sum 600.4 18151492.7 Average l Fecundity = 30230 30
Table i1. Densities per 100 m 2of water for mackeret eggs taken with 0333 and 0.202-mm mesh netting at IWPS,1994 - 1997.
0.333 0.202 Ratio 1994 May 4 8 80 13 80 1.57 5.70 7.50 132 530 8 80 1.66 7.70 6.80 0.88 2 80 3.90 1.39 I 2.70 6.70 2.48 1995 June 16 144.92 164.90 1.14
, 142.5 168.6 1.18 1473 130.5 0.89 June 26 6f>0 3 60 0.55 1996 June 19 34.70 52.70 1.52 45.20 42.50 0.94 34.70 3430 0 99 1997 June 13 12.00 16.10 1.34 10.00 11.10 1.11 13.10 5 80 0 44 Geometrie Mean 33 4 31.7 1.12 Table 12. Densities per 100 m' of water for mackeret larvae taken with 0333 and 0.202-mm mesh netting at IWPS.1994 - 1997.
0.333 0.202 Ratio 1995 June 16 1034 3 1134.2 1.10 838.2 907 3 1.08 10283 768.8 0.75 June 26 62.1 248.5 4.00 424.5 397.2 0.94 1057.1 834 4 0.79 June 28 35.5 47.7 1.34 56.5 62.8 1.11 39.7 35.1 0.88 1996 June 19 48 4.5 0.94 2.8 1.0 036 43 2.8 0.65 June 24 8.9 10.1 1.13 4.8 4.1 0 85 2.7 5.4 2.00 64 5.8 0.9I July I 32.8 31.2 0.95 22.5 23.4 1.04 41.1 32.5 0.79 July 5 5.0 1.4 0.28 9.0 7.2 0 80 5.5 3.9 0.71 1997 June 13 55.9 76.9 138 34.0 70.1 2.06 80.4 39.9 0.50 June 18 79.8 76.5 0%
23.1 91.C 3.97 65.7 54 3 0 83 July 2 563 62 0 1.10 61.5 53 4 0 87 46 7 35 1 0 75 Geometric Mean 36.1 353 0 98 31
Tabic 13. Numbers of Atlantic mackerel eggs and larvae entrained at PNPS annually,1980 - 1997. l Numbers of equivalent age I and age 3 fish are also slumn. I Total Nuanber Entrained Equivalert Adults Year Eggs Larvae Are 1 Age 3 1980 81,599,432 22,293,108 800 515 1981 183,959,791 320,135,5 % 9,240 5951 1982 108,234,931 9,388,143 504 325 1983 148,616,621 41,333,673 1,476 951 1984 22,486,619 78,315 53 34 1985 1,867,648,438 45,711,343 5,499 3541 1986 219,488,066 58,333,520 2,106 1356 1987 71,222,294 215,561 168 108 1988 3,710,026,220 3,401,489 8,514 5483 1989 6,510,097,199 65,562,469 16,583 10679 1990 3,222,258,634 4,627,282 7,441 4792 1991 668,239,740 66,009,482 3,336 2148 1992 525,957,850 8,086,393 1,417 912 1993 2,509,062,797 8,325,789 5,924 3815 1994 725,563,273 3,419,299 1,741 1121 1995 2,462,027,209 197,689,693 11,036 7107 1996 2,099,552,021 70,947,053 6,721 4328
)
1997 441,493,079 25,778,062 1,712 1103 Mesa 1,420,974,123 52,852,015 4,682, 3,015 s.e. 412,268,509 19,268,231 1,073 691 Mean 1,592,739,081 59,440,150 5,253 3,383 w/o 1984,1987 s.c. 446,092,948 21,149,950 1,128 727 l
I I
1 32 l
IMPINGEMENT OF ORGANISMS AT PILGRIM NUCLEAR POWER STATION (January - December 1997)
Prepared by: / Cd .
Robert D. Anderson Principal Marine Biologist i
Regulatory Affairs Department Boston Edison Company l
l i
l April 1998
{
i i
1
i l
Herrsng nlN NNlN::,n,,n, Alewife 8 smelt Cunner Commonly Silt >erside impinged species
\
TABLE OF CONTENTS Section Title Eage 1
SUMMARY
1
- 2 INTRODUCTION 2 l
3 METHODS AND MATERIALS 5 4 RESULTS AND DISCUSSION 7 4.1 Fishes 7 4.2 Invertebrates 17 4.3 Fish Survival 21 5 CONCLUSIONS 23 6 LITERATURE CITED 25 i
l ii
LIST OF FIGURES Figure Pace I
1 Location of Pilgrim Nuclear Power Station 3 i
i 2 Cross-Section of Intake Structure of Pilgrim 4 Nuclear Power Station l
l 3 Trends of Intake Water Temperature, and Number of Fish 12 i Captured by month from Pilgrim Station intake Screens for the Five Most Abundant Species Collected, January-December 1997 i
i l
I l
)
l I
l l !
i l
l l
iii
LIST OF TABLES Table Paae 1 Monthly Impingement for All Fishes Collected From Pilgrim Station 8 Intake Screens, January-December 1997 2 Species, Number, Total Length (mm), Weight (gms) and Percentage for 9 All Fishes Collected From Pilgrim Station Impingement Sampling, January-December 1997 3 Annual Impingement Collections (1988-1997) for the 10 Most Abundant 10 Fithes From Pilgrim Station Intake Screens During January-December 1997 4 Approximate Number and Cause for Most Notable Fish Moralities at 13 Pilgrim Nuclear Power Station, 1973-1997 5 Impingement Rates per Hour, Day and Year for All Fishes Collected 15 From Pilgrim Station intake Screens During January-December 1997 6 Impingement Rates Per Hour, Day and Year for All Fishes Collected 16 From Pilgrim Station intake Screens During 1978-1997 7 Monthly Means of Intake Temperatures ( F) Recorded During 18 Impingement Collections at Pilgrim Nuclear Power Station,1988-1997 8 Monthly Impingement for All Invertebrates Collected From Pilgrim Station 19 Intake Screens, January-December 1997 9 Survival Summary for the Fishes Collected during Pilgrim Station 22 Impingement Sampling, January-December 1997. Initial Survival Numbers are Shown Under Static (8-Hour) and Continuous Wash Cycles iv I
l l
SECTION I
SUMMARY
Fish impingement rate averaged 1.43 fish / hour during the period January-December 1997, which is considerably lower than most recent years partially because of no large impingement
! incidents. Atlantic silverside (Menidia menidia) accounted for 46.5% of the fishes collected followed by rainbow smelt (Osmerus mordax) at 12.6%. Atlantic menhaden (Brevoortia tyrannus) and butterfish (Peorilus triacanthus) represented 8.6 and 8.3%, respectively, of the fishes impinged. The peak period was November / December when fish impingement was dominated by Atlantic silversides. This time period, as well as early spring, is typical for high silverside impingement. Initial impingement survival for all fishes from static screen wash collections was approximately 28% and from continuous screen washes 61%.
l At 100% yearly (January-December) operation of Pilgrim Nuclear Power Station (PNPS) the estimated annual impingement was 12,514 fishes. The PNFS capacity factor was 73.4%
during 1997.
The collection rate (noihr.) for all invertebrates captured from January-December 1997 was l 0.45+. Jellyfish and sevenspine bay shrimp (Cranoon septemspinosa) were most numerous.
Green crab (Carcinus maenus) and common starfish (Asterias forbesi) accounted for 19.8 and 12.1%, respectively, of the invertebrates impinged and enumerated. Mixed species of algae .
collected on intake screens amounted to 1,974 pounds.
SECTION 2 INTRODUCTION Pilgrim Nuclear Power Station (lat. 41*S6' N, long. 70'34' W) is located on the northwestem shore of Cape Cod Bay (Figure 1) with a licensed capacity of 670 MWe. The unit has two circulating water pumps with a capacity of approximately 345 cfs each and five service water pumps with a combined capacity of 23 cfs. Water is drawn under a skimmer wall, through vertical bar racks spaced approximately 3 inches on center, and finally through vertical traveling water screens of 3/8 inch wire mesh (Figure 2). There are two traveling water screens for each circulating water pump.
This document is a report pursuant to operational environmental monitoring and reporting requirements of NPDES Permit No. 0003557 (USEPA) and No. 359 (Mass. DEP) for Pilgrim Nuclear Power Station, Unit 1. The report describes impingement of organisms and survival of fishes carried onto the vertical traveling water screens at Unit 1. It presents analysis of the relationships among impingement, environmental factors, and plant operational variables.
This report is based on data collected from screen wash samples during January-December 1997.
I l
i
=
=
h
=
O ,'
l ic t
- m. 8 A .
/ ' -
k . n i
o t
a t
S
~ r e
w o
.. P r
a e _
l _
y c a u B l f
W m C i r
6 g C
W l M i P
K _
N f I
o
&*S E
L n A o C i t
a c
o .
L ._
N .
/
1 e
. . "E r u
r g ma,ne,io ._*p i i
e F lylcwt i u ota d PNpS . . ,
_ ht u
o m -u Ply l-We s
tllI!!!ll l 1
i 4
l i.
} StU CE.4V TNA0WGH 3Catt g om a 31m l
I ( ( .T, \
% .___u, *dk F-'
emians , ,. . f c .* ,:'. [-
"~
141) L : @ h'- 14 k' .
9
.avinet0. '
REGI0vatti 70
- C040tt5En fIO'.Wi" 4 -
I e ,
SAA RACE $
e stor toa susot
.[L I
f s a.s. h. m' ,
'*- i
' .* X. =
i
, M.w .
we enn g y __w
- m m assim esas e = min mg[gg gg g {[glg jgg) e esse suso smuguur upmum e . misus 4eeser emise -
I 4';- i=====' e =* p elsN15 Cart 0Mel4G5 r.
,, g g ,,
.y sottom es sansmen.Att
. */d -3 .- stuits sais r- _
3= J 7; gg en siac .
$ " = inu . uin,to.
,a ' ' .m 80T10m lett
. V ~
l','T.'- -l . ?,* i.%. .. mjg.;::pyvyy; ;, p. ; iy'y. v: v,'.;.:cppic.gQ .
e l
l Figure 2: Cross-section of intake structure oi. Pilgrim Nuclear Power
- Station.
l 1
4-
1 I
SECTION 3 )
METHODS AND MATERIALS Three screen washings each week were performed from January-December 1997 to provide
! data for evaluating the magnitude of marine biota impingement. The total weekly collection time was 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> (three separate 8-hour penods: morning, afternoon and night). Two 1
collections represented dark period sampling and one represented light period sampling. At l
the beginning of each collection period, all four traveling screens were washed. Eight hours j later, the screens were again washed (minimum of 30 minutes each) and all organisms I
collected. When screens were being washed continuously, one hour collections were made at l the end of the regular sampling periods, and they represented two light periods and one dark period on a weekly basis. {
k Water nozzles directed at the screens washed impinged organisms and debris into a sluiceway j i
that flowed into a trap. The trap was made of galvanized screen (3/8-inch mesh) attached to a removable steel frame and it collected impinged biota, in the screenhouse, shortly after being washed off the screens. Initial fish survival was determined for static (8-hour) and continuous screenwash cycles.
l l
Variables recorded for organisms were total numbers, and individual total lengths (mm) and weights (gms) for up to 20 specimens of each species. A random sample of 20 fish or invertebrates was taken whenever the total number for a species exceeded 20; if the total collection for a species was less than 20, all were measured and weighed. Field work was conducted by Marine Research, Inc. Intake seawater temperature, power level output, tidal stage, number of circulating water pumps in operation, time of day and date were recorded at the time of collections. The collection rate (#/ hour) was calculated as number of organisms 1
impinged per collecting period divided by the total number of hours in that collecting period.
Beginning in 1990, if all four intake screens are not washed for a collecting period then the number of fishes collected is increased by a proportional factor to account for the unwashed screens, as requested by the Pilgrim Administrative-Technical Committee. Common and scientific names in this report follow the American Fisheries Society (1988,1989,1991a and l- 1991b) or other accepted authority when appropriate. {
'1 l
l l
l I
l l
SECTION 4 RESULTS AND DISCUSSION 4.1 Fishes In 455 collection hours,649 fishes of 25 species (Table 1) were collected from Pilgrim Nuclear Power Station intake screens during January - December 1997. The collection rate was 1.43 fish / hoar. This annual impingement rate was relatively low compared to most recent years, primarily because of large impingement incidents of Atlantic silverside (Menidia menidia) and/or rainbow smelt (Osmerus mordax) and alewife (Alosa pseudoharenaus). Atlantic silverside was the most abundant species in 1997 accounting for 46.5% of all fishes collected, followed by rainbow smelt at 12.6% (Table 2). Atlantic menhaden (Brevoortia tyrannus) and butterfish (Peorilus triacanthus) accounted for 8.6 and 8.3% of the total number of fishes j collected and identified to lowest taxon.
Atlantic silverside occurred most predominately in monthly samples from April, November and December. Hourly collection rates per month for them ranged from 0 to 3.23. Silverside impinged in April, November and December accounted for 85% of all this species captureci in impingement collections from January-December 1997. They averaged 92 mm total length and 4 grams in weight. Their impingement indicated no relationship to tidal stage or diel j factors. They are usually the dominant fish in the annual impingement catch, being the most l abundant species caught in seven of the last ten years. Impingement histories of abundant species impinged at Pilgrim Station in 1997, over the past 10 years, are documented in Table 3.
Rainbow smelt were very abundant in November and December impingement collections and have been most prevalent in the late Fall /early Winter period in the past, ranking first in 1978, 1987 and 1993 in total numbers impinged. In 1978,1993 and 1994, large impingement incidents involving smelt occurred during December.
e 40 92 33 1 1 2 D 1
. 2 v 4600 2 8 9 o 841 1 1 1 3 52 22 31 1. 2 1 N
t 7 4 n c 28 1 1 1 3 i
o O 1 1 t
t a .
t S p e 5 1 im S 1 3 3 1 1 4 2 r
g li P t s
m 7 u g 1 1 1 1 7 3 1 1 o
r 9 u 4 1 4 3 f 9 A d1 er tc e b ly 1 3 2 4 le m u 1 7 3 lo e J Cc e s
eD h -
e n 1 1 2 6 u 3 is y Fr a J l
l u An y 51 21 1 2 7 r a a 1 1 1 4 oJ f ,
M t s nn ee mer l i 80 9 7 4 ec r p 71 2 546 2 1 1 1 3 5 gS 1 n A ipk e Iman t h
lyi c r 2 h 21 1 6 4 t a 1 1 5 n M o
M 1
b 0 e 1 1 1 3 7 l
e F 1
1 2 b
a T
7 0 n 1 3 33 1 43 1 2 3 4 a 4 4 J
k c
a )
b n le e e g k r h s d d r n r ni c h s e (
e isl t a e i e d it b s i
g d n
eco r f r eh d r e ic emn n e f
f u od r in yu ph r r vo e
p l v
s e u h pe nm a e ae t ipc eofl ke i T
S is w mh sf lo k c h k e nak ntwc p k o .t as et o
r ehht a d see mpci np coh c s pr ioici f r r yaeigerh f f a ois s i t
n nbt eb bi poh r h d ne t
ekntnh et as er it s c t e epc it n la al net b ew uaol mtur ei vd nakl at r i a uo hlayuh t
l taiat t u n u r ue l t r ao t o l l
ARABWi GBALTNSRWAHLSBCNWABFW l l i t i t l t T C l'
o
,fl(
L l
Table 2. Species, Number, Total Length (mm), Weight (gms) and Percentage For All Fishes Collected From Pilgrim Station impingement Sampling, l
January -December 1997 i
Length Mean Weight Mean Percent Of l Species Number Range Length Range Weight Total Fish 1
Atlantic silverside 302 68-125 92 1-8 4 46.5 Rainbow smelt 82 57-172 109 1-25 6 12.6 Atlantic menhaden 56 38-130 66 1-21 3 8.6 1
Butterfish 54 27-89 53 1-7 2 8.3 Winter flounder 40 49-95 64 1-8 2 6.2 Grubby 23 51-93 68 1-10 5 3.5 Blueback herring 18 65-169 95 2-29 6 2.8 Alewife 15 53-272 114 - -
2.3 Lumpfish 9 27-62 46 1-10 5 1.4 Tautog 8 50-78 63 2-8 4 1.2 Norhtern puffer 5 59-88 70 4-11 7 0.8 Silver hake 5 62-96 85 1-5 4 0.8 Red hake 4 55-90 75 1-4 3 0.6 Windowpane 4 70-175 105 - -
0.6 Atlantic tomcod 3 123-145 130 13-59 31 0.5 Hake sp. 3 68-138 95 2-15 7 0.5 Little skate 3 245-483 398 - -
0.5 l Striped searobin 3 60-65 63 3-4 3 0.5 Blackspotted stickleback 2 37-41 39 1 1 0.3 Cunner 2 57-102 80 - -
0.3 Northern pipefish 2 135-152 144 1 1 0.3 White perch 2 101 101 11 11 0.3 Atlantic herring 1 40 40 0.3 0.3 0.2 Bay anchovy 1 80 80 3 3 0.2 Fourspot flounder 1 105 105 9 9 0.2 White hake 1 240 240 75 75 0.2
6
_ 7 2264038598753 9 0855421 1 053 9 3 64 1 1
imr lg i 6 5451 1 781 85562 P r 9 677 4551 21 1 9 e 9 71 2, 4 2 m b 1 1
o r m F e c
- 02322571 95677 s
e e " 067 9487 466 h D 5 1, 1 8, 4,7 5 is
- 9 1 1 30 y 9 F r 1 6 t a n u a7 n " 26440841 74099 d9i a 4 1 91 9921 1 737 9 J 1 9 1, 8 2,7 5 r m 9 3 43
' e o r
1 7
- b F
-c m d 3 0541 01 8 Mc e t
e 9 23 9 77 95253235263755 3 0 e c 7,7 2 1 D l l
e 1 1 6 e - o h y t r C
r a s 2 2520231 209645 f n ou h e 9 32 741 21 275 9 2 47
) a is 1 0 7
9 J f 4.
9 g d 9 1 i n e 9 r a 1 u n 1 51 33761 4583+7 r 8D 8
ip 9 741 9 2 1 6432 1 256
.b e 5.
9 sn m 6 2.0 8 9 1
e I
1 0 8 m9
( f 3 9 e1 s er o r
9 1 v oy n c e eNa ioS b 0 78581 931 86606 n M t
ce m 9 53463503 u6 454 J1 51 lek u 9 4 31 1 1 3,9 1 lot a N 1 1 1 5 r -
cn 9 l eh t i ibr c nn por eo t a mta i
9 0921 2958321 87 8 238 9
421 1 51 5 AcM 4O et 1 36 0 0 gS 1 1 9 3 ip n n on oo n it it it m
- 51 401 52882656 aaa r r I
8 31 1 821 r l
a u
8 9 5 0 e p e p ep n 1 ooo A
n in n ni i e
r e r e 3 )
) e e er le .r sh ww s w b
a e n
e g h#
((
r/
p sp sp T mm uum d d r n isl t a e ee u r eh emn d
n ir r
e imta TR p p p l
v s e u h nn SSS mi s
s w mhsf lo k c h s iioo WWW e c oifcir yaeigst t CCC f f cc ic e
itn n bt r eb bi net b ew mtut pol aee ooo p aiat nu t lt alt ui r lue ua oloo l l l
NNN S ARABWGBALT T CC l h8
l l
l l
Atlantic menhaden were prevalent in August samples and have been most prevalent in the
, early fall period, ranking third in 1986/1991 and second in 1989/1990. Generally, it has been one of the less abundantly impinged fish over the years.
Butterfish impingement occurred predominately in September accounting for 65% of this species total. It has been one of the least impinged fish, never dominating the annual catch.
Monthly intake water temperatures and impingement rates for the five dominant species in 1997 are illustrated in Figure 3.
There was one small fish impingement incident (20 fish or greater /hr.) at Pilgrim Station in November 1997 when mostly Atlantic silversides were recorded, but impingement rates rapidly decreased upon subsequent samplings indicating minimal impact. There were no large fish l
l impingement incidents (1,000 fish or greater) in 1997 on intake screens. Most large fish l
impingement mortalities have occurred while both circulating water pumps were operating.
l t
Fifteen large fish incidents have been documented since Pilgrim operation commenced in 1973, and most (11) have involved impingement as the causative agent (Table 4). However, at least in two of these, the possibility of pathological influence was implicated as indirectly l contributing to the mortalities. They were the Atlantic herring (tubular necrosis) and rainbow l
smelt (piscine erythrocytic necrosis) impingement incidents in 1976 and 1978, respectively.
Fish impingement rate at Pilgrim Station has been shown to be related to the number of circulating water pumps operating, in general (Lawton, Anderson et al,1984b). Reduced pump operation has lowered total impingement, particularly during the April to mid-August 1984 and portions of the mid-February to August 1987 periods when no circulating water pumps were operating for extended time frames. The significance of this relationship is supported by the
g m@fz 5 zZ ,lgg 3 ozg ,E h
, - *
- 0 0 0 0 0 0 0 3 O 4 " 0 0 o 0 5 02 0 5 4 3 6 ' 3 2 3 1 1 ll i ll R R E E B B M M E E C C E E D D R R E E B B M M E E e V V k O O a N N t n
E_
i R
E R n E io B B t O O t a
T T C C S O O m i
r R R g E E l B
i B P M M T
E s E T m7 o9 P P E E Fr91 S I S hr n be t
T T o S
U G
S U Mmyc e
< G U U Be A A d D e -
r y ur Y
_ Y t a pu L L a n U U Ca J
7 9
9
_ J h J, ids F et f
1 Oec l
E rl E
N N eo U U bC J ms A _
J ue Ni c dnpe A, S 7 Y t
A M
A e r n a M ud t
a r u n eb pA L
I L mt es R I R To P P rM A A te e av WF i
H - ee
- H kh C C aT R
+
t A - R Inr o
-- A A_T--
M M f F E N O s D g E - sn Y I S
D R d e R A E Y ner R T D R e A E E L H A r c U E N N TS R
B R
U 7_ V L M E H U U R
E T I
S S M S O B 3 A L E F
C W C F F F e R I R r Y 'R E I T O N R u R EP N B E Y ig T 2 N2 A6 R AM A -
A0 4 T0 F L5 5 N4 = A U E C L3 AI 8= T = = U WT I
N T = RN A AN AN N WN N nf J A l
J 0 0 0 0 0 0 0 0 : 0 0 ,0 5 0 0 0 0 6 5 4 3 6 3 4 ' 3 2 ' 3 1 2 1 u $ h gu. Ez
,g-
Table 4. Approximate Number and Cause for Dominant Species of Most Notable Fish Mortalities at Pilgrim Nuclear Power Station, 1973-1997 Date Species Number Cause April 9-19,1973 Atlantic menhaden 43,000 Gas Bubble Disease August / September,1973 Clupeids 1,600 Impingement April 2-15,1975 Atlantic menhaden 5,000 Gas Bubble Disease August 2,1975 Atlantic menhaden 3,000 Thermal Stress August 5,1976 Alewife 1,900 Impingement November 23-28,1976 Atlantic herring 10,200 Impingement August 21-25,1978 Clupeids 2,300 Thermal Stress December 11-29,1978 Rainbow smelt 6,200 Impingement March / April,1979 Atlantic silverside 1,100 Impingement September 23-24,1981 Atlantic silverside 6,000 Impingement July 22-25,1991 Atlantic herring 4,200 Impingement December 15-28,1993 Rainbow smelt 5,100 Impingement November 28-29,1994 Atlantic silverside 5,800 Impingement December 26-28,1994 Atlantic silverside 6,100 Impingement Rainbow smelt 5,300 impingement September 8-9,1995 Alewife 13,100 Impingement l
radmisc/ chart 98
_ _ _ _ _ _ _ . _ _ i
l fact that total fish impingement and rate of fish impingement were several times lower in 1984 /
and 1988 (low-pump operation years) than in 1989 - 1997, despite a greater number of collecting hours in 1984 and an average number of hours in 1988. In 1987, far fewer collecting hours were possible when both circulating pumps were off than in these other years which limits comparisons to them. However, total fish impingement rates in 1984,1987 and l 1988 were several times lower than in 1989-1997 when at least one circulating pump was more consistently in operation. Although there were brief periods in 1994 and 1995 when no circulating water pumps were operational, mixed results were noted regarding the effect on impingement of pump operation, possibly influenced by conditions causing large impingement
{
incidents each of these years.
Projected fish impingement rates were calculated assuming 100% operation of Pilgrim Nuclear Power Station, under conditions at the times of impingement, during the period January-December 1997. Table 5 presents hourly, daily, and yearly impingement rates for each species captured (rates are rounded to significant figures). For all fishes combined, the respective rates were 1.43, 34.29 and 12,514. The yearly rate of 12,514 is below normal and only 59% of the last 20-years' (1978-1997) mean annual projection of 21,098 fishes (Table 6).
This was considerably lower than most recent years' rates which were the highest yearly fish impingement rates since 1981 and greatly exceeded the historical annual average partially because several large impingement incidents inflated yearly projections. Relatively high impingement rate years offset low impingement years, and they may be attributed to population variances of the dominant species and/or extreme meteorological or operational conditions influencing species' behavior and vulnerability.
Over the past 20-year period (1978-1997), Pilgrim Station has had a mean impingement rate of 2.41 fishes /nr., ranging from 0.13 (1984) to 10.02 (1981) (Table 6). Anderson et al. (1975) documented higher annualimpingements at seven other northeast power plants in the early J
L
(
Table 5.. . Impingemerit Rates Per Hour, Day and Year For All Fishes Collected From Pilgrim Station-Intake Screens During January - December 1997. Assuming 100% Operation of Pilgrim Unit 1*
Rate / January- Dominant Months t
Species Rate /Hr. Rate / Day December 1997 - Of Occurrence Atlantic silverside 0.66 15.93 5,814 November / December Rainbow smelt 0.18. 4.33 1,579 November / December Atlantic menhaden 0.12 2.95 1,078 August Butterfish 0.12 2.85 1,040 September Winter Flounder 0.09 2.11 770 April
! Grubby 0.05 1.21 443 January Blueback herring 0.04 0.95 347 November Alewife . 0.03- 0.79 289 April / November Lumpfish 0.02 0.47 173 January Tautog 0.02 . 0.42 154 January / November i
' Northem puffer 0.01 0.26 96 October
' Silver hake - 0.01 0.26 96 November l Red hake 0.009 0.21 77 July / November J Windowpane 0.009 0.21 77 April Atlantic tomcod 0.007 0.16 58 October l Hake sp. 0.007 0.16 58 November l Little skate 0.007 0.16- 58 April /May
- Striped searobin _0.007 0.16 58 November Blackspotted stickleback . 0.004 0 11 39 November / December Cunner 0.004 0.11 39 October Northem pipefish 0.004 0.11 39 November White perch - 0.004 0.11 39 January Atlantic herring 0.002 0.05 19 April 4 Bay anchovy . 0.002 0.05 19 August .
Fourspot flounder 0.002 0.05 19 October !
White hake 0.002 0.05 19 June Totals 1.43 34.29 12,514 '
' Rates have been rounded to significant figures.
radmis/ chart 98
Table 6. Impingement Rates Per Hour, Day and Year For All Fishes Collected From Pilgrim Station Intake Screens During 1978-1997, Assuming 100% Operation of Pilgrim Unit 1*
l Dominant Species Year Rate /Hr. Rate / Day RateNear (RateNear) 1978 4.04 97.03 35,416 Rainbow smelt .I (29,357) 1979 3.24 77.69 28,280 Atlantic silverside (20,733) ]
1980 0.66 15.78 5,769 Cunner (1,683) 1981 10.02 240.42 87,752 Atlantic silverside (83,346) 1982 0.93 22.39 8,173 Atlantic silverside (1,696) 1983 0.57 13.65 4,983 Atlantic silverside (1.114) ,
1984+ 0.13 3.13 1,143 Atlantic silverside (185) 1985 1.14 27.46 10,022 Atlantic silverside (3,278) 1986 1.26 30.34 11,075 Atlantic herring (3,760) 1987+ 0.28 6.74 2,460 Rainbow smelt (682) 1988+ 0.27 6.48 2,372 Atlantic silverside (586) 1989 0.80 19.30 7,045 Atlantic silverside l (1,701) 1990 1.70 40.74 14,872 {
Atlantic silverside I (4,354) 1991 3.38 81,14 29,616 Atlantic Herring (22,318) i 190. 0.63 15.22 5,572 Atlantic silverside (2,633) 1993 2.78 66.78 24,375 Rainbow Smelt (9,560) 1994+ 5.97 143.18 52,259 Atlantic silverside (36,970) 1995+ 5.87 141.00 51,464 Alewife (26,972) 1996 3.11 74.64 27,318 Atlantic silverside l (16,153) 1997 1.43 34.29 12,514 Atlantic silverside ]
(5,814)
Means 2.41 57.80 21,098
- Rates have been rounded to significant figures.
+No CWS pumps were in operation 29 March - 13 August 1984,18 February - 8 September 1987, 14 April- 5 June 1988,9 October - 16 November 1994 and 30 March - 15 May 1995.
l 1970's. Stupka and Sharma (1977) showed annual impingement rates at numerous power f plant locations for dominant species, and compared to these rates at Pilgrim Station were i
lower than at most other sites. Recently, Normandeau Associates (1996) compared fish impingement at several marine power p' ant intakes which demonstrated Pilgrim rates to be among the lowest with the exception of inciderts that involve one or two species occasionally.
However, in terms of the number of fish species impinged, Pilgrim Station displays a greater variety than most other power plants in the Gulf of Maine area (Bridges and Anderson,1984a),
perhaps because of its proximity to the boreal-temperate zoogeographical boundary presented to marine biota by Cape Cod.
1 Monthly intake water temperatures recorded during impingement collections at Pilgrim Station were above normal during most of 1997 compared to the mean monthly temperatures for the 10-year interval 1988-1997 (Table 7). During all of 1997, with the exception of May, water temperatures were higher than this 10-year period.
Overall 1990/1995/1997 displayed relatively warm water temperatures, 1987/1989/1991 1994/1996 were average years, and 1988/1992/1993 were cold water years. Pilgrim Station intake temperatures approximate ambient water temperatures. However, a dominance of l
colder water species (i.e., Atlantic silverside, winter flounder, grubby and rainbow smelt) appeared in impingement collections during 1997, with the warmer water species Atlantic menhaden and butterfish also being well represented.
4.2. Invertebrates in 455 collection hours,207+ invertebrates of 15 species (Table 8) were recorded from Pilgrim St: tion intake screens between January-December 1997. The annual collection rate was 0.45+ invertebrates / hour. Jellyfish dominated, being caught only in September, 1
1 1
___________________~
0 0 0 0 9 1 3 5 6 1 7 0 8 8 0 2 3 7 2 8 7 8 3 1 9 8 6 6 6 1 8 0 2 8 6 2 7 8 9 3 3 3 4 4 5 5 5 5 5 4 3 1
2 7 3 7 0 8 7 0 2 2 0 8 9 4 9 4 3 7 3 5 8 6 9 6 5 8 8 2 8 1 8 7 1 9 8 3 5 5 9 3 4 3 4 4 5 6 5 5 5 4 3 1
0 5 5 7 3 6 3 4 1 5 3 8 6 9 4 1 8 6 8 6 2 7 3 1 8 8 t
n 9 8 8 7 6 0 3 1 4 3 5 7 2 1
e 3 3 3 4 5 5 6 6 6 5 4 4 m
e g
n ip 1 6 0 2 6 9 9 7 9 3 0 8 4 9 5 7 7 4 7 0 6 4 6 0 8 7 I
m 9 7 6 9 4 3 0 1 8 8 2 7 1 1
g 3 3 3 4 5 6 6 5 5 5 4 4 n
i7r u9 D91 2 4 2 3 2 6 1 4 0 2 3 5 6 d -
9 3 3 5 4 5 2 9 4 4 8 8 0 e 9 6 4 6 3 1 4 5 0 7 3 0 3 d8 1 3 3 3 4 5 5 5 6 5 5 5 4 r8 o9 c 1 e
R n,
) o 3 6 1 4 6 3 0 8 6 5 6 7 3 Fi t 9 3 2 2 1 3 7 7 6 5 9 9 5
- a 9 7 2 5 1 8 2 6 3 0 3 9 4
( t 1 eS 3 3 3 4 4 5 5 5 5 4 3 3 r r ue t
awo r
e pP 4 1 8 1 5 6 1 5 4 5 3 8 8 9 2 1 9 9 2 2 8 3 4 3 7 8 mra 9 8 9 0 7 4 5 6 9 0 3 5 4 ee Tl c 1
2 2 3 3 4 4 5 5 6 6 5 4 eu kNa t 5 3 1 1 7 7 3 4 1 7 3 1 3 Inmr i 9 9
1 6 5 6 7 4 1 3 3 9 6 3 f g 1
1 6 9 1 8 6 8 7 2 7 0 0 oi l 4 3 3 4 4 5 5 6 6 5 5 4 sP nt aa es 6 5 2 8 3 6 3 8 8 2 0 3 8 Mno 9 1 8 3 8 5 0 0 7 9 5 6 1 lyit 9 7 5 7 1 8 6 6 0 2 7 9 5 1
h c 3 3 3 4 4 5 5 6 6 5 4 4 t e nl oo l MC 7 8 6 0 8 2 0 9 5 6 7 1 4 7
9 7 3 2 0 8 7 5 3 6 6 8 0 e 9 8 7 3 4 7 8 0 2 1 5 0 1 l 1 3 3 3 4 4 5 6 6 6 5 5 4 b
a T 5 g
p
/
t r
r a e r r h y b e e /
c
_ h y
r ar a u h t
s m er bmm b c is t
n o
u n b a e r
a c i r
l r
p a y e n ly u u u u te b o g p e
t c o mcee ena d m
M J F M A M J J A S O N D M a r
h'
c 1
- e 3 3 1 7 3 3
2 D 0
- v. 6 o 1 6 1 1 1 2 21 6 2 4 n N 0 o
it t
a .
S t c 4 7
0 3 3 m O 4 0
ir lg i
P t p +
+ 3 5 m e
- 6 1 1 6 1 5 o S 1 2 6 f
r 0 d 7 t
e 99 t
s c1 u 3 8
g 2 1 2 1 6 1 ler e u 3 lob A 0
Cmse ec y 6 t
a e l u 7 4 5 1 2 9 4 5 rD J 1 3 b 0 t
e y-r r ea vu e 9 n na n 1 6 0 4 3 4 2 2 6 8 I u 1 1 1 3 3 l
J 0 J
lA ,
r s on y 8 f
e a 3 2 4 3 1 3 7 2 t
n r e M 1 4 0
ec mS e
gk e l i
r 3 7 p 6 4 6 na A 2 4 2 2 3 2 3 5 ipn t 0
Imi h
!y c 2 h r 8 1 1 1 1 2 4 2 t a 1 5 n M 0 o
M .
- 1 b 2 2 7 8 e 5 3 1 1 8 1 2 2 le F 0 b
a T n 2 8 a 4 4 4 1 3 2 0 4 6 1 3 4 J 0 p s.
r m ) )
e n .
s hr
. b h r m S h r n h
( #( u s y s e n e a i t b a ih e e d f s c ic e
b e
r a bo diu a h r c r u l p. a d
imt T R a e in e s i
t p n b s e o a
l q c b . a s n n m S o p e b s ea o p ip c on an s n a h r r s s a u o o e h s n n c i n a r r r c s s n c m e a s c c itt r t i
t t is mif b e e gn idk c se ei ee d y e m mla e e r
ly e e v e m f
d om l
r r l n
u e o l l t
le e r o u o o e o o o J S G C A L N R H N G L B N aS T C C r l t U
i i'
Sevenspine bay shrimp (Craaon septemspinosa) were captured in greatest numbers from January - April and were 27% of the enumernted catch. Green crab (Carcinus rnaenus) and common starfish (Asterias forbesi) represented 19.8 and 12.1%, respectively, of the total invertebrates impinged. Unlike the fishes, the 1987 and 1988 invertebrate impingement rates were comparable to 1989 - 1997 despite relatively low circulating water pump capacity available in 1987 and 1988.
A noteworthy occurrence was the collection of so many blue mussels during 1986-1989. This could be an effect of the Pilgrim Station outage during the late 1980s (reduced power level in 1989) which precluded the use of regular thermal backwashes for macrofouling control and the migratory / adhesive abilities of mussels. In 1990 - 1997 several thermal backwashes were performed and blue mussel impingement was minor for those years. During 1997 aggressive biofouling control activities included three effective thermal backwashes during the months of April, July and November.
Green crabs were the third most abundant invertebrate impinged, peaking in July. Common starfish were fourth, being most represented in June. Nineteen specimens of the commercially important American lobster were captured in 1997 ranking them fifth. This equals 366 lobsters l
impinged on an annual basis at 100% operation of Pilgrim Station, under conditions at the l times of impingement. This is considerably less than in 1991-1994 and is more comparable to the number of lobsters impinged in most previous years. The lobsters ranged in site from 28-70 mm carapace length and were impinged mostly in July.
Approximately 1,974 pounds of rnixed algae species were recorded during impingement sampling, for a rate of 4.34 pounds /hr. This equates to 19 tons of algae annually on Pilgrim in.'ake screens. This rate is considerably higher than the low flow 1984, and 1988 outage years, comparable to 1989-1992 and 1994-1996, and much lower than 1993 which experienced very adverse meterological conditions of high winds and coastal storms (panicularly in December).
t 4.3 Fish Survival Fish survival data collected in 1997 while impingement monitoring are shown in Table 9.
Continuous screenwash collections provided the fewest numbers of fishes and revealed an overall survival rate of approximately 61%. Fishes collected during static screen washes fared worse showing a r.urvival rate of 28%. The lower initial survival rate for static screen washes was influenced by the low initial survival of Atlantic silverside which were impinged in abundant numbers. As illustrated in 1993-1997, fishes have a noticeably higher survival rate during continuous screen washes because of reduced exposure time to the effects of impingement.
However, reduced intake currents in 1984, associated with limited circulating water pump operation, may have been a factor in higher static wash survival then because of less stress on impinged individuals; although this wasn't apparent from 1987 and 1988 limited pump operation results.
Among the ten numerically dominant species impinged in 1997, four demonstrated initial survival rates of 50% or greater. Grubby showed 78% survival, winter flounder 95%, alewife 67%, Atlantic silverside 40%, tautog 75%, rainbow smelt 16%, butterfish 46%, Atlantic menhaden 9%, lumpfish 22%, and blueback herring 28%. Some of these high survival percentages may be explained by the robustness and durability of some of the species that were sampled during screenwashes.
Table 9 - Survival Summary for the Fishes Collected During Pilgrim Station Impingement Sampling, January-December 1997. Initial Survival Numbers are Shown Under Static (8-Hour) and Continuous Wash Cycles Number Collected Number Survivina Total Lenath (mm)
Static Cont.
Species Washes Washes Static Cont. Mean Range Atlantic silverside 209 93 60 61 92 68-125 Rainbow smelt 43 39 0 13 109 57-172 Atlantic menhaden 45 11 4 1 66 38-130 Butterfish 16 38 1 24 53 27-89 Winter flounder 22 18 20 18 64 49-95 Grubby 18 5 13 5 68 51-93 Blueback herring 8 10 1 4 95 65-169 Alewife 5 10 1 9 114 53-272 Lumpfish 2 7 1 1 46 27-62 Tautog 3 5 3 3 63 50-78 Northern puffer 5 0 2 -
70 59-88 Silver hake 0 5 -
3 85 62-96 Red hake 2 2 0 2 75 55-90 Windowpane 1 3 1 3 105 70-175 Atlantic tomcod 1 2 1 2 130 123-145 Hake sp. 3 0 0 -
95 68-138 Little skate 2 1 2 1 398 245-483 Striped searobin 0 3 -
3 63 60-65 Blackspotted stickleback 1 1 0 1 39 37-41 Cunner 1 1 1 1 80 57-102 Northern pipefish 0 2 -
2 144 135-152 White perch 0 2 -
0 101 101 Atlantic herring 1 0 0 -
40 40 Bay anchovy 1 0 0 -
80 80 Fourspot flounder 1 0 0 -
105 105 White hake 1 0 0 -
240 240 All Speicies 391 258 111 157 Number (% Surviving) (28.4) (60.9) chart 98. doc
SECTION 5 CONCLUSIONS
- 1. The average Pilgrim impingement rate for the period January-December 1997 was 1.43 fish / hour. The impingement rates for fish in 1984,1987, and 1988 were several times lower than in 1989-1997 because of much reduced circulating water pump capacity during the former years.
- 2. Twenty-five species of fish were recorded in 455 impingement collection hours duri.1g 1997. In 1989-1997several times the number of fishes were sampled as compared to 1984 and 1988, despite more collection hours in 1984 and an average number of hours in 1988. This illustrates the importance that the number of circulating pumps operating has on the quantity of impinged organisms. Substantially less collecting hours for portions of 1987 precluded its comparison with other years.
- 3. At 100% yearly operation the estimated maximum January-December 1997 impingement rate was 12,514 fishes. This projected annual fish impingement rate was much lower than recent years' rates because of several large impingement incidents during the past few years.
- 4. The major species collected and their relative percentages of the total collections were Atlantic silverside,46.5%; rainbow smelt,12.6%; Atlantic menhaden,8.6%; and butterfish, 8.3%.
- 5. The peak in impingement collections occurred during November / December when 59% of the antiual catch of Atlantic silverside occurred.
- 6. Monthly intake water temperatures, which generally reflect ambient water temperatures, were higher for 1997 than the ten-year monthly averages for the period 1988-1997, with the exception of May, which was lower than normal.
- 7. The hourly collection rate for invertebrates was 0.45+. Jellyfish dominated in September.
Sevenspine bay shrimp were second because of relatively large winter collections. Green crab and common starfish were 19.8 and 12.1% of the enumerated catch. Nineteen American lobsters were collected which equates to a potential 1997 impingement of 366 lobsters.
- 8. Impinged fish initial survival was approximately 28% during static screen washes and 61%
during continuous washes for pooled species. Of the ten fishes imninged in greatest numbers during 1997, four showed initial survival rates of 50% or greater.
SECTION 6 LITERATURE CITED American Fisheries Society.1988. Common and scientific names of aquatic invertebrates from the United States and Canada: mollusks. Spec. Pub. No.16: 277 pp.
1989. Common and scientific names of aquatic invertebrates from the United States and Canada. decapod crustaceans. Spec. Pub. No.17:77 pp.
1991a. A list of common and scientific names of fishes from the United States and Canada. Spec. Pub. No. 20: 183 pp.
1991b. Common and scientific names of aquatic invertebrates from the United States and Canada: cnidaria and ctenophora. Spec. Pub. No. 22: 75pp.
Anderson, C.O., Jr., D.J. Brown, B.A. Ketschke, E. M. Elliott and P. L. Rule.1975. The effects of the addition of a fourth generating unit at the Salem Harbor Electric Generating Station on the marine ecosystem of Salem Harbor. Mass. Div. Mar. Fish., Boston: 47 pp.
Briges, W.L. and R. D. Anderson.1984a. A brief survey of Pilgrim Nuclear Power Plant effects upon the marine aquatic environment, p. 263-271. M: J. D. Davis and D. Merriman (editors),
Observations on the ecology and biology of westem Cape Cod Bay, Massachusetts,289 pp.
Springer- Verlag. (Lecture Notes on Coastal and Esturaine Studies, Vol.11).
Lawton, R. P., R. D. Anderson, P. Brady, C. Sheehan, W. Sides, E. Koulokeras, M. Borgatti, and V. Malkoski. 1984b. Fishes of western inshore Cape Cod Bay: studies in the vicinity of the Rocky Point shoreline, P. 191-230. b: J. D. Davis and D. Merriman (editors), Observations on the ecology and biology of western Cape Cod Bay, Massachusetts, 289 pp. Springer-Verlag. (Lecture Notes on Coastal and Esuarine Studies, Vol.11).
Normandeau Associates.1996. Seabrook Station 1995 environmental studies in the Hampton l
- Seabrook area: a characterization of environmental conditions during the operation of Seabrook Station. Section 5.0 - Fish. Northeast Utilities Service Company: 103 pp.
Stupka, R. C. and R. K. Sharma. 1977. Survey of fish impingement at power plants in the United States Vol. Ill. Estuaries and Coastal waters. Argonne National Lab.: 310 pp.
Radmisc/impgJan 1
p asar !
$be Yontniontreallb o IEadJarbadelk QT) ! 9 Uirian <f. //atine si>Letha sa c/ .h4hte %ive
's V , '
\ ;Anwf .Ac/ was.9 PHILIP G. CoATES 3
[DRECTOR Fax: 508 563 5482 MEMORANDUM To: Members of the Administrative-Technical Committee, Pilgrim Power Plant Investigations From: John Boardman, Recording Secretary
Subject:
Minutes from the 88th meeting of the A-T Committee Date: September'29,1997 The 88th meeting of the Pilgrim A-T Committee was called to order by Gerald Szal at 10:00 A.M., on September 15,1997 at Blackstone Sewage Treatment Plant.
I. Minutes of the 87th Meeting The committee had not received a fmal copy of the minutes from the 87th meeting. Bob Lawton stated that he has the final copy of the minutes, and they will be sent to the committee members. The minutes from the 87th meeting will not be accepted until the committee reviews the final draft.
II, Pilgrim Station Operational Review Bob Anderson reported on the current operational status at Pilgrim Station. A planned thirty day refueling outage was started in mid-February. However, the main transformer " blew" at the beginning of March, and this extended the shutdown by several weeks. The plant had its highest output during the period from May through August 1997. During this time the plant operated at about 96.0% capacity. Overall the plant has operated at approximately 68.0%, thus far, for 1997.
Bob also talked about the dredging project in the intake, which commenced in late June. By the end of July,21,000yd' of sediment had been removed from the intake. A total of 30,000yd' was removed by the end of August. Bob indicated that Boston Edison Company is licensed to dredge approximately 68,000yd'. The remaining sediment may be removed in the future.
III. B.cnthic Monitoring Bob Anderson presented the results of the Benthic Monitoring Program. Bob explained that the total affected area was larger than in years past. The expansion of the impacted area may be linked to an increase in plant production.
Bob Anderson told the committee that he and Jack Parr are going to meet with Isabelle
Williams of ENSR to discuss the quality of the benthic monitoring work. The committee discussed the benthic program for several minutes. Bob Maietta made a motion to leave this issue up to the Benthic subcommittee. Bob Lawton seconded this motion.
Bob Anderson put forth a request on behalf of Boston Edison Company to look at the frequency of benthic sampling. Bob suggested reducing the sampling from quarterly to twice a year. The committee talked about this proposal at length. It was decided that Jerry Szal would contact Jack Parr and discuss the possibility oflowering the number of dives to three. Jerry will contact the committee and explain Jack's recommendations. Jerry will also ask committee members to vote on the benthic monitoring program at that time. Bob Anderson told the committee that he would like to have a decision on this matter by October.
IV. Fisheries Monitoring Bob Anderson presewed the committee with a letter from Boston Edison Company regarding the fisheries contract. The committee agreed with the letter, but told Bob Anderson there would be no guarantee of ending the fisheries program in two years.
Bob Lawton then discussed the need for the committee to scope out the work to be done before any future bidding process takes place. Bob then explained the benefits cf continuing each project (winter flounder, cunner, smelt). The committee talked at length about the need to further study these species. The committee concluded that sufficient effort had been given to the cunner and smelt projects. It was determined that these studies be dropped from the fisheries contract.
However, the committee agreed that the winter flounder project should continue in an expanded fashion. Funds from the cunner and smelt projects will now be used to enhance flounder investigations.
V. Impingement and Entrainment Monitoring Carolyn Griswold made a proposal to continue both programs as before with one minor change. Mike Scherer will no longer be comparing net mesh sizes. The committee agreed with this proposal. The committee also recommended that the barrier net not be used for the rest of 1997 and all of1998.
VI. The meeting adjourned at 2:30 p.m.
+
VII. Attendees at the 88th meeting of the A-T Committee:
CZM - Rick Zeroka DEP- Gerry Szal Bob Maietta DMF- Bob Lawton Vin Malkoski John Boardman NMFS - Carolyn Griswold PNPS - Bob Anderson
PILORIM NUCLEAR POWER PLANT ADMINISTRATIVE-TECHNICAL COMMITTEE MAY, 1997 Robert D. Anderson Nicholas Prodany Boston Edison Company U.S. Environmental Protection Agency Pilgrim Nuclear Power Station Region 1, Industrial Permits Section 600 Rocky Hill Road JFK Federal Building Plymouth, MA 02360-5599 Boston, MA 02203 (508) 830-7935 (617) 565-3567 FAX (508) 830-8575 Rick Zeroka W. Leigh Bridges MA Coastal Zone Management MA Division of Marine Fisheries 100 Cambridge Street, Floor 20 State Office Building Boston, MA 02202 100 Cambridge Street (617) 727-9530 Boston, MA 02202 (617) 727-3194 Gerald Szal DEP-Division of Watershed Management Carolyn Griswold 627 Main Street, 2nd Floor National Marine Fisheries Service Worcester, MA 01608 28 Tarzwell Drive (508) 767-2789 Narragansett, RI 02882 (401) 782-3273 ------------- ---------------
John Chisholm OTHER CONTACTS MA Division of Marine Fisheries 50A Portside Drive Dr. James Blake Pocasset, MA 02559 ENSR Consulting and Engineering (508) 563-1779 89 Water Street Woods Hole, MA 02543 Robert Lawton (508) 457-7900 MA Division of Marine Fisheries FAX (508) 457-7595 50A Portside Drive Pocasset, MA 02559 Derek Mcdonald (508) 563 1779 x 118 Marine Biofouling Control Corp.
(508) 888-4431 Robert Maietta DEP-Division of Watershed Management Dr. Michael Scherer 637 Main Street, 2nd Floor Marine Research Inc.
Worcester, MA 01608 141 Falmouth Heights Road (508) 767-3793 Falmouth, MA 02540 Dg. Martha Mather Dr. Jan Praeger MA Coop Fish & Wildlife Unit U.S. Environmental Protection Agency Holdsworth Hall Environmental Research Lab University of Massachusetts 27 Tartwell Drive Amherst, MA 01003 Narragansett, RI 02882 (413) 545-4895 (401) 782-3090 Dr. Don Miller Steve Halterman U.S. Environmental Protection Agency DEP-Division of Watershed Management Environmental Research Lab 627 Main Street, 2nd Floor 27 Tarzwell Drive Worcester, MA 01608 Narragansett, RI 02882 (508) 849-4003 (401) 782-3090 Jack Paar U. S. Environmental Protection Agency New England Regional Lab Surveillance and Analysis 60 Westview Street Lexington, MA 02173 (617) 860-4604
__