Monitoring the Marine Environment of Long Island Sound at Millstone Nuclear Power Station Waterford, Connecticut THREE-UNIT OPERATIONAL STUDIES 1986 - 1987 NORTHEAST UnUTKS m WC&ni»h M»SSAf HuWTTS liCCTRiC COMHW- NU ENVIRONMENTAL LAB APRIL 1988 Monitoring the Marine Environment of Long Island Sound at Millstone Nuclear Power Station Waterford, Connecticut THREE-UNIT OPERATIONAL STUDIES 1986 - 1987 April 21, 1988 Northeast Utilities Environmental Laboratory Waterford, Connecticut Digitized by the Internet Archive in 2010 with funding from Boston Library Consortium Member Libraries http://www.archive.org/details/monitoringmarine02nort Monitoring Studies, 1986-1987 Executive Summary Rocky Intertidal Studies The rocky intertidal sampling program was designed to assess potential biological perturbation from operation of Units 1 and 2 and construction of Unit 3, and to provide base-line data that would permit prediction and assessment of potential impact from Unit 3 operation. Attached plant and animal species on local rocky shores were identified, temporal and spatial patterns of occurrence and abundance of these benthic species were examined, and physical and biological factors that induce variability were identified. This report includes data from the initial 3-unit operational period. Qualitative algal collections, quantitative studies of intertidal organisms, recolonization studies, and Ascophyllum nodosum studies were performed to EvSsess impact, and are summarized below. The local flora, as characterized by qualitative algal collections, has shown consistent spatial and temporal patterns of distribution during MNPS operation. Overall, 160 algal species have been identified since the inception of the monitoring program in 1979, consisting of 75 reds, 40 browns, and 45 greens. In the 3-unit operational period to date, 128 species have been found; 59 reds, 33 greens, and 36 browns. Divisional proportions and seasonal distributions have not changed during 3-unit operation. Anlilhamnionella floccoswn and Neinalion helminthoides have been identified only in the 3-unit operational period, but their occurrence is related to sporadic distribution rather than power-plant operation. Species composition at Fox Island-Exposed under 3-unit operating conditions continues to resemble that found under 2-unit, 2-cut conditions. Quantitative studies show intertidal zonation patterns typical of rocky shores throughout New England, with the high intertidal dominated by barnacles, the mid intertidal by barnacles and fucoids, and the low intertidal dominated by Chondrns crispus, a perennial red alga. The abundances of these major components of local rocky shore communities var>' over time and space. Variations are predictable and explainable in terms of seasonality, degree of exposure, intertidal height, inter- and intraspecific competition, and life-history of the organisms. Changes to communities have been minor, indicating stable environmental conditions during MNPS operation. An exception to the local stability was the development of a community dominated by opportunistic ephemeral algae after the opening of the second quarry cut in August 1983 at Fox Island-Exposed (I'E), the station closest to the discharges. This change was attributed to thermal incursion and water temperatures in excess of 28 °C. High water temperatures in late summer 1984 were responsible for the elimination of the perennial algae, Chondi-us rrispw;, A.scophylhtm nodosum and Funis vesiculosus from the low intertidal at FE. Codium fragile, a large green alga, became and remains a dominant component of the EE community. However, Fucus has recolonizcd the mid intertidal zone at EE; its persistence is evidence of stability at a hew level. Recolonization studies, performed during 2-utiit operation, allowed isolation and identification of some factors that influence the structure of local rocky intertidal communities. Recolonization was influenced by time of year in which denuding occurred, and related to degree of exposure and intertidal height, e.g.. Executive Summary rapid in the high intertidal of an exposed station and slow in the low intertidal of a shcUered station. Preliminary results from studies re-established during 3-unit operation corroborate those conclusions. Growth and mortality studies of Ascophyllum nodosum, a perennial brown alga sensitive to water temperature change, were included in the rocky intertidal sampling program. Tip length analyses helped distinguish between a stressed population at Fox Island and populations at two reference stations. Ascophyllum tip length data, fit to a Gompertz growth function, shows a response to elevated water temperature. With the exception of the FE intertidal community, no significant changes to the benthic shore biota were observed that could be attributed to MNPS operation. Benthic Infauna Intertidal and subtidal infaunal communities in the vicinity of MNPS were sampled during start-up of Millstone Unit 3 and over the first operational year. Power plant-induced impacts were assessed by characterizing communities in tenns of community structure and abundance and comparing these data with those collected before Unit 3 start-up. Intertidal infaunal communities exhibited spatial patterns in community abundance and composition and seasonal fluctuations in abundance that were consistent with those observed during the base-line period. Higher infaunal abundances and number of species continued to occur at .lordan C'ove (.IC), while communities at White Point (WP) and Giants Neck (CjN) generally included lower numbers of individuals and species. At the population level, the .IC community continued to be dominated by oligochaetes while Paraonis fulgens, Haploscoloplos fragilis and rhynchococis dominated the WP and GN communities. Although spatial patterns among intertidal stations were consistent between pre-operational and operational periods, there were temporal changes in community abundance and species composition evident at all sampling stations. Many of these differences appeared to be continuations of trends which were first evident before Unit 3 operation commenced, suggesting that observed declines were a response to some regional event. At JC, reductions in total abundance and species number were also evident, but they were not as extensive as those at GN and WP. There were, however, unusually low densities of Scolecolepides viridis and Hediste diversic.olor in .lune 1987. Low densities of Scolecolepides viridis were also evident at GN in .lune 1987, indicating a possible area-wide decrease. Sampling intertidal communities during 1986-87 revealed no immediate changes to these communities which could be directly attributed to Unit 3 start-up or to 3-unit operations at MNPS. Possible power plant impacts related to temperature could not be assessed in this report, given the limited 3-unit operational period. At subtidal stations. Unit 3 start-up resulted in scouring of bottom sediments at Ivfflucnt (EF), located immediately offshore from the discharge cut. Along with sedimentary changes, we observed reductions in macrofaunal density and in particular, number of species. At the species level, lowered abundances of Polycimis eximius, Aricidca c.atherinae, oligochaetes, Tcllina agilis, Tharyx spp. and Lwnhrineris tenuis were recorded and attributed to a reduction in silt/clay content. Power plant-related impacts were also evident at JC after 3-unit start-up and included a substantial increase in silt/clay attributed to the transport and settling of sediments scoured from the area of the Unit 3 discharge cut. Concurrent with this sediment change, there was a significant increase in the abundance of Mediomastus amhiseta along with a decline in the abundance of some previously dominant polychactes Ahcidea catherinae and Polycimis eximius and oligochaetes. The Intake (IN) area also exhibited changes in sediments and infaunal communities during the operational period. Silt/clay values during 1987 were generally lower than those in recent years and Monitoring Studies, 1986-1987 may be reflecting increased scour during 3-unit operations. The infaunal community in 1987' exhibited significant increases in the abundances of the amphipods, Ampelisca spp., Le.ptoc.heirus pinguis and Uniciola irrorata. These organisms were among the dominants at this station prior to the power plant-induced shifts observed following Unit 3 intake construction. Their return is believed indicative of the continued recovery of this community. Power plant-related impacts on subtidal communities during the first year of 3-unit operation appeared to be a response to scouring (at EF and IN and the transport and deposition of this sediment into JC. Infaunal habitats immediately beneath the Unit 3 discharge (and extending seaward approximately 100 m) have been eliminated due to scouring by the 3-unit discharge; however, given the limited size of this area, loss of this infaunal habitat would not significantly alter ecosystem functioning in the greater Millstone area. At .IC, the deposition of sediments apparently from the discharge area would be expected as a temporary change. Infaunal changes related to this siltation should also be temporary and cause no detectable changes in the ecology of the Millstone area. If however, the sedimentary changes are the result of some plant-induced change in the sedimentation patterns in .IC or the infaunal shifts were a response to temperature, then more widespread and ecologically significant changes in local infaunal communities might occur. Given limited operational history, impacts due to temperature could not be identified separately from those which occurred in response to sediment changes. Lobster Population Dynamics The lobster population in the Millstone Point area was sampled from May through October during 1986 and 1987 using wire pots set at three stations. The carapace length, sex, presence of eggs, missing claws and molt stage was recorded for each lobster to quantify population characteristics. I,obsters > 55 mm carapace length were tagged and released where they were caught to monitor growth and movement. Studies were also conducted on adult lobsters caught on the intake traveling screens (impingement) and on larval lobsters drawn through the plants cooling water system (entrainment). All these studies were designed to assess the impacts of 3-unit operation on the local lobster population. Total catch per unit effort (CPUE; #lobstcrs/# pots hauled) during 1986 and 1987 was 1.70 and 1.72, respectively, and within the range of values reported during 2-unit operations (0.85-2.10). I^gal CPUE was lower during 1986 and 1987 (0.097, 0.089) when compared to previous years' results and has significantly declined since 1978. The lower CPUE of legal-sized lobsters may be related to fishing pressure which has increased in Long Island Sound since 1978. A 50% decline in catch at .lordan Cove occurred from August to September 1986 and was related to 3-unit operations. Sediments in the discharge area were scoured and subsequently deposited in Jordan Cove, where lobster habitats were buried by sediments. This decline in catch was only temporary, since catches in October 1986 and throughout 1987 at .lordan Cove were normal and indicated that sediments had stabilized and lobsters had returned to the affected area. Similar impacts associated with the disruption of lobster habitat were observed in previous years following dredging in the vicinity of the intake structures. The mean size of lobsters caught during 3-unit operation was 70.1 mm in 1986 and 70.2 mm in 1987. These values were smaller than those reported in previous years (range 70.7-71.8 mm) due to the lower CPUE of legal-sized lobsters during 1986-87. Male to female sex ratios during 1986 and 1987 were 1.0:0.87 and 1.0:0.88, respectively, and within the range of values reported during 2-unit studies. The Twotree station continued to yield a higher proportion of females than the other two nearshore stations, a trend consistent since the study began. Female size at sexual maturity was similar during 2- and 3-unit operations; females began to mature between 50 and 55 mm CI, and all females were mature at sizes greater than 95 Executive Summary mm CL. The mean CL of berried females during 1986 (78.0 mm) and 1987 (76.5 mm) and the proportions of sublegal size berried females caught in 1986 (75%) and 1987 (90%) confirmed the small size at first sexual maturity of females in the Millstone area. I_x)bsters that were near molting comprised 3.2% and 3.0% of the 1986 and 1987 total catches, respectively, which were within the range of values reported during 2-unit operations. Growth per molt averaged 13.3% in 1986-87 compared to 13.9% from 1978-85. The percentage of lobsters missing one or both claws (culls) in 1986 (10.6%) and 1987 (10.3%) was lower than the average percentage culled in previous years (range 10.6%- 15.5%) due to the implementation of the escape vent regulation in 1984. The numbers of lobsters tagged in 1986 (5,698) and 1987 (5,680) were within the range of annual values for lobsters tagged in pre-operational studies. Recapture rates for 1986 (21.0%) and 1987 (23.9%) were also similar to pre-operational values (range 15.9''/o-23.9%). Lobstcrmen recaptured 20.2% of our tagged lobsters in 1986 and 17.8% in 1987. Lobster movements were localized, since 94% and 97% of all commercial recaptures were made within 8 km of Millstone Point during 1986 and 1987, respectively. Several lobsters moved outside LIS and were caught in waters off RI and MA; three lobsters moved offshore during 1986-87, where they were caught in deep water canyons on the edge of the continental shelf. Ix)bster larvae densities (number per 1000 m ) in entrainment samples were higher in 1986 (0.88) and 1987 (0.63) when compared to 1984-85 (0.42-0.43). The estimate of total lobster larvae entrainment, based on sample density and total MNPS cooling water demand, was also higher in 1986 (548,635) and 1987 (304,694) when compared to 1984-85 (79,511-138,820). Ix)bster larvae survival was 6.5% and 3.8% in 1986 and 1987, respectively. More stage IV larve were collected in 1986 compared to 1984, 1985, and 1987 when greater numbers of stage I larvae were collected. The estimated numbers of lobster impinged at Unit 2 during 1986 and 1987 were 676 and 825, respectively, these values were within the range of values reported in previous years (261-1220). Fish return systems at Units 1 and 3 improved overall survival of impinged lobsters. Based on impingement of all organisms at Unit 2 since 1972, a request made by NUSCO to discontinue impingement monitoring at Unit 2 was accepted by the CT DEP in December 1987. There is no evidence to date that MNPS has significantly affected the local lobster population. The displacement of lobsters in .lordan Cove due to scouring of sediments in the discharge area was temporary and related to the intial period of simultaneous 3-unit operation. Impacts on larval lobsters entrained tlirough the MNPS cooling water systems will not be apparent in the adult population until 4-5 years when they grow to a size vulnerable to capture in our traps. Winter Flounder Studies The life history and population dynamics of the winter flounder {Pse.udopkuronectes americanus) have been intensively studied since 1973 due to its importance to the sport and commercial fisheries of Con- necticut. Because of the localized nature of winter flounder stocks, the population closest to MNPS, which spawns in the Niantic River, has received most of our attention. Indices of abundance estimates are available from 1976 through 1987 for the adults spawning in the Niantic River. These include a Jolly composite abundance index and median trawl CPUE, both computed from data collected during annual mark and recapture surveys. The 1987 .Jolly index showed a slight increase over the 12-year low value found in 1986, but abundance remained low in comparison to earlier years. Monitoring Studies, 1986-1987 The median CPUE also increased relative to 1986 and was similar in magnitude to those for 1984 and 1985. However, this abundance index was only 30 to 60% of the estimates for 1976 through 1983. Annual median CPUE was also determined for juveniles ( < 1 5 cm) taken in the Niantic River during the adult population abundance surveys. Juvenile catches were more variable than those of adults. Compar- atively low medians were found in 1986 and 1987, which suggested poor reproductive success in recent years. However, factors such as differential distribution within the river among the years most likely affected the reliability of this index of abundance. The winter flounder was the most common fish taken in the trawl monitoring program (TMP). Annual TMP 5-mean CPUE values were computed for the first time for all winter flounder taken each year from 1976 through 1986. The pattern of annual 5-mean abundances differed from that for the Niantic River median CPUE values. The peak in S-means persisted from 1977 through 1983, but was not as pronounced as it was for the Niantic River medians, which were highest in 1981 and 1982. The §-means for 1985 and 1986 were greater than those in 1977 and 1978, while the Niantic River median CPUE in recent years indicated lower abundance than during the 1970s. About one-half to three-quarters of the fish taken by the TMP were larger than 15 cm. However, small ( < 15 cm) fish made up about two-thirds of the total from .January through April as larger fish congregated on the spawning grounds. Catch of smaller fish outside of the Niantic River as measured by the 8-mean fluctuated less than the corresponding median CPUE. This suggested that juvenile abundance may not have been as low as recent catches in the Niantic River surveys indicated. Both the armual Niantic River median CPUE and the TMP 5-means were compared to other regional indices of abundance, including two commercial fishing CPUE for Rhode Island and one for Connecticut and a University of Rhode Island research trawl time-series. With a few exceptions, most indices were significantly correlated and thus described real trends in abundance that occurred throughout Southern New England. Examination of the Rhode Island historical time-series showed that winter flounder abundances have typically fluctuated over time. Sharp increases in catches were most likely related to occurrences of particularly large year-classes. Recent declines in winter flounder populations have reduced abundance to levels at or below those found in the early 1950s and early 1970s. The sex ratio of winter flounder spawning in the Niantic River over the past 11 years was 1.33 females for every male. However, during the past 2 years more males than females were tsiken. The length of females at 50% sexual maturation was 26.8 cm, which corresponded to a fish 3 or 4 years old. Spawning in the Niantic River was usually completed by late March or eariy April and appeared to have been related to water temperature, with proportionately fewer females spawning earlier during colder years. Based on the abundance indices of females and their size distribution, annual indices of egg production were deter- mined. This index peaked in 1982 and has declined about 80% since then. However, as shown below, adult abundance and absolute egg production alone were not the only factors in determining reproductive success. A stock and recruitment relationship for Niantic River winter flounder was determined using the 1 2 years of abundance and life history data with the Ricker model. For each year, parental stock was defmed as all winter flounder age 3 and older and recruits were those fish turning 3 years old each spawning season. The two-parameter Ricker model only explained 44% of the variability seen in annual recruitment, with large differences in year-class strength seen for similar-sized parental stocks. Annual February water temperature deviations from a long-term mean were found to have been significantly and inversely correlated Executive Summary with recruitment indices. The addition of a temperature parameter to the model resulted in a much improved fit to the observed data (R = 0.78). Although the actual mechanisms affecting winter flounder recruitment were unknown, the February water temperature appeared to have been related to and ser\'ed as a good measure of those factors. Larval winter flounder studies have been conducted in Niantic River and Bay since 1983 and entrainment data are available from 1976 through the present. During 1986 and 1987, abundance peaked first in the river and then in the bay, which was similar to previous findings; the lag in dates corresponded to flushing rates in the river. Most larvae entering the bay from the river were in Stage 2 of development. Greater mortality of larvae occurred in 1987 than 1986, largely early in the season when most larvae were in the river. The effect of jellyfish predation was not as apparent during the past 2 years as it was during 1983-85. Examination of length-frequency distributions indicated that most mortality occurred during the 3 to 4-mm size-class, suggesting that this was a critical period for mortality. Annual total larval mortality rates, based on the difference in abundance between the 3- and 7-mm size-classes, ranged from 84.6 to 96.9%. Most larvae entrained were in Stage 3 of development. As expected, total entrainment estimates for 1987 following the start-up of MNPS Unit 3 were among the highest during the last 12 years, even though the median density (number per 500 m ) was among the lowest. Entrainment estimates were dependent upon plant operating conditions as well as larval densities each year. Dates of peak abundance for entrainment samples were positively correlated with March and April water temperatures. From the 12 years of entrainment data, the shape of the abundance curve, as measured by the K parameter of the Gompertz function, was found to be a good predictor of subsequent recruitment of age 3 winter flounder. The shape of the abundance curve was related to February water temperatures, with a narrow, high-peaked curve found during warmer years (low recruitment) and a broad, flatter curve during colder years (high recruitment). Laboratory studies showed that larval growth rates were dependent upon water temperature. Of the four temperature regimes examined, optimum temperatures for growth were intermediate (6.9 and 7.5°C), with decreased growth occurring at lower (5.4°C) or higher (10.8°) temperatures. Annual growth rates estimated for 1983-87 using field data collected at Station C in the Niantic River and for 1976-87 using entrainment data were consistent with the findings based on the laboratory data. Again, growth was found to be dependent upon water temperatures. Post-larval young-of-the-year winter flounder have been collected at two stations in the Niantic River since 1983. Densities at Station LR were higher in 1987 than in previous years. Smaller differences were found in growth of young at both stations in 1986 and 1987 than during 1984 and 1985. Differences among years may have been due to density-dependent growth, especially at LR. Survival rates were very similar among years, regardless of densities of young. The winter flounder was the second-most abundant fish impinged on the traveling screens at MNPS since 1976. However, relatively few specimens were taken at Unit 2 during the past 3 years because of declining abundance, varying plant operations, and possible reductions related to the construction and operation of Unit 3. The installation of fish return sluiceways at Units 1 and 3 has lessened the impact of impingement on the winter flounder because it has good ( > 85%) survival when returned to the water. Routine impingement monitoring at Unit 2 was discontinued in December 1987 upon agreement between NU and CT DEP. viii Monitoring Studies, 1986-1987 To predict the long-term effects of larval entrainment at MNPS, an impact assessment model for winter flounder is currently under development, which includes hydrodynamics and population dynamics submodels. The function of the submodels is the estimation of the fraction of total larval production lost to entrairmient at the plant and the measurement of any resulting population changes, respectively. A newer, more accurate and detailed hydrodynamics submodel is under development at the Massachusetts Institute of Technology. Larval behavior will be simulated to correspond more realistically to observations made in the field. A stochastic age- structured population submodel will incorporate the three-parameter stock-recruitment relationship, which includes a measure of compensatory mortality and the introduction of realistic environmental variability. Results from both larval analyses and the three-parameter stock-recruitment relationship showed that year-class strength was related to events in the early life history stages, with colder winters associated with better reproductive success. Greatest winter flounder mortality took place during Stage 2 of development, during which density- dependent mortality probably occurred. Exposure Panel Program The Exposure Panel Program was designed to assess the effect of Millstone Nuclear Power Station (MNPS) on the abundance and distribution of marine woodborers, and the associated rate of wood-loss. To achieve this objective, fouling organisms, woodborer densities and wood-loss arc being monitored at five dock sites and three thermal plume sites (100, 500 and 1000 m from the quarry discharge). Results presented in this report cover the initial period of 3-unit operation and provide a comparison to data collected during 2-unit operation. The fouling community on wooden exposure panels showed no clear response to 3-unit operation, but this community has correlated negatively with shipworm recruitment throughout the study. Fouling assemblages continue to be diverse, and the abundance and distribution of the component species remain patchy. Patterns of abundance and distribution of several prevalent fouling species were consistent between 2-unit and 3-unit operational periods, i.e., Cryptosula pallasiana and Laminaria saccharina at sites not influenced by the undiluted thermal plume and Mylilus edulis within tlie undiluted plume. Changes in the densities of juvenile barnacles and Balanus crenalus between the two operational periods were attributed to a sliglitly later set during the 3-unit operational period. The large percentage of cover and density of Balanus balanoides, an intertidal barnacle on subtidal panels during 1986 and 1987, were similar to incidental occurrences reported for this species in 1970 and 1971. Shipworm densities showed the most consistent differences between 2-unit and 3-unit operation. During the furst six-month collection period of 3-unit operation (May to November 1986), there was an increase in density of the shipworm, Teredo navalis, and an increase in the amount of wood-loss at the White Point and Fox Island sites. At our reference site. Giants Neck, shipworm density decreased. The largest increases in shipworm densities and wood-loss occurred in panels at our undiluted effluent sites during 3-unit operation. These increases were caused by Teredo hartschi, a non-native species collected at efQuent sites since 1975. The reason for this change is unclear. Because water temperatures at our effluent sites remain very similar between 2-unit and 3-unit operating conditions, some other mechanism was responsible for these observed increases. Possibl> , altered water circulation patterns in the effluent quarry allowed more larvae to reach these panels, which were suspended more than I m above the bottom. Panels placed at 100, 500, and 1000 m from the quarry cuts showed decreased recruitment of T. navalis at increasing distances from the quarry cuts. This was consistent with past results. Teredo bartschi was Executive Sunmiary collected in panels at 100 m during 2-unit operation, but was not sampled during initial 3-unit operation because the 100 m panels, which were exposed during May to October, were tampered with by people fishing from boats. Teredo bartschi was not collected at 500 or 1000 m during either 2-unit or 3-unit operation at MNPS. In conclusion, since Unit 3 began operation, increased shipworm abundance and increased wood-loss were observed at sites in the MNPS effluent, and during one exposure period, at White Point and Fox Island, which were potentially exposed to the 3-unit thermal plume. Further monitoring will be required to determine whether these changes are related to 3-unit operation, or are expressions of natural variability. Fish Ecology Studies The operation of MNPS could affect fish assemblages in the Millstone area by increasing mortality at various life history stages (eggs and larvae may be entrained and juveniles and adults may be impinged), or by altering the thermal environment such that the spatial distributions of some fishes change. The report this year emphasizes the comparison of data collected during 2-unit operations to those coUected since the start-up of Unit 3. Impingement monitoring at Unit 2 was discontinued on December 11, 1987 because losses were well- documented and all feasible mitigative measures had been investigated. Significant declines in total impingement were found in recent years, which were attributed to physical changes near the Unit 2 intakes and possible changes in water circulation patterns because of the operation of Unit 3. Losses due to impingement by MNPS were reduced with the installation of fish return sluiceways at Units 1 and 3. Over 100 fish taxa have been collected in the demersal trawl, shore-zone seine, impingement, and ichthyoplankton sampling programs since 1976. Eight of these taxa were selected for detailed examination due to their prevalence in entrainment or impingement collections or their abundance in the shore-zone area of Jordan Cove, an aiea which may be impacted by the thermal plume. The American sand lance was primarily collected as larvae and was a dominant entrained taxon. A decline in larval abundance has occurred since the early 1980s. This decrease was found throughout the region and included the abundance of adults. The decline in larval abundance in the Millstone area was attributed to this area- wide decrease in adult stock size. All life history stages of anchovies were very abundant in most of the sampling programs. Adults were present in impingement collections, juveniles were caught by trawl, and eggs and larvae were abundant in entrainment samples. The number impinged declined in recent years, as it has for most species. Based on differential survival from egg to larval stages which was dependent upon initial densities, compensatory mortality was apparent in its early life history. This may mitigate losses of these life stages due to entrainment at MNPS. Sticklebacks and Atlantic tomcod were primarily found in impingement samples. The impact of MNPS impingement on sticklebacks has been reduced due to high ( > 70%) survival of individuals returned by sluiceways at Units 1 and 3. There was a marked decrease in Atlantic tomcod numbers impinged at Unit 2 since the start-up of Unit 3. This species also exhibits naturally wide fluctuations in numbers because of its reproductive strategy. Monitoring Studies, 1986-1987 Silversides dominated the shore-zone area of Jordan Cove and adults were abundant in winter trawl and impingement collections. There was a recent decline in the number impinged at Unit 2 that was not evident in the number caught by trawl. No apparent changes occurred in length-frequency distribution or seasonal abundance in Jordan Cove related to the 3-unit thermal plume. Grubby larvae were present in ichthyoplankton collections and juveniles and adults were taken in trawl and impingement collections. Larval abundance declined in recent years to levels similar to those of the late 1970s. Numbers impinged at Unit 2 have decreased in recent years, most likely due to physical changes near the intake and because of the start-up of Unit 3. Individuals returned by the Units 1 and 3 sluiceways had higli survival ( > 74''/o). Except for recent years at the Niantic River station, no long-term changes were found in the mean length or in abundance indices of adults collected by trawl. The tautog is an important recreational fish in the Millstone area and the greatest potential impact of MNPS on it is through the entrainment of eggs. Egg abundance, the best index of adult stock size, increased in recent years, but larval densities declined to levels similar to the late 1970s. An apparent low egg to larval survival was found in all years examined. Because the tautog takes 2 to 4 years to reach maturity, the possible impact of entrainment by 3-unit operations on adult stock size will not be evident for several years. Abundances of all life history stages of cunner collected near MNPS have recently declined. In 1986, the numbers of eggs and larvae were among the lowest found since 1976. Decreases in the trawl catch were evident at stations closest to MNPS. Part of the declines in both impingement and in juvenile and adult abundance at a station near the intakes could have been caused by physical changes to the habitat, but reasons for the decline at other stations are not known. The apparent low egg to larval survival in all years was significantly correlated with that for the tautog, suggesting similar factors affecting their repro- ductive success. Monitoring will continue to determine if the recent decrease in the cunner population was due to a natural fluctuation in abundance or from the operation of MNPS. Hydrothermal Studies During 1987 the extent and configuration of the 3-unit thermal plume were determined during various tidal regimes. A dye survey was conducted on 26 August 1987 and supplemental temperatures were recorded by automatic data loggers, which were deployed during October. The configuration and extent of the thermal plume produced during 3-unit operation, as measured by dye concentration, generally matched predictions during all four tidal regimes. Water temperatures recorded continuously at selected locations were also generally what was expected, based on predictions. The plume is highly dynamic and those regions influenced by increased water temperatures generally experienced a respite from warm water for at least several hours during a tidal cycle. Based on both the survey and supplemental temperature data, at no time during the study period did the 4°F (2.2°C) isotherm appear to extend past the 8,000-ft (2,439 m) limit imposed by the NPDES permit. Based on dye concentrations at Unit 1 intakes, very little recirculation of discharge water occurs. Executive Summary xii Monitoring Studies, 1986-1987 Preface This report was prepared by the Northeast Utilities Service Company Environmental Laboratory (NUEL) staff. All contributors are acknowledged according to their respective disciplines: Laboratory Manager Benthic Ecology Paul M. Jacobson Dr. Milan Keser, Supervisor Fish Ecology Bette Fields James F. Foertch Donald F. Landers Douglas E. Morgan Joseph M. Vozarik Dr. Linda E. Bireley, Supervisor Robin E. Field Raymond O. Heller Richard A. Larsen John T. Swenarton Statistical Support NUEL Mailing Address John A. Castleman Donald J. Danila David G. Dodge JoAnne Konefal Dr. Ernest Lorda Northeast Utilities Environmental Lab. P. O. Box 128 Waterford, Connecticut 06385 David P. Colby Gregory C. Decker Christine P. Gauthier J. Dale MiUer Special thanks are extended to Dr. WtUiam C. Renfro, Director, Environmental Programs Department (NUSCO) and the following members of the Millstone Ecological Advisory Committee for their critical review of this report: Dr. John Tietjen, Dr. Nelson Marshall, Dr. Saul Saila, and Dr. William Pearcy. Reviews provided by Dr. Robert Wilce, Dr. Robert Whitlatch, and Dr. L^nce Stewart are also gratefully acknowledged. We would also like to thank the following people for their contributions to the annual report: James Vouglitois, of GPU Nuclear Corporation, for providing publications and information on the bay anchovy; Mark Gibson and Chris Powell, of the Rhode Island Division of Fish and Wildlife, who provided data on the regional abundance of winter flounder; Eric Smith, of the Connecticut Department of Environmental Protection Marine Fisheries Laboratory', for supplying current landings data for winter flounder and lobster; Frank Almeida and Paul Kostovich, of the National Marine Fisheries Service Woods Hole Laboratory, who supplied abundance indices which were requested for selected species and specific areas covered as part of the Northeast Fisheries Center Tnshore and Offshore Bottom Trawl Surveys during 1975-87. This report is dedicated to the memory of Bette Fields. Preface xiv Monitoring Studies, 1986-1987 Contents Introduction 1 Rocky Intcrtidal Studies 11 Benthic Infauna Studies 59 lobster Population Dynamics 121 Winter Moundrr Studies 149 Exposure Panel Program 227 Fish l'>ology Studies 255 The Usage and Estimation of Delta Means 311 Hydrothcrmal Studies 323 Introduction In accordance with Section 221-430 of Chapter 446k of the Connecticut General Statutes and Section 402b of the Federal Water Pollution Con- trol Act, Northeast Nuclear Energy Company (NNECO) was issued a National Pollution Dis- charge Elimination System (NPDES CT0003263) permit regulating the discharge of cooling water to Long Island Sound (LIS) from Millstone Nuclear Power Station (MNPS). Para- graph 5 of this permit, issued by the Connecticut Department of Environmental Protection (C"T DEP) states that The permittee shall conduct or continue to conduct biological studies of the supplying and receiving waters, entrainment studies, and in- take impingement monitoring. The studies shall include studies of intertidal and subtidal benthic communities, fmfish communities and entrained plankton and shall include detailed studies of lobster population's and winter floun- der populations. Further, paragraph 13 of the permit requires that On or before April 30, 1986 and annually there- after, submit for the review and approval of the Commissioner a detailed report of the on- going biological studies required by paragraph 5 and as approved under paragraph 12. The present report satisfies this NPDES reporting requirement, and provides summaries of the on- going biological monitoring studies conducted by Northeast Utilities Service Company (NUSCO) and contractor personnel on behalf of NNECO at MNPS during calendar years 1986 and 1987. During this period combined three-unit operations occurred for the fu^st time. Referenced frequently throughout this report are data collected prior to 1986 (see NUSCO 1987), which serve as the base- line against which potential effects of three-unit operation can be compared. The goal of the MNPS monitoring program, established in 1968, has been to characterize the various estuarine communities in the vicinity of MNPS and determine if station construction and operation have resulted in changes beyond those that would be expected to occur naturally. To accomplish this goal, various investigations have been conducted. Early biological investigations included exposure panel monitoring of fouling communities and surveys of intertidal sand and rocky shore communities and shore-zone fish as- semblages. The scope was expanded between 1970 and 1973 to include impingement and en- trainment monitoring, surveys of pelagic and demersal fish assemblages, plankton and subtidal benthic communities and studies of lobster and Niantic River winter flounder population dynam- ics (NUSCO 1987). In addition, numerous hydrographic studies have been conducted for the purpose of thermal plume mapping and predictive modeling. Tidal circulation models were also de- veloped to refme thermal plume predictions and to model dispersal of larval winter flounder. A detailed discussion of early studies was provided by NUSCO (1987). The Study Area The MNPS is located on the southeastern coast of Connecticut in the Town of Waterford, approximately 8 km west southwest of New I^n- don (Fig. 1). The approximately 500-acre site is situated on Millstone Point, a peninsula in the eastern part of LIS that is bounded on the west by Niantic Bay, on the east by Jordan Cove and on the south by Twotree Island Channel. The I IS estuary is approximately 82 km long and 40 km wide with an average depth of 19 m (Nixon 1983). Because the average tidal excursion in LIS Introduction is about 1.5 m, the tidal exchange is about 2 x 10^ m''/s, which over a 12.5-hr tidal cycle, pro- duces strong tidal currents (3 - 5 knots) in the Race, the geographic feature through which most tidal flushing of LIS occurs (NO A A 1987). The MNPS monitoring program covers an ap- proximately 50-km'^ study area 2 km west of Black Point, 2 km south of Twotree Island and 2 km east of White Point (Fig. 1). In this area, water depth varies, reaching 15 m in Twotree Island Channel and up to 20 m in one area south- west of Twotree Island. The bottom throughout the study area is generally composed of fine to medium sand but includes some rock outcrops and muddy-sand in nearshore areas (NIJSCO 1975c). The tides in the study area have a mean and spring range of 0.8 and 1 m, respectively (NOAA 1987). Because of the Station's proximity to the Race, tidal currents dominate natural water move- ment in the area. In particular, the flow into and out of Niantic Bay forms a strong current past MNPS along a line running from the site through Twotree Island Channel. Currents in Twotree Island Channel are on the order of 1 to 1.8 knots and currents passing the Station are on the order of 1 to 1.5 knots (NUSCO 1975b) which produces a tidal exchange in Niantic Bay of 2.8 x 10" m /s. In contrast, currents in .Jordan Cove, even during the strength of ebb and flood tides are relatively weak (NUSCO 1975b). The salinity in LIS ranges from 26 to .30%o and water temperature can vary from less than PC to more than 23''C (Nixon 1983). NUSCO (1975c) reported that salinities near MNPS range from 26 to 30%o and water temperatures range from less than 0 to 25°C; since 1976, water temperature and salinity values near MNPS have continued in these ranges (Fig. 2). Thermal and salinity Fig. 1. The area where biological monitoring studies are conducted to assess (he efTccts of the operation of MNPS. Fig. 2. Water temperature and salinity histories from three areas near MNPS. The sources of these histories are as follows; the Niantic Bay series contains a monthly minimum and maximum water temperature based on measurements taken during ichlhyoplankton, trawl, and lobster sampling in the Niantic Ray, including the intake area; the Jordan Cove series contains a monthly minimum and maximum based on measurements taken during trawl and lobster sampling in Jordan Cove; the offshore series contains a monthly minimum and maximum selected from all temperatures measured during trawl and lobster sampling in the Twotree Island and Barllelt Reef areas. induced stratification occurs in regions unaffected by the strong tidal currents and considerable nat- ural temperature variation is observed in near shore areas (Stoltzcnbach and Adams 1979). Introduction Millstone Nuclear Power Station The MNPS complex consists of three nuclear power plants (Fig. 3). Unit 1, a 660-MWe boiling water reactor, first generated electricity on October 26, 1970 and began commercial operation No- vember 29, 1970. Unit 2, an 870-MWc pressurized water reactor, first generated electricity on No- vember 7, 1975 and began commercial operation in December 1975. Unit 3, a 1150-MWe pres- surized water reactor, first generated electricity on February 12, 1986, and began commercial oper- ation April 23, 1986. All three units use once- through condenser cooling water systems. The rated circulating flows for Units I, 2 and 3 are 26.5, 34.6 and 56.6 m' /s, respectively. Cooling water is drawn from depths greater than 1 m below mean sea level by separate shoreline intakes lo- cated on Niantic Bay (Fig. 3). The intake struc- tures, typical of shoreline installations, have coarse bar racks and traveling screens. The cooling water from Units 1, 2, and 3 (noininally heated to a maximum of 13.9, 12.7 and 9.5°C above ambient, respectively) flows from discharge structures and combines in an abandoned granite quarry. The wanned water (about 1 1°C warmer than ambient when all three units arc operating near maximum capacity) exits the quarry at high velocity througli two quarry cuts equipped with fish baniers. Once in LIS the effluent mixes with ambient water so that the surface water temperature of the plume has cooled to 4°C above ambient within about 1,100 m of the cuts; the configuration and extent of the thermal plume is highly dynamic and varies with tidal currents (see Ilydrothermal Studies sec- tion of this report). An operational history in- cluding cooling water flow and discharge temper- ature is provided in Figure 4. Because the Millstone site has been under de- velopment since the early 1960's, a chronology of \ Millstone Nuclear fUnitS \ /-^J Power Station ( a — - "r'Cy \unit2^ bAunitK^ ^ discharges / Jordan Cove CofferH-l!^ Dam 1 . / r Niantic Bay 200- intakes T^M \ m / \ \ \ 1 x\ n1 \-<;:^^^^^^^^^lsland 0 - - \\v quarry ^^econd^t N original cut Twotree Channel rig. 3. The MNPS site. Construction of the cofTer dam around the Unit 3 intake structure was completed during March 1976. Its removal began during April 1983 and was completed during September 1983. The second cut was opened in August 1983. JAN86 JAN87 Date I'lg. 4. Slalion operating conditions January I9S.S tlirough December 1987. On the top graph, the lower dashed hnc is average (Units 1 and 2) intake temperature; upper dashed line is the design efTluenl temperature; the solid line is tlic actual elTluent temperature (average of measurements made at both quarry cuts). Station cooling water usage (m'/s, "cms") is plotted on the middle graph and station power production (MWe) is plotted on the bottom graph. construction and operation events with possible ecological impact implications has been prepared as a reference (Table 1, Fig. 5). Briefly, some of the more important events include construction and removal of cofferdams around the intakes, dredging activities, construction of each of the cuts from the quarry into LIS, initial operation of each unit, and the installation of mitigative devices including the fish barriers at the quarry cuts, and fish return sluiceways at Units 1 and 3 intakes. Demonstrable effects were noted in con- junction with several of these events, most notably construction of the second quarry cut which changed near-field thermal plume dispersal pat- terns such that higher water temperatures were experienced by the adjacent rocky shore commu- nity. In addition, removal of the Unit 3 cofferdam changed sediment characteristics and correspond- ing infaunal assemblages in the intake area. Si- multaneous operation of three units for the fu-st time also increased the volume of cooling water discharged and resulted in sediment scouring im- mediately downstream of the quarry cuts. These changes and the corresponding ecological effects arc discussed further in this report. A separate report section is included for each continuing monitoring program: rocky intertidal communities, benthic infaunal communities, lob- ster population dynamics, woodborers, finfish as- semblages and winter flounder population dynam- ics. In addition, two special sections are included, one discussing the theory and application of means calculated from a delta distribution and the other containing a summary of the results of hydro thermal studies done in 1987 during three- unit operation. Introduction 1 ^1 Fig. 5. Major construction and operation events at MNPS lasting longer than 2 weeks: a) number or units operating; b) periods of coffer dam construction and removal; c) presence of coffer dams at Unit 2, then Unit 3 intakes; d) second quarry cut open; e) presence of fish barriers at quarry cul(s); f) presence of fish return system at Unit 1 intake; g) periods when a surface boom was present at Unit I intake; h) periods when a surface boom was present at Unit 2 intake; i) period when a bottom boom was present at Unit 1 intake. TABLE 1. Choronology of major construction and operation events at the Millstone Nuclear Power Static Date Activity December 1965 November 1969 October 26, 1970 November 29, 1970 December 28, 1970 January 15, 1971 - February 22, 1971 August - December 1972 November 1972 September 3, 1972 - March 20, 1973 November 1972 April 18 to July 28, 1973 August - December 1973 July - December 1974 September 1 to November 5, 1974 July - October 1975 July 1975 August, 5 1975 September 10 to October 20, 1975 October 7, 1975 November 7, 1975 November 13, 1975 December 1975 March. 19 1976 June - October 1976 October 1 to December 2, 1976 Decrmber 20, 1976 to January 20, 1977 May 6 to June 25, 1977 June ■ October 1977 November 20, 1977 to May 1, 1978 March 10 to April 15, 1978 March 10 to May 21, 1979 April 28 to June 27, 1979 August 10 to 25, 1979 November 1 to December 5, 1979 May 7 to June 19, 1980 June 1 to June 18, 1980 August 15 to October 19, 1980 October 3, 1980 to June 16, 1981 January 2 to 19, 1981 December 5, 1981 to March 15, 1982 March 1981 September 10 to November 18, 1982 March 2 to 18, 1983 April - September 1983 May 28, 1983 to January 12, 1984 December 1983 August 1983 April 13 to June 29, 1984 February 15 to July 4, 1985 June 1985 September 28 to November 7, 1985 October 25 to December 22, 1985 November 1985 February 12, 1986 April 23, 1986 July 25 to August 17, 1986 September 20 to December 18, 1986 December 1 to IS, 1986 January 30 to Februai^ 16, 1987 March 14 to April 10, 1987 June 5 to August 17, 1987 November 1, 1987 Construction for MNre 1 began Construction for MNPS 2 began MNPS 1 initial criticality; producpd first thermal effluent MNPS 1 initial phase to grid MNPS 1 began commercial operation MNPS 1 shutdown Surface boom at MNPS 1 Fish barrier installed at quarry cut MNPS 1 shutdown MNPS 2 coffer dam removed MNPS 1 shutdown Surface boom at MNPS 1 Surface boom at MNPS I MNPS 1 shutdown Surface boom at MNPS I Bottom boom installed at MNI^ 1 MNPS 3 coffer dam construction began MNPS 1 shutdown MNPS 2 produced first effluent MNPS 2 initial criticality; produced firrit thermal effluent MNPS 2 initial phase to grid MNPS 2 began commercial operation MNPS 3 coffer dam construction finished Surface boom at MNPS 2 MNPS I shutdown MNPS 2 shutdown MNPS 2 shutdown Surface boom at MNPS 2 MNPS 2 shutdown MNPS 1 shutdown MNPS 2 shutdown MNPS I shutdown MNre 2 shutdown MNPS 2 shutdown MNPS 2 shutdown MNPS I shutdown MNPS 2 shutdown MNPS 1 shutdown MNPS 2 shutdown MNPS 2 shutdown Bottom boom removed at MNPS 1 MNPS 1 shutdown MNPS 2 shutdown MNPS 3 coffer dam removed, intake maintenance dredging MNPS 2 shutdown Fish return system Installed at MNPS 1 Intake Second quarry cut opened MNPS I shutdown MNPS 2 shutdown Intake maintenance dredging MNPS 2 shutdown MNPS 1 shutdown MNPS 3 produced first effluent MNPS 3 produced first thermal effiuent MNPS 3 began commercial operation MNPS 3 shutdown MNPS 2 shutdown MNPS 1 shutdown MNPS 2 shutdown MNPS 3 shutdown MNPS 1 shutdown MNPS 3 shutdown began NUCSO NUSCg 1973a DNGL DNGL DNGL DNGL NUSCO 1978a DNGL NUSCO 1973a DNGL NUSCO 1978a NUSCO 19783 DNGL NUSCO 1978a NUSCO 1978a DNGL^ EDAN F.DAN DNGL NUSCO 1986b NUSCO 1978a DNGL DNGL DNGL NUSCO 1978a DNGL DNGL DNGL DNGL DNGL DNGL DNGL DNGL DNGL DNGL DNGL DNGI(. NUEL DNGL DNGL NUEL DNGL NUSCO 1986b NUSCO 1986a DNGL DNGL NUEL DNGL DNGL ' EDAN EDAN DNGL DNGL DNGL DNGL DNGL DNGL DNGL I, DNGL is the daily net generation (MWe) log ^ EDAN is the environmental data aquisilion network NUEL is the NU Environmental Lab records Introduction References Cited National Oceanic and Atmospheric Administra- tion (NOAA). 1987. Tide Tables 1988, High and Low Water Predictions, East Coast of North and South America, Including Greenland. U.S. Department of Commerce. Nixon, S. W. 1983. Estuarine ecology--a com- parative and experimental analysis using 14 estuaries and the MERL- microcosms. Einal report to the U.S. Environmental Protection Agency, Chesapeake Bay Program under Grant No. X-003259-ni. University of Rhode Island, Kingston, RI. NUSCO. 197.3a. Environmental effects of site preparation and construction. Pages 4.4-1 to 4.5-1 in Millstone Nuclear Power Station, Unit 3, Environmental report, construction permit stage. . 197.5a. Operational data. Pages 1.2-1 to 1.2-2 in Summary Report, Ecological and I lydrographic Studies, May 1966 through De- cember 1974, Millstone Nuclear Power Station. . 1975b. Hydrography. Pages 3.1-1 to 3.6-1 ill Summary Report, Ecological and Hydrographic Studies, May 1966 through De- cember 1974, Millstone Nuclear Power Station. . 1975c. Water quality. Pages 5.1-1 to 5.8-3 in Summary Report, Ecological and Hydrographic Studies, May 1966 through De- cember 1974, Millstone Nuclear Power Station. . 1978a. Impingement Studies, Millstone Units 1 and 2, 1977. Pages 1-1 to 4-2 in Annual report, ecological and hydrographic studies, 1977. Millstone Nuclear Power Station. . 1986a. Rocky intertidal studies. Pages 1-66 in Monitoring the marine environment of IjOng Island Sound at Millstone Nuclear Power Station, Waterford, Connecticut. Annual re- port, 1985. . 1986b. Nuclear Power at Northeast Util- ities. Internally published brochure. . 1983. Effects of operation of heat dissi- pation system. Pages 5.1-1 to 5.1-79 in Mill- stone Nuclear Power Station Unit 3, environ- mental report, operating license stage. Vol 2. . 1987. Monitoring the marine environment of I^ng Island Sound at Millstone Nuclear Power Station, Waterford, Connecticut. Sum- mary of studies prior to Unit 3 operation. Stoltzenbach, K.D. and E.E. Adams. 1979. Thermal plume monitoring at the MilKstone Nuclear Power Station. Report prepared for NUSCO, Hartford, CT. Contents Rocky Intcrtidai Studies 11 Introduction 11 Materials and Methods 11 Qualitative Collections II Undisturbed Transects 12 Recolonization Studies 12 Transects 12 Exclusion cages 13 Ascophyllum nodosum Studies 13 Temperature 13 Data Analysis 14 Results and Discussion 15 Temperature 15 Qualitative Studies 16 Undisturbed Transects 22 Recolonization Studies 42 Ascophyllum nodosum Studies 45 Growth 46 Mortality 49 Summary 51 References Cited 52 Rocky Intertidal Studies Introduction Rocky shores, and the communities which de- velop upon them, are important components of the marine ecosystem. The intertidal community is among the most productive of the world (Mann 1973). The plants and animals provide food di- rectly and indirectly to fish, birds, invertebrates, even man (Edwards et al. 1982; Mcnge 1982); they are involved in complex patterns of energy and nutrient transfer (Paine 1966, 1980). Rocky intertidal communities are particularly suitable for environmental impact assessments. The shores are biologically productive but phys- ically stable; the same area, in many cases the same individual plants and animals, may be sam- pled over time. Perennial algae and sessile invertebrates integrate the effects of long-term ex- posure to potential impacts, while motile species and ephemeral algae respond to quickly changing conditions. These characteristics have led to the use of intertidal communities throughout the world for assessment of impacts associated with, e.g., oil spills (Southward and Southward 1978), sewage (Murray and Littler 1978), and thermal pollution (Vadas et al. 1976). Specifically, eco- logical monitoring programs at every ocean-sited nuclear power plant in New England have in- cluded rocky shore studies (MYAPCO 1978; Wilce et al. 1978; NAI 1984). Rocky shores in the vicinity of MNPS have been subjected to potential impacts resulting from construction and operation of the power station since 1965. To assess these impacts, the Rocky Intertidal Studies were designed and implemented with the following objectives: to identify the attached plant and animal spe- cies found on nearby rocky shores. 2. to identify and quantify temporal and spatial patterns of occurrence and abundance of these species, and 3. to identify the physical and biological factors that induce variability in local rocky intertidal communities. To achieve these objectives, the rocky intertidal studies include qualitative algal collections, abun- dance measurements of intertidal organisms (per- centage of substratum coverage), measurement of rates and patterns of recolonization following small-scale perturbation, and growth studies of Ascophyllum nodosum. This report will discuss results of studies performed during the Unit 3 operational period to date (March 1986-Septem- ber 1987), and compare them to data collected from March 1979 to February 1986 (i.e., the pre- operational period, "pre-op") and summarized in NUSCO (1987). We will assess whether differ- ences exist among communities near MNPS and those removed from potential impact, and whether the magnitude of those differences has changed since 3-unit operations began. Materials and Methods Qualitative Collections The benthic algal flora at nine rocky intertidal stations (Eig. 1) was monitored qualitatively on a monthly basis. These stations are, in order of most to least exposed: Bay Point (BP), Fox Island-Exposed (EE), Millstone Point (MP), Twotree Island (TT), White Point (WP), Seaside Exposed (SE), Seaside Sheltered (SS), Giants Neck (GN), and Fox Island-Sheltered (FS). Qualitative collections were made over an area sufficiently wide to characterize the flora at each site. Samples were identified fresh or after short- Rocky Intertidal Studies Fig. 1 . Location of rocky intertidal sampling sites. GN = GianLs Neck, DP = Bay Point, MP = Millstone Point, FE = Fox Island-Exposed, rS = Fox Island-Sheltered, TT = Twotree Island, WP = White Point, SE = Seaside Exposed, SS = Seaside Sheltered. term freezing. Voucher specimens were preserved using various methods, depending on the material: in 4% formalin/seawater, as dried herbarium mounts, or on microscope slides. Undisturbed Transects At each qualitative collection station except TT (because of insufficient exposed bedrock), five permanent transects were established perpendicu- lar to the water-line, one-half meter wide and extending from Mean High Water to Mean Low Water levels. Each transect, composed of 0.5 m X 0.5 m quadrats, was non-destructively sampled six times per year, in odd numbered months (or a total of ten times in the Unit 3 operational period to date). The percentage of substratum cover of all organisms and remaining free space in each quadrat was subjectively determined and recorded. Understory organisms, or species that were partially or totally obscured by the canopy layer, were assigned a percentage that reflected their true abundance. Recolonization Studies Transects Rates and patterns of recolonization following substratum denudation were determined in recolonization transect experiments at four sta- tions: FE, PS, WP, GN. Sample design included two pairs of stations with similar degrees of ex- posure: exposed at FE and WP, and sheltered at GN and FS. The Fox Island stations, because of their proximity to the MNPS discharge, were considered potentially impacted, while WP and GN were identified as reference stations. Three vertical transects were established at each station; each transect was scraped free of attached algae and invertebrates and burned with a liquid petro- 12 leum gas torch. All recolonization transects were sampled monthly in the same manner as described for undisturbed transects. In the pre-op period, denudings were performed in April 1979, and again in September 1981 at the same transects, to determine the effect that seasonality of denuding would have on recolonization. Autumn denudings (September 1986) were re-established in the Unit 3 operational period, to assess possible 3-unit ef- fects on recolonization. Exclusion cages To investigate the effects of grazing and preda- tion on recolonization rates and patterns, nine areas were selected at each of the recolonization stations, three areas in each of three tide zones. In each area, two 20 cm x 20 cm patches were cleared and burned; one was covered with a stain- less steel mesh cage (20 cm x 20 cm x 5 cm, 3 mm mesh), the second left as a control. Each month the percent cover of colonizing organisms was determined. The effect that season of denud- ing had on rates and patterns of recolonization was also determined. The pre-op series of exclu- sion cage experiments began in April 1979, June 1980, September 1981, and December 1982; each area was re-burned 15 months after the previous denuding. The exclusion cage studies were re- established in December 1987 to determine the effects of grazing and predation on recolonization under 3-unit operating conditions. Results from these studies will be presented in future armual reports. Ascophyllum nodosum Studies Growth and mortality of populations of the perennial brown alga, Ascophyllum nodosum, were studied at two control stations (GN, 5.5 km west of the discharge and WP, 1.5 km east of the discharge. Fig. 1) and an experimental station (FL, ca. 75 m east of the original Millstone quarry cut. Fig. 2) from 1979-1984. Ascophjllum was eliminated from FL in summer 1984, its loss at- tributed to elevated water temperatures resulting from the thennal plume of two operating units discharging through two quarry cuts (NUSCO 1985). In spring 1985 a second experimental Ascophyllum station (FN) was established between FE and FS (Fig. 2, ca. 250 m from the quarry discharges, northeast of the Fox Island-Exposed sampling site). Following the loss of plants from FL, FN supported the Ascophyllum population nearest the discharges. Ascophyllum plants were measured at monthly intervals from April, after the onset of new vesicle formation, until the following April. Fifty plants at each station were marked with a numbered plastic tag at the base of each plant, and five apices were marked on each plant with colored cable ties. Linear growth was determined by mea- surements made from the top of the most recently formed vesicle to the apex of the developing axis, or apices if branching had occurred. Vesicles were not large enough to be tagged in April or May, so five tips were measured on each of 50 randomly chosen Ascophyllum plants, and monthly mea- surements of tagged plants began in June. Lost tags were not replaced, and the pattern of loss was used as a measure of mortality. Loss of the entire plant was assumed when the base tag and tip tags were missing. Tip survival was determined in terms of remaining tip tags. Temperature Water temperatures were obtained from the EDAN (Envirormiental Data Acquisition Net- work) system which continually records a variety of environmental parameters and reports at 15-minute intervals. Ambient water temperatures were recorded by sensors in Unit 1 and 2 intake bays, and effluent water temperatures by sensors in the quarry cuts. Temperatures at FE, MP, and the experimental Ascophyllum stations (Fl, and FN) were measured over several tidal cycles with a portable thermistor and strip chart recorder. During 1987, solid-state data loggers were de- ployed at several sites in the Millstone area to record water temperatures under differing power plant operating levels; these data were incorpo- rated into models of regimes to which rocky intertidal stations in the vicinity of the MNPS discharge were exposed. Rocky Intertidal Studies 13 Fig. 2. Detail map of MNPS vicinity. FL = original experimental Ascophyllum site (1979-1984), FN = new experimental Ascophyllum site (1985-present). Data Analysis Relative abundance of intertidcil organisms was estimated on the basis of percentage of substratum covered by each taxon. Unoccupied substrata were classed as free space. Similarity between communities was determined by a standardized percentage form of the Bray-Curtis coefficient (Sanders 1960), calculated as: Sjk= Y.™^^PijPik) where P(i]') is the percentage of species (i) at sta- tion (j), P(ik) is the percentage at station (k), and (n) is the number of species in common. A flexible-sorting, clustering algorithm was applied to the resulting similarity matrix. The calculations were performed on untransformed percentages. Comparisons of pre-op and 3-unit operational abundances were based on the 5-mean as an index of abundance. This method utilizes the assump- tion that the non-zero values of percentage cover are log-normally distributed; a complete descrip- tion of the 5-distribution and rationale for its application is included in the Delta Distribution section of this report. 14 A Gompertz growth curve was fitted to Ascophyllum length data using non-linear regres- sion methods (PROC NLIN, SAS Institute Inc. 1985). The growth curve parameters were com- pared among years and stations using 2-sample t-tests (a = 0.05). Ascophyllum mortality data are presented as number of surviving base tags (plants) and tip tags (apices); Unit 3 operational data (1986-87) are plotted against the mean and range of pre-op data (1979-1986). Because of the total elimination of Ascophyllum from the original Fox Island Ascophyllum station in 1984 and establish- ment of a new study site, pre-op data from Fox Island excludes the 1984-85 season, and separates 1985-86 (FN) from 1979-84 (FL). Results and Discussion Temperature Since Unit 3 began operation, ambient water temperatures (measured at the MNPS cooling water intakes) have ranged from 2.0 °C (16 Feb- ruary 1987) to 22.1 °C (17 and 18 August 1987). These values are typical of those reported in past years, when minima generally occurred in January- February (1-3 °C) and maxima in August- September (20-22 °C). These temperatures were recorded at a depth of about 3 m below Mean Low Water; insolation of shallow water near the rocky intertidal sampling stations raised summer maxima 2-3 °C, and in winter, slush/ice formed in near-shore shallows. Effluent water temperatures, measured at the discharge quarry cuts, were dependent on reactor power level and cooling water flow (see Introduc- tion to this report). The designed temperature rise above ambient AT was 12 °C for 3-unit, full power operation, but AT was less when a unit was shut down and its unheated effluent diluted the discharge of the operating units. The hydrodynamics of the 3-unit thermal plume, and its behavior at various tidal stages, are de- scribed more fully in the Hydrothermal Studies section of this report. Briefly, the 3-unit plume extends into Twotree Island Channel, where it is subject to tidal flushing. This is different from the 2-unit/2-cut plume, that produced elevated water temperatures along the shore between the cuts and the southwest tip of Fox Island, regardless of tidal stage. The effects of the 2-unit effluent were summarized in NUSCO (1987). The exposure of local rocky shores to the full power 3-unit plume, therefore, varies with tidal stage, as well as distance from the discharges. On an ebbing tide, as water moves out of Long Island Sound, the plume is deflected to the east, across Fox Island. At the original experimental Ascophyllum station (FL, Fig. 2), ca. 75 m east of the discharges, the plume elevated water tem- perature 7-9 °C; temperatures remained elevated for 10-11 hours per tidal cycle. However, tem- peratures dropped close to ambient levels for 1-2 hours during maximum tidal flooding, as the plume was deflected to the west, and the heated water was displaced. At FE, ca. 100 m from the discharges, water temperatures were elevated for 9-10 hours per tidal cycle and peaked at 6-8 °C above ambient; ambient water temperatures occurred for 2-3 hours, around the time of high tide. At the new experimental Ascophyllum station (FN, Fig. 2), 250 m from the discharges around the tip of Fox Island, maximum water temperature elevation was 4-5 °C above ambient, only during the ebbing tidal stage. Maximum flood tide deflected the full power, 3-unit plume to the west. At high tide, water temperatures 4 °C above ambient were recorded at MP, ca. 250 m west of the discharges. At this station, there was also a 2-3 °C increase above ambient at the time of low slack water, as the plume spread laterally (cf. Hydrothermal Studies). The plume characteristics described above are representative of full power, full cooling water flow; because of scheduled and unscheduled shut- downs, these conditions existed for less than 50% of the Unit 3 operational period to date. Specif- ically, we have not seen consistent full power Rocky Intertidal Studies 15 operation during periods of maximum ambient water temperature. Various combinations of op- erating units, varying water flow, and changing meteorological conditions affect the behavior of the thermal plume, and the degree to which the plume affects intertidal communities. The gener- alized temperature regimes provide a physical framework for interpreting the biological responses of rocky intertidal plants and animals discussed in the following sections. Qualitative Studies NIJEL qualitative studies were designed to iden- tify algal species present in intertidal and shallow subtidal areas in the vicinity of MNPS throughout the year, and to characterize their spatial and tem- poral distribution patterns. Changes in these pat- terns, i.e. differences in species composition among stations or years, may indicate environmental changes and require assessment of whether the changes were related to construction or operation of MNPS. Floristic analyses have been used in similar environmental impact assessments, e.g., Wilce et al. (1978) and NAI (1984). A rich and diverse flora occupies the rocky intertidal monitoring area, relative to other areas of Long Island Sound. Overall, 128 species (ex- cluding blue-greens and diatoms) have been iden- tified in the Unit 3 operational period, but not all species were found at any one station, nor were they found in any one collection period. Qualitative algal collections for the 3-unit moni- toring period are presented as number of stations at which each species was found in any given month, and as number of months each species was found at any given station (Table 1). Fucua vesiciihxus is the only alga that was col- lected in every month at every station during 3-unit operation, although Chondnis c.rlspus and Ascophyllum nodosum were ubiquitous at all sites except Fox Island-Exposed (dissimilarities be- tween the flora at FE and those at other stations will be discussed in a later section). Other species (e.g., Ceramium ruhnim, Uka lactuca, Codium fragile), while not ubiquitous, were common throughout the area, throughout the sampling pe- riod. Of the 128 bentliic algal species found in the 3-unit period, several were site-specific (Table 1), such as Laminaha digitata at TT, Fucus spiralis at BP, and Agardhiella subulata at FEi. In addi- tion, some species were characteristically rare at a particular station, such as Corallina officinalis at ON. The local flora also showed seasonal trends. Some examples of temporal differences include Bangia atropurpurea, Dumontia contorta, and Monostroma spp. as most common in winter- spring, Leathesia diffornns, Pctalonia fascia, and Scytosiphon lomenlaria in spring-summer, Champia parvula and Giffordia milchelliae in summer-autumn, and Spcrmothamnion repens and Sphacelaria cirrosa in autumn-winter (Table 1). Spring collections were typically richest and winter collections were poorest. These patterns were noted in the pre-op flora as well (NUSCO 1987). Some local species exhibit no apparent spatial or temporal pattern of occurrence, i.e., they are not characteristic of a particular station or season but occur sporadically among sampling sites. Sporadic occurrence of many algal species ac- counts for the fact that algae collected in any one year will comprise only a portion of the total flora, represented by aggregated collections since 1979. The presence of infrequent or occasional species also explains why new species are added to the overall species list each year. For example, A ntithamnionella floccosum and Nemalion hehninlhoidcs were newly recorded in the Unit 3 period. Antithamnionella was found througliout the period at 6 out of 9 stations (generally at exposed sites) and Nemalion only once, in .luly at HP. (^lonversely, some rare and small epiphytic algal species that had been reported between March 1979 (when this sampling program began) and February 1986 (the last sampling month of the pre-op period) have not yet been found in the 3-unit operational period. 16 ^ ^ OO>J'OWi-t-HOf0Cjr0*0fvJOCJOC0OKIOOO'-«^O-^OeuO*-tOOh- li.^ ^^-^^ ^^ lu ^<0r'>cJo^^^•-*^-«e^J^<^floc0'^cJ^oooor^(^■-^<0cu^ooKtcJO«^■•-<^c^ H lflOCJOO«0'0-5"OOO>J^ON0f^K»OCJOC0*^-0-
    oK».j-0'^m-Noooo^^r^oO'>^oroooocjoooors. < o•-4K>oooooo^^^^mom•-•^o•-Hoc^'^^^■^o^J'OKloooK1oc^Jo-H(0 ~> i^rom-*oooootf»>3"^o^^>3-cjoKtoocja'mor^cas^ooO'd'OKii-io(> ^ »-l^0OOOOO•-•-«^*C^J«^•-H•-^^O-^•-^O<^CJtfD^ON..-^C^JOOO'^■-^O-^OC0 CO a^< if(m-^ooooojr^roooococj»3''*---^0'a"ooo^oooo^ E r^.^cjooiri'ooomoifiocj'd'^ooococj-^cjooo^ooo^ooooifl U- ^iricvji-40<— •O'HOU^oe^JO^^HKto-^o9''^^^c^JOOOoooooo^^ooc0 ^ 0'00<-HOOOOOLn'«CJOLA— •r^OOO<0(MOC\JO-40rvJOOOCJO>^00>0 D cust'-^oooooocj— ^K^omo•-•ooo^-^^*'^oe^JocJoooKlo■--oojooo»-«ooooin O ^-^^oooooocO<\(ifioror*c\joooO*cjcOinO'a"OCJOO-*roooo^'»o to o^Jooooooo»^'^^^o•^lnooo-H(^-^^^■^olf^oK>oooroo-^ooco ^o^ ocjooooooO'O'^ojoininifto-^o^— <^'>^oKioK»-HOoroooooo^ CO o^~) cjCNJoo-^oo-^O'^cu-^oiAifiCj^ooocJcocsjor^O'-^ooorooKtoor** ~) c^J<^JOo>-^oocvJ-HlftOh^olr^NcJocJoCT'CJa^p^^o^J'«vJOKl^rooo-*o^* £ ^H•-4C^J•-«•-4^f>0U^>0^^J^^•-4C^J^»OOOO(0C^JU>C^JO^OOO'H^4C0OOOO<^ < r^*j'ooor*cocjrOLnocjOsOOKiooosO-*cjc\i-^ooe\JooocJoooom 31 'O'OO— •o^ocf'i^^'OOsTO'.fift^^oo— ^cOO' 0 [. SI E t. C 1. 0 re 1- 0 m -H I. a -< ra ra 0 01 a. t. 01 ■'- E-t ■- 1. 3 C a C t. 0 > CI Q. nj n re re re re re re re re — ui 1- 3 t- re re tn •- 01 '*- 01 reEre \fl-^tBCi-mt~C tn -H30)reooo»'— -HCrei- OJ racu-ooreoxin— aopreTS ■H ■- 01 — X I. •- o 01 X — ji; in — re •--^re^o)H-^.•^-•aoI. oocti 0. -<-H0)0)333 HEC'--1 ._ - - . _ . - re •- ra re re t- I. in X 1. t. E 3 3 Q. t. •- Z re ^aoo xxEXXXXxxxxreo-HXX ^•-•Hj.raoi ^"^ ■** H-Q.3 rererecQ. 0.0.0.0.0.0.0- o."- 3ino.Q.rerererererere''-c G.X E re > re'-EEl-l-reI.l.l.l.t-i.I.;.L.-H UUI.l.l-L.I.t.I.CO L, w^ X xreret."DoooO'-ooooooooraoreoooooooooO'Hininre'»- a ■-t.t.Orei.l.EEOiEESEEEEEinre-4EEXXXXXXX-HO ■-•-•- U o <- o o a-H -H-Hoocoooooooo ----------.- _j 1- XQ.Q.inUininD) □!•- i.i-i-i.i-t.i.i._ __ _ S! O ■HinuiOOOOCCT30l4IO)OiaO>OIO)OrevnaaT3"DT3"DT3TI"ON'-OOX"- ™ -H oooi--HCCOo---H-f*^-H»^-i-'+'+'i.>rererarareraren>rere--X>>i-T3 < X -a3NO o^ooorOoooocj OO^OCjKlCOOKlCdin t 2 'T-^^OOmcJinwOOOOOOOOOsONO— ^•J'CU-^OCJC0CJOOOOC^ KlMooocJ^J■«d■'J■ooooooooooc^ocJ-^c^ic^<^JCJcJ(^ooo<^ Na^ocjinrort*3-inoto-^oooooopjo^tooooK>oroc>ocMO(> -^wo^cjw^u^LftoKiooo o o oc>ja'00-^oooor^ ^wcjo-^oroo'— ^^a>oooooooc^JKlo-^oooo^roo*oe^JO(^ <^lf^ooco— •Ktojr'^— »ro>ooo— •ooo-Hco— *o— ^cJooc\JOC^orJmc^ £> o <\J o Q- - o o o o -0 •^3 o - o -o -» lO CM o o CM - - •3- h- o <\J CJ o o o o o o -1 o -. o o o m in -. CM o m -> * - o - ■-• ro o ro c o -3- o o o o o o o m CO o o - o o - <> o CO r o o o o C\i CJ -o o -< - ~< r CM •a- o o o -3- o - ro o CM « o o -. o o o r» CO o -H o CM o o <0 — o- < o o o o o CJ sO -< — — CM < CM ro -• o o >* CM CJ r^ o - o o o o o o o CO CO o -• o - o ro ro o r o o o o o — in o o CJ M r CM CM o o -■ CM o o o o - - •o « o o (0 -> o -< o o CM - - ro - o o o o o o o o o o -3- o o o o o - ro CM o ~l o -< o o - m o - o 0^ a -< o - o CM - CM ■o CM o o o o o o o o — m o o o o o o - ro * z ~l o -• o o ~> -o o -< o c z m o o - CM - o •o to o C3 o o o o o o o <3- o o o o o o o m CJ o CJ o CJ ~> o m f^ o -« o l> o — o o o [^ ro CM CO -O o o o o o o o o o CM o o o o o o -• in -< in -< o - C^J o It) -o o o o c^ in 1*1 o o o •3- o o ro -• o o o o o o o o '- o o o o o o o -< CM CM - o - - o -3- r^ o in o o o o o o o in in o o ro o o « 1^ o -t o o o o o - 00 o - m 9- -> * -o -" — - CJ o o r«. o -3- - o o CM o -• o c> c^ o m CM — o CM 1^ CJ r o o o o o o •o o o - ■f £ in ro o CM CJ in CM m o " in CJ o o o o o o o- o o * o - - •3- — r o o o o o o r-~ - o o r» z o t«l -H o o ro o m -o -• o o o o o o o o o o- - CM o -i o CJ ro CJ CO O — t -4 Ch CO o o o <^ CO o ^ '^ o^ ff) o o -* <^ CO O O O C" CO -^ CJ O Ch CO O CJ o O^ CO o o o o^ <0 o o o o^ CO o o o O^ CO o o o a- (O o o o ^ CO o o o ^ CO o -^ w (^ CO o CJ o ff* CO o ^ o o* CO o o o (^ to ^ ^ ^ o^ CO O CJ O 0< i a t> 4^ c 5 •¥* u re re c n > ra in 1 n n a -D cr in n 01 o 01 u c c 3 > 0 1. 1- 01 in a a C u c Ol □1 u > z C > a a rs I. 0 a u "> u ■□ X E E 3 c 6 SI 3 a O ra a n re re re re re a cH J3 "c C "c c 'E E 'E V\ o o o o 0 o 0 ■H > ra nixxxxxxx 1. ^ a. a a a Q. a a 01 -a 1- c 0 U "0 in in in in in in in c 0 > c > > > > > > >. > in o — I i-< ^ ■ i-xrexooooooo CDa.auQ.a.o.[La.a.a. re •♦- -H -f in E U 3 E ,^ rg 01 E in t. — m 3 re — in re in 3 0 I. O -H in 01 1. re 0 re •.- — in in t- re -H ra c ra c ra in o re re .- ._ > 3 -< ra 3 0 -H re re u E -< ■- ■H 3 01 c 01 -a U -t 01 I. c o I.-i-D m ~a 0 re ra 01 3 X O 01 o u I. 0 a 0 *-.- E 3 1. CO*" E u Ji >t- ci ra o c u o o Q. re -H -H 0 3 u E M- re — re in — 1 re > in in 1- 3 3 — -1 X in E c ra p-H re E Ol E -< 1- TOt- o 3 5 o at- C ra re 5 in in re 01 O X o 3 a 1- 3 a 1. re re re -5 ^ -( E OJ c > re in X ra ra 1. u u o — U 1. o u re re re X in 01 ■M — in >>- u re 1- m u 0 1. T 'E -S- re 1- 1. 01 o X 01 re 1. 0 *- in ra re o in L. u re o o o o > Ol in X X T3 01 i. o 0 ■f —1 o ra re TJ o o ■*->(- re c u o 01 E 01 o o re -^ E E 1. •H 0 ra o X Q. in re 5 01 o in n o o o a ra ra OJ X a in 01 X 01 0) X 5 S a. in a. lU X -1 O 10 < a a. a. 0. c in a o u I. c re E 01 01 re 3 •>- in 3 'fi.i.oininin in f- rareurei.0 3 in-H--xt.tJinin in 0'-DIO—033inO •H oic o o cxx — -H r — o ra o o 01 T3 in re e — — F 0 ra re ra t. -H in in L I. -< >T3 -a ra ra re 0) X •n r r c o a in in — re 0 3 3 O E E E X o o o — 1 .- .- a 01 Xrarereain3333 (J— I— '-JV)rno^J■0'-*cJoo«0'-*ocjooiriv0^oo oc\joo-^r<«(^rooLnoooooroooO'^orOcJoor>«Osft/iooou^oooomooooo-^ro^o-H o^a■ooc>Joo(Oo<^olf^ooor^oo^or^oC'cocJC^oolrlmo-*^Dooo-^oo-^o^ocu-^lftr*.^oo^ OK>ooo^J■^co-^moooo^oooooc^Jo^nooo^^ooo^ooo^*'*'ooo<^^ocJooo^Lrloo OC0CJ"-«Kl^O-HOOONl'OOOob^O-4^OOOr0OO»0OOO-^h*^OOO0^Oe\i'^OO^0'»3'O-* oooocJ^*o^<^oroorooooooomK^ootft•^c^c^-^lfiC^olf^^*sOO-^oooo-4s0^oo•£lC^o•-*c^J ocjoocga^cj-^^r«.w-Moooooo-^oorse\joCT'inocoiA— •oolr^o^oo-^o^a•^ooo-^K>o<^ ^»oooro>or»«ifioi^ooj^o-NooooKtor^— *oor^-:r'a'0-* o o o -» .H O O -H -H o eg o o o o o o o -. * o K) o « O o o CJ o o o o o -< o CJ -> o o in o in o o o -H to o o O o o -1 o o o -H o o - o o o o -r o -i- o o — -■ M o — o o o CJ - K) o o o - o o o o IT- ■f o in o O Kl CVJ lO o — o o o -H o Kl -< o o CJ o Kl — - CO o o to o CJ o - o - o o o o ro - o o -■ o o o o o CJ o -3- o o CO m -• o CO o o -• o CO in o in o o — o CJ o -< o o o CJ o o o o — ■-> o - CJ o CO 1^ o r^ o -. -H o K1 o CO t^ o •* o _■ o » o — o ^ — 1^ o CJ - -4 CO CO o in CO O ^ --4 CVI o ^ •^ o o r^ - o o O .3- r~ CJ o o o ^ CJ o o o O Kl CJ o o o o -i' o o - o o o o o - o o o o o o o -H O - o o — o o Kl o o o o m •C o o o O sO m — o o O CO •o Kl o o o CO CO o — o o o <3- o o o o -3- o M o o O CJ •H O O O CJ o o o o CJ o o o o o — o -^ o o o -^ -^ o o o -< -^ o ■-■ O M M o O O Kl Kl O O O >^ CJ o o •-• •* o -< O O CJ o o o o o -^ o O O "f o o •^OCJOO-HOOCJOO^^-^ OOCJOOOOO-HOOCJCJO •t o €3 -a- o o O O ^ CJ o o o < -• -O -< O <3- -3- I^ CJ -H Kl CJ (O CJ O p^ O 1^ CJ o -< O 0 O O '^ o .J- -» o — o m CJ o CJ O CO CJ O -4 o -O •3- O -4 o CO r^ -4 ^ o r-- in -< o Kl a- r* o o m o* K» o -4 m CO in o -4 Kl r« .3- o o o CO in o o O m Kl O O o r^ -• o o o -O -3* ^ o ro -^ >.- C ra •— tt D ■- D +• O 1(1 "D Q. U) — I. o -< ra in 3 o — — H- o -H I. ra o — — n 3 — X E (- -H in ax in "□ in in c E n i-§- n in c 3 1. 1. -4 «i ■H-HOQ.nJurabcuaciaiEC-'- ■ -H .- (J *- e -H ai c — I (5 o (t -a o 4^ I- c C3 in SI >. ■H 3 in 13 0 — n ts a 3 -< u u I. -i t. re -t O in I. a-c Dl — O o a ui ai ■♦^ a. a E r -I a fi rs 1. in E O o X -a (fl i- -4IU!.0(BU1-H-- re o 13 u a ■H jj c ra E O X 3 C E -H C c C^ a ) _ OOCCCOO0inraJ3 — - > > > C (3 >. > >■ XXX en a a a o - •- — — •- E -< '- 01 -4 >3 ^7J O^. ■XXinC>Cr30i.rB O o o 3 I tu (Q 0. CL a. < tl 13 -D ^ E 3 3 3 U 01 O I < +■ o ' — X ■H a-< — — C C C E E E L.I.l.EEErai3raE u X "- >3 ra ra cB ♦^ O C (-•-♦"■w-HXi caioiraaccc-H*'*'- ^ 4- ._ .^ .„ O -D X ra^inraC>»>o I. E o -H • . T) E E OJ 01 U • E E E 6 — • ra E E E-^-^+'-H^-H < a. Q. u (S Q: u I ra X o o o c ^--hpH^-^ ra ra ro i. !-•— 0) >-H raraoaioiacLC intn< I u u u u (J I Rocky Intertidal Studies 19 Comparing the flora represented by the 1986 collections to those collected in each of the pre-op years (Table 2), there was a tendency at most stations for red algae to comprise a smaller pro- portion of the total flora found since Unit 3 began operation than in the pre-op period. In most cases, the relative decrease in the number of red algal species was attributed to the absence (in the 3-unit operational period, to date) of some small, rarely found plants. At all stations except FE, the proportion of reds in 1986 was within one percentage point of previous years' values, and the decrease is not regarded as a major community change. At FE, however, the proportion of red algae in 1986 (35%) was 10 percentage points lower than the pre-op summary, and the number of red algal species (22) was at least 7 less than in any previous year except 1985 (Table 2). These patterns represent a continuation of spa- tial and temporal trends that have occurred at FE since the opening of the second quarry cut in August 1983 (NUSCO 1985), and are different from those identified at the other sampling sites. Many perennial species and associated epiphytes at FE were eliminated in the summer of 1984 when water temperatures exceeded 28 °C. Com- munity changes resulting from elevated water tem- peratures included the loss of established popula- tions of perennial macroalgae such as Chondrus crispus, Ascophyllum nodosum, and Fucus vesiculosus, and increased abundance and persis- tence of opportunistic species such as the greens Codir^m fragile and Enteromorpha flexuosa. Ele- vated temperatures at FE also caused a decrease in the number of brown algal species, thereby decreasing the proportion of browns relative to reds and greens (cf. Schneider 1981; Quarry Study in NUSCO 1987). Changes at FE were identified both as a decrease in species number (from a total of 80 species in 1982 to 50 in 1985), and as a shift in divisional proportions (more greens, fewer browns and reds). Local spatial and temporal distribution patterns are also apparent when the qualitative algal col- lections are presented as number of species in each division (Table 2). Number of species in each division and total number of species at each station during the Unit 3 operational period gen- erally fell within the range of previous years. Fewer species have been collected in the 19 months since Unit 3 began operation than were found in the 7 years of pre-op studies. Continued collection during the 3-unit operational period will augment the 3-unit species list. Relationships between the pre-op and 3-unit operational floras may also be represented graph- ically (Fig. 3). When division proportions of the overall flora were analyzed, proportions were sim- ilar and independent of species number. Relative species proportions m the first 12 months of the 3-unit operational period (45:26:29) were virtually identical to those of the pre-op summary (Fig. 3). The local flora proportions continue to be similar to those of other researchers in the north- west Atlantic (Vadas 1972; Wilce et al. 1978; Schneider et al. 1979; Mathieson et al. 1981; Mathieson and Hehre 1986). In brief, the algal flora of the Millstone area, as represented by collections in the 3-unit operational period to date, was similar to that reported for 2-unit operating conditions. The community changes described at Fox Island-Exposed were attributed to elevated water temperatures resulting from opening the second quarry cut, not from start-up of Unit 3. However, most of these changes have persisted during 3-unit operations, as water temperatures close to 28 "C have occurred at FE in summer. The FE community has shown some response to the periodic incursion of ambient-temperature water that occurs near the time of high tide (see Temperature section); e.g., Fucus persists throughout the year, and isolated Chondrus plants have been collected. However, Ascophyllum has not recolonized, and Codium and Enteromorpha remain the most abundant spe- cies at FE. Continued monitoring at FE and at nearby stations (e.g., FS, MP) will allow us to determine whether the observed thermal effects will remain within present bounds. 20 TABLE 2. Number of species by station, year, and division (percentages are in parentheses). Pre-op (3-79 to 2-85) and 3-unit operational (3-86 to 9-87) summaries are included. Each year represented by collections from March to following February. Station Division 1979 1980 1981 1982 1983 1984 1985 pre-op summary 1986 3-unit summary BP reds 31(43) 33(47) 47(50) 40(46) 34(43) 39(46) 33(46) 57(44) 33(43) 39(46) browns 16(22) 18(26) 24(26) 24(28) 18(22) 21(25) 20(27) 36(27) 21(27) 21(25) greens 25(35) 19(27) 23(24) 22(26) 28(35) 25(29) 20(27) 38(29) 23(30) 25(29) FE reds 33(43) 33(46) 31(40) 33(41) 29(42) 29(45) 21(42) 50(45) 22(35) 28(39) - browns 18(24) 16(22) 22(29) 17(21) 17(24) 14(21) 10(20) 26(23) 16(26) 16(22) greens 25(33) 23(32) 24(31) 30(38) 24(34) 22(34) 19(38) 36(32) 24(39) 28(39) FS reds 32(45) 29(42) 28(39) 39(49) 39(49) 37(45) 28(42) 54(44) 23(38) 23(35) browns 16(23) 15(22) 21(29) 16(20) 16(20) 22(27) 15(23) 32(25) 16(27) 17(26) greens 23(32) 25(36) 23(32) 25(31) 24(31) 23(28) 23(35) 38(31) 21(35) 25(39) GN reds 28(42) 33(45) 37(44) 38(45) 37(45) 36(43) 34(44) 56(45) 27(41) 28(40) browns 17(25) 18(24) 23(27) 20(23) 17(21) 20(24) 19(25) 30(24) 17(26) 18(25) greens 22(33) 23(31) 25(29) 27(32) 28(34) 28(33) 24(31) 39(31) 22(33) 25(35) MP reds - - - 30(40) 33(42) 31(40) 32(44) 51(45) 29(43) 34(47) browns - - - 20(27) 22(29) 23(30) 18(24) 30(27) 16(24) 16(22) greens - - - 25(33) 22(29) 23(30) 24(32) 32(28) 22(33) 22(31) SE reds 26(45) 24(44) 34(48) 33(46) 32(46) 29(41) 28(44) 48(44) 24(41) 25(40) browns 15(26) 17(31) 19(27) 20(27) 15(22) 20(29) 15(24) 29(27) 15(25) 16(2.5) greens 17(29) 14(25) 18(25) 20(27) 22(32) 21(30) 20(32) 32(29) 20(34) 22(35) ss reds 29(43) 32(48) 39(47) 40(45) 40(47) 43(51) 30(44) 57(46) 29(45) 34(46) browns 16(24) 13(19) 23(27) 21(24) 23(27) 19(22) 17(25) 31(25) 17(26) 18(24) greens 22(33) 22(33) 22(26) 27(31) 22(26) 23(27) 21(31) 36(29) 19(29) 22(30) TT reds 39(45) 37(46) 37(48) 40(51) 54(50) 32(44) 40(45) browns - - - 26(30) 23(29) 22(29) 20(26) 28(25) 22(30) 27(30) greens - - 22(25) 20(25) 18(23) 18(23) 27(25) 19(26) 22(25) WP reds 32(44) 38(47) 44(46) 42(44) 39(43) 43(48) 41(48) 58(46) 36(44) 39(44) browns 18(25) 20(24) 25(26) 24(25) 23(26) 22(24) 23(26) 33(26) 20(25) 20(23) greens 22(31) 24(29) 27(28) 29(31) 28(31) 25(28) 23(26) 36(28) 25(31) 29(33) Total reds 44(44) 47(46) 60(47) 58(44) 57(46) 61(45) 54(46) 73(46) 54(45) 59(46) browns 26(26) 26(25) 35(27) 35(27) 32(25) 34(26) 29(25) 40(25) 31(26) 33(26) greens 30(30) 30(29) 34(26) 38(29) 37(29) 38(29) 34(29) 45(29) 35(29) 36(28) Rocky Intertidal Studies 21 reds □ browns ireens Fig. 3. Relative species proportions of major algal divisions under pre-operational and 3-unit operation (March 1986-February 1987) conditions. Numbers in bars represent percentage of total flora; height of bars represents numbers of species in each division. Undisturbed Transects Zonation is a universal feature of rocky intcrtidal communities (Stephenson and Stephenson 1949, 1972; Lewis 1964). Local rocky intertidal com- munities are separated into horizontal bands rep- resenting the high, mid, and low intertidal zones, where the high intertidal (Zone 1) consists mostly of bare rock and barnacles {Balantis balanoides), the mid intertidal (Zone 2) is dominated by a canopy of the perennial brown alga Fucus vesiculosus over an understory of barnacles, and the low intertidal (Zone 3) is dominated by the perennial red alga Chondnts crispus. Although zonation appears stable on a localized scale, it is subject to natural influences such as degree of exposure, desiccation, temperature, storms, ice- scour, predation, and competition, as well as man- induced influences, which can alter the abundance of intertidal populations and increase the com- plexity of the conununity. Similarity Dendrogram To discern pattern in this complexity, the Bray- Curtis similarity index has been utilized to char- acterize the community at each station in terms of average annual abundance of each mid and low intertidal taxon (measured as percentage of coverage on permanently marked, undisturbed transects, sampled six times per year). The sim- ilarity index was calculated for each station/year combination, and a clustering Jilgorithm was ap- plied to the resultant similarity matrix. Exami- nation of the hierarchial dendrogram (Fig. 4) al- lows generation of hypotheses concerning the re- lationships among stations and among years. First, the general patterns of similarity among stations and years (including 1986 data, marked with asterisks in Fig. 4) were identical to those 22 e:^ e s = c^ X4UDIILUIS Rocky Intertidal Studies 23 reported for the 2-unit operational period (NUSCO 1987), indicating that the local commu- nities, and the factors responsible for structuring them, have remained relatively stable since Unit 3 began operation. Second, the community at Fox Island- Exposed continued to appear dissim- ilar to those at other rocky intertidal stations. These hypotheses can be tested by analyzing lo- cally prevalent species whose patterns of abun- dance are important in structuring local intertidal communities (e.g., Chondrus, Fucus, Balanus, Mytilus), and those which, although not dominant in terms of percentage of cover, interact with and exert influence on other components of the com- munity (keystone species sensu Paine (1966)), e.g., predatory and grazing snails. Barnacles and predatory snails Barnacles have been shown to influence the structure of intertidal communities by providing surface heterogeneity and spatial escapes from consumers for settling plants and animals that would otherwise be quickly eliminated from smooth substrata (Lubchenco 1983). Under 3-unit operational conditions, barnacles (mostly Balanus halanoides) exhibited a characteristic pat- tern of abundance. Balanus settlement on local shores occurred as early as December- January, but the period of maximum settlement was February- March, earlier than reported for north- ern New England (Grant 1977). Growth of newly settled barnacles continued through summer, when barnacles occupied almost all available pri- mary space; barnacle abundance ranged from 55% to 95% in Zone 2 in May (Fig. 5). A decline in abundance began in late summer and barnacle coverage was lowest in winter (e.g., 6-45% cover in November, Zone 2). In most cases, the rock surfaces exposed by the mortality of barnacles remained free of macroscopic cover until the fol- lowing spring. This pattern of barnacle abun- dance was similar to that observed by Katz ( 1 985) in local populations. Barnacle abundance varied among stations and zones (Fig. 5). The lowest abundance of barnacles was evident in the high intertidal zone, because of reduced settlement and slower growth due to physical factors such as decreased immersion time and probable desiccation (Menge 1976; Wethey 1985). Sheltered stations such as FS and GN had a lower abundance of barnacles in Zone 1 ( < 10%) than at exposed stations such as FE and BP (ca. 30%) where the spray zone increased available moisture for settlement and survival, es- pecially in cracks or crevices that retained mois- ture. Competition for space was not the limiting factor for barnacle abundance in the high intertidal zone. - Conversely, at times of maximum cover (early summer), barnacles covered virtually all available surfaces in the mid and low intertidal zones, al- though competition for space with the perennial alga Chondrus crispus was responsible for the lower absolute abundance of barnacles in Zone 3 as compared to Zone 2. Particularly in Zone 2, intraspecific competition for space was a source of barnacle mortality. When barnacles settle densely and grow rapidly, they crowd and cannot expand laterally. The consequent upward expan- sion of the individuals results in relatively small basal areas for attachment, and the hummocks (cf. Connell 1961; Grant 1977) were lost during storms. Predation by carnivorous snails (most com- monly Urosalpinx cinerea and Thais lapillus) also influences barnacle abundance, especially in the low intertidal zone (Connell 1961; Menge 1976). In the 3-unit operational period to date, Urosalpinx (primarily) and Thais (to a lesser ex- tent) were present throughout the year, but were most abundant and active in late summer, and in Zone 3. Maximum coverage was less than or equal to 3%, but this population was sufficient to reduce average barnacle cover to ca. 3% by November. Similar predator abundances, and similar control of barnacle populations, have been reported by others (e.g., Menge 1976; Katz 1985). In general, abundances of barnacles and preda- tory snails and their seasonal cycles during 3-unit operation were similar to those of the pre-op pe- riod. The single exception to these basic trends 24 y3A00 lN30M3d ySAOO XN3oy3d Rocky Intertidal Studies 25 y3AOO ±N33!J3d y3AOO J.N30a3d 26 MBAOO XN30a3d yBAOO XN30a3d CLT-J O 3 i s n & -^l a-^ 5* "r § i 1 ii ySAOO XN30!J3d ySAOO ±N30a3id Rocky Intertida] Studies 27 y3A00 lN30y3d WHAOO ±N30a3d I I i -- 9r i5 3 -^ a3AOO ±N30a3d g S S S S 2 a3/>oo XN30y3d 4i 1 28 in local barnacle and predatory snails abundances was the coverage and periodicity observed at FE since September 1984. These changes were related to the opening of the second quarry cut, and attributed to elevated water temperatures (NUSCO 1985, 1986, 1987). Mussels Mussels (mostly Mytilus edulis) were present at aU rocky shore sampling sites during the 3-unit operational period, usually occupying < 5% of the available space (Fig. 6). Mussels typically occurred as clumps of adults in rock crevices where they could persist for long periods (in ex- cess of 20 years, Bayne 1976) or as juveniles that settled among barnacles in mid-summer, but were eliminated by late summer by Urosalpinx and Thais. One exception to this generality occurred at Seaside Sheltered in July 1986, when 8% of the low intertidal (Zone 3) was covered by a mass of adult mussels that washed into one transect of the study area, presumably after being dislodged from a nearby population. These mussels did not firmly attach themselves and were washed away by the next sampling period. Similar events occurred at SS and SB in 1982 (NUSCO 1983); however, these mussels persisted for as long as two years. Mussels have not been a dominant component of local rocky shore communities; in the pre-op period, mussel cover exceeded 10% only at Bay Point and Giants Neck. Other researchers in New England have reported that Mytilus is supe- rior to Balanus as a competitor for available space (e.g., Menge 1976; Grant 1977; Lubchencho and Menge 1978), and as a result dominates intertidal areas. We have also noted the competitive dom- inance of mussels in exclusion cage studies (e.g., NUSCO 1985, 1987). A common fmding of these studies was that mussels could exclude barnacles only when predation pressure was reduced, by physical conditions (e.g., wave shock) or experi- mental manipulation; these conditions are not representative of the Millstone area. An atypical pattern of mussel abundance was noted at Fox Island-Exposed (Fig. 6). In both summers of the 3-unit operational period to date, mussels settled densely among barnacles in the mid and low intertidal zones. These mussels grew rapidly and increased coverage to ca. 30%, outcompeting barnacles as they did so. However, the mussels were almost totally eliminated by au- tumn. Both the initial rapid growth and subse- quent high mortality were attributed to elevated water temperature. Incursion of warm water pro- duced optimum growth conditions in early sum- mer; however, maximum temperature for adult survival is reported to be ca. 27°C (Read and Cumming 1967; Bayne 1976; Gonzalez and Yevich 1976). This temperature occurred in the quarry (Schneider 1981; Johnson et al. 1983) and at FE under 3-unit operating conditions. FuctJS The perennial brown alga, Fucus vesiculosus, is found throughout intertidal communities locally, but is especially abundant in the mid intertidal zone (Fig. 7). Growth conditions are optimal in Zone 3, but Fucus is usually outcompeted by Chondrus (cf. Lubchenco 1980). Station- to-station variability exists in Fucus abundance, related to degree of exposure. Consistently low Fucus abundance was typical of BP (Zone 2, 0-3% in the pre-op and 3-unit operational peri- ods), an exposed sampling site prone to physical stress caused by storm activity and ice damage. The other stations, however, were more moder- ately exposed, and reflected conditions suitable for Fucus growth (Topinka et al. 1981; Keser and Larson 1984). Overall, average Fucus canopy in Zone 2 was about 32%, excluding BP data. Temporal variability is also evident in local Fucus populations. Fucus occupies new substrata following settlement of zygotes in spring (Knight and Parke 1950; Keser and Larson 1984), and growth of germlings which achieve peak abun- dance in late summer. Typically, Fucus settles between barnacles and utilizes the spatial escape until it is large enough to be unpalatable to graz- ing snails (Keser 1978; Geiselman and McCoimell Rocky Intertidal Studies 29 y3A03 INHOMHd ,;.a J a y3A00 iN30a3d y3A03 lN33a3d 30 ^-2 yBAOO iN30y3d o o O O O O a3A0D iN33H3d a3A03 iN30M3d a3A03 XN30y3d Rocky Intertidal Studies 31 y3A03 XN30y3d ysAoo XN3oyad Q_ QQ / / >/ a, \ \ >< -V <■> "4-4 > < J- < liJ S; S i ! -6 2l 2 1 r? y3A00 lN30a3d a3A00 XN30a3d CL- OD 1: s g 1 ojo a l^r^ I I 2i I Q-n i i E: E § i E J § « « g O V) M OS 1 E lS5 re ySAOO XN30y3d y3A00 XN30y3d 32 a3A03 lN33H3ci a3AOO iN33y3d tn 3 O 5 ! I — -i5 I y3A03 xN3aa3d y3A00 XN33y3d tn y3AOO XN33y3d ysAOO j>J3oy3d Rocky Intertidal Studies 33 M3A03 lN30y3d a3Aoo iN3oaad 34 ysAoo jj^i3oy3d 00 CO y3A00 XN33a3d 00 OO ySAOO J.N30a3d S 1 i I & H ! I I yiAOO lN30y3d ^ s n 1 \/' lu'^ 1 \/ ft ^i o 9 • 1- xfr i 1 § / A 1 1 1 \ / i .^s \ i 4— »K \ \ \ 1 --< ft m S 1 ^— .s / ,.l. 5 CLT-l 1 ! o — Lj 9 — g a g s 2 s s ^ O 2 * c- - o 1 i o-r-n «- ySAOO iN30y3d y3A03 JJMSOyjd Rocky Intertidal Studies 35 1981; Lubchenco 1983). The seasonal cycle is seen most clearly at Seaside Exposed. As Fucus plants mature, they become increasingly suscep- tible to epiphytism (Menge 1975), storm damage and ice-scouring (Mathieson et al. 1982; Chock and Mathieson 1983) in autumn and winter. These processes tend to remove many plants at once, opening new substrata for colonization and perpetuating the cycle of Fucus abundance (cf. Schonbeck and Norton 1980; Keser and Larson 1984). If these processes operate on a small scale, removal of old plants from one area will be offset by growth of young plants in a nearby area, and the average Fucus cover at the station will be relatively stable over time (e.g., SS and WP). If, however, the removal process clears a large area, many zygotes will settle, grow, and senesce in synchrony, and produce a long-term cycle in Fucus abundance, based on its 3-5 year lifespan. This phenomenon has been observed locally (e.g., FS and GN) and by other researchers (Niemeck and Mathieson 1976; Keser and Larson 1984). Pre-op and 3-unit operational Fucus populations were similar at most stations in the monitoring program. The mid intertidal zone was dominated by a Fucus canopy, and similar spatial and tem- poral patterns were observed for each operational period. Exceptions to these general trends in Fucus abundance were observed at MP and FE in the 3-unit operational period. Fucus abundance in Zone 2 at MP steadily de- creased from 45% in 1982 to 1% just before Unit 3 began operation in 1986; coverage has ranged from 1% to 4% in the 3-unit operational period (Fig. 7). This decrease in Fucus abundance was initially interpreted as the descending portion of a long-term Fucus cycle, like those seen at FS and at other stations, but more protracted at MP than elsewhere. The delay in increased Fucus abundance may be related to grazing pressure; other researchers have shown that high grazer densities can retard Fucus recolonization for sev- eral years (Lubchenco 1983; Keser and Larson 1984). However, the abundance of grazers was stable for most of the study's duration and similar to that at other stations, despite high densities of Littorina littorea at MP in November 1984 (over 50% cover in some quadrats). Continual low Fucus abundance at MP may be related to water temperature; however, Kanwisher (1966) showed that adult Fucus can tolerate temperatures up to 30 "C without thermal injury, and maximum wa- ter temperatures measured at MP were ca. 25 °C. To more fully understand the processes occurring at MP, a schedule of temperature measurement will be developed and a series of exclusion cages is planned in spring 1988. As this site is the second-closest station to the discharges, continued monitoring is needed to determine if MP could be subjected to thermal effects during 3-umt op- eration. Fucus abundance at FE has undergone substan- tial changes since the inception of the monitoring program. Fucus coverage was very high from 1979 to 1981 (e.g., averaged ca. 60% cover in Zone 2; Fig. 7) and gradually decreased to low abundance (ca. 15% cover in July 1983, ca. 6% cover in July 1984), suggestive of the descending portion of the local 3-5 year Fucus abundance cycle. The thermal impact resulting from 2-unit/ 2-cut operation caused elimination of the Fucus population at FE in September 1984, conse- quently interrupting the Fucus population cycle; the expected increase in Fucus cover following settlement of zygotes in spring 1984 was not seen. The abundance of grazers at FE prior to thermal impact was similar to that measured at other rocky intertidal stations, but grazers have been virtually nonexistent at FE since the time of im- pact; therefore, the failure o{ Fucus to recover was not due to grazing pressure. Lethal effects of high water temperatures on Fucus populations have been discussed in past annual reports (NUSCO 1985, 1986, 1987) and by other researchers (e.g., Kanwisher 1966; Vadas et al. 1976). Fucus settled in spring 1985, reached about 6% cover, then was eliminated in late summer. Fucus zygotes settled the following spring (1986, beginning of the 3-unit operational period) and germlings grew to achieve a higher abundance (ca. 20% in Zone 2) in sum- mer. These plants survived and eventually reached a peak abundance of over 60% in summer 1987. It is apparent that summer conditions at FE, as 36 well as the shoreline between FE and the dis- charges, approach the physiological limits for sur- vival of Fucus. It is likely that slight changes in summer maximum ambient water temperatures, and plant operating levels during critical periods, will affect the Fucus population at FE. Chondrus Chondrus crispus is a bushy, perennial red alga that is the dominant species in the low intertidal zone at most stations. Chondrus contributes to community stability by maintaining extensive populations through time. In both pre-op and 3-unit operational periods, Chondrus occupied from 4-80% of the Zone 3 substrata (average ca. 45%, Fig. 8 a,c), excluding FE which is discussed separately. Station-to-station variability of Chondrus abundance existed; MP and WP showed consistently high (70-80%), and FS consistently low (ca. 10%) Chondrus abundance in Zone 3 for both operational periods. Chondrus grows as a clump of upright stalks from a basal crust; longevity of the stalks is 2-3 years, and the crust may live 6 years or longer (Ring 1970; Taylor and Chen 1973). Young stalks continually grow up to replace old ones that are lost. Occasionally, processes occur that remove stalks of all ages. A period of extremely low tides concurrent with extremely cold temper- atures in February 1980 exposed Chondrus to le- thal conditions (NUSCO 1982, 1983), and was responsible for the general decline in local Chondrus abundance in the following spring. Similar events, on a smaller scale, have occurred in nearly every winter of our study. Regrowth from surviving crustose holdfasts occurs relatively quickly (Prince and Kingsbury 1973), maintaining Chondrus as the dominant species in the low intertidal zone at most rocky intertidal sampling stations. If the basal crust of Chondrus is daTiaged or removed, recovery of the population is much slower, on the order of 3-5 years (Ring 1970; Lubchenco 1980; NUSCO 1987). This extent of damage was seen only at Fox Island-Exposed. When temperatures exceeded 28 °C at FE in Sep- tember 1984, Chondrus was eliminated from the community and was replaced by opportunistic, ephemeral algae; the subsequent community was described in detail in past aimual reports (NUSCO 1985, 1986, 1987). CAortJ30a3d a3A00 lN30M3d I o I § M3AOO lM30y3ci O O Q O O < id3A00 XN30y3d § 1 S i, 3 e- Rocky Intertidal Studies 39 yBAOO lN30y3d g s s s e a3A03 J>130y3d (/I o -5 6 I I _ M3A00 iN30a3d O O Q O O C y3A03 lN30a3d 40 a3AOO iN30y3d aSAOO XN30y3=! Rocky Intertidal Studies 41 November 1983, and has maintained relatively high abundance since. The seasonal periodicity of Polysiphonia noted throughout the area, includ- ing FE prior to the opening of the second quarry cut, has not been seen at FE subsequently. Monostroma coverage at MP decreased from spring 1984 through winter 1986, which may have indicated a thermal effect of the 2-umt/2-cut plume since MP is the second closest station to the MNPS discharge. Also, Monostroma abun- dance at WP was atypically low in spring 1987 possibly indicating influence of the 3-unit plume. Further monitoring will allow us to determine whether these events represent natural variability or an indication of thermal incursion. To summarize, rocky intertidal quantitative data collected during 3-unit operation were within pre- op ranges at most stations. Spatial and temporal abundance patterns of local species have been established and are similar to those observed by other researchers in New England (e.g., Menge 1975; Menge 1976; Grant 1977). Seasonality, de- gree of exposure, and competition induced vari- ability in community parameters. The structure of the Fox Island-Exposed community changed after September 1984 when water temperatures exceeded 28 °C, the upper physiological limit of most species present. The FE community had not re-established itself to pre-impact levels by September 1987 (3-unit operation), and it is not expected to under existing thermal fluctuations described in the Temperature section of this report. Recolonization Studies Natural perturbations to established communi- ties of attached plants and animals result in free space for recolonization, and are an on-going pro- cess in intertidal communities. Factors that de- termine rate of recovery following perturbation include physical or physiological stress, grazing and predation, species competition, and temporal and spatial heterogeneity (Dayton 1975). In ad- dition, life-history stages, degree of exposure, and time of denuding can determine rate of recolonization in intertidal communities. These recolonization studies, therefore, simulate natural processes, and allow examination of the factors that influence community recovery. Previous studies (e.g., NUSCO 1985) have shown that rates and patterns of recolonization, especially in the first year following denuding, may be characterized by patterns of recovery of two major groups that distinguish local intertidal communities: Balanus balanoides and Fucus vesiculosus. Chondrus crispus, the dominant alga in the low intertidal zone of undisturbed areas, has not recolonized the denuded transects ade- quately ui the first 12 months to warrant inclusion in this report. Preliminary data from the 1986 autumn denuding are compared to the 1981 au- tumn denuding; these data will be updated in subsequent reports. Recolonization transects data arc compared to undisturbed transects data for each site to show rate and extent of recovery. Rates of community recovery are related to intertidal height, and to the complexity of species assemblages found in each zone. The high intertidal zone consists mostly of barnacles on otherwise bare rock (or occasionally, blue-green algae and ephemerals, which can appear in high intertidal areas within days of denuding) and there- fore appears recovered, or similar in appearance to undisturbed areas, by the end of the first bar- nacle set. On the other hand, the low intertidal zone is dominated by Chondrus and associated epiphytes; as noted earlier, Chondrus recovery on denuded substrata is slow, and it may take at least several years before Zone 3 recolonization quadrats resemble nearby undisturbed areas. Rates of recovery in the mid intertidal zone are intermediate; therefore, this report will emphasize data only from Zone 2. Barnacles Recolonization of barnacles was observed in spring in all recolonization transects after both autumn denuding experiments. Annual peak bar- nacle abundance, occurring in early summer, was very similar in undisturbed and recolonization transects at the four recolonization sites; peak 42 CLOS.-O -^ ^ v ~-^. "'^^ ~'^ Z> y^ ^ *\ ^^^r\. \ J X^**^»v c^ ^ ,^ CO y , CT> \ ^ o> \ c c N z h s TJ 3 > 3 O o TJ couiQ-car^ 3.^'' ^f X" ^J:- ^ .-'^' ^^M b) Pre-op — --'iS'^'^ 1 O-h — APR JUN AUG OCT Date DEC FEB APf Fox Island Giants Neck White Point Fig. 11. Ascophyllum growth under (a) 3-unit operational and (b) pre-operational conditions. Curve corresponds to the Gompertz growth model fitted to the data, and including inflection points (i-P-) as vertical lines. Error bars represent monthly mean lengths ±2 SE. Rocky Intertidal Studies 47 E npre a 1984 FN 8586 3-unit a 110.00 83.66 90.68 98.26 r^ .31 .74 .69 .71 i.p. OSjul I4inay ISjul 20iui a) Fox Island -U ■a 1984 •FN 1985-36 pre-op 3-unit a 34.40 86.25 .- .77 .75 I.p. /3jul 04 .30 .82 i.p. 31jul Olaug c) White Point • pre-op — 3-unit Fig. 12. Ascophyllum growth, under pre-operatJonal and 3-unit operational conditions: (a) Fox Island, (b) Giants Neck, and (c) White Point. Curve corresponds to the Gompertz growth model fitted to the data, and including inflection points (i.p.) as vertical lines. Error bars represent monthly mean lengths ±2 SE. 48 atures 3-4 °C above ambient for 4-5 hours per tidal cycle. Ascophyllum growth patterns at the control sites have remained consistent throughout the pre-op and 3-unit operational periods (Figs. 12b and c). Total lengths (84-89 mm) and periods of peak growth (late July) were similar between stations and between operational periods. In summary, local Ascophyllum populations have not been affected by operation of Unit 3 to date. Growth rates were similar to those reported for Ascophyllum throughout its geographical range (cf. Vadas et al. 1976, 1978; Stromgren 1977; Wilce et al. 1978). Mortality Ascophyllum mortality, determined as thaUus breakage, is a result of mechanical and environ- mental stress. Thallus breakage could occur be- low the base tag, between the base tag and the colored tie wrap used as a tip tag, or between the tip tag and the growing apex. Our measure of tip mortality represents either loss of the tip tag, or loss of all viable apices on a tagged tip. Loss of tip tags implies mechanical removal and im- mediate loss of plant material. Loss of viable apices and/or damage to the apical cell implies a potential loss of biomass due to lack of growth. Locally, mortality is variable from year to year. Factors that contribute to mortality include degree of exposure, grazing, wave-force and movement, temperature, competition for space, increased drag due to epiphytization, ineffective reproduction, and thallus breakage (Vadas et al. 1976, 1978; Bokn and Lein 1978; Seip et al. 1979). Mortality at each Ascophyllum site is illustrated by plots of the number of remaining tips (Fig. 13) and number of remaining plants (Fig. 14). Means of monthly values from 1979-86 (pre-op) are plotted with their ranges, together with 1986-87 (3-unit operation) data, for the control stations WP and GN. Fox Island Ascophyllum mortality data are divided into three periods, corresponding to the temperature regimes specified in the Ascophyllum growth section: 1979-84 (FL, pre- op) with means of monthly values plotted with their ranges; 1985-86 (FN, pre-op); and 1986-87 (FN, 3-unit operational). The 1984-85 pre-op data are excluded because of the elimination of Ascophyllum at FL in September 1984. Under 3-unit operational conditions, the control stations had very similar mortality rates, and the experimental station showed a different, higher rate of mortality than the control stations for both tips and plants. Tip mortality was 94% at FN, and 59% and 73% at WP and GN, respec- tively; plant mortality was 86% at FN, and 40% and 46% at WP and GN (Figs. 13 and 14). Mortality rates at FN were more precipitous than at the control stations, but at all stations, greatest loss in tips and plants occurred in early autumn. Similar patterns of Ascophyllum mortality were evident in pre-op mortality data. The greatest loss of tips and plants was seen at Fox Island, while WP and GN had mortality rates similar to each other under pre-op conditions. Approximate tip and plant losses, respectively, at Fox Island were 80% and 60%, at both WP and GN 75% and 50% (Figs. 13 and 14). Mortality rates at Fox Island were more sudden than at the control stations. Greater mortaUty is typical of exposed stations, such as FL and FN (cf. Jones and Demetropoulos 1968; Baardseth 1970b; Seip 1980). Average Ascophyllum plant and tip losses in the 3-unit operational period are similar to losses re- corded in the pre-op period, excluding 1984-85 data from FL (during and after elimination of Ascophyllum at FL). I^cal plant loss data are similar to those of Chock and Mathieson (1983) who reported 50% of the autumn's Ascophyllum standing crop in New Hampshire was lost to storms and ice-rafting. Other researchers have noted extensive losses of Ascophyllum axes (e.g., Vadas et al. 1978; Wilce et al. 1978; Topinka et al. 1981; Mathieson et al. 1982) suggesting that decomposition of the fragments provides an im- portant source of nitrogen to the detrital pool. Rocky Intertidal Studies 49 JUN AUG OCT DEC FEB APR Date •pre-op — 3-unit -FN 1985-86 250 225 200 175 150 125 100 75 50 25 0*— APR b) Giants Neck JUN AUG OCT DEC FEB APR Date — pre-op — 3-unit 250 N~1^^ 225 \X^ 200 \ 175 \ --^v..^ 150 \ ^s^ 125 \ ~~~~\^ 100 75 c) White Point \ -._ 1 . 50 -- 25 0 Al "R JUN AUG OCT DEC ■ FEB APR Date — pre-op -3- unii 50 45 "^ i 40 \\ 35 ^ 30 \ \- 25 \ \ ■'-'^ 20 \ ^^ s \ -■- 15 a) Fox Island ^ 10 ■\ N 5 APR JUN AUG OCT DEC FEB Dote • pre-op — 3-unlt ■FN 1985-86 APR APR b) Giants Neck JUN AUG OCT DEC FEB Dote — pre-op — 3-unit APR bU 45 ■^ X 40 \ \_ 35 \ N 30 '*• ^ 25 ^-.. --. 20 15 c) White Point 10 ' 5 n APR JUN AUG OCT DEC Date — pre-op — 3-unit FEB APR Fig. 13. /(i-co/^/z^/Zum mortality, as number of remaining tagged tips. Unit 3 operational data (1986-87) are plot- ted against the mean and range of pre-operational data (1976-86). Fig. 14. /(.rcop/y/Zum mortality, as number of remaining tagged plants. Unit 3 operational data (1986-87) are plotted against the mean and range of pre-operational daU (1976-86). 50 Vegetative propagation and lateral proliferation from surviving holdfasts replace lost plant material (Printz 1956; Baardseth 1970a; Keser et al. 1981), and these replacement processes maintain the A scophy Hum populations at consistent levels. The effects of thermal stress were seen only at FL, less than one month after the opening of the second quarry cut in August 1983 and in the following summer. Before the second quarry cut was opened, Ascophyllum mortality was not as- sociated with proximity to the discharge, but rather to the increased physical stress associated with a higher degree of exposure. Ascophyllum is not expected to recolonize at FL, even though conditions may be lethal only for a short time in summer. The critical phase for Ascophyllum appears to be the establishment and growth of germlings on the substrate (Rueness 1973), and these stages are more susceptible to environmental impact than are adults (Bird and McLachlan 1974). Since repopulation involves long-term survival of individuals, even short-term exposure to lethal water temperatures in summer will prevent Ascophyllum recovery. The substra- tum previously occupied by Ascophyllum at FL will continue to be dominated by ephemeral algae, notably Codium fragile. The localized scale of this impact (150 m) must be emphasized. Sam- pling of control populations will continue to pro- vide information to the rocky intertidal monitoring program, as to whether thermal effects may be seen over a larger area (FS, MP, WP), and as to whether trends seen to date will continue during extended 3-unit operation. Summary 1. Exposure of local rocky shores to the full power 3-unit plume varies with tidal stage as well as distance from the discharges. On an ebbing tide the plume was deflected to the east, across Fox Island, and elevated water temperatures at FL and FE (75 m and 100 m east of the discharges, respectively) for 9- 1 1 hours per tidal cycle. During maximum tidal flooding, temperatures were close to ambient levels for 1-2 hours, as the plume was deflected to the west and the heated wa- ter was displaced. About 250 m west of the discharges, at MP, water temperatures were 4 °C above ambient at high tide and 2-3 °C above ambient during low slack water as the plume spread laterally. Consistent full power, fuU cooling water operation has not existed during periods of maximum ambient water temperature; it is during these conditions that intertidal communities at nearby stations may be affected by the 3-unit thermal plume. 2. Overall, 128 benthic algal species have been identified in the 3-unit operational period. Relative species proportions in the first 12 months of the 3-unit operational period were virtually identical (45:26:29) to those of the pre-operational summary. Spatial and tem- poral patterns of occurrence were evident in the local intertidal qualitative collections. Community changes at FE (i.e., a decrease in species numbers, shift in divisional pro- portions, loss of perennial species, and in- creased abundance and persistence of oppor- tunistic species) were attributed to elevated water temperatures resulting from opening the second quarry cut, not from Unit 3 start- up. Most of these community changes have persisted during 3-unit operation, as water temperatures approaching 28 °C have oc- curred at FE in summer. 3. Local communities, and the factors respon- sible for structuring them, have remained rel- atively stable since Unit 3 began operation and were similar to communities under pre- operational conditions. Seasonality, degree of exposure, and competition induced vari- ability in community parameters. 4. Recolonization experiments were undertaken to isolate factors that influence the structure of rocky intertidal communities. Locally, recolonization was influenced by time of year in which denuding occurred and it was related to degree of exposure and intertidal height. After the 1986 autumn denuding, barnacles Rocky Intertidal Studies 51 recovered after spring barnacle set; Fucus also recolonized the following spring. Oslofjord, Norway. Norw. J. Dot. Vol. 25, pp. 9-14. Oslo. ISSN 0300-1156. 5. Ascophyllum growth rate has been shown to be sensitive to water temperature changes, especially increases to ambient temperature. Ascophyllum growth at FN in the 3-unit op- erational period was similar to that seen in the single pre-op growing season for which there is data (1985-86); growth patterns at the control sites have remained consistent throughout the pre-op and operational peri- ods. Under 3-unit operational conditions, the control sites had very similar mortality rates, and the experimental station showed a different, higher rate of mortality than the control stations for both tips and plants. Lo- cal Ascophyllum populations have not been affected by operation of Unit 3 to date. Sampling of control populations will continue to provide information as to whether thermal effects may be seen over a larger area, and as to whether trends seen to date will continue during extended 3-unit operation. References Cited Baardseth, E. 1970a. Synopsis of biological data on knobbed wrack Ascophyllum (Liimaeus) Le.Tolis. FOA Fisheries, Synopsis #38, Rev. 1. . 1970b. A square-scanning, two-stage sam- pling method of estunating seaweed quantities. Report No. 33, NITT, University of Trondheim, Trondheim, 41 pp. Chock, J.S., and A.C. Mathieson. 1983. Varia- tions of New England estuarine seaweed biomass. Bot. Mar. 26:87-97. Connell, J.II. 1961. Effects of competition, pre- dation by Thais lapillus and other factors on natural populations of the barnacle, Balanus balanoides. Ecol. Monogr. 31:61-104. . 1975. Some mechanisms producing struc- ture in natural communities: a model and ev- idence from field experiments. Pages 460-490 in M.L. Cody and J.M. Diamond, eds. Ecology and evolution of communities. Belknap Press, Cambridge, Mass. David, n.M. 1943. Studies in the autecology of Ascophyllum nodosum LeJol. J. Ecol. 31:178-199. 706 pp. Dayton, P.K. 1975. Experimental evaluation of ecological dominance in a rocky intertidal algal community. Ecol. Monogr. 45:137-159. Draper, N., and 11. Smith. 1981. Applied Re- gression Analysis, 2nd edition. Wiley- Interscience, New York. 709 pp. Edwards, D.C., D.O. Conover, and F. Sutter, III. 1982. Mobile predators and the structure of marine intertidal communities. Ecology 63:1175-1180. Bayne, B.L. (ed.). 1976. Marine mussels: their ecology and physiology. Cambridge University Press. Cambridge. 506 pp. Bird, C.J., and J. McLachlan. 1974. Cold- hardiness of zygotes and embryos of Fucus (Phaeophyceae, Fucales). Phycologia 13:212-225. Bokn, T., and T.E. I^in. 1978. Long-term changes in fucoid association of the iimer Fralick, R.A., and A.C. Mathieson. 1972. Winter fragmentation of Codium fragile (Suringar) llariot ssp. tomentosoides (van Goor) Silva (Chlorophyceae, Siphonales) in New England. Phycologia 11:67-70. Gciselman, J.A., and O.J. McConnell. 1981. Polyphenols in brown algae Fucus vesiculosus and Ascophyllum nodosum: chemical defenses against the marine herbivorous snail, Littorina lillorea. J. Chem. Ecol. 7:1115-1133. 52 Gonzalez, J.G., and P. Yevich. 1976. Responses of an estuarine population of the blue mussel Mytilus edulis to heated water from a steam generating plant. Mar. Biol. 34:177-189. Grant, W.W. 1977. High intertidal community development on a rocky headland in Maine, U.S.A. Mar. Biol. 44:15-25. Johnson, G., J. Foertch, M. Keser, and B.R. Johnson. 1983. Thermal backwash as a method of macrofouling control at MiUstone Nuclear Power Station, Waterford, CT, USA. Section 25 in I.A. Diaz-Tous, M.J. Miller, and Y.G. Mussalli, eds. Symposium on condenser macrofouling control technologies - the state of the art, June 1-3, 1983. Hyannis, MA. Jones, W.E., and A. Demetropoulos. 1968. Ex- posure to wave action: Measurements of an important ecological parcimeter on rocky shores on Anglesey. J. Exp. Mar. Biol. Ecol. 2:46-63. Kan wisher, G.W. 1966. Photosynthesis and res- piration in some seaweeds. Pages 407-420 in M. Barnes (ed.) Some Contemporary Studies in Marine Science. George Allen Unwin Ltd., Lx)ndon. Katz, C.H. 1985. predator-prey 66:1426-1438. A nonequilibrium marine interaction. Ecology Keser, M. 1978. Ecolo^cal effects of harvesting on the growth of Ascophyllum and the growth dynamics of FwcMj. Ph.D. thesis, Univ. Maine, 138 pp. Keser, M., R.L. Vadas, and B.R. Larson. 1981. Regrowth of Ascophyllum nodosum and Fucus vesiculosus under various harvesting regimes in Maine, U.S.A. Bot. Mar. 24:29-38. Keser, M., and J. Foertch. 1982. Co'onization and growth of Ascophyllum nodosum in New England. Presented at First Intemat. Phycol. Congr., St. John's, Newfoundland. August 9, 1982. Keser, M., and B.R. Larson. 1984. Colonization and growth dynamics of three species of Fucus. Mar. Ecol. Prog. Ser. 15:125-134. Knight, M., and M.W. Parke. 1950. A biological study of Fucus vesiculosus L. and F. senatus L. J. Mar. Biol. Assoc. U.K. 29:437-514. Lewis, J.R. 1964. The Ecology of Rocky Shores. English Univ. Press, London. 323 pp. Littler, M.M., and S.N. Murray. 1974. The pri- mary productivity of marine macrophytes from a rocky intertidal community. Mar. Biol. 27:131-135. . 1975. Impact on sewage on the distribution and community structure of rocky intertidal macro-organisms. Mar.-Biol. 30:277-291. Littler, M.M., and D.S. Littler. 1980. The evo- lution of thallus form and survival strategies in benthic marine macroalgae: Field and labora- tory tests of a functional form model. American Naturalist 116:25-44. Lubchenco, J. 1980. Algal zonation in the New England rocky intertidal community: an experimental analysis. Ecology 61:333-344. . 1983. Littorina and Fucus: effects of her- bivores, substratum heterogeneity, and plant escapes during succession. Ecology 64:1116-1123. , and B.A. Menge. 1978. Community de- velopment and persistence in a low rocky intertidal zone. Ecol. Monogr. 59:67-94. Mann, K.IL 1973. Seaweeds: their productivity and strategy for growth. Science 182:975-981. Mathieson, A.C., J.W. Shipman, J.R. O'Shea, and R.C. Hasevlat. 1976. Seasonal growth and reproduction of estuarine fucoid algae in New England. J. Exp. Mar. Biol. Ecol. 25:273-284. Rocky Intertidal Studies 53 Mathieson, A.C., N.B. Reynolds, and E.J. Hehre. 1981. Investigations of New England marine algae II: the species composition, distribution and zonation of seaweeds in the Great Bay estuary system and the adjacent open coast of New Hampsliire. Hot. Mar. 24:533-545. Mathieson, A.C., C.A. Penniman, P.K. Busse, and E. Tveter- Gallagher. 1982. Effects of ice on Ascophyllum nodosum within the Great Bay estuary system of New Hampshire-Maine. J. Phycol. 18:331-336. Mathieson, A.C., and E.J. Ilehre. 1986. A syn- opsis of New Hampshire seaweeds. Rhodora 88:1-139. Menge, B.A. 1976. Organization of the New England rocky intertidal community: role of predation, competition and environmental het- erogeneity. Ecol. Monogr. 46:355-393. . 1982. Reply to a comment by Edwards, Conover, and Sutter. Ecology 63:1180-11? Niemeck, R.A., and A.C. Mathieson. 1976. An ecological study of Fucus spiralis L. J. Exp. mar. Biol. Ecol. 24:33-48. NUSCO (Northeast Utilities Service Company). 1982. Rocky Shore. Pages 1-41 /« Monitoring the marine environment of Long Island Sound at Millstone Nuclear Power Station, Waterford, Connecticut. Arvnual report, 1981. . 1983. Rocky Shore. Pages 1-39 in Mon- itoring the marine environment of Long Island Sound at Millstone Nuclear Power Station, Waterford, Connecticut. Annual report, 1982. . 1984. Rocky Shore. Pages 1-31 in Mon- itoring the marine environment of Long Island Sound at Millstone Nuclear Power Station, Waterford, Connecticut. Annual report, 1983. . 1985. Rocky Shore. Pages 1-41 in Mon- itoring the marine environment of Long Island Sound at Millstone Nuclear Power Station, Waterford, Connecticut. Annual report, 1984. Menge, J. 1975. Effect of herbivores on commu- nity structure on the New England rocky intertidal region: distribution, abundance, and diversity of algae. Ph. D. thesis, Harvard Univ. 164 pp. . 1986. Rocky Intertidal Studies. Pages 1-38 in Monitoring the marine environment of Long Island Sound at Millstone Nuclear Power Station, Waterford, Connecticut. Annual re- port, 1985. Murray, S.N., and M.M. Littler. 1978. Patterns of algal succession in a perturbated marine intertidal community. J. Phycol. 14:506-512. MYAPCO (Maine Yankee Atomic Power Com- pany). 1978. Final Report, Environmental surveillance and studies at the Maine Yankee Nuclear Generating Station 1969-1977. 672 pp. . 1987. Rocky Intertidal Studies. Pages 1-66 in Monitoring the marine environment of Long Island Sound at Millstone Nuclear Power Station, Waterford, Connecticut. Summary of studies prior to Unit 3 operation. Paine, R.T. 1966. Food web complexity and species diversity. Amer. Natur. 100:65-75. NAI (Normandeau Associates, Inc.). 1984. Seabrook Environmental Studies, 1983. A characterization of baseline conditions in the Ilampton-Seabrook area, 1975-1983. A pre- operational report for Seabrook Station. Pre- pared for Public Service Company of New Hampshire, pp. 211-237. . 1980. Food webs: linkage, interaction strength and community infrastructure. J. Anim. Ecol. 49:667-685. Prince, J.S. and J.M. Kingsbury. 1973. The ecol- ogy of Chondms crispus at Plymouth, Massachusetts. I. Ontogeny, vegetative anat- 54 omy, reproduction, and life cycle, studies. Am. J. Bot. 60:956-975. II. Field Printz, H. 1956. Recuperation and recolonization in Ascophyllum. Second Int. Seaweed Symp. Braarud, T. and N.A. Sorenson (eds), Pergamon Press London, pp. 194-197. . 1959. Investigations of the failure of re- cuperation and repopulation in cropped Ascophyllum nodosum. Norske Vidensk. Akad. K. Mat. Nat. Kl. 3:1-15. Ramus,.!. 1972. Differentiation of the green alga Codium fragile. Amer. J. Bot. 59:478-482. Read, K.R.II., and K.B. Gumming. 1967. Ther- mal tolerance of the bivalve molluscs Modiolus modiolus L. and Brachidontes demissus Dillwyn. Comparative Biochemistry and Physiology. 22:149-155. Ricker, W.E. 1975. Computation and Interpre- tation of Biological Statistics of Fish Popula- tions. Bull. Fish. Res. Board Can. 191:382 pp. Ring, P.D. 1970. Developmental and ecophysiological studies of Chondrus crispus (L.) Stackh. M.S. thesis, Univ. Maine. 73 pp. Rueness, J. 1973. Pollution effects on littoral algal communities in the inner Oslofjord, with special reference to Ascophyllum nodosum. Helgolander wiss. Meeresunters 24:446-454. Sanders, H.L. 1960. Benthic studies in Buzzards Bay. III. The structure of the soft bottom community. Limnol. Oceanogr. 5:138-153. SAS Institute Inc. 1985. SAS user's guide: sta- tistics, 1985 edition. SAS Institute Inc., Gary, NC. 956 pp. Schneider, C.W. 1981. The effect of elevated temperature and reactor shutdown on the benthic marine flora of the Millstone thermal quarry, Connecticut. J. Therm. Biol. 6:1-6. , M.M. Suyemoto, and C. Yarish. 1979. An Annotated Checklist of Connecticut Sea- weeds. State Geol. and Nat. Hist. Surv. CT Dept. of Envu-on. Prot., Bull. 108, 20pp. Schonbeck, M.W., and T.A. Norton. 1980. Fac- tors controlling the lower limits of fucoid algae on the shore. J. Exp. Mar. Biol. Ecol. 43:131-150. Seip, K.L. 1980. A computational model for growth and harvesting of the marine alga Ascophyllum nodosum. Ecological Modelling 8:189-199. , G. Lunde, S. Melsom, E. Mehlum, A. MeUiuus, and H.M. Seip. 1979. A mathemat- ical model for the distribution and abundance of benthic algae in a Norwegian fjord. Ecol. Model. 6:133-166. Southward, A.J., and E.G. Southward. 1978. Recolonization of rocky shores in Cornwall af- ter use of toxic dispersants to clean up the Torrey Canyon spill. J. Fish. Res. Board Can. 35:682-706. Stephenson, T.A., and A. Stephenson. 1949. The universal features of zonation between tide- mjirks on rocky coasts. J. Ecol. 38:289-305. . 1972. Life between Tide Marks on Rocky Shores. Academic Press, London. 383 pp. Stromgren, T. 1977. Short-term effects of tem- perature upon the growth of intertidal fucales. J. Exp. Mar. Biol. Ecol. 29:181-195. Sundene, O. 1973. Growth and reproduction in Ascophyllum nodosum (Phaeophyceae) Norw. J. Bot. 20:249-255. Taylor, A.R.A., and L.G.-M. Chen. 1973. The biology of Chondrus crispus Stackhouse: systematics, morphology and life history. In Harvey, M.J., and J. McLachlan (eds). Chondrus crispus. Nova Scotian Inst. Sci., Halifax, pp. 1-33. Rocky Intertidal Studies 55 Topinka, J., L. Tucker, and W. Korjeff. 1981. The distribution of fucoid macroalgal biomass along central coastal Maine. Bot. Mar. 24:311-319. Vadas, R.L. 1972. Marine algae. Pages 250-310 in Third Ann. Rpt. on Environ. Stud. Maine Yankee Atomic Power Company. , M. Keser, and P.C. Rusanowski. 1976. Influence of thermal loading on the ecology of intertidal algae. In Esch, G.W. and R.W. MacFarlane (eds). Thermal Ecology II. ERDA Symposium Series, Augusta, GA p. 202-251. . 1978. Effect of reduced temperatures on previously stressed populations of an intertidal alga. In Thorp, .T.H. and G.W. Gibbons (eds). DOE Symposium Series, Springfield, VA. p. 434-451. (CONF-771114, NTIS). Wethey, D.S. 1985. Catastrophe, extinction, and species diversity: A rocky intertidal example. Ecology 66:445-456. Wilce, R.T., .1. Foertch, W. Grocki, J. Kilar, H. Levine, and J. Wilce. 1978. Flora: Marine Algal Studies. In Benthic Studies in the Vicinity of Pilgrim Nuclear Power Station, 1969-1977. Summary Rpt., Boston Edison Co. p. 307-656. 56 Contents Benthic Infauna 59 Introduction 59 Materials and Methods 60 Data Analyses 60 Regression Analyses 61 Biological Index Value 63 Species Diversity 63 Numerical Classification and Cluster Analyses 63 Intertidal Results 64 Sedimentary Environment 64 General Community Composition 64 Community Abundance 66 Number of Species 69 Community Dominance 69 Dominant Taxa 73 Species Diversity 81 Cluster Analysis 81 Discussion 84 Conclusions 85 Subtidal Results 86 Sedimentary Environment 86 General Community Composition 86 Community Abundance 89 Number of Species 89 Community Dominance 89 Dominant Taxa 96 Species Diversity 108 Cluster Analysis 108 Discussion 110 Conclusions 113 References Cited 113 Benthic Infauna Introduction Benthic infauna are relatively small organisms inhabiting intertidal beach and subtidal bottom sediments. The many species of worms (polychaetes and oligochactes), clams (molluscs) and crustaceans (amphipods), which comprise infaunal communities, are important components of marine ecosystems. Many of these organisms are prey species for demersal fishes (Woodin 1982; Moelleret al. 1985; Witman 1985; Le Mao 1986). In Long Island Sound (LIS), annelids are the main food item of winter flounder, Pseudopleuronectes americanus (Richards 1963). In addition, many studies have described the in- fluence of infaunal feeding, burrowing and tube building activities on nutrient recycling in marine ecosystems (Goldhaber et al. 1977; Alter 1978; Ilylleberg and Maurer 1980; Raine and Patching 1980; Zeitzschel 1980). Infaunal organisms are also useful environmen- tal monitoring tools because they are relatively sedentary and respond to stress in a predictable manner (Boesch 1973; Reish 1973; Sanders et al. 1980; Boesch and Rosenberg 1982; Young and Young 1982). For example, a physically stressed community will typically exhibit lower numbers of species and high densities of a few characteristic species (e.g., Polydora ligni, Capitella spp. and Mediomaslus amhisetd), which are capable of tol- erating environmental stress or can rapidly colo- nize an area following disturbance (McCall 1977; Reish et al. 1980; Sanders et al. 1980). Impact studies based on observed changes in infaunal community abundance and species com- position assume that the structure of the unimpacled community can be described, and that any trends evident before impacts occur can be identified and removed (Nichols 1985). Be- cause naturally occurring variations in physical factors such as cold winters (Beukema 1979), storms (Boesch et al. 1976), heavy rainfall (Flint 1985; .lordan and Sutton 1985) or changes in the level of competition and predation (L^vinton and Stewart 1982; Woodin 1982; Moeller et al. 1985) can strongly affect infaunal communities, only long-term data can adequately describe the range and overall trends in community abundance and composition required for impact assessment stud- ies (Holland 1985). Benthic infaunal communities in the vicinity of Millstone Nuclear Power Station (MNPS) have been sampled since 1969, as part of a larger mon- itoring program designed to assess the potential impacts of construction and operation of Mill- stone Units 1 and 2. Plant-related changes in benthic communities during this period have been attributed to construction and opening of the Unit 3 discharge cut and to construction and dredging near the intake structures (NUSCO 1987). These data also provide the baseline con- ditions against which potential changes associated with Unit 3 start-up will be evaluated (NUSCO 1987). This is the fu^st report to include data collected after commercial start-up of Millstone Unit 3 (April 1986). Environmental changes, which could possibly impact infaunal communities, in- clude doubling of the heated effluent discharged into I ,IS and the enlargement of the area subjected to the scour produced by the additional intake and discharge of the 3-unit (combined) cooling waters. In addition, changes might occur in re- sponse to chemical additions associated with the increased volume produced during 3-unit opera- tions. To assess changes in local benthic com- munities that might occur during 3-unit operation at MNPS, the infaunal sampling program was designed to address the following objectives: Benthic Infauna 59 1 . characterize the abundance and species com- position of infaunal communities inhabiting shallow subtidal and intertidal areas both within and beyond the influences of 3-unit operation at MNPS, 2. describe spatial and temporal patterns in these communities and identify the extent and direction of changes attributable to 3-unit operation and 3. assess the ecological significance of power plant-induced changes should they occur. For the purpose of assessing 3-unit operational impacts, the pre-operational period was considered to extend from March 1979 - June 1985. Collec- tions made from September 1985 - .lune 1986 and from September 1986 - .Tunc 1987 (the 1986 and 1987 reporting years, respectively) will be consid- ered operational years. The 1986 sampling year included three sampling periods before commercial start-up; however, this was considered an opera- tional year because certain conditions (i.e., oper- ation of Unit 3 cooling water pumps), which could influence infaunal communities, existed dur- ing at least 2 of the 4 quarters considered the 1986 sampling year. Materials and Methods Infaunal communities were sampled at four subtidal and three intertidal stations (Fig. 1). The Giants Neck subtidal (GN-S) and intertidal (GN- I) stations are located 5.5 km west of the power plant and serve as reference stations because both are beyond any physical influences of the power plant discharge. Data collected at these stations provide the baseline for assessment of naturally- induced changes which might occur on a regional scale (e.g., hurricanes, severe winters, heavy pre- cipitation). The intake subtidal station (FN-S) is located 0. 1 km seaward of the Millstone Unit 2 and Unit 3 intake structures and is exposed to the scouring (if any) produced by intake of cooling water. The effluent subtidal station (EF-S) is lo- cated approximately 0. 1 km offshore of the Unit 3 discharge. This area is subjected to plant-induced changes related to temperature, scour, and possible chemical additions associated with power plant operations. The .Jordan Cove subtidal (.TC-S) and intertidal (.IC-I) station.^, located 0.5 km east of the power plant and the White Point (WP-I) intertidal station, 1.6 km east of the power plant, are all within the area influenced by the thermal plume during some tidal stages (see Hydrothermal Studies section of this report). At each subtidal and intertidal station, ten 0.0078 m cores (10 cm diameter x 5 cm deep) were collected quarterly (September, December, March and .June). For reporting purposes, a sam- pling year begins in September and ends in June with the year of the June sample assigned as the sampling year. Subtidal samples were taken within 3 m of each station marker by SCUBA divers. Each sample was placed in a 0.333 mm mesh Nitex bag and brought to the surface. Intertidal samples were collected at approximately 0.5 m intervals parallel to the water line at mean low water. Samples were brouglit to the laboratory and fixed with a 10% buffered formalin/rose bengal solution. After a minimum of 48 h, organisms were floated from the sediments onto a 0.5 mm mesh sieve and the float and residue were preserved separately in 70% ethyl alcohol. Organisms were removed under dissecting microscopes, sorted into major groups (annelids, arthropods, molluscs, and others), identified to the lowest possible taxon and counted. Organisms that were too small or otherwise not quantitatively sampled by our meth- ods (e.g., nematodes, ostracods, copepods, and foraminifcra), were not removed from samples. At the time of infaunal sampling, a 3.5 cm di- ameter X 5 cm deep core was taken for sediment analysis which was performed using the dry sieving method (Folk 1974). Data Analyses Impacts on infaunal communities attributable to start-up of Millstone Unit 3 and the subsequent operation of 3 units at MNPS were assessed by 60 Fig. 1. Map of the Millstone Point area st^iovving Ihc location of intertidal (GNI = Giants Neck, JCI = Jordan Cove and WPI = White Point) and siibtidal (GNS = Giants Neck, INS = Intake, FFS = Emiicnt, JCS = Jordan Cove) inlauiial monitoring stations. comparing data obtained during the pre- operational and operational periods. Power-plant related impacts should be evidenced as changes in infaunal abundance, species number, and changes in dominance structure (Boesch 1973; Oden 1979; .Jordan and Sutton 1985). In addition, impacts might result in high densities of species considered "opportunists". (McCall 1977; Sanders et al. 1980). However, in LIS, many of these species are considered typical of soft-bo'tom com- munities (Rhoads and Young 1970; McCall 1977), thus, their appearance may not necessarily reflect power plant-induced impacts. The data analysis techniques used in this report were selected to allow for identification of power plant impacts within the naturally variable framework charac- teristic of temperate intertidal and shallow-water macrobenthic communities (Eagle 1975; Livingston 1982; Tlint and Younk 1983; Holland and Mountford 1977). Regression Analyses Multiple regression techniques were used to identify and remove significant temporal variation in community abundance, numbers of species and Benthic Infauna 61 dominant, species abundance related to abiotic factors, reproductive/recruitment cycles, and long- term climatic conditions. Removal of this varia- tion was performed to improve the sensitivity of statistical analyses used to compare data collected during the pre-operational and operational peri- ods. Regression analysis techniques were applied to log-transformed (l,N(X+l)) quarterly abun- dances (no. /core for total community and popu- lation abundances) and quarterly mean numbers of species (no. /core) collected from March 1979 through .June 1987. The following factors were used as explanatory variables: Precipitation Daily precipitation records compiled by the IJ.S Weather Bureau at the Ciroton Filtration Plant were obtained from June 1976 through .Tune 1987. Values to the nearest 0.01 inch were used as our "rain" data for the regression model. Water and Air Temperature Ambient water temperatures (at the intake struc- tures) and air temperatures (at the 33-foot level of the Millstone mcteorolo^cal tower) were ob- tained from the Northeast Utilities Environmental Data Acquisition Network (EDAN). Daily av- erages, based on observations made at 15-minute intervals, were calculated for the period .Tunc 1976 to .Fune 1987. Wind Speed and Direction Wind speed and direction (at the 33-foot level of the Millstone meteorological tower) were ex- tracted from the EDAN database at 15-minute intervals from .Tune 1976 to .June 1987. Ihese values were used to calculate a Wind Index which was wind speed weighted according to wind di- rection. A NOAA navigational chart of the sam- pling area was used to calculate site-specific, wind directional weighting coefficients. The directional weight ranged from 0, when wind could not in- fluence the station, to 1, when the wind could result in waves directly affecting the area. The Wind Index was then computed by multiplying the directional weight by the wind speed. Because the effect of wind was assumed to be cumulative, daily averages were derived using only Wind Index values greater than 0 (that is, when the wind was from a direction which could produce wind ef- fects). Sedimentary Parameters Sedimentary parameters including mean grain size and silt/clay content were obtained as part of the monitoring studies and the quarterly values included as explanatory variables in the regression analyses. Seasonal Reproduction-Recruitment Component Infaunal organisms in the Millstone area exhibit annual peaks in abundance, which often reflect the seasonal nature of reproduction and recruit- ment cycles or periods of favorable climatic con- ditions. Spectral analyses of quarterly data showed that annual cycles in community abundance and numbers of species were present. To account for this periodicity, harmonic terms having a period of 1 year were included as explanatory variables in the regression models. Climatic Extremes (Deviations) Additional variables were created to represent periods of extreme climatic conditions which have occurred during the sampling period. High or low deviations (i.e., extremes) were derived for each abiotic factor as the difTerence between the quar- terly mean or daily value and the 1 1 year mean for that quarter. Deviations based on quarterly means reflected the effects of longer term extremes (e.g., an unusually cold winter), while those based on daily values were intended to remove the ef- fects of shorter-term episodic events (e.g., storms). Daily deviations were averaged and summed (for cumulative effects) over each sampling quarter. 62 In all, 32 variables were used during initial mul- tiple regression analyses. These included 2 sedi- mentary parameters, 2 seasonal/reproductive com- ponents and 7 climatic variables, each of which had 4 values representing daily and quarterly high and low extremes. Model Selection Procedure The quarterly mean values were first detrended using a polynomial regression equation to create the residuals needed for regression analysis. If no significant long-term trend was evident, residuals were created by subtracting the quarterly mean from the 1 1 year mean. A step-wise multiple regression was then used on these residuals to select explanatory abiotic variables and combina- tions of variables that were significantly different from zero, at a probability level of p < 0.05. 1 his probability was deemed sufficient to guard against fitting more parameters than can be reliably esti- mated, given the sample size. The model that minimized the mean square error and maximized the R was selected as best model describing observed variability in community abundance, numbers of species and abundances of selected species. Analyses of covariance were then conducted to test for annual differences in abundance and spe- cies numbers using significant explanatory vari- ables as covariates. Results of these analyses and pair-wise t-tests on adjusted means (least square means) were used to identify significant (p < 0.05) interannual differences in the abundance and spe- cies number. In this report, pair-wise differences among years involving only 1986 and 1987 will be emphasized. Biological Index Value The Biological Index Value (BIV) of McCloskey (1970), an index of dominance, was calculated for the 1 0 most abundant taxa at each statios ■ collected from 1986-1987. To provide comparisons with the pre-operational data, a BIV was also calculated for 1980-85. To calculate the BIV, the top ten numerically abundant species in each sampling year are ranked from highest to lowest and ranlcs summed for each year. The BIV is sum of the ranks across all years for each taxon expressed as a percentage of a theoretical maximum sum that occurs if a species ranked first in all sampling years. For example, the BIV would be equal to 100% and the theoretical maximum equal to 60 when a species ranks first in abundance in each of six years and a total of 10 species are collected. Species Diversity Species diversity at each station was calculated using the Shannon information index: 1^' = Z^l"g2 5 {Pielou 1977) where n \ = number of individuals of the i species, N = total number of individuals for all species and S = number of species. An evenness component of diversity was calculated as: //' {Piehu 1977) where II max ~'og 2 ■'' ^^^ represents the theo- retical maximum diversity when all species are equally abundant. Evenness ranges from zero to one and increases as the numbers of individuals among species become more evenly distributed. Diversity calculations excluded oligochaetes and rhynchocoels (groups that sometimes accounted for over 80% of the total organisms collected) because they were not identified to species. Sim- ilarly, other organisms that could not be identified to species, either because they were juveniles or in poor physical condition, were excluded from this analysis. Numerical Classification and Cluster Analyses (Cluster analyses, based on annual abundances of organisms, were performed using the Bray- Curtis similarity coefficient. This coefficient is cal- Benthic Infauna 63 culated using the formula: ^2min(-?,^.^,.*) Z(^y + ^«) (Clifford and Stevenson 1975) where X jj = abundance of attribute i at entity j and X ik = abundance of attribute i at entity k. Based on these similarities, cluster analyses incor- porating a flexible sorting strategy (P = -0.25) was used to form station groups (L,ance and Williams 1967). Intertidal Results Sedimentary Environment Intertidal beach sediments at GN during the 1986-87 operational period were comprised of medium sands (0.3 - 0.4 mm) which consistently contained low amounts ( < 1%) of silt/clay (Fig. 2). At JC, grain size in 1986-87 ranged from 0.5 - 1 .0 mm (coarse- very coarse) and silt/clay content from 0.5 - 3.3% . Medium sands also predomi- nated during this period at WP ranging from 0.4 - 0.8 mm; silt/clay content at this station was also low throughout the sampling period ( < 1%). In the operational period, sediments were gen- erally coarser at both potentially impacted stations (JC and WP) than during prc-operational years. Higher grain size was first observed in .June 1985, prior to any possible 3-unit environmental changes that might influence grain size at our stations. Sedimentary characteristics at GN from 1986-87 were consistent with previous years results, and with no discemable increase over 1986-87 com- parable to that observed at JC and WP. General Community Composition Intertidal infaunal communities during 1986 and 1987 were dominated by oligochaetes at JC (68% and 87%), rhynchocoels at WP (68% and 59%) and polychaetes (70% and 60%) at GN, respec- tively (Table 1). Polychaetes were the second most abundant group, contributing 12-32% of the fauna at JC and 26-34% at WP. Oligochaetes accounted for only 5-6% of the total at WP and rhynchocoels < 1 % of the total collected at JC. At GN, oligochaetes were the second most abun- dant group (21%) followed by rhynchocoels (8-9%). Polychaetes generally accounted for over half of the total species collected each year. Arthropod species were nearly as numerous at JC as polychaetes, although abundances of individuals were usually low. Other major groups at this station (e.g., molluscs, rhynchocoels) v/ere repre- sented by few species. At both WP and GN, numbers of mollusc and arthropod species were low in 1986 and 1987. The general community composition at JC dur- ing the 1986-87 operational period was typical of that observed during the pre-operational study. Oligochaetes dominated this community and ac- counted for over 80% of the total organisms; polychaetes were usually the second most abun- dant group and accounted for most of the species. At WP, total abundances of polychaetes in 1986 and 1987 were below the 1980-85 range; the num- ber of oligochaetes at this station was also low while that of rhynchocoels was near the high end 64 1.50 1.2S STATION: GIANTS NECK GRAIN SIZE (-.-.-.) SILT/CUY (+-+-+) ■ ■^ 1.00 3 ^0.75 z f\ " o.so fvA^-J^ jK 0.25' », 0.00- MAR79 UAR80 MARSI MARa2 MARS3 MARa4 MAR8S UAR86 MAR87 MARSS 1.50 1.25 STATION: JORDAN COVE GRAIN SIZE (-.-.-.) SILT/CLAY (+-+-+) t 1 1.00 '^ 1 n 0.75 A U \ M/ I 0.50 0.25 O.DO' 1 I. .A v-v \ UAR79 VMRSO UAR81 UAR82 MAR83 UAR84 UAR85 MAR86 UAR87 UARS8 STATION: WHITE POINT GRAIN SIZE (-.-.-•) SILT/CLAY (+-+-+) ..»^,-»»-^-^«^^,..^^.-' S— ^.-»^.-»— ... MAR79 UARaO MARSI UARa2 MARB3 MARa4 MARaS MARBB MAR87 MARBB Fig. 2. Quarterly mean grain size (mm) and silt/clay content (%) of sediments sampled at Millstone intertidal stations from March 1979 - June 1987. Benthic Infauna 65 TABLE I. Number of species (S), number of individuals (N) and percentage of the total (%) for each major taxon collected at Millstone intcrtidal stations September 1985 - June 1987, with ranges from 1980 -1985. STATION Range 1980- ■1985 1986 1987 S N % S N % S N % GIANTS NECK Polychaela 13-24 613-2222 40-82 16 1040 70 27 1408 60 Oligochaeta 79-626 4-40 306 21 486 21 Mollusca 0-8 0-14 0-11 1 1 < 1 4 8 <1 Arthropoda 5-14 12-59 <1-2 3 3 < 1 3 19 1 Rhynchocoela 109-932 7-31 131 8 210 9 JORDAN COVE Polychaeta 12-22 1055-4120 12-30 23 3551 32 19 1651 12 Oligochaeta - 86-15012 40-86 7809 68 12365 87 Mollusca 6-10 10-711 "l 8 f^J r T , a - / 4 2 1980 1981 1982 1983 1984 1985 1986 1987 12- STATION; WHTC POINT INTERTIDAL 10 8 a: » a. 6 ^ 3 « " '"-1 ,- ' ■ ' 2 Fig. 6. Annual means of number of species of Millstone inlertidal infaunal communities from March 1979 - June 1987. (Annual means were adjusted using analysis of covariance which included abiotic and climatic conditions as covariates). Error bars represent ±2 SE. Benthic Infauna 71 TABLE 2. Mean number of individuals per core and Biological Index Value (BIV) of the ten most numerically abundant taxa collected at the Millstone intertidal stations September 1985 - June 1987, with annual range and mean (no./core) for 1980-1985. 1980-85 1986 1987 1980-85 1986-87 Station Range Mean Mean Mean BIV BIV GIANTS NECK Oligochaeta 2-16 6 8 12 83.3 91.7 Haploscoloplos fragilis 4-9 5 7 10 89.2 83.3 Scolecolepides viridis 2-1! 7 17 5 85.3 83.3 Paraonis fulgens 1-19 8 1 16 84.3 79.2 Rhynchocoela 3-15 8 3 5 90.2 75.0 Capitella spp. 80%) during either the pre- operational or the operational period. Quarterly mean abundances are presented with multiple re- gression models and adjusted annual means. Val- ues presented in the following figures are log transformed (LN(X + I)) and those in the text are the exponentials of these values. Oligochaetes These deposit-feeding annelids commonly in- habit the littoral and shallow subtidal marine hab- itats in areas of high organic content and feed on the bacterial populations that colonize organic detritus (Soulsby et al. 1982; IIuU 1987). In the Millstone area, oligochaetes have been among the more dominant intertidal organisms. This group ranked first in abundance at JC in both the pre- operational and operational periods. Although oligochaetes were consistently among the domi- nants at WP and GN, average densities were fre- quently an order of magnitude less than those at .IC. Quarterly oligochaete abundance from Septem- ber 1985 - .lune 1987, ranged from 0 - 30/core at GN, 9 - 666/core at .IC, 1 - 3/core at WP (Fig. 7A-C). Densities at JC varied widely during the last two years and values near the upper and lower extremes were obtained. Seasonal peaks occurred in .lune of 1986 and 1987. Large seasonal fluctuations also occurred at GN during the 1986-87 period, but unlike .IC, peaks occurred in September. The oligochaete population at WP Benthic Infauna 73 1 \ / / 9 \^ z o 111 z 5 1 Vi U) o \ I / (L+X30"l) 3y00 y3d AilSN3a NV3H (l+XOOl) 3MO0 M3d AllSN3a NV3H S E ^ re «— . c- t^ — o E ;L.^ E + c X ^ Z — , c V E E m o •u j:; C! ■o c F _ ,o •o t 5 o ^ cy-c = (l+X 6o-t) 3H00 a3d/UJSN3a NV3H (L+x 6o-i) 3aoo aad xiisNaa nv3h 74 station WHITE POINT INTERTIDAL Oligochaetes R^ =0.54 UAR79 MAR80 MAR81 MAR82 MAR83 MAR84 MARB5 MAR86 MAR87 MAR88 STATION; WHITE POINT INTERTIDAL Oligochaetes c. Fig. 7. Continued. during the last two years was much less variable than at other stations. After multiple regression analysis removed 50% (GN), 17% (.IC) and 54% (WP) of the temporal variation, no significant trends were evident at .JC or GN, while at WP, there was a significant de- creasing trend fiom 1982-1985 that continued through the operational period (1986-87). Oligochaete abundance at GN in 1987 was sig- nificantly liigher than abundances in 1980-1983. At JC, annual means for 1986 and 1987 (also 1985) were higher than most previous years, but significant from only 1984. Mean abundance at WP in 1986 and 1987 was significantly lower than only 1982, when a peak in abundance occurred. Scolecolepides viridis Scolecolepides viridis is most common intertidally, burrows in a variety of substrata and is particularly abundant in sand (Wells and Gray 1964). This polychaete is most frequently found in areas of reduced salinity (Smith 1964). Adults Benthic Infaima 75 inhabit a mucus-lined burrow and feed on surface deposits, detritus, diatoms, filamentous algae and nematodes (Sanders et al. 1962). At Millstone, this species consistently ranked among the nu- merical dominants at GN and JC and although frequently among the top ten at WP, abundances at this station were usually very low ( < 1/core). During 1986-87, quarterly densities of Scokcolepides vlridis exhibited wide seasonal fluc- tuations, with abundances ranging from 0 - 65/core at GN and from 2 - 105/core at JC (Fig. 7D-E). Although most seasonal values were in the range of previous values, at both stations the June 1986 density was higher than all previous values; the June 1987 density, in contrast, was the lowest density observed during .Tune since 1980. Multiple regression models accounted for 59% (GN) and 58% (JC) of the variation since 1980, much of which was attributable to the strong seasonal reproductive/recruitment cycles evident in abundances at both stations. After this known variation was removed, no significant trends were evident at either station. In addition, annual ad- justed means over 1986 and 1987 were within previous ranges at both stations and there were no significant diiferences among any of the sam- pling years. Paraonis fulgens Paraonis fulgens is a deposit-feeding polychaete that typically inhabits sandy intertidal beaches (Whitlatch 1977; Strelzov 1979) from Maine to North Carolina (Gosner 1971). This species has typically been among the dominants at only ex- posed sandy beach stations (GN and WP) and has been found only once at JC. At both GN and WP, Paraonis fulgens is generally most abun- dant in June although peaks have sometimes been recorded in September. During the last two years, quarterly abundances at WP ranged from 2 - 12/core and from 0 - 20/core at GN (Fig. 7F-G). Abundances at WP in 1986 and 1987 were within the range of past years, although the value obtained in September 1986 was lower than in any previous September. At GN, the range of quarterly values in 1986 was low relative to values in past years; an increase occurred in 1987 and all quarterly values were near the upper end of the range. Multiple regression analyses indicated that no significant long-term trends occurred in the abun- dances of this species at either WP or GN. At WP, despite the consistent decline in annual abun- dances, there were no significant differences be- tween any sampling years. At GN, the 1987 annual mean was significantly higlier than 1986 and also 1980, 1981 and 1984. The 1986 annual mean was significantly lower than 1982. Hediste diversicolor Hediste diversicolor can be found in near-shore waters from the North Atlantic and North Sea to the Mediterranean (Gosner 1971). This omnivo- rous polychaete is frequently abundant in nutrient rich areas and has been considered an "opportun- ist" and an "indicator of pollution" (Hull 1987). Hediste diversicolor was a consistent dominant at JC, ranking 3rd and 2nd in abundance in 1986 and 1987, respectively. From 1980 through 1985, this species ranked 3rd according to the BIV. Quarterly density during 1986-87 ranged from 5 - 62/core and most values were within the range of previous years (Fig. 7H). In both years, den- sities of this species exhibited large seasonal fluc- tuations with the highest values occurring in Sep- tember. A similar pattern was observed in 1984 and 1985. The June 1987 mean was the lowest June density observed for this species since 1983. The multiple regression analysis identified a sig- nificant trend in the abundance of Hediste diversicolor due to increased abundance from 1982 through 1985. This trend continued through 1986. The increased annual means were primarily due to the high seasonal peaks which became evident in 1984. Annual abundances in 1986 and 1987 were significantly higher than 1982 and 1983; 1986 was also significantly higher than 1980. 76 (1+X301) 3aoo aid ajjsnbo nvsh ( I+X301) 3^00 a3d AJJSN3a NV3I^ (i+x 6on) 3yoo a3d AjjsNsa n¥3m (l+x 6oi) ayoo aad Alls^Ga nv3m Benthic Infauna 77 (i+xoon) 3yoo a3d AiisNaa msn (1.+X901) 3yoo a3d jujsnso uvin ( L+X 601) 3yoO y3d AilSNSa NV3M (l+X 6o"l) 3aOO H3d AilSN3a NV3W 78 Fig. 7. Continued. station: JORDAN COVE INTERTIDAL Hedxste diversicoLor R ^ = 0.47 UAR79 MAR80 MAR81 MAR82 MAR83 MAR84 MARB5 MAR86 UAR87 MAR88 , STATION: oORDAN COVE INTERTIDAL Hediste diversicoloT Rhynchocoela The majority of rhynchocoel species either bur- row in the sand or live among seaweeds on rocky shores. Although common in all seas, this group is most common in colder waters. Species which live in mud and sand are excellent burrowers and coupled with their ability to stretch, they are able to escape the beating surf (MacGinitie and MacGinitie 1968). These carnivores have been among the numerical dominants at both GN and WP throughout the monitoring program. They were the most abundant taxa collected at WP during 1986 and 1987. At GN, this taxon was less abundant in 1986-87 (ranking 4th) compared to the 1980-85 period; when it ranked 1st. During 1986-87, quarterly abundance ranged from 1 - 43/core at WP and from 2 - 10/core at GN (Fig. 7I-J). Seasonal patterns in abundance were evident only at WP, where densities were highest in June in each of the last two years, a pattern consistent with previous years. Abun- dances at WP during 1986-87 fluctuated widely Benthic Infauna 79 (i+x30"i) ay 00 aid ajjsnbq nv3h (i+xoon) 3aoo a3d AusNaa nv3h (l+x 6oi) 3aoo a3d xusNoa Nvaw (l+X 601) 3aO0 a3d AJJSN3a NV3W 80 with values near the extremes (high or low) in each sampling quarter. Abundances at GN during this period were well within the range established in previous years, and exhibited much less tem- poral variation. After regression analysis removed variation at- tributable to abiotic factors (23% at WP and 45% at GN), no significant trends in rhynchocoel abundance were evident. In addition, no signifi- cant differences were evident among years at either sampling station. Haploscoloplos fragilis This burrowing deposit-feeder is capable of in- gesting large sand grains and deriving nutrients from the bacterial flora developed upon them. In addition, morphological adaptations allow this species to rapidly burrow through loose uncon- solidated sands so that this species can inhabit areas of high sediment movement (Myers 1977). Haploscoloplos fragilis is a typical dominant of the clean sandy beach communities found at GN and WP where it ranked either 2nd or 3rd in terms of abundance during the last two years. Quarterly densities from September 1985 to June 1987 ranged from 0 - 26/core at GN and 0 - 8/core at WP (Fig. 7K-L). At both stations, this species was most abundant in September and least abundant in March or ,Iune. Peak abun- dances at GN during 1986 and 1987 were generally higher than previous years (except 1985), while at WP they were below those obtained in previous years. After removing natural variation (72% at GN and 86% at WP), there were no significant long- term trends in density at either station, nor were there any significant differences among sampling years at GN. At WP, the 1986 and 1987 means were significantly different from only 1984. Species Diversity Annual mean species diversity (H') during the past two years ranged from 1.1 - 2.0, evenness (J) from 0.4 - 0.6, and species numbers from 6-17 (Table 3). Lowest diversity of aU stations occurred at GN in 1986 (11'= 1.1) reflecting the generally lower numbers of species and low evenness (due to high densities of Paraonis fulgens). All indices at GN for 1987 were within the range of previous years. At JC, H', S and N were generally higher than other stations in both 1986 and 1987 with 1987 totals reflecting high abundances of Scolecolepides viridis, He.disle diversicolor and Fahrica sabella. Overall there were no major shifts in parameters used to describe diversity of intertidal communities in 1986 or 1987 relative to the pre-operational period. Althougli the values of IT, S, and N were lower at WP in both 1986 and 1987, they were still within the historical range. Cluster Analysis Cluster analysis produced a dendrogram which showed the very low similarity (-26.0 %) of the JC community to that of WP and GN (Fig. 8). This spatial pattern is consistent with previous years and is due to differences in both patterns of species abundance and composition (NUSCO 1987). During pre-operational and operational years, similarity among the WP (potentially im- pacted) and GN (reference) communities (Group I) was higher with all years grouping at 45%. This group further divided into three subgroups: Group A contained most GN years; Group B most WP years; and Group C, the 1986 GN and WP years. Within the WP cluster (Group A), the 1987 collection linked with 1982-84, and separated from other years within the group due to similarly low numbers of Scolecolepides viridis, Polydora ligni and Capitella spp.. Within the GN cluster, 1987 chained onto collections made in 1984 and 1985; these shared similarly high numbers of oligochaetcs and low numbers of rhynchocoels. Group C contained 1986 GN and WP collections Benthic Infauna (t+XOOl) 3y00 y3d AilSN3a NV3W (i+xooi) 3yoo a3d AiisN3a mifi LiJ Z t. < i^ (5 o t^ y o o ^ II « J^ a: ^ §•,0 ■4 w o o •2 II "5 s-'- en :q oi. (l+x 6o-i) 3aoo y3d AiisNsa hjvan (L+X 6o-i) 3yoa y3d AllSN3a NV3N 82 TABLE 3. Annual mean species diversity (H'), evenness (J), species number (S) and total individuals (N) (± 1 standard error) for Millstone intertidal stations September 1979 - June 1987. STATION 1980 1981 1982 1983 1984 1985 1986 1987 GIANTS NECK H' 1.9 ±0.3 1.7 ±0.2 1.5 ±0.4 1.3 ±0.2 2.2±0.5 1.7 ±0.4 1.1 ±0.7 1.9 -'-0.3 J 0.7±0.1 0.6 ±0.1 0.4 ±0.1 0.5±0.1 0.6±0.1 0.5±0.1 0.4 ±0.2 0.5±0.1 S 9±3 10±1 13±1 6±1 13±3 9±2 7±2 14±2 N 322 ±202 202 ± 73 368±61 162±87 220±112 149 ±54 261 ±168 351 ±104 JORDAN COVE H' 1.5±0.5 1.7±0.5 1.6 ±0.5 1.7 ±0.5 1.5 ±0.4 2.3±0.1 1.8 ±0.2 2.0 ±0.3 J 0.4 ±0.1 0.4 ±0.1 0.5 ±0.2 0.4 ±0.1 0.5±0.1 0.5±0.I 0.4 ±0.04 0.5 ±0.05 S 12±2 15±5 9±2 17±4 12±4 21±3 17±3 14±3 N 267 ±127 421 ±160 309 ±230 558 ±162 581 ±249 580 ±237 917±385 4,34 ±279 WHITE POIN I H' 2.2 ±0.1 2.1 ±0.2 1.2 ±0.2 1.5 ±0.1 1.4 ±0.2 1.7 ±0.4 1.3 ±0.2 1.3 ±0.4 J 0.6 ±0.1 0.7 ±0.1 0.4±0.1 0.6 ±0.1 0.4 ±0.1 0.5±0.1 0.6±0.1 0.5 ±0.1 S 12±2 11±3 10±2 , 7±2 10±1 11±2 6±1 7±2 N 193 ±53 27S±101 337 ±148 97 ±39 181 ±63 182±58 62±22 100 ±37 -30- -20- -10- ^ °- ^ 10- 2 20- -30- Lj 40 - 50- A 1 60- 1 ^ 70- 80- 90- 1 ± -1 *>W*^ *$.i> *^ */ *o * */ *o * Fig. 8. Dendrogram resulting from the classification of annual intertidal collections at Millstone intertidal stations September 1979 - June 1987. Benthic Infauna 83 which shared high densities of Scolecolepides viridis and low densities of Paraonis fu/gens, oligochaetes and Capitella spp. In addition, low abundances of Paraonis fulgens at GN in 1986 were more comparable to those collected typically at WP in past years (NUSCO 1987). The J" group (Group II) was subdivided into two smaller groups with 1982 and 1984 collections chaining onto these groups at lower similarity. Collections in the first group, which included 1987, shared relatively low numbers of Capitella and Scolecolepides viridis and higher numbers of oligochaetes than the second group (1986, 1985 and 1983 collections). The separation of 1984 and 1982 from other sampling years was due to low oligochaete abundances. Discussion Intertidal communities sampled during start-up and through the first year of 3-unit operation exhibited spatial patterns in community abun- dance and composition and seasonal fluctuations in abundance that were consistent with those ob- served during the baseline period (NUSCO 1987). Higher infaunal abundances and number of species continued to occur at JC, the most sheltered of our sampling stations. Communities at WP, the most exposed station, generally included lower numbers of individuals and species. A similar pattern of increasing density and number of spe- cies along gradients of decreasing exposure to wind and wave-induced scour have been reported in many studies (Holland and Dean 1977; Withers and Thorpe 1978; Maurer and Aprill 1979; Tourtellotte and Dauer 1983). At the population level, the .IC community continued to be domi- nated by oligochaetes and other surface deposit- feeding forms whose abundance can be enhanced by the presence of algal and eelgrass detritus (Soulsby et al. 1982; Hull 1987). In contrast, mo- bile carnivores and large, burrowing deposit- feeding feeding organisms {Haploscoloplos fragilis, Paraonis fulgens, rhynchocoels, and SlreptosylUs arenae) which are more typical of exposed habitats (Dexter 1969; Maurer and Aprill 1979; Tourtellotte and Dauer 1983), were consistent dominants at WP and GN. Spatial distributional patterns and seasonal fluctuations evident in Mill- stone intertidal communities are typical of tem- perate intertidal beaches, where physical factors such as wind and wave-induced beach scour or the resulting habitat characteristics (e.g., sediment size, silt/clay content, sediment stability and po- rosity) strongly influence the structure of macrofaunal communities (Green 1969; Holland and Polgar 1976; Croker 1977). Although spatial patterns were consistent be- tween pre-operational and operational periods, there were temporal changes in community abun- dance and species composition evident at all sam- pling stations. Many of these differences appeared to be continuations of trends which were first evident before 3-unit commenced. For instance, increased macrofaunal abundance was observed at JC since 1985, before 3-unit operation started as the result of higher densities of principally oligochaetes. At WP, lower annual community abundance in 1986 continued a trend first evident before 3-unit operation and reflected unusually low numbers of both oligochaetes and polychaetes. At both GN and WP, lower abun- dances during the 1986-87 were observed in oligochaetes, Paraonis fulgens, rhynchocoels and Haploscoloplos fragilis. These reductions were evident in December 1985 and March 1986, before 3-unit operation and were evident at our reference station, suggesting that the declines were a re- sponse to large-scale regional events. At .IC, re- ductions in total abundance and species number in December 1985 and March 1986 were also evident, although they were not as extensive as those at GN and WP. There were, however, un- usually low densities of Scolecolepides viridis and Ifediste diversicolor in .lune 1987. Low densities of Scolecolepides viridis were also evident at GN in June 1987, further indicating a possible area- wide decrease, independent of power plant oper- ations. During the operational period, several events occurred which might have induced the temporal changes in intertidal community structure and 84 abundance. First, at JC, and particularly at WP, there was an mcrease in sediment grain size start- ing in June 1985, and continuing through March 1987. Since no corresponding increase occurred at GN, this change was not attributable to a re- gional climatic event. Neither was it likely to be power plant-related because the change was evi- dent before operation of Unit 3. Increased grain size at WP was accompanied by a reduction in oligochaetes, although few other taxa seemed af- fected. Lower oligochaete abundances in 1986 may have been a response to reduced food avail- ability given the increased grain size. At .IC, al- though sediment size increased, levels of silt/clay content were witliin the historical range and no changes in macrofaunal communities could be attributed to the observed changes in sediment characteristics. The second event that occurred during this study was the passage of Hurricane Gloria (Sep- tember 1985). This storm passed within 72 km of the Millstone sampling area, producing winds nearing 161 km/h. Following the storm, lower abundances and species number were observed (December 1985 and March 1986) at all monitor- ing stations. This contrasts however, with other studies reporting surprisingly minor effects on interidal communities after hurricanes (Croker 1968; Saloman and Naughton 1977). Studies in the Millstone area (Dobbs and Vozarik 1983) identified only minor rearrangements in commu- nity dominance patterns in the shallow subtidal area adjacent to our JC intertidal station associated with the passage of Hurricane David. Conclusions Sampling of intertidal communities during 1986-87 revealed no immediate changes to these communities which could be directly attributed to Unit 3 start-up or to 3-unit operations at MNPS. Spatial distribution and abundance pat- terns of dominant species at the intertidal stations were consistent with those observed during base- line studies. Major temporal shifts observed over 1986-87 appeared to be mediated by natural shifts in grain size and passage of Hurricane Gloria. Given the limited 3-unit operational history, ad- ditional data will be needed before subtle impacts (like those due to plant-related temperature in- creases) can be assessed. Benthic Infauna 85 Subtidal Results Sedimentary Environment Sediments at subtidal stations from September 1985 - June 1987 were comprised of very fme to coarse sands which ranged in size from 0.07 - 0.55 mm ana contained from < 1 - 44% silt/clay (Fig. 9). During this period, sediments were generally coarsest at EF, where medium sands (0.31 - 0.50 mm in size) of low silt/clay (0.7%- 5. 17%) content predominated. At tliis station, there was a trend for increasing grain size and decreased sUt/clay content beginning in September 1985 and contin- uing through .lune 1987. Grain size at GN ranged from 0.21 - 0.55 mm and although fme-to-medium sands predominated at this station, coarse sediments were obtained in September 1985. The seasonal trend of coarse sediments during September also occurred in 1983 and 1984, but not during 1986 or 1987. SiU/clay at this station ranged from 10 to 20% with no consistent seasonal periodicity. During 1986-87, grain size and silt/clay content at this station were consistent with past results (NUSCO 1987). Sediments at IN ranged from 0.07 - 0.21 mm (very fme - fme samd) since September 1985, and were within the range of those collected in previous years (NUSCO 1987). Silt/clay content at this station remained high, relative to other stations, and ranged from 5.8 - 44.5%. Higher and more variable values were obtained during 1986 than in 1987. Temporal fluctuations in sedimentary character- istics also occurred at JC during the 1986 - 1987 sampling period. Quarterly mean grain size ranged from 0.10 mm - 0.52 mm (very fme to coarse sands) and silt/clay from 3.9 - 38%. From Sep- tember 1986 through .lune 1987, values for grain size at this station were lower than all previous observations. Silt/clay values during 1986 were consistent with those observed in past years (NUSCO 1987). During 1987 sampling, levels of silt/clay within JC sediments generally exceeded those obtained in previous years. During the operational period, changes in sed- imentary characteristics at all stations, except GN, were apparently related to start-up of Millstone Unit 3. At EF, the increased grain size and de- creased silt/clay content was attributable to scour- ing produced by the 3-unit discharge. At JC. un- usually high silt/clay content was detected in Sep- tember 1986, although divers reported reduced visibility and unusually turbid conditions at this sampling station as early as June. Reduced silt/ clay at IN also corresponded to power plant start- up. There was no comparable shift at the GN reference station, which might indicate that a nat- ural event was responsible for the changes which occurred at other stations. General Community Composition The general composition of subtidal communi- ties sampled during 1986 - 1987 is presented in Table 4. PolychaeteS were most abundant, and during 1986 and 1987 accounted for 66 - 56% of the total individuals collected at GN and JC, re- spectively. Oligochaetes were the second most abundant group at these stations in 1986. In 1987, total arthropod abundance doubled at GN and increased by an order of magnitude at JC. Oligochaetes dominated at EF (46% and 50%) followed by polychaetes (36% and 28%). The large increase in arthropods observed at GN and JC during 1987 was not evident at EF where total numbers collected were near the lower end of the range in both 1986 and 1987. Arthropods were the major component of IN communities during both 1986 and 1987, accounting for 54% and 67% of the total individuals, respectively. Polychaetes were the second most abundant group, followed by molluscs. Polychaetes were the most diverse group and accounted for the majority of infaunal species 86 / S } s { »<:i^^^ < J), N t • t; < (^ "■"' ^ (j<) AVio/nis (Hw) 3ZIS Nivyo W AVio/nis (Hw) 3ZIS Nwya ,^^ ^ } < \ '> 1 > < \, /^ ^ \ \ ^ ■> V > s \ < 3 U- Ll. ' / r } g \ ^,==:ZZ!Z^^ S ^ '\ c^ (x) A\no/nis (WH) 32is Nivyo (x) Avio/nis (hh) 3Z1S Nivyo Benthic Infauna 87 TABLE 4. Number of species (S), number of individuals (N) and percentage of the total (%) for each major taxon collected at Millstone subtidal stations September 1985 - June 1987, with ranges from 1980-85. STATION Range 1980-1985 1986 1987 S N % S N ' % s N % EFFLUENT Polychaeta 62-75 2464-7849 39-73 57 2213 36 58 2079 28 Oligochaeta 1470^496 18^7 2848 46 3815 50 Mollusca 25-34 140-934 2-8 30 429 7 28 638 8 Arlhropoda 32^5 318-1412 4-12 26 558 9 36 580 8 Rhynchocoela - 47-245 1-2 92 1 195 3 ■Others' 0-5 0-35 0-1 5 75 1 6 240 3 GIANTS NECK Polychaeta 57-74 4262-9527 58-79 75 6590 66 69 8946 66 Oligochaeta - 962-2658 13-30 1896 19 2116 16 Mollusca 12-27 49-443 1-5 30 332 3 29 215 2 Arlhropoda 31-42 535-761 5-12 45 1029 10 42 2154 16 Rhynchocoela 28-107 o^) aaoo tod AusNsa nv3h 90 \ ' \ \ i • f ' :^ ^ t UJ i § 1 5 1 z I / 1 / / / \ \ \ — *-< • / / / / "^ V CO \ 1 ■ I ' z -I-, 1 / &5 / ( i+xoon) 3aoo aad xusNaa win (1+X301) 3y00 a3d AilSMBQ NV3H (i+x 001) 3yoo a3d AiJSN3a nvsw (l+X 60-1) 3yOD a3d UiSN3a NV3H Benthic Infauna 91 S S 8 3^00 a3d S3l03dS NV3M 3^3 3MO0 a3d S3l03dS NV3N S S S 3^03 U3d S3l33dS NV3n ■ ^ 5 i?-S V- E 2 £ c E 'o ■— -n 1) K " O C K ^ I ? Sol 3 O CB f - T3 O C -o c (2 " ^ "^ -g ^2 •= a •£ "5 *- ^ m 4J H S '^ll 3il03 ^d S3l33dS NV3r< 92 - X \ r 1 — 1 - / _ / / ^ / i / / \ 1 \ , \ „. o / t; 3a00 a3d S3IC3dS NV3W 3M00 H3d S3l03dS NV3N 3d03 »3d S3l33dS NV3n 3a00 H3d S3l03dS NV3n Benthic Infauna 93 TABLE 5. Mean number of individuals per core and Biological Index Value (BIV) of the ten most numerically abundant taxa collected at the Millstone subtidal stations September 1985 - June 1987, with annual range and mean (no. /core) for 1980-1985. 1980-85 1986 1987 1980-85 1986-87 Station Range EFFLUtiNT Oligochaeta Mediomastus ambiseta Protodorvillea gaspeensis Tharyx dorsobranchlalis Ampelisca verrilli Rhynchocoela Tellina agilis Owenia fusiformis Mytilus edulis Archiannelida Spiophanes bomhyx Ampelisca vadorum Polycirms eximius Aricidea catherinae Tharyx acutus Exogone hebes Capitella spp. Eumida sanguinea Caulleriella spp. Lumbrmetis tenuis Polydora caulleryi Pagurus acadianus Microphthalmus aberrans Leptocheirus pinguis Prionospio steenstrupi Gammants lawrencianus GIANTS NECK Mediomastus ambiseta Oligochaeta Aricidea catherinae Tharyx dorsobranchialis Protodorvillea gaspeensis Lumbrineris tenuis Leptocheirus pinguis Exogone dispar Ampelisca abdita Ampelisca vadorum Spio setosa Unciola irrorata Tharyx acutus Polycirrus eximius Phoxocephalus holbolli Polydora caulleryi Prionospio steenstrupi Capitella spp. Gammarus lawrencianus Eumida sanguinea Lumbrineris impatiens Polydora quadrilobala 37 - 112 72 71 95 99.0 100.0 <1 - 26 7 14 11 61.5 92.3 2 - 15 8 8 9 86.8 80.7 <1 - 11 4 9 2 67.3 57.7 <1 - 5 1 7 3 46.6 57.7 1 - 6 3 2 5 69.2 53.8 1 - 12 6 6 3 71.4 50.0 0 - <1 <1 I 6 - 46.1 0 -1 < 1 <1 6 - 42.3 <1 - 3 2 <1 6 42.3 0 - < 1 < 1 <1 4 34.6 <1 - 4 1 2 1 26.9 8 - 100 31 2 < 1 82.2 23.1 2 - 21 8 <1 < 1 62.3 2 - 76 21 0 0 54.6 - 2 - 4 3 1 <1 53.6 - <1 - 4 2 2 1 52.6 - <1 - 7 3 <1 <1 49.5 <1 - 3 2 2 2 48.1 - <1 - 6 2 1 1 46.2 - 0 - 7 3 I 1 44.0 - <1 - 6 2 1 2 41.3 - <1 - 2 1 1 2 40.6 - <1 - 5 2 1 2 39.4 . <1 - 3 1 <1 2 36.8 - <1 -4 1 <1 <1 25.7 - 2 - 73 19 35 77 60.2 91.7 24 - 66 48 47 53 94.4 87.5 27 - 88 47 24 59 94.4 83.3 20 - 48 31 58 32 90.7 83.3 4 - 11 7 8 11 68.5 58.3 2 - 8 5 8 7 53.2 45.8 <1 - 4 1 1 37 17.6 45.8 1 - 12 5 S 8 48.6 41.7 <1 - 2 1 1 9 - 41.7 <1 - 1 <1 4 4 25.0 0 - <1 < 1 6 <1 - 25.0 0 - < 1 <1 0 4 . 20.8 4 - 37 15 0 0 75.9 1 - 13 7 3 1 58.3 - 2 - 9 5 3 2 51.4 - < 1 - 10 4 2 2 42.6 - 1 - 6 3 2 2 35.2 - 1 - 4 2 1 3 34.7 - 0 - 6 2 <1 2 33.8 - <1 - 3 2 1 1 32.4 - 0 - 8 2 0 0 32.4 - <1 - 6 2 < 1 2 25.5 - = not among the ten most numerically abundant taxa 94 TABLE 5. cont'd 1980-85 1986 1987 1980-85 1986-87 Station Range INTAKE Mediomastus ambisela Ampelisca abdita Ampelisca verrili Ampelisca vadorum Leptocheirus pinguis Owenia fusiformis Oligochaeta Nucula proximo Tellina agilis Unciola irrorata Lacuna vincta Tharyx dorsobranchalis Pagurus longicarpus Prionospio sleensturpi Aricidea catherinae Capitella spp. Exogone hebes Tharyx aaitus Spiophanes bombyx Vnicola serrata Protodorvillea gaspeensis Gammarus lawrencianus Clymenella torquata Polydora quadriiobata Pygospio elegans Polydora ligni Phyllodoce mucosa Sabellaria vulgaris Crangon septemspinosus JORDAN COVE Mediomastus ambiseta Oligochaeta Aricidea catherinae Lumbrineris tenuis Tharyx dorsobranchialis Leptocheirus pinguis Polycirrus eximius Microphthalmus aberrans Prionospio steenstrupi Tellina agilis Polydora caulleryi Capitella spp. Pholoe minuta Mltrella tunata Spio setosa Tharyx acutus Rhynchocoela Eumida sanguinea Lumbrineris impatiens Parapionosyliis longicirrata Exogone hebes Gammarus lawrencianus <1 - 10 3 34 31 60.1 92.9 <1 - 25 5 54 18 49.0 89.3 <1 - 11 4 8 12 70.5 75.0 <1 - 8 2. 11 11 45.5 71.4 0 - 17 4 1 125 50.0 62.5 0 - <1 < 1 1 22 - 55.4 2 - 9 6 4 4 93.1 53.6 <1 - 4 1 6 4 52.1 53.6 1 - 5 2 4 4 67.0 42.9 0 - 1 <1 0 12 - 37.5 < 1 - <1 <1 3 2 - 32.1 < 1 - <1 <1 3 3 . 32.1 0 - I < 1 0 10 30.4 <1 - 3 1 1 <1 49.0 21.4 <1 - 7 4 1 3 79.2 - <1 - 6 2 1 2 67.4 - <1 - 4 2 <1 1 64.9 - <1 - 5 2 0 0 62.8 - <1 - 4 1 <1 2 57.6 - <1 -4 1 1 <1 57.3 . <1 - 2 1 <1 <1 46.2 - <1 - 4 1 <1 <1 45.4 - 0 - 5 1 0 1 41.7 - <1 - 2 1 < 1 1 39.9 - 0 - 2 1 0 0 38.9 - 0 - 9 2 <1 <1 36.1 - 0 - 1 < 1 <1 <1 29.5 - 0 - 3 10/core) at IN in 1987 were: the tube building polychaete, Owenia fusiformis the amphipod, Uniciola inorala and the hermit crab, Pagurus longicarpus. When compared to 1980-85, there were changes in the dominance structure over 1986-87 relative to 1980-85; most of these were first apparent in 1987. A regional change occurred in the abun- dance of arthropods, particularly Leptocheirus pinguis. Abundance of this species, as well as other arthropods {Uniciola irrorata, Ampelisca abdita and A. vadorum), at GN, JC and IN during 1987 were well above the 5-year average. A sec- ond area-wide change in dominance occurred in densities of Mediomastus ambiseta. This species had a high BIV ( > 90%) during 1986-87 at all stations compared to the range for 1980-85 (60.1-77.6%). In 1986-87, 3 taxa at EF, 2 at GN and IN and 1 at .IC were among the top ten for the first time. All newly ranked species were generally abundant in only one of the last two years and had low 1986-87 BIV's. The number of species included among the dominants in 1980-85, but not during 1986-87 were greater, although the majority of these had low BIV's for 1980-85 or were collected in low abundances. Those species with high 5-year BIV's ( > 70%) or high mean abundance ( > 10/core) and not among the top ten in 1986-87 were: Tharyx acutiv; and Poly cirrus eximius at GN and EF and Aricidea catherinae at IN. Polycirrus eximius was aslo found in lower density at .IC in the last two years, although this species remained among the top ten. Changes in abun- dance of dominant taxa will be described in more detail in the following section. Dominant Taxa Temporal patterns in abundance of selected subtidal taxa were examined using the same mul- tiple regression analysis procedure used to remove natural sources of variation in abundance and number of species. Taxa selected ranked among the top four numerical dominants (BIV's > 80%) during either the 1980-85 pre-operational period, or the 1986-87 operational period. In some cases, taxa were included to allow comparison at all stations, even though the BIV may not exceed 80% (e.g., Aricidea catherinae at EF). Quarterly and annual means plotted in the figures are log- transformed; those described in the text are the exponentials of plotted values. Oligochaetes These annelids occupy a variety of habitats in the Millstone area, and feed on the fine deposits incorporated into the sediments. Oligochaete den- sities can increase both quickly (Gierc 1975) and markedly (Price and Ilylleberg 1982) as the amount of detritus incorporated in sediments in- creases. Oligochaetes were a dominant taxon at 3 of the 4 subtidal stations; at EF this group ranked 1st in both 1986 and 1987; 3rd at GN and 2nd at JC in terms of the 2-year BIV. Oligochaetes ranked 7th at IN over 1986-87. Throughout the operational period, oligochaete abundance ranged from 35 - 56/core at GN, 23 - 123/core at JC, 23 - 127/core at EF and from 1 - 6/core at IN (Fig. 12A-D). Oligochaete abun- dance at GN, EF and IN was within the range of previous years while at JC, densities from March 1986 - June 1987 were at their lowest level since June 1981. 96 < \ . / / - a it 1 Ld 3 CO 5« q: (l+X 6o1) 3H00 a3d AilSSQQ NV3W (l+X &01) 3800 a3d AilSNSa NV3M 100 \ • \ o "\ \ 4- ^ a / / 51 , I , 1/) s i| \ ^ g ^ 5 1 / ti :5 ► / \ . \ IS \ /"^ - -<^ F m =3 \ STATION: JORDAN COV Uediomastus ambist , \ / / (1+X301) 3^00 y3d AJJSN3a NV3I^ (L+XOOl) 3^00 a3d AUSN3a NV3H 30T) 3800 a3d XJJSN3a NV3W (L+X 601) 3aO0 a3d AlJSN3a NV3W Benthic Infauna 101 (1+X901) 3^00 y3d AllSN3a NV31^ (1+X301) 3a03 a3d AilSN3a NV3n (l+X 60-1) 3ao3 y3d AlISNaa Nvsn (l+X 6o-l) 3aoO y3d AilSN3a NV3W 102 (1+X301) 3^00 y3d JUJShGa NV3I1 ( H-X301) 3yoo y3d AUSNaa nv3i«i (i+x son) 3aoo M3d Aushaa NV3H (L+X Boi) 3aOO a3d AUShGa NV3M Benthic Infauna 103 annual mean was significantly higher than 1981, 1983, 1984 and 1986. Polycirrus eximius Polycirrus eximius is a deposit-feeding polychaete common in shallow subtidal marine and brackish temperate zone waters along the east coast (Gosner 1971). Polycirrus eximius was among the ten most abundant taxa at EF and .IC during 1986-87. Quarterly values at EF ranged from 1 - 7/core and from 2 - 13/core at .IC (Fig. 12 M-N) from September 1985 - June 1987. Since December 1985, densities at EF have been well below those obtained in prior sampling periods. Densities at JC were within the range of previous years, although those recorded in September and December 1986 and March 1987 were among the lowest since 1980. Regression analysis, which removed 68% (EF) and 36% (.IC) of the variation over years indicated that decreases in abundances have occurred since 1984 at both stations. At EF, abundances for 1986 and 1987 were significantly lower than 1980, 1981, and 1983-1985. Abundances at JC for 1986 and 1987 were significant lower than 1984, when peak densities occurred. Protodorvillea gaspeensis This small motile polychaete is considered a facultative carnivore (Fauchald and Jumars 1979) and common in near-shore sublittoral environ- ments from the Gulf of St. I^wrence to LIS (Pettibone 1963). Protodorvillea gaspeensis was a dominant member of the EF community during 1986 and 1987, ranking 3rd in terms of the BIV and 3rd or 4th in terms of average density. 'Phis species was also among the dominants at this station from 1980-85, although large year-to-year shifts in abundance occurred. At GN, this species was among the top ten numerical dominants since 1980; however, densities were consistently below 10/core in all but the 1987 sampling period (Fig. 12 O-P). From September 1985 - June 1987, average quarterly abundance of Protodorvillea gaspeensis ranged from 3 - 1 3/core at both stations. All values obtained since September 1985 were within the range of those obtained in past sampling periods. Multiple regression analysis removed 53% and 46% of the variation at EF and GN, respectively, and revealed that significant year-to-year varia- tions occurred at both stations. At GN there has been a significant increasing trend in the abun- dances of this species since 1980. In addition, densities in 1987 were significantly higher than those obtained from 1980-83. An increasing trend was not evident at EF; where significant differences occurred between the 1980 low and all other years. In addition, 1986 was significantly lower than 1981. Lumbrineris tenuis Lumbrineris tenuis is a burrowing deposit-feeding omnivore which can consume a variety of food items ranging from sediments, algae and eelgrass detritus to other infauna (Pettibone 1963). Com- mon in subtidal areas from Maine to the Gulf of Mexico, this species is found in muds, sands and eelgrass beds to depths of 128 fathoms (Pettibone 1963). This species was a consistent dominant at JC during 1986-87 and also during 1980-85. Al- though this species was less abundant and exhib- ited large temporal fluctuations at GN, analysis was performed to provide comparison to results obtained at JC. Quarterly densities during the 1986 and 1987 sampling periods ranged from 3 - 18/core at JC and from 4 - 14/core at GN (Fig. 12 Q-R). Values at both stations were within the range of previous years, although March and June 1986 means were among the lowest recorded at this station. Multiple regression analysis indicated that there were no long-term trends at cither station; how- ever, significant interannual differences occurred. At JC, the 1987 mean was significantly higher than 1983; also 1986 was significantly lower than 1984. Annual abundances were less variable at GN and 1986 and 1987 were significant from only 1980 and 1981. 104 ► /• ' ^ / s / . / ■ 1 / ^ / - ^ \ a \ F \ 3 m 1- 1 H- LU 3 1 / t UJ ^ / z f o ■J? E ti ^ r^ (D m * K, CM ^ o — 1 s 1 ' / ' I" / / ^7^ ^ / s \ ■ — ^ — 1 ■=> \ en \ \ > o 5 3 . \ . / Q O a / z g o ?A / EH ^ (l+XSOl) 3^00 y3d AUSNQQ NV3H ( i+xoon) 3yo3 y3d Ajjshaa nv3w (l+X 6oi) 3aoO H3d AUShGQ NV3W (l+X 6oi) 3ao3 a3d AUSN3a NV3H Benthic Infauna 105 (i+x3on) 3M0O a3d xiJSN3a Nvan ( 1.+X001) 3aoo asd Aiishoa nv3h (l+x 6on) 3aoo a3d axisngq man (l+x 6o-|) 3aoa a3d alisnsq nv3w 106 (l+XSOl) 3y00 MHd AJJShQO NV3H ( l+XOOl) 3aO0 a3d ^MSH3a NV31«( ( L+x 6oi) 3aoo y3d Ajjshaa win (l+X 601) 3ij00 d3d AiJSN3a NV3W Benthic Infauna 107 Atnpelisca spp. These suspension-feeding amphipods inhabit sediments from fme sands to muds and are com- mon to shallow subtidal communities from the Gulf of St. Lawrence to the Gulf of Mexico. Atnpelisca spp. (including Ampelisca ahdita, A. vadorwn, A . verrilli) are common members of topographic depressions of the Middle Atlantic Continental Shelf, and do not enhance materials exchange because they live at the sediment surface and actually tend to reduce sediment resuspension and hence geochemical exchange (Boesch 1979). Ampelisca spp. were dominant components of the IN community during 1986-87, and were also present in previous years, although densities were typically low ( < 10/core) and large year-to-year fluctuations in density were common (low BIV's). In 1986 and 1987, quarterly densities of this taxon ranged from 2 - 228/core (Fig. 12 S). Except for March 1986, abundances over the last two years were consistently higher than those obtained from March 1979 - June 1984 and of similar magnitude to those obtained from September 1984 through June 1985. Species diversity values during 1986 and 1987 generally fell within the range established by pre- vious studies (NLISCO 1987). In 1986 and 1987, diversity was generally lower than 1985, a year during which relatively higher species number and evenness occurred. At three of the monitoring stations (GN, JC, and IN), lower H' and J over the last two years were attributable to the large increase in the abundance of Mediomastus ambiseta and Leptochcirus pinguis. Diversity at EF was within the range observed in previous years, but density and number of species in 1986 and 1987 were low compared to 1984 and 1985. Cluster Analysis Cluster analysis of annual subtidal species abun- dances (Fig. 13) showed two major station/groups: Group I included all Intake collections and Group II contained all remaining samples. The low sim- ilarity of IN to other sampling stations is due to large differences in species composition and abun- dance. For example, the infaunal communities collected in 1986 and 1987 at IN included high numbers of ampelisciid amphipods, species which are present, but are not usually among the nu- merical dominants at other stations. Over the eight year sampling period, a significant increasing trend has occurred in the abundance of these species principally due to the large in- creases observed in each of the last three years. Annual mean abundances in 1986 and 1987 were significantly higher than those obtained from 1980-1984. However, neither year was significantly different from 1985. Species Diversity Mean species diversity (IF) of subtidal commu- nities during 1986 and 1987 ranged from 2.9 - 4..3, evenness (J) from 0.5 - 0.7 and total number of species (S) from 46 - 82 (Table 6). In both sampling years, highest diversity and evenness and lowest number of individuals were collected at EF. Highest species number occurred at GN in both sampling years. The IN grouping was further divided into a group containing 1984-87 collections and one in- cluding 1980-8.3 collections. This separation re- flected a change in the species composition which occurred after 1983. The 1984-87 collections shared similarly high densities of amphipods, molluscs, Mediomastus ambiseta and Owenia fusiformis, and overall lower numbers of other annelids, particularly oligochaetes. Group II included all other station/years which separated into spatial groups (Subgroups A, B, C). The 1986 and 1987 collections at GN, EF and JC formed couplets within each of their re- spective subgroups which then linked to remaining years. The lower similarity between 1986-87 col- lections and those of other years reflected not only interannual shifts in population abundance but changes in dominance structure as well. 108 STATION: INTAKE SUBTIDAL Aynpelisca spp. R ^ = 0.73 UAR79 MAR80 MAR81 MARS2 MAR83 MAR84 MAR85 UAR86 MAR37 MAR88 STATION: INTAKE SUBTIDAL ATTvpelisca. spp. s. / Fig. 12. Continued. The GN collections in 1986 included higher numbers of Mediomastus ambiseta, Tharyx dorsobranchialis, Ampelisca vadorum, Ampelisca verrilli and lower numbers of Tharyx aculus and Polycirrus eximius i\y3si ^TQwxons y&zis. The 1987 collection also included large numbers of Leptocheirus pinguis. At EF, 1986 and 1987 ex- hibited lower similarity to other years due to the overall decline in the abundance of Tharyx aculus, Poly cirrus eximius, Aricidea catherinae and the increase in oligochaetes, Mytilus edulis, rhynchocoels and Owenia fusiformis. In 1986 Benthic Infauna 109 samples at JC, lower numbers of Mediomastus ambiseta, oligochaetes, Aricidea catherinae, Lumbrinerls tenuis were collected relative to pre- vious years, while in 1987, a large increase in Mediomastus ambiseta and Leptocheirus pinguis and continued decreases in oligochaetes, TABLE 6. Annual mean species diversity (IT), evenness (J), species number (S) and total individuals (N) (± 1 standard error) for Millstone subtidal stations September 1979 - June 1987. STATION 1980 1981 1982 1983 1984 1985 1986 1987 EFFLUENT H' 2.7 ±0.2 3.9 ±0.3 4.4 ±0.2 4.6±0.1 3.6 ±0.2 4.8±0.1 4.3 ±0.2 4.3 ±0.2 J 0.4±0.1 0.7±0.1 0.7 ±0.1 0.7±0.1 0.6 ±0.1 0.7±0.1 0.7±0.01 0.7 ±0.01 S 63 ±6 66±13 63±9 75 ±8 84 ±4 86 ±5 60 ±6 68 ±6 N 1583 ±29 1324 ±364 689 ±144 809 ± 88 2333 ±211 1667 ±228 778 ±134 875 ±153 GIANTS NECK H' 3.6±0.1 3.6 ±0.2 3.4 ±0.1 3.4±0.1 3.5 ±0.2 4.2 ±0.1 3.8±0.1 3.3±0.1 J 0.6 ±0.1 0.6±0.1 0.6±0.1 0.6±0.! 0.6 ±0.1 0.6±0.1 0.6 ±0.02 0.5 ±0.02 S 68±4 57 ±10 69±6 53 ±5 73 ±8 82±4 82 ±4 74 ±10 N 2080 ±46 11 77 ±394 1975±456 1230±175 2549±217 1824±28! 1994 ±279 2826±419 INTAKE H' 4.1 ±0.1 3.8±0.1 3.9±0.1 3.4 ±0.3 3.4 ±0.2 3.8 ±0.3 3.0 ±0.4 3.2 ±0.3 J 0.8 ±0.1 0.7±0.1 0.7 ±0.1 0.7±0.I 0.7 ±0.1 0.7±0.1 0.6±0.1 0.5±0.1 S 44±4 37±6 45±3 37±3 30±8 5I±3 46 ±3 60±3 N 389 ±797 301 ±102 474±71 369 ±81 445±335 907 ± 293 1392 ±589 2790 ±92 1 JORDAN COVE H' 3.6 ±0.2 3.7 ±0.4 3.0 ±0.4 3.0 ±0.2 2.6 ±0,2 3.8 ±0.2 3.6±0.1 2.9 ±0.5 J 0.7±0.1 0.6 ±0.1 0.6 ±0.1 0.5±0.1 0.4 ±0.1 0.6 ±0.1 0.6 ±0.01 0.5±0.1 s 66 ±1 66 ±9 44 ±6 55 ±4 67 ±9 72 ±9 68 ±10 70±5 N 1694 ±480 1202 ±449 724 ±145 1477 ± 174 3561 ±523 1949±7n 1489 ±3.54 3455 ±890 Aricidea catherinae, and Poly cirrus eximius oc- curred. Discussion The most immediate environmental impacts as- sociated with Unit 3 start-up were attributable to the increased volume of cooling water discharge and to sedimentary changes caused by the scouring action of the discharge. Infaunal community changes that might have been related to elevated temperatures could not be distinguished from those due to sediment shifts which occurred after start-up. The progression in sediments below the discharge from medium sands to coarse gravels, to cohesive clays (glacial lake bed deposits) and finally to bedrock was similar to that obser\'ed by Saenger et al. (1982) and in Maine following power plant start-up (Maine Yankee Atomic Power Co. 1978). The EF station, located approximately 100 m offshore, was subjected to less severe scour than areas immediately below the discharge; however, plant-induced scour and sedimentary changes (in- creased giain size and decreased silt/clay content) were evident after Unit 3 start-up. Along with 110 ' \ \ \ \ \ \ % % % % % % % % '^p'f','^c•t','t',•1p^'t^^'t'^x<>^^ <>;<>;<>; Fig. 13. Dendrogram resulting Trom the classification of annual subUdal collections at Millstone subtidal stations September 1979 - June 1987. sedimentary changes at EF, we observed reduc- tions in macrofaunal density and species number. The reduced species richness ended the increasing trend evident at this station since 1980. There was also a trend for lowered abundances of Poly cirrus eximius, Aricidea catherinae, oligochaetes, Tellina agilis, Tharyx spp. and Lumbrmeris tenuis. These deposit-feeding species became abundant during 1984 and 1985 following an increase in silt/clay due to construction of the Unit 3 Intake structure and to the completion of the Unit 3 discharge cut (NUSCO 1987). Power plant-related impacts were also evident at JC after 3-unit start-up and were evidenced by both sedimentary and infaunal community changes. Increased sUt/clay content was observed at this station in September 1986, and was prob- ably due to the transport and settling of sediments scoured from area of the Unit 3 discharge cut. Although the Unit 3 discharge cut was completed in August 1984, the bottom area was not subjected to scour until April 1986, when 3-unit operation began. Concurrent with sediment changes at .IC, there was a significant increase in the abundance of Mediomastus ambiseta along with a decline in the abundance of many polychaetes and oligochaetes. Although higher densities of Mediomastus ambiseta, were apparent at other stations during the operational period, increased dominance by Benthic Infauna 111 this species was probably enhanced by the higher sih/clay content, given its affinity for inhabiting silty areas that have been recently disturbed (e.g., Boesch 1982; Sanders et al. 1972). Abundances of other species, including Aric.idea catherinae and Polycirrus eximius, significantly declined following siltation of JC. As mentioned previously, these organisiHs are deposit-feeders and thus their de- cline was probably not food-related. However, since they feed near the sediment surface they might be subject to burial. In addition, high sediment load can cause direct mortality in resident populations (Turk and Risk 1981). Changes in sediment parameters similar to those evident after start-up of Millstone Unit 3 have been observed at other power stations (Dean and Ewart 1978; Saenger et al. 1980) The IN area also exhibited changes in sediments and infaunal communities during the operational period. Silt/clay values during 1987 were generally lower than those in recent years, but similar to values obtained prior to 1984 (NUSCO 1987). The infaunal community in 1987 exhibited signif- icant increases in the abundances of the amphipods, Ampelisca spp., Leptocheinis pinguis and Uniciola irrorata. These organisms were among the dominants at IN prior to the construc- tion activities and are the 'background' compo- nents of the Nucula-Nepthys community found in Niantic Bay, the area adjacent to the IN station (NUSCO 1980). Prior to Unit 3 construction activities, IN was characterized as a dynamic area subjected to strong tidal currents, and the infaunal communities were dominated by suspension-feeding arthropods (NUSCO 1984). Since then, construction impacts (see NUSCO 1987 for summary) have been evi- denced by increased silt/clay content and reduc- tions in infaunal abundance and number of species and the increased dominance of opportunistic deposit-feeding polychaetes such as Polydora ligni, Capilella spp. and Mediomastus amhiseta, which become abundant following stress (Grassle and Grassle 1974; McCall 1977; Swartz et al. 1980; Flint and Younk 1983; Nichols 1985). A reduction in silt/clay content in 1987 may have been a re- sponse to increased currents produced by start-up and continuous operation of Unit 3 circulating water pumps. The return to an amphipod- dominated community at IN is believed indicative of recovery from plant-related impacts which oc- curred in previous years. Power plant-related impacts during the first year of 3-unit operation appeared most related to scouring (at EF and IN) and to the transport and deposition of this sediment Into .IC. Infaunal habitats immediately beneath the Unit 3 cut have been eliminated due to scouring by the 3-unit discharge; however, given the limited size of the area, loss of this habitat would not significantly alter ecosystem functioning in the greater Mill- stone area. At JC, the deposition of sediments apparently from the discharge area would be ex- pected as a temporary change. Once all available sediment is scoured from the EF area, no addi- tional siltation of JC should occur. Infaunal changes related to this siltation should also be temporary and cause no detectable changes in the ecology of the Millstone area. If however, the sedimentary changes are the result of some plant- induced change in the sedimentation patterns in JC or the infaunal shifts were a response to tem- perature, then more widespread and ecologically significant changes in local infaunal communities might occur. These possible impacts can not be addressed until a more comprehensive operational database is established. Conclusions Subtidal infaunal communities exhibited changes in abundance, number of species and population abundances following start-up of MiU- stone Unit 3. Infaunal changes appeared most related to shifts in sedimentary characteristics at- tributed to scouring in the Unit 3 intake and discharge areas and to transport of these sediments to adjacent subtidal areas. Changes which might have been mediated by increased water tempera- ture could not be identified separately from power plant-induced sediment changes. 112 References Cited AUer, R.C. 1978. Experimental studies of changes produced by deposit-feeders on pore water, sediment, and overlying water chemistry. Am. J. Sci. 278:1185-1234. Beukema, J.J. 1979. Biomass and species richness of the macrobenthic animals living on a tidal flat area in the Dutch Wadden Sea: Effects of a severe winter. Neth. J. Sea Res. 13:203-223. Croker, R.A. 1968. Distribution and abundance of some intertidal sand beach amphipods accompanying the passage of two hurricanes. Ches. Sci. 9:157-162. , R.A. 1977. Macrofauna of Northern New England marine sand: Ix)ng-term intertidal structure. Pages 439-450 in B.C. Coull, ed. Ecology of Marine Benthos. University of South Carolina Press, Columbia, South Caro- lina. Boesch, D.F. 1973. Classification and community structure of the Hampton Roads area, Virginia. Mar. Biol. 21:226-244. . 1979. Benthic ecological studies: macrobenthos. Middle Atlantic Outer Conti- nental Shelf Environmental Studies. Virginia Inst. Mar. Sci. Spec. Sci. Rept. 194. 301 pp. . 1982. Ecosystem consequences of altera- tions of benthic community structure and func- tion in the New York Bight Region. Pages 543-568 in G.F. Mayer, ed. Ecological Stress and the New York Bight: Science and Manage- ment. Proceedings of a symposium on the Eco- logical Effects of Environmental Stress, New York, New York, June 10-15, 1979. Estuarine Research Federation, Columbia, South Caro- lina. , R.J. Diaz, and R.W. Vimstein. 1976. Ef- fects of tropical storm Agnes on soft-bottom macrobenthic communities of the James and York estuaries and the lower Chesapeake Bay. Chesapeake Sci. 17:246-259. , and R. Rosenburg. 1982. Response to stress in marine benthic communities. Pages 179-200 in G.W. Barrett and R. Rosenburg. eds. Stress Effects on Natural Ecosystems. John Wiley, New York. Clifford, H.T., and W. Stephenson. 1975. An in- troduction to numerical classification. Academic Press, New York. 229 pp. Dean, D., and J. Ewart. 1978. Benthos. Final Report. Environmental surveillance and studies at the Maine Yankee Nuclear Generating Sta- tion 1969-1977. Section 9:1-37. Dexter, D.M. 1969. Structure of an intertidal sandy-beach community in North Carolina. Estuar. Coast. Mar. Sci. 9:543-558. Dobbs, F.C, and J.M Vozarik 1983. Immediate effects of a storm on coastal infauna. Mar. Ecol. Prog. Ser. 11:273-279. Eagle, R.A. 1975. Natural fluctuations in a soft- bottom benthic community. J. Mar. Biol. Ass. U.K. 55:865-876. Fauchald, K.P., and P.A. .lumars, 1979. The diet of worms: a study of polychaete feeding guilds. Oceanogr. Mar. Biol. Annu. Rev. 17: 193-284. Flint, R.W. 1985. I^ng-term estuarine variability and associated biological response. Estuaries 8:159-169. , and J. A. Younk. 1983. Estuarine benthos: long-term community structure variations. Cor- pus Christi Bay, Texas. Estuaries 6:126-141. Folk, D. 1974. Petrology of Sedimentary Rocks. Ilempshill Publishing Company, Austin, Texas. 182 pp. Giere, O. 1975. Population structure, food rela- tions and ecological role of marine oligochaetes, Benthic Infauna 113 with special reference to meiobenthic species. Mar. Biol. 31:139-156. Goldhaber, M.B., R.C. Aller, .I.K. Cochran, J.K. Roscnfield, C.S. Martens and R.A. Bemer. 1977. Sulfate reduction, diffusion bioturbation Long Island Sound sediments: Report of the FOAM Group. Am. J. Sci. 277:193-237. .Jordan, R.A., and C.E. Sutton. 1985. Oligohaline benthic invertebrate communities at two Chesapeake Bay Power plants. Estuaries 7:192-212. Kinner, P.C., and D. Maurer. 1978. Polychaetous annelids of the Delaware Bay region. Fishery Bull. 76:209-224. Gosner. K.L. 1971. Guide to identification of ma- rine and estuarine invertebrates. Wiley-Interscience. John Wiley and Sons, Inc. 693 pp. Grassle, J.F., and J. P. Grassle. 1974. Opportun- istic life histories and genetic systems in marine benthic polychaetes. J. Mar. Res. 32:253-284. Green, R.II. 1969. Population dynamics and en- vironmental variability. Am. Zool. 9:393-398. Holland, A.F. 1985. Long-term variation of macrobenthos in a mesohaline region of Chesapeake Bay. Estuaries 8:93-113. , and J.M. Dean. 1977. The community biology of intertidal macrofauna inhabiting sandbars in the North Inlet area of South Car- olina, pages 423-438 in B.C. Coull ed. Ecology of Marine Benthos. University of South Car- olina Press, Columbia, South Carolina. , and N.K. Mountford. 1977. Temporal variation in upper bay mesohaline benthic communities. I. The 9-m mud habitat. Chesapeake Sci. 18:370-378. , and T.T. Polgar. 1976. Seasonal changes in the structure of an intertidal community. Mar. Biol. 37:341 -.348. IIuU, S.C. 1987. Macroalgal mats and species abundance: a field experiment. Estuar. Coast, and Shelf Sci. 25:519-532. Hylleberg, J., and K. Maurer. 1980. The central role of bioturbation in sediment mineralization and element recycling. Ophelia (Suppl. 1):1-16. lance, G.N., and W.R. Williams. 1967. A general theory of classificatory sorting strategies, I. Hierarchical systems. Comput. J. 9:373-380. !.£ Mao, P. 1986. Feeding relationships between the infauna and the dominant benthic fish of the Ranee Estuarjf (France). .1. Mar. Biol. Ass. U.K. 66:.391-401. Levinton, J.S., and S. Stewart. 1982. Marine suc- cession: The effect of two deposit-feeding gastropod species on the population growth of Paranais litoralis Mutter 1784 (Oligochaeta). J. Exp. Mar. Biol. Ecol. 59:231-241. Livingston, S. 1982. L/Ong-term variability in coastal systems: background noise and environmental stress. Pages 605-620 in G.F. Mayer, ed. Ecological Stress and the New York Bight: Science and Management. Proceedings of a symposium on the Ecological Effects of Environmental Stress, New York, New York, .lune 10-15, 1979. Estuarine Research Federa- tion, Columbia, South Carolina. Maine Yankee Atomic Power Co. 1978. Final Report. Environmental surveillance and studies at the Maine Yankee Nuclear Generating Sta- tion 1969-1977. MacGinitie, G.E. and N. MacGinitie. 1968. Nat- ural History of Marine Animals. MaGraw-Hill, New York 473 pp. Maurer, D., and G. Aprill. 1979. Intertidal benthic invertebrates and sediment stability at the mouth of Delaware Bay. Int. Rev. ges. Hydrobiol. 64:379-403. 114 Maurer, D., and W. Leathern. 1980. Dominant species of polychaetous annelids of Georges Bank. Mar. Ecol. Prog. Ser. 3:135-144. McCall, P.L. 1977. Community pattern and adap- tive strategies of the infaunal benthos of I^ong Island Sound. J. Mar. Res. 35:221-226. Pettibone, M.H. 1963. Marine polychaete worms of the New England region. I. Aphroditidae through Trochocaetidae. BuU. U.S. Natn. Mus. 227:1-356. Pielou, E.C. 1977. Mathematical Ecology. Wiley- Interscience, New York. 385 pp. McCloskey, L.R. 1970. The dynamics of the com- munity associated with a marine scleractinian coral. Int. Rev. ges Hydrobiol. 55:13-81. Moeller, P., L. Pihl, and R. Rosenberg. 1985. Benthic faunal energy flow and biological interaction in some shallow marine soft bottom habitats. Mar. Ecol. Prog. Ser. 27:109-121. Myers, A.C. 1977. Sediment processing in a ma- rine subtidal sandy bottom community: I. Phys- ical Aspects. J. Mar. Res. 35:609-632. Nichols, F.I I. 1985. Abundance fluctuations among benthic invertebrates in two Pacific estuaries. Estuaries 8: 136-144. NUSCO (Northeast Utilities Service Company). 1980. Infauna. Pages 143-194 in Monitoring the marine environment of Long Island Sound at Millstone Nuclear Power Station, Waterford, Connecticut. Annual Report 1979. . 1984. Benthic Infauna. Pages 1-58 /« Mon- itoring the marine envirorunent of Long Island Sound at Millstone Nuclear Power Station, Waterford, Connecticut. Annual Report 1983. . 1987. Benthic Infauna. Pages 1-59 in Mon- itoring the marine environment of Lx)ng Island Sound at Millstone Nuclear Power Station, Waterford, Cormecticut. Summary of studies prior to Unit 3 operation. Annual Report 1987. Oden, G.E. 1979. The freshwater littoral meiofauna in a South Carolina reservoir receiv- ing thermal effluents. Freshwater. Biol. 9:291-304. Poore, G.C.B., and S. Rainer. 1979. A three-year study of benthos of muddy environments in Port Phillip Bay, Victoria. Estuar. Coast. Mar. Sci. 9:477-497. Price, L.II., and .1. Hylleberg. 1982. Algal-faunal interactions in a mat of &ufun. in False Bay, Washington. Ophelia 21:75-88. Raine, R.C.T., and J.W. Patching. 1980. Aspects of carbon and nitrogen cycling in a shallow marine environment. J. Exp. Mar. Biol. Ecol. 47:127-139. Rcish, D.J. 1973. The use of benthic animals in monitoring the marine envirormient. J. Environ. Plann. PoUut. Control 1:32-38. , D.F. Soule, and .I.D. Soule. 1980. The benthic biological conditions of Los Angeles - Long Beach harbors: Results of 28 years of investigations and monitoring. Helgolander. Meeresunters. 34:193-205. Rhoads, D.C., and D.K. Young. 1970. The in- fluence of deposit-feeding organisms on sedi- ment stability and community trophic structure. J. Mar. Res. 28:150-178. Richards, S.W. 1963. The demersal fish popula- tion of Long Island Sound. Bull. Bingham Oceangr. CoU. 8:1-101. Saenger, P., W. Stephenson, and J. Moverley. 1982. Macrobenthos of a cooling water dis- charge canal of the Gladstone Power Station, Queensland. Aust. J. Mar. Freshw. Res. 33:1083-1095. Benthic Infauna 115 Saloman, C.H., and S. P. Naughton. 1977. Effect of Hurricane Elloise on the benthic fauna of Panama City Beach, Florida, USA. Mar. Biol. 42: 357-363. Sanders, H.L., E.M. Goudsmit, EL. Mills, and G.R. Hampson. 1962. A study of the intertidal fauna of Barnstable Harbor, Massachusetts. Limnol. Oceanogr. 7:63-79. , J.F. Grassle, and G.R. Hampson. 1972. The West Falmouth oil spiU. I. Biology. Woods Hole Oceanogr. Inst. Tech. Rpt. WHOl 72-20. 48 pp. , .IF. Grassle, G.R. Hampson, L.S. Morse, S. Gamer- Price, and C.C. .lones. 1980. Anat- omy of an oil spill: Long-term effects from the grounding of the barge Florida off West Falmouth, Massachusetts. Mar. Res. 38:265-380. Smith, R. 1964. Keys to marine invertebrates of the Woods Hole region. No. 11. Systematics- Ecology Program, Marine Biolog- ical laboratory, Woods Hole, Massachusetts. Soulsby, P.G., D. Lx)wthion, and M. Houston. 1982. Effects of macroalgal mats on the ecology of intertidal mudflats. Mar. Poll. Bull. 13:162-166. Strelzov, V.E. 1979. Polychaete worms of the family Paraonidae Cerruti, 1909 (Polychaeta, Sedentaria). Amerind Publishing Co. Pvt. Ltd., New Delhi. 212 pp. Swartz, R.C., W.A. DeBen, F.A. Cole, and L.C. Bentsen. 1980. Recovery of the macrobenthos at a dredge site in Yaquina Bay, Oregon. Pages 391-408 in R.A. Baker, ed. Contaminants and Sediments, Vol. 2. Arm Arbor Science Pub- lisher, Inc., Ann Arbor, Mich. Taghon, G.L., ARM, Nowell, and P.A. .lumars. 1980. Induction of suspension feeding in spionid polychaetes by high particulate fluxes. Science 210: 562-564. Tourtellotte, G.H., and DM. Dauer. 1983. Macrobenthic communities of the lower Chesapeake Bay. II. Lynnhaven Roads, Lymihaven Bay, Broad Bay and Linkhorn Bay. Int. Rev. ges. Hydrobiol. 68:59-72. Turk, T.R, and M.J. Risk. 1981. Effects of sedi- mentation on infaunal invertebrate populations of Cobequid Bay, Bay of Fundy. Can. J. Fish. Aquat. Sci. 38:642-648. Wells, H.W., and I.E. Gray. 1964. Polychaetous annelids of the Cape Hattcras area. J. Elisha Mitchell Sci. Soc. 67:1.33-161. Whitlatch, R.B. 1974. Food resource partitioning in the deposit-feeding polychaete Peclinaria gouldii Biol. Bull. Mar. Biol. Lab. Woods Hole. 147:227-235. . 1977. Seasonal changes in the community structure of the macrobenthos inhabiting the intertidal sand and mud flats of Barnstable Har- bor, Massachusetts. Biol. Bull. 152:274- 294. Withers, R.G., and C.H. Thorpe. 1978. The macrobenthos inhabiting sandbanks in Langstone Harbour, Hampshire. J. Nat. Hist. 12:445-455. Witman, .I.D. 1985. Refuges, biological distur- bance and rocky subtidal community structure in New England. Ecol. Monogr. 55:421-445. Woodin, S.A. 1982. Browsing: important in ma- rine sedimentary environments? Spionid polychaete examples. .1. Exp. Mar. Biol. Ecol. 60:35-45. Young, M.W., and D.K. Young. 1982. Marine macrobenthos as indicators of environmental stress. Pages 527-539 in G.F. Mayer, ed. Eco- logical Stress and the New York Bight: Science and Management. Proceedings of the sympo- sium; 1979 June 10-15; New York, New York. Estuarine Research Federation, Columbia, S.C. 715 pp. 116 Zeitzschel, B. 1980. Sediment-water interactions TenoreandB.C. Coull,eds. Benthic Dynamics, in nutrient dynamics. Pages 195-218 in K.R. University of South Carolina Press, Columbia, South Carolina. Benthic Infauna 117 Contents Lobster Population Dynamics 121 Introduction 121 Materials and Methods 121 Results and Discussion 123 Abundance and Catch Per Unit Effort 123 Population Characteristics 126 Sb.e Frequencies 126 Sex Ratios 126 Reproductive Characteristics 130 Molting and Growth 133 Claw Loss 135 Tagging Program 136 Movement 136 Entrainmcnt 137 Impingement 139 Summary 140 Conclusion 141 References Cited 142 Lobster Population Dynamics Introduction The American lobster, Homarus americanus, is the most valuable commercial species in Long Island Sound (LIS) (Blake and Smith 1984). Commercial landings of 1.57 and 2.03 million pounds in 1986 and 1987, respectively, were val- ued in excess of 5 million dollars. The proportion of the total catch landed in New London county, which includes the Millstone Point area, was 27% in both years (CT DEP Marine Fishery Statistics for 1986 and 1987). Because lobsters are an im- portant commercial resource in the Millstone Point area, dynamics of the local lobster popula- tion have been studied to determine if construction and operation of the Millstone Nuclear Power Station (MNPS) have caused changes in the local population beyond those expected from natural variation. Potential effects of MNPS operations on the lobster population are entrainment of larvae througli the cooling water systems, impingement of lobsters on the intake traveling screens, thermal effects of the discharge and disruption of lobster habitat caused by routine dredging in the vicinity of the intakes. While mortality due to entrainment and impingement may alter recruitment patterns of lobsters by reducing the survival of lobster larvae and juveniles, the thermal plume may affect the behavior of adults which may result in a de- cline in the local inshore fishery. Dredging may temporarily disrupt lobster habitat (shelters) and thereby displace lobsters from the affected area until sediments stabilize. The lobster studies at MNPS were designed to evaluate year-to-year, seasonal, and between sta- tion changes in catch per unit effort and in pop- ulation characteristics such as size frequency, growth rate, sex ratio, female size at sexual ma- turity, characteristics of egg-bearing females and lobster movements. Lobster larvae studies were conducted to assess potential impacts of entrain- ment on the larval stage of lobsters. The results of these studies are compared to other studies conducted throughout the range of the American lobster. In order to assess potential impacts on the local lobster population associated with 3-unit opera- tion, this report compares data collected during 3-unit operation (1986-87) to data collected before Unit 3 became operational. Materials and Methods A detailed description of methods used to con- duct lobster population studies from 1969 to 1985 can be found in NUSCO (1987a). Since 1978, four pot trawls consisting of five double entry wire pots (76x51x30 cm; 2.5 cm mesh) equally spaced along a 50-75 m line bouyed at both ends were used to collect lobsters from May througli October at three stations (.lordan Cove, Intake, and Twotrce; Fig. 1). Pots were individually num- bered beginning in 1984 to determine the variabil- ity in catch among pots, and provided more ac- curate values for catch-per-pot than an average catch-per-pot based on the 20 pots of each sam- pling location. Pots were checked three times each week, rebaited with flounder carcasses and reset in the same area. lobsters > 55 mm carapace length were banded to restrain chelipeds, brought to the lab, and kept in a tank supplied with a continuous flow of seawater. On Fridays, lobsters caught that week were examined and the following data were recorded: sex, presence of eggs (berried), carapace length (CL), crusher claw position, missing claws and molt stage (Aiken 1973). Lobsters were then tagged with a serially numbered international orange sphyrion tag (Scarratt and Elson 1965; Scarratt 1970), and re- leased at the site of capture. Recaptured tagged LxDbster Population Dynamics 121 Fig. 1. Location of the Millstone Nuclear Power Station (MNPS) and the three lobster sampling stations (•). lobsters, severely injured or newly molted (soft) lobsters, and those < 55 mm CL were released untagged after recording the above data. Beginning in 1981, data were collected to deter- mine the size at which females become sexually mature by measuring the maximum outside width of the second abdominal segment of all females to the nearest millimeter. Female size at sexual maturity was estimated by calculating the ratio of the abdominal width to the carapace length and plotting that ratio against the carapace length (Skud and Perkins 1969; Krouse 1973). Catch per unit effort (CPUE), calculated as the number of lobsters caught per pothaul was aver- aged by computing the arithmetic mean. Because the CPUE data are ratios, they are not additive and have an asymmetric distribution about the arithmetic mean. Therefore, we computed the geometric mean which is the best statistic for con- structing asynmietric confidence intervals for log- normal data (Snedecor and Cochran 1967). Since 1984, in addition to counting the number of lob- sters caught in each pot we counted the number of other organisms caught, to examine the influ- ence of competing species on lobster catch. The amount of time between pothauls (soaktime or sct-over-days) also influences lobster catch, and CPUE data are weighted to account for varying soaktimes. As a result, catch per unit effort was adjusted by covariance analysis for the effect of soaktime and the catch of competing species that significantly affected CPUE. Methods for the collection of lobsters on the intake traveling screens are described in the Fish Ecology section of this report under Methods and Materials- Impingement. In December 1987, based on historical impingement data collected since 1972, NUSCO and the CT DEP agreed to 122 TABLE 1. Catch statistics for lobsters caught in pots from May through October (1978-87). Total Number Number Pots Total Percent Caught Tagged Recaptured Hauled CPUE Recaptured 1978 3578 2768 521 4232 0.85 18.8 1979 5037 3732 722 4086 1.23 19.4 1980 4268 3634 522 4182 1.02 14.4 1981 5110 4246 704 4375 1.17 16.6 1982 9109 7575 1278 4340 2.10 16.9 1983 6376 5160 936 4285 1.49 18.1 1984 7587 5992 1431 4550 1.67 23.9 1985 7014 5609 1235 4467 1.57 22.0 1986 7211 5740 1204 4243 1.70 21.0 1987 7280 5681 1356 4233 1.72 23.9 discontinue impingement monitoring at Unit 2 (NUSCO 1988). Lobster larvae were sampled from 1984 to 1987 during their period of occurrence (May through July) at Units 1 and 2 discharges. Sampling was not conducted at Unit 3 because of design prob- lems with the gantry system used to collect sam- ples at the Unit 3 outfall. Cooling water flows for combined 3-unit operation were used for cal- culating total entrainment estimates for the larval season. Samples were collected with a 1.0 x 6.0 m conical plankton net of 1 .0 mm mesh. Sample volumes were averaged from those calculated from four General Oceanic flowmeters; about 4000 m of cooling water was filtered in each sample by deploying the net for 45-60 minutes. Four day and four night samples were collected weekly (1 day, 1 night on each of 4 days). Each sample was placed in a large 1.0 mm mesh sieve and kept in tanks supplied with a continuous flow of sea- water. Samples were sorted shortly after collection in a white enamel pan and larvae were examined for movement and classified as either live or dead. Lobster larvae were also classified by stage ac- cording to the criteria established by Ilerrick (1911). Results and Discussion Abundance and Catch Per Unit Effort The total numbers of lobster caught during 1986 and 1987 were 7,211 and 7,280, respectively, and both were within the range of values reported since 1978 (Table 1). Total catch per unit effort values for 1986 (1.70) and 1987 (1.72) were also within the range of wire pot values (0.85-2.10). The lower number cauglit and smaller total CPUE from 1978-81 was due to the use of wood pots (NUSCO 1987a). In wood pots, lobsters were more vulnerable to capture at sizes greater than 76 mm and in wire pots at sizes greater than 70 mm (Keser et al. 1983). The performance of wood and wire pots was evaluated (NUSCO 1987a) and provided the basis for using wire pots exclusively in our studies. Figure 2 presents geometric mean CPUE for all sizes and legal-sized lobsters caught from 1978 to 1987. The total catch mean CPUE for 1986 and 1987 (1.585, 1.633) was within the range of values reported from 1978 to 1985 (0.904-2.006). No significant trends were identified for total CPUE (test for significance of slopes; r=0.25, p=0.48). Lobster Population Dynamics 123 1978 1979 1980 1981 1982 ■ 1983 1984 1985 1986 1987 Fig. 2. CPUE geometric means with 95% C.I. for all sizes and only legal size lobsters caught in wire pots from 1978 through 1987. however, the legal CPUE has steadily declined from 0.196 in 1978 to 0.089 in 1987 (r = 0.98, p < 0.05). This decline is most likely due to an increase in fishing pressure in LIS. Over 90% of legal-size lobsters in our catch were recruited from sublegal-size classes. Increased fishing pressure on lobsters in NY and CT waters was recently reported by Colvin (1987) and Smith (1987). The variation in monthly CPUE values during 1986 and 1987 was typical of the seasonal abun- dance of lobsters (Table 2). The increase in CPUE during the spring and early summer months (May- July) was related to the seasonal variation in water temperature, because lobster activity (e.g., feeding, movement, and molting) is at a maximum after water temperatures rise above 10 °C (Mcl^eese and Wilder 1958; Dow 1966, 1969, 1976; Flowers and Saila 1972). A 50% decline in total CPUE at .Jordan Cove from August to September 1986 represented the greatest monthly decline in CPUE at any station since the study began and it was related to scouring at the dis- charge due to 3-unit operation. Beginning in August 1986, a fine sUt material covered traps and lobster shelters in Jordan Cove (500 m from dis- charge). Analysis of sediment samples collected at the Jordan Cove and Effluent subtidal sand stations indicated changes in sediment composi- tion during the same period (see Benthic Infauna section). SCUBA surveys of the discharge area revealed that fine sediment had been scoured dur- ing a period of 3-unit simultaneous operation (May-August 1986). The decline in lobster catches from August (351 lobsters) to September (165 lobsters) was temporary, and catches in- creased in October and continued to be normal through 1987, indicating that sediments in Jordan Cove had stabilized. A similar short-term impact associated with the disruption of lobster habitat 124 TABLE 2. Monthly catch statistics for lobsters caught at each station during 1986 and 1987. 1986 Number of Total number Mean CPUE Total legals Legal Month pots hauled caught Arithmetic Adjusted caught CPUE JORDAN COVE MAY 217 466 2.11 2.09 9 0.04 JUN 260 568 2.19 2.20 32 0.12 JUL 240 501 2.03 2.03 17 0.07 AUG 238 351 1.47 1.48 13 0.06 SEP 220 165 0.75 0.75 11 0.05 OCT 220 218 0.99 INTAKE 0.99 7 0.03 MAY 219 343 1.56 1.79 6 0.03 JUN 260 538 2.05 2.07 29 0.11 JUL 240 448 1.87 1.87 21 0.09 AUG 240 381 1J9 1.52 16 0.07 SEP 220 264 1.20 1.10 18 0.08 OCT 218 239 1.09 TWOTREE l.OI 13 0.06 MAY 220 477 2.16 2.21 24 O.I I JUN 260 594 2.35 2.34 36 0.14 JUL 240 518 2.19 2.19 46 0.19 AUG 240 445 1.84 1.79 33 0.14 SEP 220 309 1.41 1.42 11 0.05 OCT 216 330 1.50 1.51 18 o.os 1987 Number of Tola! number Mean CPUE Total legals Legal Month pots hauled caught Arithmetic Adjusted caught CPUE JORDAN COVE MAY 220 387 1.76 1.76 16 0.07 JUN 257 493 1.89 1.89 20 0.08 JUL 260 458 1.76 1.76 33 0.08 AUG 240 346 1.44 1.44 12 0.05 SEP 220 250 1.14 1.14 10 0.05 OCT 219 262 1.19 INTAKE 1.19 7 0.03 MAY 219 329 IJI 1.58 14 0.06 JUN 260 420 1.58 1.63 30 0.12 JUL 259 425 1.64 1.63 25 0.10 AUG 240 295 1.22 1.14 14 0.06 SEP 220 308 1.40 1.32 17 0.08 OCT 215 228 1.04 TWOTREE 1.10 7 0.03 MAY 215 438 1.99 2.00 17 0.08 JUN 260 662 2.48 2.47 27 0.10 JUL 256 640 2J5 2.54 36 0.14 AUG 240 465 1.94 1.94 29 0.12 SEP 214 439 2.00 2.00 20 0.09 OCT 219 435 1.98 1.98 16 0.07 CPUE value 5 adjusted for the significant eftects of soaktime and the incidental catches of competing specie s presented in Table 3. Lobster Population Dynamics 125 was reported at the Intake station following dredg- ing in 1985 (NUSCO 1987a). Dredging in the vicinity of the intake structures removed existing habitat (shelters), and catches at Intake during 1985 were the lowest reported for that station since wire pots were first used. After the dredged area stabilized, lobsters returned to the area and catches at Intake increased in 1986 and 1987. Pottle and Elner (1982) reported that smothering of habitats occupied by juvenile lobsters may re- sult in increased spatial competition among lob- sters for shelter in remaining gravel habitats. In addition to lobsters, trap catches contained several species of vertebrates and invertebrates. In 1984, we began analyzing the effects of these species on lobster catch using covariance analysis (NUSCO 1985). The species caught in pots at each station were Initially used as covariates to identify species that significantly (p < 0.05) influ- enced lobster catch (Table 3). During 1986 and 1987, the influence of catches of competing species on lobster CPUE was the same as in previous years. Five species were identified as significantly influencing lobster CPUE (spider crabs, hermit crabs, whelks, summer and winter flounder). The whelk or conch {Busycon spp.) was the only spe- cies that had a positive influence on lobster catch. Soaktime was another parameter used in the model as a covariate to examine the significance of varying set-over-days on lobster catch at each station. Soaktime had a significant influence on lobster catch only at Intake; during 1986 and 1987, however, soaktime significantly influenced lobster catch at all stations during 1984. Our soaktimes were not very different; the majority of pots were hauled after a two or three day set. Commercial lobstermen, however, vary the num- ber of days between pot hauls to maximize their catch; therefore, catch statistics for the commercial lobster fishery are often based on catch per trap haul set-over-day to account for the variability in soaktime. The average soaktime for the (Connect- icut fishery has ranged between 3.2 and 4.5 days since 1978, whereas our soaktime averaged be- tween 2.4 and 2.6 days over the same period (CT DEP Marine Fishery Statistics; NUSCO 1987a). The mean CPUE adjusted for the significant covariates (least square means) are presented with its unadjusted arithmetic mean in Table 2 for each station. Given the similarity between the arithmetic and adjusted mean, the incidental catch of competing species and the variability of our soaktimes did not influence the reliability of our CPUE. Population Characteristics Size Frequencies Annual size frequency distributions for male and female lobsters caught in wire pots from 1979 to 1987 are shown in Figure 3. The mean carapace lengths (CL) of lobsters caught during 1986 (70.1mm) and 1987 (70.2mm) were smaller than previous years' range 1978-85 (70. 7-7 1.8mm) (Table 4). When 3 stations were compared, the mean CL's of lobsters caught at .lordan Cove and Intake were within the range of values reported since 1978, however, the mean CL at Twotree during 1986-87 was smaller than previous years (Table 5). From 1978 to 1984, catches at Twotree contained larger lobsters and greater proportions of legal-size lobsters than nearshore catches; since 1985, the mean CL of lobsters and percentage of legals caught at Twotree have been similar to the nearshore stations. The mean size and proportion of legal-size lobsters caught in our studies since 1978 were lower than that reported by other stud- ies in LIS (Smith 1977; Briggs and Mushacke 1979; Marcello et al. 1979). Sex Ratios Sex ratios of males to females were 1.0:0.87 and 1.0:0.88 during 1986 and 1987, respectively, which were within the range of values reported in pre- vious years (Table 6). The Twotree catch had a consistently higher proportion of females than Intake and Jordan Cove catches, which contained more males. This trend in sex ratios has been consistent at the three stations since the study began. Smith (1977) found male to female ratios of commercial catches ranging from 1.0:1.06 to 1.0:1.81 in four different areas of LIS, which is similar to ratios at Twotree 1.5 km offshore. Sex 126 TABLE 3. Total numbers of lobster and incidental catch of other species and the type of influ- ence that competing species had on lobster catches at each station from 1984 to 1987. 1984 1985 1986 1987 Lx)bster Rock crab Jonah crab Spider crab* Hermit crab* Blue crab Winter flounder* Summer flounder* Skates Oyster toadfish Scup* Cunner Tautog Sea raven Wlielks* 7587 7014 7211 7280 391 145 121 37 74 32 37 71 3237 1950 1344 1754 428 496 435 721 40 21 26 44 45 40 19 30 60 24 38 35 15 17 33 14 76 67 58 39 27 90 288 169 141 207 206 167 39 250 196 208 20 19 6 2 66 78 164 132 JORDAN COVE 1984 1985 1986 1987 Spider crab Hermit crab Winter flounder Summer flounder Scup Whelks ( + ) (--) (--) INTAKE 1984 1985 1986 1987 Spider crab Hermit crab Winter flounder Summer flounder Scup Whelks (--) (--) (--) (--) (--) (--) ( + ) TWOTREE 1984 1985 1986 1987 Spider crab Hermit crab Winter flounder Summer flounder Scup Whelks (--) (--) ( + ) (--) ( + ) (--) (*) Species having a significant (p<0.05) effect on CPUE. (--) Significant (p < 0.05) negative or ( + ) positive effect on CPUE. Lobster Population Dynamics 127 y 100- 50 75 50 75 50 75 50 75 50 75 50 75 50 75 50 75 50 75 100 CARAPACE LENGTH (MM) Fig. 3. Size frequency distributions of male and female lobsters caught at all stations from 1979 through 1987. TABLE 4. Lx)bster population carapace-length statistics for wire pot catches from May through October 1978-1987. Range (mm) Mean Carapace Length ± 95% CI Percent Legals 1978 1508 53-111 1979 2846 44-100 1980 2531 40-96 1981 1983 43-96 1982 7835 45-103 1983 5432 40-121 1984 6156 45-107 1985 5723 38-101 1986 5961 36-107 1987 5924 36-99 71.4 ± 0.33 71.2 ± 0.26 70.7 ± 0.27 71.0 ± 0.33 70.8 ± 0.15 71.7 ± 0.19 71.8 ± 0.18 71.3 ± 0.17 70.1 ± 0.17 70.2 ± 0.17 8.7 8.2 7.2 9.6 7.3 10.1 9.6 6.6 5.1 4.4 Recaptures not included 128 TABLE 5. Lobster population carapace-length statistics for wire pot catches at each station from May through October 1978-1987. JORDAN N^ Range Mean Carapace Percent COVE (mm) Length ± 95% CI Legals 1978 489 54-111 70.3 ± 0.54 4.9 1979 1138 46-96 70.7 ± 0.39 7.6 1980 831 40-93 70.2 ± 0.45 5.2 1981 556 45-93 70.6 ± 0.64 7.7 1982 2323 49-96 69.8 ± 0.26 5.1 1983 1965 40-100 71.0 ± 0.32 9.0 1984 1999 52-107 70.7 ± 0.29 6.7 1985 1722 48-96 71.1 ± 0.32 6.4 1986 1748 38-99 69.8 ± 0.31 4.0 1987 1690 44-95 70.2 ± 0.32 4.1 INTAKE 1978 645 55-110 71.8 ± 0.50 10.4 1979 1087 50-100 71.4 ± 0.41 8.3 1980 855 46-95 70.6 ± 0.45 6.2 1981 686 43-95 69.2 ± 0.53 5.0 1982 2402 51-103 70.2 ± 0.27 5.8 1983 1436 52-110 71.2 ± 0.37 7.5 1984 1830 45-105 70.5 ± 0.32 6.7 1985 1215 44-99 71.2 ± 0.37 6.0 1986 1888 50-107 69.3 ± 0.31 4.9 1987 1687 47-94 70.2 i 0.32 5.2 TWOTREE 1978 374 53-94 72.2 ± 0.67 10.7 1979 621 44-94 71.8 ± 0.58 9.3 1980 845 40-96 71.3 ± 0.49 10.1 1981 741 48-96 73.0 ± 0.54 15.3 1982 3110 45-102 72.0 ± 0.25 10.2 1983 2031 43-121 72.8 ± 0.32 12.9 1984 2327 50-105 73.7 ± 0.29 14.4 1985 2786 38-101 71.5 ± 0.25 7.1 1986 2325 36-97 71.0 ± 0.27 6.1 1987 2547 36-99 70.2 ± 0.27 4.1 Recaptures not included Lobster Population Dynamics 129 TABLE 6. Male to female sex ratios of lobsters caught in wire pots from May to October, 1978 to 1987. Jordan Cove Intake Twotree All Stations 1978 1.0 : 0.79 1.0 0.97 1.0 1.02 1.0 : 0.92 1979 1.0 : 0.68 1.0 0.83 1.0 1.15 1.0 : 0.82 1980 1.0 : 0.66 1.0 0.90 1.0 1.15 1.0 : 0.88 1981 1.0 : 0.70 . 1.0 0.71 1.0 1.19 1.0 : 0.86 1982 1.0 : 0.62 1.0 0.66 1.0 1.09 1.0 : 0.79 1983 1.0 : 0.72 1.0 0.67 1.0 1.25 1.0 : 0.87 1984 1.0 : 0.60 1.0 0.71 1.0 1.22 1.0 : 0.82 1985 1.0 : 0.70 1.0 0.67 1.0 1.38 1.0 : 0.97 1986 1.0 : 0.64 1.0 0.73 1.0 1.26 1.0 : 0.87 1987 1.0 : 0.71 1.0 0.63 1.0 1.24 1.0 : 0.88 Recaptures not included ratios close to 1:1 were also reported by other researchers working in waters close to shore (Ilerrick 1911; Templeman 1935a; Ennis 1971, 1974; Stewart 1972; Krouse 1973; Thomas 1973; Cooper et al. 1975; Briggs and Mushacke 1980). Reproductive Characteristics Female size at sexual maturity, development and fullness of egg masses carried by females and the percentage of berried females caught, were compared for preoperational and operational study periods. Female size at sexual maturity was determined by measuring the second abdom- inal segment and calculating the ratio of the ab- dominal width to the carapace length, and plotting that ratio against the carapace length (Templeman 1935b; Skud and Perkins 1969; Krouse 197.3) (Fig. 4). Data for 1986 and 1987, represented as (*) in Figure 4, were compared to mean values collected from 1981 to 1985. Female size at sex- ual maturity was similar under 2- and 3-unit op- erating conditions. Females began to mature be- tween 50 and 55 mm CL and all females were mature at sizes greater than 95 mm CL. The smallest berried females collected in our studies (62 mm CL) were between 54-56 mm CL when oviposition fu^st occurred assuming 14% growth per molt and thus confirms the small size at which females mature in our area. Briggs and Mushacke ( 1 979), working in western LIS, found that females began to mature at about 60 mm CL and most were mature at about 80 mm CL. In contrast to the LIS lobster population, females in northern waters (Maine) mature at a substantially larger size (80 mm CL, Krouse 1973). This was attrib- uted to low water temperatures which retard re- productive maturation, whereas warmer summer water temperatures of LIS favor early maturation of females (Smith 1977; Aiken and Waddy 1980). The sexual maturity of males has been well doc- umented and therefore was not investigated in our studies. Briggs and Mushacke (1979) reported that males in western L.IS first reached maturity (i.e., produced mature spermatozoa), at 40 to 44 mm CL and over half were mature at 50 to 54 mm CL; in Maine, male lobsters also began ma- turing at relatively small sizes (50% mature at 44 mm CL, Krouse 1973). Egg masses were examined from 1984 to 1987 to determine if the complement of eggs carried by females was normal. Smith (1977) expressed concern about the fecundity of berried females in western LIS when he found 10-14% of the berried females carried abnormally low numbers of eggs 130 ^ 0.75 ^ 9 0.70 0.50 Ht All females are mature 60 70 80 £ CARAPACE LENGTH (MM) Fig. 4. IVIorphometric relationship between the ratio abdominal width/carapace length and the carapace length for data collected Prom 1981 to 1987 for female lobsters. (*) mean value for each 5 mm size for 3-unit data; (- — ) y = a + bx + cx +dx for 2-unit data; ( ) upper and lower 95% C.I. in 1976. Over 90% of the berried females in 1986 and 1987 had 1/2 or more the full complement of eggs (Table 7). This compares to 89% and 86% in 1984 and 1985. Only 3.6% in 1986 and 1.4% in 1987 of the berried females had less than 1/4 the normal complement of eggs. This com- pares to 3.7% in 1984 and 7.7% in 1985. From the above data it is apparent that the complement of eggs carried by berried females in our area is very good, and there has been no change following 3-unit operation. The numbers of berried females caught during 1986 (134) and 1987 (158) were within the range of values repo'ted since 1978 (58-171), as was the proportion caught at each station (Table 8). Twotree catches continued to yield a higher pro- portion of berried females (8.0-9.6%) than either Jordan Cove (3.0-3.2%) or Intake (2.3-1.9%). The mean size of berried females collected during 1986 was 78.0 mm, which was within the range of pre-operational values (77.0-81.2 mm). The 1987 mean size of 76.5 mm was the smallest value reported for berried females and was due to the fact that 90% of the berried females collected in 1987 were of sublegal size, whereas only 75% of the berried females caught in 1986 were sublegal. The proportion of sublegal-size berried females has, in general, been increasing since 1981 and may be related to the high exploitation rate of lobsters in LIS. The number of legal-size females that contribute to reproduction is limited by the high level of fishing pressure which removes most females shortly after reaching legal size. The fact that females become sexually mature and bear eggs at sizes well below the legal size is important because these individuals are able to spawn before growing to marketable size and thereby sustain Lobster Population Dynamics 131 TABL.E 7. The number of berried females examined for egg mass fullness and egg development from May through October 1984-1987. Month Number of Berried Females Examined Number with Complement Number with 1 4 Complement Number with 1 2 Complement Number with 3 4 Complement Number with Full Complement Developmental Stage 1984 MAY JUN JUL 28 16 4 0 4 0 1 1 0 4 2 1 11 1 1 12 8 2 Ripe eggs Lt. Green with optical disks AUG 18 0 0 0 4 14 New eggs SEP 48 1 5 16 10 16 Black-dark green OCT 50 1 5 7 16 21 lOTAL 164 6 12 30 43 73 1985 MAY 34 3 4 8 3 16 Ripe eggs JUN 19 6 1 0 5 7 Lt. Green with optical disks JUL 7 2 0 0 4 1 " AUG 17 0 1 1 3 12 New eggs SEP 56 2 3 4 14 33 Black-Dark Green OCT 89 4 6 8 23 48 TOTAL 222 17 15 21 52 117 1986 MAY 21 0 1 7 5 8 Ripe eggs JUN 6 1 0 2 0 3 Lt. Green with optical disks JUL 0 0 0 0 0 0 AUG 21 0 0 1 2 18 New eggs SEP 49 1 1 4 10 33 Black-Dark Green OCT 67 4 8 13 19 23 TOTAL 164 6 10 27 36 85 1987 MAY 38 0 5 5 8 20 Ripe eggs JUN 11 1 0 I 3 6 Lt. Green with optical disks JUL 2 0 0 0 1 1 New eggs AUG 14 0 0 0 3 11 Black-Dark Green SEP 61 1 3 11 18 28 " OCT 81 1 2 6 14 58 '■ 132 TABL,E 8. Carapace-length statistics and percentage of berried females caught at each station from 1978 to 1987. Jordan Cove Intake Twotree N^ Length Length % Range (mm) Mean±95% CI Sublegal 58 74 - 88 80.1 ± 1.04 67 70 64 - 93 80.5 ± 1.28 54 71 66 - 93 79.1 ± 1.27 64 82 69 - 97 81.2 ± 1.35 52 108 64 - 99 80.0 ± 1.08 58 123 66 - 103 80.5 ± 1.04 60 173 62 - 95 79.1 ±0.87 67 171 63 - 94 77.0 ± 0.81 82 134 65 - 94 78.0 ± 0.95 75 158 62 - 90 76.5 ± 0.67 90 1978 1.4 2.6 5.3 1979 1.9 2.7 7.2 1980 3.5 1.8 5.6 1981 1.6 2.7 7.1 1982 0.8 0.9 6.1 1983 2.1 3.2 8.5 1984 3.6 3.5 10.6 1985 3.5 4.5 8.5 1986 3.0 2.3 8.0 1987 3.2 1.9 9.6 ^ Recaptures not included the population. In contrast, females in northern populations (Maine) begin to mature at sizes close to the legal size and only a small percentage of these females are able to spawn prior to reaching marketable size (Aiken and Waddy 1980). Molting and Growth The proportions of near-molting lobsters in the 1986 and 1987 total catches were 3.2% and 3.0% respectively, which were within the range of pre- operational values reported in NUSCO (1987a, range 2.5%-6.4%). The timing of peak molting was examined to determine the influence of water temperature during pre-operational and opera- tional studies. The cumulative percentage of molt- ing lobsters was compared for pre-operational and operational study years. The Gompertz growth function was then fitted to these data to estimate the time of peak molting as the time at which the inflection point of the growth curve occurred (Fig. 5). The inflection point of a Gompertz growth curve occurs at t = log(k/b), where (k) and (b) are parameters of the Gompertz function. During years when May temperatures were warmer than average, molting peaks occurred earlier, and con- versely, they occurred later in the season when May temperatures were colder than average (Fig. 6). Templeman (1936) correlated the timing of molts with summer water temperatures in the Ca- nadian Maritimes. He found that for each degree (C) of lower water temperature the first molting period was postponed for a week or more. The influence of varying water temperature on the molt cycle was also documented by Aiken and Waddy (1980) who found that lobsters exposed to lO'C, after a normal winter period, quickly entered pre-molt cind progressed through to ecdysis. Secondary molts were not observed in the 1986-87 studies, although spring and fall molts were observed in some years when sampling was conducted from May to November and January to December (NUSCO 1987a). I/obster growth was determined for lobsters that molted between the times of tagging and recapture. Carapace lengths at recapture (post-molt size) were related to carapace lengths at tagging (pre- molt size) using a simple linear regression which best describes growth per molt in crustaceans (Wilder 1953; Kurata 1962; Mauchline 1976). Lobster Population Dynamics 133 100- 90- CO a: LJ 1- 80- tn OD S 70- E 60- 3-unit /° y Li O 2 50- 1_ O (1986-87) P / D/' j)>' 2-unit □ / */ (1980-85) K 40- '' */ o 30- D/' / Q-' / 20- /'/ 10- / a/* n- ^''C^ * JUL AUG MONTH Fig. 5. Cumulative percentage of moiling lobsters caught in preoperational (*= 1980-85) and 3-unit studies (D= 1986-87) and the Gompertz growth function that was fitted to the 2-unit (— ) and 3-unit (- -) data. UJ 10.0 S5 9-0 1986 ^^=0.85 p=0.006 1987 1 980 ^"~~----,^^___^ 1983 1985 1981 ""~"---.^^. T984 1982 ^"^^ 17JUN 20JUN 23JUN 28JUN OIJUL 04JUL DATE OF PEAK MOLTING Fig. 6. Linear relationship between the date of peak molting (log b/k, from Gompertz growth function) and annual mean May bottom water temperature at all stations 1980-87. 134 Regression plots, equations and growth parame- ters for all lobsters (n= 987), males (n= 381) and females (n = 606) are presented in Figure 7. Growth per molt for all lobsters averaged 13.3% from 1986-87 and 13.9% from 1978-85. Female growth per molt averaged 13.9% during 3-unit operations and 13.7% prior to Unit 3 start up. Males had lower growth per molt during the 1986-87 studies (12.3%) compared to the pre- operational studies ( 14. 1 %). However, lower val- ues for male growth (12.3%) had been reported in previous years (NUSCO 1980). In eastern LIS, Stewart (1972) reported 15.8% growth per molt for males, and 15.4% for females. Briggs and Mushacke (1984) reported 14.5% growth per molt for males and 12.5% for females in western LIS. Higher growth increments were reported by Cooper and Uzmann (1980) for an offshore lob- ster population, 18.7% for males and 16.7% for females. They attributed the lower growth of the inshore population to lobster inactivity during the colder months of the year. Claw Loss The percentages of lobsters caught missing one or both claws (culls) in 1986 (10.6%) and 1987 (10.3%) were lower than the average percentage culled in previous years (10.6%-15.5%) (Table 9). The proportion of culls at each station was similar in 1986 and 1987. The highest percentage of culls occurred at Intake, 14.7% in both 1986 and 1987, followed by Jordan Cove, 10.9%-11.9% and Twotree 6.8%-6.5%. The percentage of culls at Twotree has been lower than at the nearshore Jordan Cove and Intake stations since this study began (NUSCO 1987a). Working in eastern LIS, Smith (1977) reported 26.4% claw loss, whereas Briggs and Mushacke (1979) reported claw loss varying between 7.4 and 22.8% in western LIS. Since 1984, the percentage of culls in our catch was lower than in previous years due to the im- plementation of the escape vent regulation. This trap regulation requires that pots contain an open- ing to allow sublegal-sized lobsters to escape thereby reducing injuries and mortality associated with overcrowded pots (Landers and Blake 1985). Studies in Maine and Massachusetts by Pecci et Aa LOBSTERS 2-unit Nav Size ■= 15.B33 + (0.896) Old Ska (^=0.76. nB734 3-Lnit Neif Sat - 13.365 + (0.927) OldX"« r^O.BO. n-253 CARAPACE LENGTH AT TAGGING (MM) CARAPACE LENGTH AT TAGGING (MM) MALES 2-unrt Now Siza - 1 9.706 + (D.S42) Old Size r«0.71 . ,n-296 3~unJt Nov Size = 1+.623 + (0.906) Old Size r=0-e3. n=ia5 CARAPACE LENGTH AT TAGGING (MM) rig. 7. Linear regressions for carapace lengths at tag- ging and recapture times for all lobsters, males and females for data collected during pre-operational studies 1978-85 ( ) y = a + bx; ( ) upper and lower 95% C.I.; data for 3-unit studies (1986-87) represented as individual points (*). Lobster Population Dynamics 135 TABLE 9. Claw loss for lobsters caught in wire pots from 1978 to 1987. Year Percent Cull Percent Missing Percent Missing One Claw Two Claws 1978 14.9 14.0 0.9 1979 15.5 14.4 1.2 1980 13.6 12.2 1.5 1981 12.0 11.1 1.0 1982 11.1 10.4 0.7 1983 12.4 11.6 0.8 1984 10.6 9.8 0.7 1985 11.1 10.4 0.7 1986 10.6 9.8 0.8 1987 10.3 9.5 0.7 al. (1978) reported that trap-related injuries re- sulting in claw loss were often associated with water temperature, fishing pressure (i.e., handling by lobstermen), trap soaktime, and physical con- dition of the lobster (i.e., its nearness to moUing). Tagging Program The numbers of lobster tagged in 1986 and 1987 were 5,698 and 5,680 respectively. These values were within the range of values established in pre-operational studies (Table 10). Recapture percentages during the same period were 21.0% and 23.9%, and also within the range of pre- operational percentages (15.9-23.9%). The per- centage of commercial recaptures was 20.2% in 1986 and 17.8% in 1987. These values were lower than in previous years (range 21.1-47.6%) due to the implementation of a new trap regulation in 1984 which required escape vents in commercial traps. Lobstermen recaptured 27.7 to 47.6% of our tagged lobsters from 1978 through 1983, an overall average of 37.4%. When the escape vent went into effect in 1984, lobstermen recaptured fewer of our tagged lobsters (21.5-17.8%), an overall average of 18.8%. In contrast, our traps do not have escape vents and subsequently our rates of recapture increased from 17.0% to 22.7% after the regulation went into effect. I ,andcrs and Blake (1985) demonstrated the retention rate of tagged lobsters in vented and unvented traps. With the regulation in force, the mean size of tagged lobsters caught in commercial traps fitted with escape vents was larger than the mean size caught in unvented traps (i.e., a number of the tagged sublegal-sized lobsters were escaping from pots which had escape vents). Movement Movement patterns of the local lobster popu- lation were assessed using recapture data from our sampling efforts and those of commercial lob- stermen. Because lobsters were tagged and re- leased at the station where captured, any move- ment between stations could be detected at recap- ture. During 1986 and 1987, 97% of the lobsters were recaptured at the release station compared to 95% in pre-operational study years indicating that movement between stations continues to be minimal. Of the exchanges that did occur, most were between the nearshore Jordan Cove and In- take stations. Tagging studies conducted in east- em LIS by Stewart (1972) demonstrated the hom- ing behavior of nearshore lobster populations. 136 TABLE 10. Summary of tag and recapture studies from 1978 through 1987. NUSCo Commercial Year No. tagged Mean CL (mm) Returns No. % Mean CL (mm) Returns No. % Mean CL (mm) 1978 3193 73.6 508 15.9 75.5 884 27.7 81.1 1979 3732 72.8 722 19.4 75.1 1776 47.6 77.5 1980 3634 75.5 522 14.4 75.7 1363 37.5 76.4 1981 4246 72.4 707 16.7 74.8 1484 35.0 76.3 1982 7575 70.9 1278 16.9 73.2 2518 33.2 75.5 1983 5160 71.8 936 18.1 73.6 2266 43.9 76.9 1984 5992 71.9 1431 23.9 73.0 1289 21.5 78.7 1985 5609 71.3 1235 22.0 73.1 1185 21.1 78.3 1986 5698 70.2 1194 21.0 72.3 1153 20.2 78.3 1987 5680 70.4 1356 23.9 72.8 1010 17.8 78.6 Tag returns from commercial lobstermen fishing within 8 km around MNPS accounted for 94% and 97% of all commercial returns during 1986 and 1987, respectively, which was similar to re- turns during pre-operational studies (93%). Of the 103 lobsters that were recaptured outside the study area (>8 km) during 1986 and 1987, 98% moved to the east. Several of these lobsters trav- eled a great distance; 2 were recaptured off Point Judith, and 1 off Watch Hill, RI, 1 tag was re- turned from Buzzards Bay, and 1 from Nantucket Shoals, MA. Based on tag returns from the deep water canyons on the edge of the continental shelf (Hudson Canyon, n = 2; Veatch Canyon n=l), some of our lobsters moved offshore during the 1986-87 studies. Similar offshore movement pat- terns were established in pre-operational studies (NUSCO 1987a). Results from other tagging studies in LIS indicated a similar easterly trend in lobster movement (Lund and Rathbun 1973). Other researchers working in waters off New Eng- land and on the continental shelf deriionstrated similar exchanges between the inshore and off- shore populations (Saila and Flowers 1968; Uzmann et al. 1977; Fogarty et al. 1980). Based on our sampling efforts and commercial returns around MNPS, lobster movements in our study area were typical of nearshore populations and agreed with other tagging studies conducted in coastal waters of eastern North America which indicated localized lobster movement (Templeman 1940; Wilder and Murray 1958; Wilder 1963; Cooper 1970; Stewart 1972; Cooper et al. 1975; Fogarty et al. 1980; Krouse 1980, 1981; Campbell 1982; Ennis 1984; Campbell and Stasko 1985). Entrainment Lobster larvae were collected from mid-May to early-July during 1986 and 1987. Stage I lobster larvae predominated in samples collected in 1984, 1985, and 1987, and stage IV in 1986 (Table 11). Seventy-seven percent of these larvae (Stage IV) were collected in 4 night samples during the last week of June 1986. More larvae were collected in night than in day samples from 1984-1986. However, during 1987 more larvae were caught in day samples, when a single sample contained 52 Stage I larvae, representing 50% of all Stage I larvae collected in 1987. Variability in both numbers and stages of larvae collected in our Lobster Population Dynamics 137 TABLE 11. Summary of lobster larvae entrainment data from 1984 to 1987. Stage I Stage II Stage III Stage IV Total Day Night lota! Day Night Total Day Night Total Day Night Total 1984 15 0 0 1 16 73 1 1 11 86 88 1 1 1985 12 102 56 0 3 2 61 69 1 2 10 82 125 1 5 1986 12 143 33 1 4 8 46 54 10 11 111 186 87 11 15 1987 119 232 104 4 5 3 116 56 6 3 4 69 160 10 8 7 185 samples was high, similar to findings by other researchers working with lobster larvae (Bibb et al. 1983; Lux et al. 1983; Blake 1984). The con- tagious distribution of lobster larvae has been as- sociated with surface water circulation patterns (Fogarty 1983), which generate sea surface "fronts". These fronts, visible on the surface wa- ters as "scum", "foam", or "slick" lines, occur in the vicinity of MNPS, and were reported to con- tain high densities of planktonic organisms, in- cluding lobster larvae (Cobb et al. 1983; M. Blake personal communication). Night samples contained more larvae than day samples in 3 of the 4 study years. This may be due to the design of the intake structures, which have curtain walls extending about 2 m below MLW. This means that cooling water is drawn from 2 m below the surface. Since the early stages of lobster larvae are photopositive and dis- perse from surface waters during darkness (Templeman 1939), they are more susceptible to entrainment at night regardless of tidal stage. However, during daylight hours lobster larvae predominate in surface samples (Fogarty 1983; Fogarty and l^wton 1983). Therefore larvae avoid entrainment during the day due to their photo-behavior and the intake structure design. The density of lobster larvae in the MNPS cool- ing waters was estimated as the 5-mean (see Delta Distribution Section of this report and also Pennington 1983). The annual 5-mean density (number per 1000 m ) of lobster larvae collected in entrainment samples was higher in 1986 (0.88) and 1987 (0.63) than 1984-85 densities (0.42-0.43, Table 12). The estimated number of lobster larvae entrained through the MNPS cooling water systems in 1986 and 1987, was 548,635 and 304,694, respectively. The higher estimates in 1986-87, relative to 1984 and 1985 values of 79,511 and 138,820, respectively, were related to the fact that Unit 3 began operating in 1986. Fntrainment estimates were based on both the 138 TABLE 12. Annua] mean densities (number per 1000 m ) of lobster larvae in entrainment samples during their season of occurrence and annual entrainment estimates with 95% C.I. for MNPS from 1984 through 1987. Year Dates Found Mean Density^ (n per 1000 m'^) 95% CI Estimate 95% CI 1984 21May-10Jul 0.42 0.19-0.65 79,511 35,983-123,100 1985 15May-29Jul 0.43 0.22-0.64 138,820 71,024-206,615 1986'' 14May-14Jul 0.88 0.43-1.33 586,226 286,451-886,000 1987 18May-7Jul 0.63 0.29-0.97 304,695 140,256-469,134 Mean densities are calculated as the S-mean (see Delta Distribution Section and Pennington 1983). Unit 3 began commercial operation density of larvae in the cooling waters and total cooling water volumes; since Unit 3 requires al- most double the cooling water demand of Units 1 and 2 combined, a doubling of the estimated number of lobster larvae entrained was expected when all units were operating during the larval season. Projecting the impact associated with lobster larvae entrainment to the population level (adult lobsters) was difficult due to the lack of knowledge regarding larval and post-larval stages of lobsters (Phillips and Sastry 1980). Given that lobsters require several years of growth to reach a trappable size, our studies of adults would not detect 3-unit entrainment effects until 5-6 years after an impact occurred. During 1986, 6.5% and in 1987, 3.8% of the lobster larvae survived after passing through the plant's cooling water system suggesting that en- trainment mortality may be lower than the as- sumed 100%. Similar findings at other power stations have been reported. ColUngs et al. (1981) reported 14"/o survival for lobster larvae (Stage II) collected at the Canal Electric Company, Sand- wich, MA. Impingement Impingement of lobsters on the Unit 1 and 2 intake screens has been summarized for the period 1975 to 1985 in NUSCO (1987a). Throughout these studies several measures were investigated to mitigate impingement losses, including the use of underwater barriers, acoustic and light deter- rents and more recently fish return systems (sluiceways). In 1983, a sluiceway system was installed in the Unit 1 intake structure which re- turned 100% of the lobsters caught on the screens back to Niantic Bay. A sluiceway was also con- structed at Unit 3 and operated during 1986 and 1987 (NUSCO 1987b). The estimated number of lobsters impinged at Unit 2, which does not have a sluiceway, was 676 and 825 in 1986 and 1987, respectively (Table 1 3). These values were within the range of values reported for impingement at Unit 2 (261-1220). The impingement of lobsters was highest during the summer months and coincided with peak catch in traps (NUSCO 1987a). The mean sizes of lobsters impinged during 1986-87 were 55.7 and 55.8 mm CL, respectively, which were within the range of pre-operational values 48.6 to 64.9 mm CL and continued to be Lobster Population Dynamics 139 TABLE 13. Annual impingement estimates for lobster collected at Units I and 2 from 1978 to 1987. Unit 1 Unit 2 Both Units 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 245 323 368 665 938 999 a a 261 426 405 1009 1041 497 1220 480 676 825 506 749 773 1674 1979 1496 1220 480 676 825 Total 3538 6840 10378 Unit 1 sluiceway began operating December 1983. smaller than the trap catch values (NUSCO 1987a). Smaller lobsters enter the intake through the course bar racks (bar spacing 6.4 cm) more readily than larger lobsters which are seldom im- pinged. Male to female sex ratios of impinged lobsters during 1986 and 1987 were 1.0:0.46 and 1.0:0.38, respectively. These values were slightly lower than the range of pre-operational sex ratios reported from 1982 to 1985 (1.0:0.47 to 1.0:0.58) and reflected the higher abundance of male lobsters nearshore at the Jordan Cove and Intake stations (1.0:0.6-1.0:0.7, NUSCO 1987a). The percentage of impinged lobsters missing one or both claws (culls) during 1986-87 (27%) was lower than pre- operational values (30-50%). Impinged lobsters suffered greater claw loss when compared to trap catch values (wire pots, 10-16%) due to the high pressure (80 psi) wash used to remove debris from the traveling screens (NUSCO 1987a). Survival of lobsters impinged at Unit 2 during 1986 (97%) was higher than the ten year range of survival values reported from 1975-85 (64-80%). During 1987 survival was slightly lower (62%), due to higher impingement of lob- sters during the summer; more than half of all lobsters were impinged from May through Sep- tember in 1987. Historically, mortality of im- pinged lobsters was highest during the peak molt- ing period (May-June) when lobsters were soft and easily injured, and during the later summer months when water temperatures were highest (August-September; NUSCO 1987a). Summary 1. Total CPUE during 1986 and 1987 was 1.70 and 1.72, respectively, within the range of values reported during 2-unit operations (0.85-2.10). Legal CPUE was lower during 1986 and 1987 (0.097, 0.089) when compared to previous years' results and may be related to increased fishing pressure. A 50%> decline in catch at Jordan Cove occurred from August to September 1986 and was related to 3-unit operations. Sediments in the discharge area were scoured and subsequently deposited in Jordan Cove, where lobster habitats were buried by sediments. This decline in catch was only temporary, since catches in October 1986 and througliout 1987 at Jordan Cove were normal and indicated that sediments 140 had stabilized and lobsters had returned to the affected area. 2. The mean size of lobsters caught during 1986 (70.1 mm) and 1987 (70.2 mm) was smaller than values reported in previous years (range 70.7-71.8 mm) due to the lower CPUE of legal-sized lobsters during 1986-87. 3. Male to female sex ratios during 1986 and 1987 were 1.0:0.87 and 1.0:0.88, respectively, within the range of values in previous years. The Twotree station continued to yield a higher proportion of females than the other two nearshore stations, a trend consistent since the study began. 4. Female size at sexual maturity was similar during 2- and 3-unit operations; females be- gan to mature between 50 and 55 mm CL and aU females were mature at sizes greater than 95 mm CL. The mean CL of berried females during 1986 (78.0 mm) and 1987 (76.5 mm) and the proportions of sublegal size berried females caught in 1986 (75%) and 1987 (90%) confirmed the small size at first sexual maturity of females in the Mill- stone area. 5. Lobsters that were near molting comprised 3.2% and 3.0% of the 1986 and 1987 total catches., respectively, which were within the range of values reported during 2-unit oper- ations. Growth per molt averaged 13.3% in 1986-87 compared to 13.9% from 1978-85. 6. The percentage of culls in 1986 (10.6%) and 1987 (10.3%) was lower than the average percentage culled in previous years (range 10.6%- 15.5%) and due to the implementa- tion of the escape vent regulation in 1 984. 7. The number of lobsters tagged in 1986(5,698) and 1987 (5,680) was witliin thi- range of annual values for lobsters tagged in pre- operational studies. Recapture rates for 1986 (21.0%) and 1987 (23.9%) were also similar to pre-operational values (range 15.9%-23.9%). Lobstermen recaptured 20.2% of our tagged lobsters in 1986 and 17.8% in 1987. 8. Lobster movements were localized, since 94% and 97% of all commercial recaptures were made within 8 km of Millstone Point during 1986 and 1987, respectively. Sev'eral lobsters moved outside LIS and were caught in waters off R.I. and MA.; 3 lobsters moved offshore during 1986-87, where they were caught in deep water canyons on the edge of the continental shelf. 9. Lobster larvae densities (number per 1000 m ) in entrainment samples were higher in 1986 (0.88) and 1987 (0.63) when compared to 1984-85(0.42-0.43). The estimate of total lobster larvae entrainment, based on sample density and total MNPS cooling water de- mand, was also higher in 1986 (548,635) and 1987 (304,694) when compared to 1984-85 (79,511-138,820). Lobster larvae survival was 6.5% and 3.8% in 1986 and 1987, re- spectively. More stage IV larvae were col- lected in 1986 compared to 1984, 1985, and 1987 when greater numbers of stage I larvae were collected. 10. The estimated numbers of lobster impinged at Unit 2 during 1986 and 1987 were 676 and 825, respectively, these values were within the range of values reported in previous years (261-1220). Fish return systems at Units 1 and 3 improved overall survival of impinged lobsters. Based on impingement of all or- ganisms at Unit 2 since 1 972, a request made by NUSCO to discontinue impingement monitoring at Unit 2 was accepted by the CT DEP in December 1987. Conclusion Our studies indicate that the local lobster pop- ulation is heavily exploited; more than 90% of legal sized lobsters are removed by fishing. The commercial and recreational catches are highly dependent on the number of lobsters in the Lobster Population Dynamics 141 prerecruit size class. Legal CPUE values over all stations during 1986-87 were the lowest reported since this study began, although legal CPUE has declined since 1978 and may be related to the higli fishing pressure in LIS. During simultaneous 3-unit operation, a short term impact on lobsters occurred at Jordan Cove in August 1986 due to scouring of fine silt in the discharge area. Silt originating from the discharge area fouled lobster habitat in Jordan Cove and temporarily displaced lobsters. Once sediments stabilized, lobsters reinhabited the area, and catch rates at Jordan Cove in October 1986 and throughout 1987 were normal and typical of previous study results. A similar short term impact occurred following dredging activities in the vicinity of the intakes during 1985; dredging removed lobster habitat (shelters) and thereby displaced lobsters from the area. Lobsters returned to that site soon after the sediments stabilized and CPUE values for 1986-87 were comparable to the other stations. , and S.L. Waddy. 1980. Reproductive Biology. Pages 215-276 in J.S. Cobb, and B.F. Phillips, eds. The Biology and Management of Lobsters, Vol. I, Academic Press, Inc. New York. Bibb, B.G., R.L. Ilersey, and R.A. Marcello Jr. 1983. Distribution and abundance of lobster larvae {Homarus americanus) in Block Island Sound. U.S. Dep. Commer. NOAA Tech. Rep. NMFS SSRF-775: 15-22. Blake, M.M. 1984. Annual progress report Con- necticut lobster investigations, January- December 1983. NOAA-NMFS Project No. 3.374-R. 47pp. , and E.M. Smith. 1984. A marine resources plan for the state of Connecticut. Connecticut Dept. of Environ. Protection, Mar. Fish. 244pp. Because lobsters require at least 4-5 years of growth before they are vulnerable to our traps, and an additional 2 years to reach marketable size, there is a lag of about 6-7 years between the time of a potential impact on larvae and the time at which we can detect that impact. The sensitivity of our program in defining population trends (i.e., observing the strong prerecruit class in 1982) and impacts (displacement of lobsters as a result of scour and dredging) is vital to the evaluation of impacts associated with the operation of three units at Millstone Point. If changes occur in the local lobster population, they will be detected by the alteration of the basic population parameters now being monitored. Changes in these param- eters during 3-unit operations will demonstrate the effects (if any) of MNPS operations on the local lobster population. References Cited Aiken, D.E. 1973. Proecdysis, setal development, and molt prediction in the American lobster, {Homarus americanus). J. Fish. Res. Board Can. 30:1337-1.344. Briggs, P.T., and F.M. Mushacke. 1979. The American lobster in western Long Island Sound. New York Fish and Game J. 26:59-86. . 1980. Tlie American lobster and the pot fishery in the inshore waters off the south shore of Long Island, New York. New York Fish and Game J. 27:156-178. . 1984. The American lobster in western I^ng Island Sound: Movement, growth and mortality. New York Fish and Game J. 31:21-37. Campbell, A. 1982. Movements of tagged lob- sters released off Port Maitland, Nova Scotia, 1944-80. Can. Tech. Rept. Fish. Aquat. Sci. No. 1136 41pp. , and A.B. Stasko. 1985. Movements of tagged American lobsters, Homarus americanus, off southwestern Nova Scotia. Can. J. Fish. Aquat. Sci. 42:229-238. Cobb, J.S., T. Gulbransen, B.F. Phillips, D. Wang, and M. Syslo. 1983. Behavior and 142 distribution of larval and early juvenile Homarus americanus. Can. J. Fish. Aquat. Sci. 40:2184-2188. Collings, W.S., C.C. Sheehan, S.C. Hughes, and J.L. Buckley. 1981. The effects of power generation on some of the living marine re- sources of the Cape Cod Canal and approaches. Mass. Dept. of Fish., Wildlife, and Rec. Veh., Div. Mar. Fish., 212pp. Colvin, G.C. 1987. A view of the problems and opportunities from the Long Island side of the Sound. Pages 97-109 in V.R. Gibson, and M.S. Conner, eds. Long Island Sound: Issues, Resources, Status and Management Seminar Proceedings. NOAA Estuary of the month Seminar Series No. 3. Cooper, R.A. 1970. Retention of marks and their effects on growth, behavior and migrations of the American lobster, Homarus americanus. Trans. Amer. Fish. Soc. 99:409-417. , R.A. Clifford, and CD. Newell. 1975. Seasonal abundance of the American lobster, Homarus americanus, in the Boothbay Region of Maine. Trans. Amer. Fish. Soc. 104:669-674. , and J.R. Uzmann. 1980. Ecology of Juvenile and Adult Homarus americanus. Pages 97-142 in .I.S. Cobb, and B.F. Phillips, eds. The Biology and Management of Lobsters, Vol II, Academic Press, Inc. New York. Dow, R.l. 1966. The use of biological, environ- mental and economic data to predict supply and to manage a selected marine resource. The Amer. Biol. Teacher 28:26-.30. . 1969. Cyclic and geographic trends in seawater temperature and abundance of Amer- ican lobster. Science. 164:1060-1063. . 1976. Yield trends of the American lob- ster resource with increased fishing effort. Mar. Technol. Soc. 10:17-25. Ennis, G.P. 1 97 1 . Lobster {Homarus americanus) fishery and biology in Bonavista Bay, New- foundland. 1966-70. Fish. Mar. Serv. Tech. Rep. 289, 46pp. . 1974. Observations on the lobster fishery in Newfoundland. Fish. Mar. Serv. Tech. Rep. 479, 21pp. . 1984. Small-scale seasonal movements of the American lobster, Homarus americanus. Trans. Am. Fish. Soc. 113:336-338. Flowers, J.M., and S.B. Saila. 1972. An analysis of temperature effects on the inshore lobster fishery. J. Fish. Res. Board Can. 29:1221-1225. Fogarty, M..I. 1983. Distribution and Relative Abundance of American Lobster, Homarus americanus Larvae: New England Investigations during 1974-79. NOAA Tech. Rep. NMFS SSRF-775, 64pp. , D.V.D. Borden, and H.J. Russell. 1980. Movements of tagged American lobster, Homarus americanm, off Rhode Island. Fish. BuU. U.S. 78:771-780. Fogarty, M.J., and R. Lawton. 1983. An over- view of larval American lobster, Homarus americanus, sampling programs in New England during 1974-79. U.S. Dep. Commer. NOAA Tech. Rep. NMFS SSRF-775:9-14. Ilerrick, F.H. 1911. Natural history of the Amer- ican lobster. Bull. U.S. Bureau Fish. 29:149-408. Keser, M., D.F. Landers, Jr., and J.D. Morris. 1983. Population characteristics of the Amer- ican lobster, Homarm americanus, in eastern I^ng Island Sound, Coimecticut. NOAA Tech. Rep. NMFS SSRF-770, 7pp. Krouse, J.S. 1973. Maturity, sex ratio, and size composition of the natural population of American lobster, Homarus americanm, along the Maine coast. Fish. Bull. 71:165-173. Lobster Population Dynamics 143 . 1980. Summary of lobster, Homarus americanus, tagging studies in American waters (1898-1978). Can. Tech. Rep. Fish. Aquat. Sci. 9.12:135-140. _. 1981. Movement, growth, and mortality of American lobsters, Homarus americanus, taggeJ along the coast of Maine. U.S. Dep. Commer. NOAA Tech. Rep. NMFS SSRF-747:12pp. Kurata, 11. 1962. Studies on the age and growth of Crustacea. Bull. Hokkaido Reg. Fish. Res. Ub. 24:1-115. Landers, D.F., .Jr., and M.M. Blake. 1985. The effect of escape vent regulation on the American lobster, Homarus americanus, catch in eastern Long Island Sound, Connecticut. Trans. 41st Annual N.E. Fish and Wildlife Conference, 9pp. Lund, W.A., Jr., and C.J. Rathbun. 197.3. In- vestigation on the lobster. NOAA Tech. Rept. NMFS Project No. 3-130-R, 189pp. Lux, F.E., G.F. KeUy, and C.L. Wheeler. 1983. Distribution and abundance of larval lobsters {Homarus americanus) in Buzzards Bay, Mas- sachusetts, in 1976-79. U.S. Dep. Commer. NOAA Tech. Rep. NMFS SSRF-775:29-33. Marcello, R.A., Jr., W. Davis III, T. O'Hara, and J. Hartley. 1979. Population statistics and commercial catch rate of American lobster {Homarus americanus) in the Charlestown- Manatuck, Rhode Island region of Block Island Sound. Yankee Atomic Electric Company. Submitted to New England Power Company YAEC1175, 40pp. Mauchline, J. 1976. The Hiatt growth diagram for Crustacea. Mar. Biol. 35:79-84. McLeese, D.W., and D.G. Wilder. 1958. The activity and catchability of the lobster (&homa) in relation to temperature. J. Fish. Res. Board Can. 15:1.345-1354. NUSCO (Northeast Utilities Service Company). 1980. Lobster Population Dynamics. Pages 1-22 in Monitoring the Marine Environment of Long Island Sound at MiUstone Nuclear Power Station, Waterford, Connecticut. Annual Re- port 1979. . 1985. Lobster Population Dynamics. Pages 1-27 in Monitoring the Marine Environ- ment of Long Island Sound at Millstone Nu- clear Power Station, Waterford, Connecticut. Annual Report 1984. . 1987a. Lx^bster Population Dynamics. Pages 1-42 in Monitoring the Marine Environ- ment of Long Island Sound at Millstone Nu- clear Power Station, Waterford, Connecticut. Summary of Studies Prior to Unit 3 Operation 1987. . 1987b. The Effectiveness of the Unit 3 Fish Return System 1987. 20pp. . 1988. Proposed ecological studies. Mill- stone Nuclear Power Station, 1988. Letter to Connecticut Department of Environmental Protection. Pecci, K.J„ R.A. Cooper, CD. Newell, R.A. Clifford, and R.J. Smolowitz. 1978. Ghost fisliing of vented and unvented lobster, Homarus americanus, traps. Mar. Fish. Rev. 40:9-43. Pennington, M. 1983. Efficient estimators of abundance, for fish plankton surveys. Biometrics, 39:281-286. Phillips, B.F., and A.N. Sastry. 1980. Larval Ecology Pages 11-57 in J.S. Cobb, and B.F. Phillips, eds. The Biology and Management of Lobsters, Vol II, Academic Press, Inc. New York. Pottle, R.A., and R.W. Finer. 1982. Substrate preference behavior of juvenile American lob- sters, Homarus americanus, in gravel and silt- clay sediments. Can. J. Fish. Aquat. Sci. 39:928-932. 144 Saila, S.B., and J.M. Flowers. 1968. Movements and behavior of berried female lobsters displaced from offshore areas to Narragansett Bay, Rhode Island. J. Cons. Int. Explor. Mer. 31:342-351. Scarratt, D.J. 1970. Laboratory and field tests of modified sphyrion tags on lobsters (Homarus americanus). J. Fish. Res. Board Can. 27:257-264. , and P.F. Elson. 1965. Preliminary trials of a tag for salmon and lobsters. J. Fish. Res. Board Can. 22:421-423. Skud, B.E., and H.C. Perkins. 1969. Size com- position, sex ratio and size at maturity of off- shore northern lobsters. U.S. Fish Wildl. Spec. Sci. Rept. Fish. 598, 10pp. Smith, E.M. 1977. Some aspects of catch /effort, biology, and the economics of the I^ng Island lobster fishery during 1976. U.S. Dept. Commer., NOAA-NMFS, Commer. Fish. Res. Dev. Act, Project No. 3-253-R-l, 97pp. . 1987. A view of the problems and op- portunities from the Connecticut side of the Sound. Pages 111-123 in V.R. Gibson, and M.S. Conner, eds. Lx)ng Island Sound: Issues, Resources, Status and Management Seminar Proceedings. NOAA Estuary of the month Seminar Series No. 3. Snedccor, G.W., and W.C. Cochran. 1967. Sta- tistical methods. The Iowa State University Press, Ames, I A. 593pp. Stewart, L.L. 1972. The seasonal movements, population dynamics and ecology of the lobster, IlomatTis americanus (Milne-Edwards), off Ram Island, Connecticut. Ph.D. thesis. University of Coimecticut, Storrs, CT. 112pp. Templeman, W. 1935a. Lobster tagging in the Gulf of St. I^wrence. J. Biol. Board Can. 1:269-278. . 1935b. Local differences in the body proportions of the lobster, Homarus americanus. J. Biol. Board Can. {-.in-llb. . 1 936. Local differences in the life liistory of the lobster {Homarus americanus) on the coast of the maritime provinces of Canada. J. Biol. Board Can. 2:41-88. . 1939. Investigations into the life history of the lobster {Homarus americanus) on the west coast of Newfoundland, 1938. Newfound- land Dep. Nat. Resour. Res. Bull. (Fish) 7, 52pp. . 1940. Lobster tagging on the west coast of Newfoundland 1938. Newfoundland Dep. Nat. Resour. Res. Bull. (Fish) 8, 16pp. Thomas, J. C. 1973. An analysis of the commer- cial lobster {Homarus americanus) fishery along the coast of Maine August 1966 through De- cember 1970. NOAA-NMFS Tech. Rept. SSRF-667, 57pp. Uzmann, J.R., R.A. Cooper, and K.J. Pecci. 1977. Migrations and dispersion of tagged American lobsters, Homarus americanus, on the southern New England Continental Shelf. U.S. Dep. Commer. NOAA Tech. Rep. NMFS SSRF-705 92pp. Wilder, D.G. 1953. The growth rate of the American lobster {Homarus americanus). J. Fish. Res. Board Can. 10:371-412. . 1963. Movements, growth and survival of marked and tagged lobsters liberated in Egmont Bay, Prince Edward Island. J. Fish. Res. Board Can. 20:305-318. , and R.C. Murray. 1958. Do lobsters move offshore and onshore in the fall and spring? Fish. Res. Board Can. Atl. Prog. Rep. 69:12-15. Lobster Population Dynamics 145 Contents Winter Flounder Studies 149 Introduction 149 Materials and Methods 150 Adult and juvenile studies 150 larval studies 154 I'ost-larval young-of-thc-ycar studies 157 Impingement of winter flounder at MNPS 157 Results and Discussion 158 Adult and juvenile studies 158 Abundance of winter flounder near Millstone 158 Regional and historical trends in abundance 164 Reproduction 170 Stock and recruitment 172 larval studies 177 Abundance and distribution 177 F-ntrainment of larvae at MNPS 183 Growth 187 Mortality 197 Post-larval young-of-thc-ycar studies 197 Abundance 197 Growth 198 Mortality 203 Impingement of winter flounder at MNPS 207 Impact assessment 207 Approaches to impact assessment 207 larval dispersal and entrainmcnt model 208 Population dynamics model 210 I-ong-tcrm impact assessment 211 Conclusions 213 Summary 216 References Cited 218 Frratum 224 Winter Flounder Studies Introduction This section summaiizes recent research on the winter flounder {Pseudopleuronecte.s americanm) and builds upon the data base previously assem- bled by Northeast Utilities (NU) as part of envi- ronmental studies for the Millstone Nuclear Power Station (MNPS). Due to the local abundance and importance of the winter flounder to the Connecticut sport and commercial fisheries, this species has been intensively studied since 1973 and considerable data have been collected on its life history and population dynamics. A summa- rization of the data, results, and conclusions for all aspects of the work from 1 973 through early 1986 was included in NUSCO (1987). This time frame represented the pre-operational period for MNPS Unit 3, which began commercial operation in late April of 1986. The population of winter flounder is composed of reproductively isolated stocks which spawn in specific estuaries and coastal areas from Nova Scotia to New Jersey (Ix^bcll 1939; Perlmutter 1947; Saila 1961; Leim and Scott 1966). Most adults enter natal estuaries in fall and early winter and spawning occurs in late winter and early spring. Females usually mature at age 3 and 4 and males at age 2 and average fecundity is about 500,000 eggs per female. Eggs are demersal and hatch in about 15 days, depending upon water temperature. The larval stage lasts ab(5ut 2 months, also depending upon water temperature. Small larvae are planktonic and remain in natal estuaries to a great extent, although some may be carried out into open waters by tidal currents. Some of these larvae may return to the estuary on subsequent incoming tides, but the rest are lost from the system. larger larvae maintain some control over their position by vertical move- ments and also may spend considerable time on the bottom. Following metamorphosis, most demersal juveniles remain in the estuary in shallow waters. Immature yearling (age 1) winter flounder become photoncgative and though many remain within the estuary, are usually found in deeper water than age 0 young-of-the-year (Pearcy 1962; McCracken 1963). Many adults stay in estuaries following spawning, while others disperse into deeper waters. By summer, most have left shallow waters as their preferred temperature range is 12-15''C (McCracken 1963). However, some re- main inshore and may escape temperatures above 22.5°C by burying themselves in cooler bottom sediments (Olla et al. 1969). Adults are omnivores and as opportunistic feeders eat a wide variety of algae and benthic invertebrates. They are sight feeders and are usually active only during the day. Additional details regarding winter flounder hfe hi.story, physiology, behavior, and population dy- namics may be found in Klein-MacPhce (1978). Because winter flounder stocks are localized, our studies have focused on the population spawn- ing in the Niantic River to determine if MNPS impacts of impingement and entrainment have caused or would cause changes in local abundance beyond those expected from natural variation. Operation of Unit 3 approximately doubled the volume of cooling water used by MNPS and some impacts could be expected to increase com- mensurately. In assessing the impact of winter flounder larval entrainment, our efforts have been aimed at estimating the number of individuals annually entrained and measuring changes at the adult population level. The approach chosen to carry out these two basic tasks consisted of a combination of sampling programs and analytical methods designed to provide a preliminary short- term assessment capability and, ultimately, a long- term assessment tool in the form of a compre- hensive simulation model, which includes hydro- dynamics and population dynamics submodels. Short-term empirical assessment methods (Ilorst 1975; Goodyear 1978) and a preliminary Winter Flounder Studies 149 deterministic population model (I less et al. 1975; Saila 1976) were used when, only 3 to 4 years into the winter flounder research program, com- plete data specific to the local winter flounder population were not yet available (NUSCO 1983). In recent years, the work has increasingly been directed towards more specific and detailed studies of earl> life history and on the critical stock- recruitment relationship. This has been reflected by studies designed to address specific concerns of NU and the Connecticut Department of En- vironmental Protection (CT DEP) (NUSCO 1988a, 1988b). The information will be incorpo- rated into the models currently under develop- ment, which should be available for use within about a year. This report discusses the results of ongoing winter flounder research and provides brief descriptions of the models that will be used for impact assessment. Materials and Methods Adult and juvenile studies Abundance estimation of the Niantic River spawning population of aduh winter flounder has been based on mark and recapture methodologies and details concerning annual surveys from 1976 through 1986 are summari7,ed in NUSCO (1987). Since 1982, each survey started after ice-out in the river in mid to late February and ended in early April, when the proportion of reproductively active females decreased to less than 10% of all females examined for two consecutive weeks. The Niantic River was subdivided into a number of areas (stations) for each survey (I'ig. I); no sam- ples were taken outside of the navigational channel in the lower portion of the river since 1979 be- cause of an agreement with the liast I,yme- Waterford Shellfish Commission to protect bay scallop [Argopecten irradiam) habitat. Collections were made during 2 or 3 days each week and were usually allocated to a station according to the expected abundance of winter flounder with more tows taken tn areas where fish were most numerous. Winter flounder were captured with a 9.1-m otter trawl (6.4-mm bar mesh codend liner), which has usually been towed at a standardized distance of 0.55 km since 1983; tows of variable length were taken previously. The standard distance was chosen because it represented the maximum tow length at station 1 and because using the same tow length at all stations was expected to reduce variability in calculating catch-pcr-unit-effort (CPUE), used as an index of abundance. In 1987, tows one-half or two-thirds of this length were frequently taken in the upper river (mostly at station 51) to avoid overloading the trawl with algae and detritus. Tow time and numbers of fish from tows of shorter distances in 1987 were compared to those from tows of the standard distance (0.55 km) to determine if the catches were proportional to distance of tow for compu- tation of CPUE. Because catch data from station 2 were also used for the trawl monitoring program, hauls at that station were maintained at the stan- dard distance of 0.69 km for that sampling pro- gram. Mostly because of differences in tidal currents, wind, and amounts of extraneous material col- lected in the trawl, tow times for the standardized distances varied and were usually greater in the lower than in the upper river. For 1983-87, the mean duration for standardized tows at stations I and 2 was 15.7 min and at stations 4, 51, 52, 53, and 54 was 12.5 min. Tows from 1976-82, when tow distance varied, that had shorter or longer duration compared to the distribution of tow times from 1983-87, when tow distance was uniform, were excluded from data analyses and calculation of CPUE. For comparisons among years, all catches of winter flounder larger than 15 cm made during a 4-weck period from mid- March through early April were standardized to cither 15-min tows (stations 1 and 2) or l2-min tows (all other stations) and annual median CPUE values were determined. A 95% confidence in- terval was calculated for each median using a distribution-free method, where the order statistics for the upper and lower confidence limits were plus or minus the square root of the sample size from the order statistic describing the median 150 NIantIc River Fig. 1. Location of Niantic River adult and juvenile winter flounder sampling stations used from 1983 through 1987 (Snedecor and Cochran 1967). The CPUE of juveniles smaller than 1 5 cm taken in the Niantic River during the spawning season was uetermined in a manner similar to that for adults. The catch of fish from stations 1 and 2 (Fig. 1) was used, as in most years juveniles were scarce in the upper river. The winter flounder caught in each tow during the adult abundance survey were placed immedi- ately into water-filled containers aboard the survey vessel. At least 200 randomly selected fish were measured to the nearest 0.1 cm in total length during each week of the population abundance survey in all years. Since 1983, all winter flounder Winter Flounder Studies 151 larger than 20 cm have been measured and sexed. Non-measured fish were classified into various length and sex groupings, depending upon the year; at minimum, all fish caught can be classified as smaller or larger than 15 cm. Since 1977, the sex and reproductive condition of the larger winter flounder were determined either by observing eggs or milt or by the presence (males) or absence (females) of ctenii on the caudal peduncle scales of the left side (Smigielski 1975). Following mea- surement or classitication, all fish 15 (1977-82) or 20 (1983-87) cm or larger were marked with a number or letter made by a brass brand cooled in liquid nitrogen; the mark was changed weekly. Fish recaptured were noted and remarked with the brand designating the week of their recapture. In 1 976, fish were fin-clipped and recaptures were marked with a numbered spaglietti tag. The log- likelihood ratio test (G-test of Sokal and Rohlf 1969) was used to compare the proportions of winter flounder marked and recaptured in each category of sex, length interval, and station. The probability level chosen to reject the null hypoth- esis in this and other statistical tests ^ven in this section was p^0.05. Estimates of abundance of all winter flounder 1 5 or 20 cm and larger in the Niantic River during the spawning season were obtained from the mark and recapture data using the Jolly (1965) model. The actual computations were done using a com- puter program (Davies 1971) of Jolly's model. As a result of a comprehensive review of our mark-recapture methodology, a composite index of relative abundance was developed to describe the number of adult winter flounder spawning in the Niantic River each year (NUSCO 1986a, 1987). This index was computed by averaging the central weekly estimates of N (scaled as thou- sands) from the Jolly model, with the first £ind last estimates excluded when sufficient data were available. These Jolly estimates are less reliable and were eliminated from the computations in all years except when the number of values used would have been less than three. The standard error of N (a) was determined as: (i)7Var of(Ar,) -I- Var of(Arj) + Var o^Wj) (1) where N is the weekly Jolly estimate of population size Using observations of reproductive condition from 1981-87, probit analysis (SAS Institute Inc. 1985) was used to estimate the length at which 50% of all females were mature. An index of the number of females reproducing in the Niantic River each year since 1977 was created by esti- mating their abundance in each 1-cm length in- crement starting with 26 cm. Fecundity (annual egg production per female) was estimated from length-frequencies and a length-fecundity relation- ship determined for Niantic River winter flounder (NUSCO 1987). Annual average fecundity was determined from the sum of all individual egg production estimates divided by the number of spawning females for each year. A relative index of annual egg production (in billions) was found by summing all individual estimates. The catch of winter flounder taken in the trawl monitoring program from 1976 through 1987 (see Fish Ecology section for methods) was also used as an index of abundance. CPUE was expressed as the annual 5-mean (Pennington 1986; see the Delta Distribution section of this report for details concerning this abundance index) during a calen- dar year to facilitate comparisons with other re- gional abundance indices. For other analyses, length-frequency information was used to parti- tion the catch into size groups smaller and larger than 15 cm. Both the annual median CPUE of adult winter flounder taken in the Niantic River during the spawning season (1976-87) and the trawl moni- toring program annual 5-mean (1976-86) were 152 compared to other indices of winter flounder abundance to determine if similar trends were ap- parent. These included National Marine Fisheries Service (NMFS) statistics for Rhode Island and a University of Rhode Island (URI) long-term (1966-86) trawl survey in Narragansett Bay (M. Gibson, Rhode Island Department of Environ- mental Management, pers. comm.), and Connecticut-licensed trawler catches from 1979 through 1987 (E. Smith, CT DEP, pers. comm.). The NMFS statistics were CFUE expressed as the number of pounds of winter flounder per 50-ton unit for Rhode Island vessels (1947-87) and as the number of pounds per directed trip in Statistical Area 539 (Narragansett Bay and Block and Rhode Island Sounds), available from 1964 through 1987. The Connecticut commercial CPUE was calculated as the number of pounds per hour of trawling. The URI data are annual 5-means determined from monthly sums of winter flounder taken during weekly sampling by trawl. Due to an apparent preponderance of juveniles in the latter time-series (Gibson 1987), the annual CPUE values were lagged 1 year before making comparisons. Additional information concerning the URI time-series may be found in Jeffries and Terceiro (1985) and Jeffries et al. (1986). Abun- dance indices were correlated using the nonparametric Spearman rank-order correlation. Historical records of winter flounder catches given by Perhnutter (1947) and NMFS (1987) were also examined. The annual abundance data for the Niantic River spawning stock from 1976 through 1987 were used to develop a stock and recruitment model. All fish 21 .4 cm and larger were considered to be adults (age 3 and older). Annual median CPUE values of winter flounder 15 cm and larger were adjusted to account for fish larger than this minimum length to estimate relative parental stock size for each year. Similarly, the CPUE were partitioned to determine a value for age 3 (21.4-28.0 cm) fish each year. The latter value was multiplied by 1.938 in order to account for life-time reproductive contribution and to scale the values into similar units for comparison with parental stock size estimates. This scaling factor was determined from estimates of maturity and mortality (NUSCO 1987) and an assumed max- imum life span of 12 years for Niantic River win- ter flounder. The adjusted CPUE values of the age 3 recruits were plotted against the parental stock size CPUE for 3 years prior. As a dome- shaped recruitment curve with reduced recruit- ment at high parental stock sizes was suggested by this procedure, a Ricker (1954, 1975) stock and recruitment model was fit to these data using nonlinear regression methods (SAS Institute Inc. 1985) where: R = aPexpC-pP) (2) where R = CPUE index of the number of age 3 recruits P = CPUE index of parental stock size a = dimensionless parameter associated with density-independent mortality P = parameter describing the instantane- ous rate of density-dependent mor- tality with dimension of 1/P Water temperature, believed to be an important environmental variable in the early life history of winter flounder, was considered as an additional parameter in order to improve the fit of the ob- served data to the model. Continuous water tem- perature records were available from the intakes of MNPS Units 1 and 2 and mean water temper- atures during 1976 through 1987 were found for individual and for various combinations of months encompassing the January through May spawning and larval recruitment period. Means over an 1 1-year (1976-86) period were also determined for the same months and groups. For each time period, deviations from the long-term mean were computed and compared (Spearman rank- correlation) with annual estimated age 3 recruit- ment. Water temperature, expressed as a positive or negative deviation from the long-term mean, having the highest correlation with recruitment was used as a third parameter in the Ricker model: R = aPexp(-pp)exp(-(pT) (3) Winter Flounder Studies 153 where T = annual temperature deviation from a long-term mean (p = dimensionless parameter that relates log-recruitment to annual water tem- perature deviations Larval studies The history and development of the current ichthyoplankton sampling programs were given in NUSCO (1987). Samples for winter flounder larvae in 1986 and 1987 were taken in Niantic River at stations A, B, and C and in Niantic Bay at EN (entraiimient sampling) and NB (Fig. 2). Fig. 2. Location of stations sampled for larval winter flounder Collections in the river and at NB were made with a 60-cm bongo sampler with 3.3-m long nets towed at approximately 2 knots and weighted with a 28.2-kg oceanographic depressor. Volume of water filtered was determined using a single General Oceanics (GO) flowmeter (model 2030) mounted in the center of each bongo opening. A stepwise oblique tow pattern was used with equal sampling time at surface, mid-depth, and near bottom. The length of tow line necessary to sample the mid-water and bottom strata was based on water depth and the tow line angle as measured with an inclinometer. Winter flounder larvae entrained by MNPS were collected at Units 1 and 2 discharge (station EN) using a gantry system to deploy a 1.0 x 3.6-m plankton net. Four GO flowmeters were positioned in the mouth of the net to account for horizontal and vertical flow variation; sample volume was deter- 154 mined by averaging the four volume estimates from the flowmeters. All sampling at EN was conducted with 333-[im mesh nets. On the bongo sampler, 202-[im mesh nets were used from February through the first week of April and 333-|im mesh nets were used during the remainder of the season. The bongo sampler was towed for 6 min at station A, B, and C (filtering about 120 m ) and for 15 min at station NB (filtering about 300 m ). Generally, the net was deployed at EN for 5 to 6 min (fil- tering about 400 m ), but this varied depending upon plant operations (number of circulating pumps). All ichthyoplankton samples were pre- served with 10% formalin. At the three river stations, jellyfish medusae were sieved (1-cm mesh) from the sample and measured volumetri- cally (ml). During the occurrence of larval winter flounder in 1986 and 1987, sampling time and frequency varied with station, season and year. At EN, collections were taken during the day and night (two replicates during each) once a week in Feb- ruary and June and similarly during 4 days each week from March through May. Single bongo tows were made at NB biweekly (day and night) in February and March, and weekly in April through the end of the larval winter flounder sea- son, except that weekly day and night samples were collected in March 1987. Preliminary tows were made during the daytime in February in the upper portion of the Niantic River (ice permitting) to determine when larval winter flounder were present. From the first larval occurrence through the first week of April, single daytime tows at each station were made twice weekly within an hour of low slack tide. During the last 3 weeks of April, single day and nighttime bongo tows were made twice weekly. The day samples were collected within an hour of low slack tide and the niglit samples during the second half of a flood tide. During the remainder of the seison until the disappearance of larvae at each station, tows were made twice a week only at night during the second half of a flood tide. This varying sampling scheme, based on information from previous sam- pling, was designed to increase efficiency in data gathering and reduce sampling biases (NUSCO 1987). Only one of the replicates from a bongo tow or entrainment collection was processed in the laboratory. Samples were split to at least one-half volume and larvae were identified and counted using a dissecting microscope. Up to 50 randomly selected winter flounder larvae were measured to the nearest 0. 1 mm in standard length (snout tip to notochord tip). The developmental stage of each measured larva was recorded using the fol- lowing identification criteria: Stage 1 . The yolk-sac was present or the eyes were not pigmented (yolk- sac larvae) Stage 2. The eyes were pigmented, no yolk-sac was present, and no fin ray development Stage 3. Fin rays were present, but the left eye had not migrated to the mid-line Stage 4. The left eye had reached the mid-line, but juvenile character- istics were not present Stage 5. Transformation to juvenile was complete and intense pigmen- tation was present near the caudal fm base Larval data analyses were based on standardized densities per 500 m of water sampled. An av- erage of weekly densities was used in analyses because weekly sampling frequencies varied due to sampling design and weather conditions. The geometric mean was chosen because these data generally followed a lognormal distribution (see the Delta Distribution section of this report). Previously, weekly means were determined arith- metically and the results reported in NUSCO (1987) may differ from those herein. For com- parisons of river and bay data and also for some Winter Flounder Studies 155 previous years, daytime samples after April were excluded, except for estimating larval entrainment with station EN data. During this period, daytime samples were not collected in the river because these samples underestimated abundance due to the fact that older larvae apparently remained near the bottom during the day and were not suscep- tible to the bongo sampler (NUSCO 1987). Typically, the temporal distribution of larval abundance was skewed, with a rapid increase to a maximum followed by a slower decline. This skewed density distribution results in a sigmoid- shaped cumulative distribution where the time of peak abundance is the time at which the inflection point of the sigmoid occurs. The Gompertz growth function (Draper and Smith 1981) was chosen to describe the cumulative distribution of the abundance data because the inflection point of this function is not constrained to the central point of the sigmoid curve. The form of the Gompertz function used was: (abundance curve). This density function has the form: C, = a(exp[ - Pe "']) (4) where Q = cumulative density at time t a = total or asymptotic cumulative den- sity P = location parameter K = shape parameter t = time in days since February 15 The origin of the time scale for our data was set to the 15th of February, which is when winter flounder larvae generally fu-st appear in the Niantic River. Least-squares estimates and asymptotic 95% confidence intervals for these parameters were obtained by fitting the above equation to the cumulative abundance data (based on the weekly geometric means) using nonlinear regres- sion methods (SAS Institute 1985). The derivative of the Gompertz function with respect to time yields a "density" function which directly describes the larval abundance over time 4 = aPK(exp(-K?{-|3e "'}]) where d, = density at time t (5) and all the other parameters are the same as in Equation 4, except for a, which was rescaled by a factor of 7 because the cumulative densities were based on weekly geometric means and thus accounted for a 7-day period. Time of peak abundance was estimated as the date tj corre- sponding to the inflection point of the function defmed by its parameters P and k as: (log«P) (6) The a parameter was used as an index to compare annual abundances. The k parameter was used to compare the steepness of the abundance curve (Equation 5), where k increases as the peak of the curve increases. Winter flounder larvae were reared in the lab- oratory during 1986 to determine growth rates at various temperature regimes. Eggs were stripped from a female and fertilized with milt from two males. larvae that hatched within 24 hours of each other were placed in 39- L aquaria held in a water bath. The water temperature in each regime was gradually increased during the holding period to mimic the seasonal temperature increase during larval winter flounder development. Photoperiod was similar to natural conditions. Larvae were fed rotifers {Brachionus plicatilus) and brine shrimp nauplii {Artemia salina) ad libitum. Known-age larvae were routinely sacrificed and measured to the nearest 0.1 -mm standard length to obtain in- formation on growth rate. For comparisons with other laboratory growth studies on larval winter flounder, length was converted to weight (|ig) by the length- weight relationship of Laurence (1979): weight = 0.045( length) (7) 156 and daily specific growth rates (SGR) were deter- mined by: (log w^ - loggwr,) SGR = 100 ,; _ , , (8) with t\ and f2 "= first and last days of observation wt\ and u'?2 = weight at days t\ and tj, respectively Annual entrainment estimates were calculated from data collected at station EN in addition to using these data to describe the abundance of winter flounder in Niantic Bay. The estimates were computed as the median density (number per 500 m ) during the larval season times the total number of 500 m units of seawater with- drawn by MNPS during the same period of time. A nonparametric method (Snedecor and Cochran 1967) was used to construct a 95% confidence interval around each median and corresponding entrainment estimate. Post-larval young-of-the-year studies A quantitative study of post -larval young- of-the-year winter flounder in the Niantic River began in 1983 (NUSCO 1987). Station LR has been sampled every year and WA since late 1984 (Fig. 1). Each station was visited once every week from late May through late September or early October during daylight within about 2 hours before to 1 hour after high tide. A 1-m beam trawl with interchangeable nets of 0.8- , 1.6-, 3.2-, and 6.4-mm bar mesh was used to catch young winter flounder. Two tickler chains were added in late June of 1983 to increase catch effi- ciency as older and larger young apparently were able to avoid the net without them (NUSCO 1987). In 1983, triplicate tows were made using one of the nets, which was changed as young grew during the season. Since 1984, two nets of successively larger mesh were used during each sampling trip to collect the entire available size range of young. A change to the next larger mesh in the four-net sequence was made when young had grown enough to become susceptible to it; the larger meshes reduced the amount of detritus and algae retained. Two replicates with each of the two nets were made at both stations and the nets were deployed in a random order. Distance was estimated by letting out a measured line at- tached to a lead weight as the net was towed at about 25 m per min. Tow length was increased from 50 to 75 to 100 m as the number of fish decreased throughout the summer of each year. For data analysis and calculation of CPUE, the catch of both nets used at each station was summed and standardized to give a density per 100 m of bottom covered by the beam trawl. For comparisons among years, a moving average of three weekly density estimates was used to smooth the trends in abundance over time. The young winter flounder collected were mea- sured in the field or laboratory to the nearest 0.5 mm in total length (TL). During the first few weeks of study, standard length (SL) was also measured because many of the specimens had damaged caudal fin rays and total length could not be taken. A relationship between the two lengths determined by a functional regression (NUSCO 1987) was used to convert SL to TL whenever necessary. To calculate mortality rate, aU young were as- sumed to comprise a single cohort. A catch curve was constructed with the natural logarithm of density plotted against time in weeks. The slope of the descending portion of the curve provided an estimate of the weekly rate of instantaneous mortality (Z). Once Z was determined, weekly survival rate (S) was estimated as exp(-Z) and monthly as exp((-Z)(30.4/7)). Impingement of winter flounder at MNPS The number of winter flounder impinged on the traveling screens of MNPS Unit 2 was esti- mated using techniques described in the Fish Ecology section of this report. Length-frequency Winter Flounder Studies 157 TABLE 1. Annual mark and recapture data from Niantic River adult winter flounder abundance studies during the spawning season from 1976 through 1987. Dates sam- Number of Number Number % Year pled weeks marked recaptured recaptured 1976 March 1 - April 13 7 6,479 453 7.0 1977 March 7 - April 12 6 3.737 257 6.9 1978 March 6 - April 25 8 4,417 360 8.2 1979 March 12 - April 17 6 4,067 241 5.9 1980 March 17 - April 15 5 4,313 433 10.0 1981 March 2 - April 14 7 6,726 469 7.0 1982 February 22 - April 6 7 5,795 270 4.7 1983 February 21 - April 6 7 5,196 363 7.0 1984 February 14 - April 4 8 3,740 197 5.3 1985 February 27 - April 10 7 3,024 170 5.6 1986 February 24 - April 8 7 2,790 175 6.3 1987 March 9 - April 9 5 2,334 133 5.7 Minimum size for marking was 15 cm during 1976-82 and 20 cm thereafter. data of impinged fish were also available. Routine impingement monitoring was discontinued in mid- December 1987, upon agreement between NU and CT DEP. Results and Discussion Adult and juvenile studies Abundance of winter flounder near Millstone The Niantic River winter flounder population is deraographically open and therefore subject to immigration, emigration, natural death, and re- moval by fishermen (White et al. 1982). Mark and recapture surveys designed to estimate abun- dance of open populations using the stochastic model of Jolly began in 1976 (NUSCO 1987). The Jolly model is an extremely powerful general formula that uses all the information provided by the mark and recapture experiment and provides the most efficient abundance estimates for open populations as long as basic assumptions are ap- proximately met (Cormack 1968; Southwood 1978; Begon 1979). Application of the Jolly model to the Niantic River winter flounder pop- ulation was discussed previously in NUSCO (1986a, 1987). The 1987 survey had the latest start since 1980 due to extended ice cover in the river, which re- sulted in only 5 weeks of sampling (Table 1). Although the percentage of recaptures (5.7%) was similar to the range observed during 1982-86 (4.7-7.0%), fewer fish were marked than in any previous year. A condition peculiar to 1987 was the large amount of kelp and detritus found in the mid to lower river chaimel (stations 2 and 4), which entirely precluded sampling at the latter station after the first week of study. During most of the year, few winter flounder were found in the navigation channel (stations 1 and 2) and catch in the basin of the upper river (station 51) was also less than in previous years; most fish were in the western arm of the river (stations 52-54). However, during the fourth week of the survey, many fish withdrew from the upper river arm into the basin following a storm. For the 1987 survey, log-likelihood ratio tests indicated no significant differences in the propor- tions of marked and recaptured fish classified by 158 sex or length. The percentages of fish recaptured at stations 51 (9.6%) and 52 (8.1%) were signif- icantly greater than those at the other stations (2.5-3.9%). In most other years when significant differences were found, greatest percentages of re- captures were from the lower river stations ( 1 or 2), most likely because any marked winter flounder moving out of the river would have had a greater probability of being caught near its mouth. TABLE 2. Weekly catch data used for estimating the Jolly index of winter flounder abundance during the spawning season in the Niantic River. Week no. Date- week of Total catch No. unmarked No. marked No. removed No. examined Recaps. 1981-86 Recaptures week marked) Total recaps 1 2 3 4 5 1 3-9 990 269 720 1 721 37 - 2 3-16 686 231 455 0 455 19 22 22 3 3-23 804 200 604 0 604 26 21 13 34 4 3-30 941 385 555 1 559 27 13 13 22 - 48 5 4-6 661 661 - 0 372 14 3 2 7 17 - 29 Total 4,082 1,746 2,334 2 2,711 123 59 28 29 17 - 133 Using the methodology previously described, annual .lolly composite abundance indices were calculated to describe relative abundance of winter flounder in the Niantic River during the spawoiing season. The weekly catch data (Table 2) were used with the Jolly model and the computed 1987 index of abundance was 10.0 ± 3.8 (Table 3). This represents a slight increase over the 1986 TABLE 3. The Jolly index of abundance for winter flounder larger than 20 cm during the 1987 spawning season in the Niantic River. Date- week of Jolly estimate (N) Standard error of N Probability of survival Standard error of15 cm) -i- ~Y 79 80 Fig. 3. Jolly index of abundance (±2 standard errors) for Niantic River winter flounder larger than 15 cm from 1976 through 1982 and larger than 20 cm from 1983 through 1987. For comparisons, the CPUE index w^s adjusted upwards by adding fish between 15 and 20 cm during 1983-87 (shown by a *). 160 within ± 50% of actual values. As population size decreases, as it has in recent years, accuracy decreases and abundance estimates may become negatively biased. A second, and perhaps less biased, measure of abundance for winter flounder was the CPUE during a 4-week period from mid-March through early April, the only period which included com- parable data for all annual surveys. The median CPUE was used as the most appropriate catch statistic because the trawl catch data were not normally distributed and were positively skewed. A 5-mean CPUE was used as an index of abun- dance for other trawl data sets where the data series had zero observations. However, the Niantic River abundance surveys had only a few tows without fish throughout the past 12 years. A comparison of the two indices showed that they were highly correlated for winter flounder both larger and smaller than 15 cm (Spearman rank-correlation, r = 0.96 and 0.98, respectively). Therefore, the median CPUE was retained as the best measure of abundance for the Niantic River spawning stock. The 1987 median CPUE of 13.7 was similar to the 1986 value of 12.0 (Table 5; Fig. 4). Mean tow duration and catch of winter flounder from tows of one-half (n = 2; 6.6 min; 6.0 fish) and two-thirds (n = 79; 7.8 min; 1 1.3 fish) of the stan- dard tow were reasonably proportional to those for the regular distance of 0.55 km (n= 137; 12.1 min; 18.2 fish) and shorter tow length did not affect the calculation of median trawl CPUE in 1987. Annual trends in median CPUE generally corresponded with the JoUy composite index of abundance until 1982. The CPUE in 1982 (42.6) was nearly the same as in 1981 (43.4), but the Jolly abundance index increased 72% (28.7 to 49.4). However, the latter 1982 estimate had a large confidence interval (± 16.7). The decline in CPUE for fish larger than 20 cm from 1983 (22.1) to 1984 (12.8) and 1985 (12.6) was greater than for the Jolly abundance index (29.9 to 29.3 and 21.6). The Jolly index for 1986 (8.3) decreased more than 60% relative to 1985, but the CPUE decreased by only about 20%. Both the JoUy index and CPUE showed similar increases (22%i and 14%, respectively) in 1987. The CPUE for 1984-87 indicated population levels about one-half of those during 1976-80, which also contradicted the Jolly abundance indices. These differences between abundance indices and biases of both were discussed at length in NUSCO (1987). TABI/E 5. Median CPUE of Niantic River winter flounder larger than 15 cm from 1976 through 1987 during the period of mid-March through mid-April. Total tows Tows used % of tows Median 95% confidence Coeff. of Year made for CPUE used CPUE interval skewness^ 1976 112 85 76 28.0 22.5-37.0 2.33 1977 154 123 80 24.0 20.0-30.0 1.45 1978 106 88 83 19.6 16.2-25.0 1.18 1979 93 77 83 26.8 22.4-38.4 1.67 1980 112 91 81 31.5 26.1-42.5 1.54 1981 109 97 89 43.4 36.2-51.4 1.24 1982 90 87 97 42.6 35.2-48.8 1.13 1983 135 134 99 30.8 24.1-33.9 0.96 1984 145 143 99 15.0 13.6-16.6 1.48 1985 156 155 99 14.7 12.7-15.0 1.13 1986 179 173 97 12.0 10.6-14.6 1.25 1987 198 197 99 13.7 12.4-15.3 0.59 Zero for symmetrically distributed data. Winter Flounder Studies 161 70 60 50- 40 30- 20 10 CPUE (>15 cm) 76 77 78 79 80 n 82 YEAR 83 84 85 86 87 Fig. 4. Median trawl CPUE (±2 standard errors) for Niantic River winter flounder larger than 15 cm from 1976 through 1987. For comparisons with the Jolly index, the CPUE was adjusted downwards by deleting fish between 15 and 20 cm during 1983-87 (shown by a *). The median CPUE of winter flounder smaller than 15 cm. was calculated for fish taken during the adult winter flounder surveys in the Niantic River from 1976 through 1987. Nearly all of the fish in this size group were age 1 yearlings and represented the year-class spawned during the pre- vious year. Data were restricted to the mid- March to mid-April period for comparability among years and to stations 1 and 2 because small winter flounder were less abundant than adults in the upper river. Inclusion of data from upper river stations could have biased inter-year comparisons because few or no tows were made there prior to 1981. Juvenile catches were more variable than those of adults (Table 6). Abundance reached a peak during 1981 through 1983 (50.1-87.2), but fell to previous levels in 1984 (16) and 1985 (27.7). CPUE declined to 3.6 in 1986 and 5.5 in 1987. The small numbers of juveniles suggested poor reproductive success in recent years. However, this measure of winter flounder abundance may be affected by differential distribution of the ju- veniles. Unlike adults, juveniles do not necessarily enter the river during the spawning season and other factors, such as water temperature, may in- fluence their movements. Although temperature is an important factor in winter flounder distri- bution, annual differences in late winter through early spring water temperature were not signifi- cantly correlated with abundance inside or outside of the river. During the past several years, juve- niles have been found in greater numbers through- out the entire river. As their distribution in area increased, concentrations in the lower river most likely decreased. This confounded the use of an abundance index based on tows from only the lower river channel stations. Using data from the trawl monitoring program, a comparison between the number of juveniles inside and outside the river during the spawning period was made and is presented below. 162 TABLE 6. Median CPUE of Niantic River winter flounder smaller than 15 cm from 1976 through 1987 during the period of mid-March through mid-April (stations 1 and 2 only). Total tows Tows used % of tows Median 95% confidence Coeff. of Year made for CPUE used CPUE interval skewness^ 1976 80 64 80 18.0 13.5-25.0 0.81 1977 143 116 81 25.5 18.0-30.9 1.10 1978 100 84 84 16.1 10.2-25.0 1.81 1979 79 71 90 27.0 17.4-42.3 1.88 1980 101 90 90 48.7 33.8-60.0 1.14 1981 47 45 96 87.2 61.0-120.9 0.67 1982 39 39 100 61.0 46.5-86.3 1.00 1983 44 44 100 50.1 32.8-61.2 0.58 1984 41 41 100 16.0 9.9-20.4 2.71 1985 48 48 100 27.7 20.9-41.1 1.00 1986 37 35 95 3.6 2.5-8.7 1.51 1987 33 33 100 5.5 3.4-7.4 2.00 Zero for symmetrically distributed data. An annual 5-mean CPUE was computed for winter flounder of aU sizes taken throughout the year (January- December) at all stations sampled by the trawl monitoring program. This period represented a change from previous reports, where an October-September year was defined. Using a calendar year allowed for direct comparisons with other regional indices of abundsmce. Neither period had particularly strong biological meaning; the winter flounder was ubiquitous in the Mill- stone area and made up about 40% of the catch (ranked first) in the trawl monitoring program. The 5-mean CPUE index showed a pattern of fluctuating abundance (Fig. 5). However, it dif- fered in several respects from the median CPUE for the Niantic River spawning stock. The peak in 5-means persisted from 1979 through 1983 and was not as pronounced as it was for the Niantic River medians, which were highest in 1981 and 1982 (Fig. 4). Abundance, as measured by the 5-mean for 1985 and 1986 was greater than that for 1977 and 1978, whereas the median CPUE for the Niantic River spawning stock in recent years has been smaller than that during the 1970s. Although the trawl monitoring program catch included winter flounder of all sizes, fish larger than 15 cm made up about two-thirds of the catch; annual percentages ranged from 55% in 1983 to 75% in 1976. On a monthly basis, larger fish comprised 76 to 92% of the catch from June through November. Small ( < 1 5 cm) fish made up two-thirds of the area-wide total from January through March as larger fish congregated on the spawning grounds. Approximately equal numbers of small and large fish were taken in April, May, and December. Based on elcctrophoretic studies, there was most likely a mixture of stocks present at the trawl stations outside of the Niantic River, except for the spawning season (NUSCO 1987). A comparison of the trends in abundance of small and large winter flounder taken by the trawl monitoring program during January through April (except station NR) was made with the catch of similar-sized fish during the annual population surveys in the river. This period was chosen because it overlapped the spawning period and the Niantic River station was eliminated because catches in the river were used for the determination of the winter flounder survey median. On the basis of the overlapping 95% confidence intervals, there was little dtfierence among the numbers of larger winter flounder at stations outside the river; catches remained at a low and stable level of Winter Flounder Studies 163 50 45 40 35 30- 25- 20 15 10 5 OH 76 77 78 79 80 81 82 83 YEAR (JANUARY-DECEMBER) 84 85 86 Fig. 5. Annual 8-mean CI'UE (±2 standard errors) for winter flounder taken by the trawl monitoring program (TMP) from January 1976 through December 1987. about 5 fish a tow (Fig. 6). This was in contrast to the large fluctuations in abundance seen within the river. Catch of juveniles (< 15 cm) in winter and early spring also fluctuated less outside than inside the river. As the number of small fish in the river declined to low levels in 1986 and 1987, the num- ber outside increased in respect to 1984 and 1985 and was at levels of abundance seen from 1976 through 1982 (Fig. 7). This was an important finding, as the very low abundance of juvenile fish in the river in recent years seemingly pointed towards continued declines in an already reduced adult stock. However, the greater catches in the much larger area outside of the river suggested that the year-class strength for 1986 and 1987 in the general area may not have been as low as catches during the winter flounder survey would have indicated, assuming that many of these fish were produced in the Niantic River. The differ- ential distribution and abundance of juveniles during the trawl surveys has made it difficult to predict future adult population size in the Mill- stone area. Regional and historical trends in abundance The abundances of winter flounder for the Niantic River population surveys and from the trawl monitoring program was compared to other regional indices (Table 7). With a few exceptions, most indices were significantly correlated and thus described real trends in abundance that occurred throughout Southern New England. One excep- tion was the previously mentioned lack of corre- spondence between the two Millstone series. However, the Niantic River adult median CPUE was significantly correlated with several other measures of adult stock size. These included two commercial fishing CPUE indices for Rhode Is- land and one for Connecticut (Fig. 8) In addition, the Niantic River annual medians were correlated with the URI trawl survey annual 5-mean, lagged 164 60 50 40 30 20 10 CPUE of winter flounder > 1 5 cm TMP (5-MEAN 76 77 78 79 80 81 82 83 YEAR (JANUARY-APRIL) 84 85 86 87 Fig. 6. Comparison or annual January-April 5-mean CPUF. for the trawl monitoring program CrMP) with the Niantic River survey median (WPS) for winter flounder > 1 5 cm from 1976 through 1987. 125 100 75 50 25 CPUE of winter flounder < 15 cm TMP (5-MEAN T 1 1 1 1 1 1 1 1 1 1 76 77 78 79 80 81 82 83 84 85 86 87 YEAR (JANUARY-APRIL) f'ig. 7. Comparison of annual January-April 8-mean CPUli for the trawl monitoring program (TMP) with the Niantic River survey median (WFS) for winter flounder < 15 cm from 1976 through 1987. Winter Flounder Studies 165 TABLE 7. Matrix of Spearman's rank-order correlations for various Southern New England winter flounder abundance indices from 1964 through 1987. NMFS NMFS Connecticut URI trawl Millstone trawl Source 'Red Book' (1964-87)^ Area 539 (1964-87)*' DEP (1979-87)'^ survey (1966-86)'^ survey (1976-86)^ Niantic River 0.6853 ^ 0.7552 0.8034 0.6224 0.5000 adult survey 0.0139 ♦ 0.0045 •• 0.0091 ♦• 0.0307 * 0.1173 NS (1976-86)*' 12 12 9 12 11 NMFS 0.7332 0.5858 0.6844 0.7546 'Red Book" 0.0001 ** 0.0974 NS 0.0006 ** 0.0073 ** landings 24 9 21 11 NMFS 0.7364 0.8347 0.8000 Area 539 0.0237 • 0.0001 ♦* 0.0031 •• landings 9 21 11 Connecticut 0.5105 0.5749 DEP commercial 0.1603 NS 0.1361 NS trawl landings 9 8 University of 0.6364 Rhode Island 0.0353 * trawl survey 8 National Marine Fisheries Service 'Red Book' commercial landings annual CPUE for Rhode Island (lbs/50-ton unit). National Marine Fisheries Service annual CPUE for statistical area 539 (lbs/directed trip). Connecticut DEP-licensed commercial trawler CPUE (Ibs/lrawl-hr). University of Rhode Island annual 8-mean trawl CPUE (advanced 1 year for comparisons). NU trawl monitoring program annual 5-mean trawl CPUE for all winter flounder. Niantic River winter flounder survey median trawl CPUE for adult (> 15 cm) winter flounder. Shown for each Spearman rank-order correlation: correlation coefficient probability level, where NS - not significant, * - significant at p<0.05, ** - significant at p<0.01 number of observations 1 year. Similarly, the annual 5-means for the trawl monitoring program were significantly cor- related with the three Rhode Island indices. Since larger fish made up one-half to three-quaiters of the catch in the Millstone trawl survey, no lags were used for the comparisons. 166 130- 120- 110- llJ u 7' 100- < (-1 90- -1 m 80- < Lu O 70- X III 60- n ■ z 50- 1 ■ ^ 40- z ■ < 30- 20- 10 A. Lbs per trawl— hr for Connecticut— licensed vessels B. Lbs (X 10) per directed trip in Statistical Area 539 C. Median trawl CPUE of Niantic River flounder > 15 cm 76 77 78 79 80 82 83 84 85 86 87 YEAR Fig. 8. Comparison of annual winter flounder CPUF, for Connecticut-licensed trawlers, NMFS Statistical Area 539, and Niantic River abundance surveys from 1976 through 1987. The Rhode Island indices represented the long- est time-series of available data, with one set of statistics going back to 1947. These data illustrated the inherent variability typical for winter flounder abundance (Fig. 9). Numbers were relatively high from the mid-1950s through the early 1970s, with several sharp increases in commercial catches most likely related to the occurrence of particularly large year-classes. Abundance declined in the mid and late 1970s, but another large year-class was produced in 1978. This year-class, along with better-than-average recruitment in 1979 and 1980, resulted in the winter flounder abundance peak seen during the early 1980s. The recent declining trend has reduced winter flounder to levels at or below those found in the early 1950s and early 1970s. Perlmutter (1947) presented a brief history of the winter flounder fishery in New England and New York from its beginning in the late 1800s to the 1940s. His work was prompted by a de- cline in catches in the late 1930s and 1940s, com- pared to peak years of the fishery in 1928-31. He noted considerable fluctuation in commercial fish- ing CPUE from 1910 through 1947. For example, catch-per-fyke-net at both Boothbay Harbor, ME and Woods Hole, MA was very high during the startup of the fishery in the early part of this century and relatively high from 1925 through 1933. However, catches decreased about 30-40% during 1934 to 1940. Introduction of more effi- cient gear (change from fyke nets to beam trawls to otter trawls) and vessels (sail to engines) as well as increased market demand allowed for the full development of the fishery and increasing ex- ploitation of the stocks. The increase in relative fishing power has undoubtedly continued through the present with the addition of electronics and other fishing aids in response to market demand and high prices. Perlmutter (1947) also provided logbook data for a Connecticut trawler working in Fishers Is- land Sound. The catch-pcr-trawl-day (average daily hours fished not given) from 1930 through Winter Flounder Studies 167 175- Lbs (X (X 6- 1000) 10) pe -mean per r di 50 —ton unit fnr Rl hont?; Lbs 'ectf'f^ tfin in Arekn fi.-^Q (UR )/2 for — -• r — Narraaansett Bqv — — 150- 125- A 100: \ 75- [ / i /I / / ^ \ 1 1 / l\ 1 I 50: /^\ J U — / / / I fs ' / \ 1 ' 1 / ^ 11 i 25- n- y \ V'"^\--'' / A) 47 52 57 62 67 YEAR 72 77 82 87 Fig. 9. Comparison of annua! winter flounder CPUE for Rhode Island trawlers, NMFS Statistical Area 539, and the URI Narragansett Bay trawl survey from 1947 through 1987. 1941 ranged from 402 to 853 lb, with largest catches made in the early 1930s when winter flounder were most abundant. In recent years (1979-87), catch-per-trawl-day for all Connecticut- licensed trawlers was calculated (from CT DEP, unpublished data) to have been from 315 to 521 lbs (average of 5 hours fished per day). The highest daily rate occurred in 1983, when winter flounder were most available, and lowest rates were in 1986-87. Despite the recent decline in winter flounder abundance and commercial trawler CPUE, landings have remained relatively high in Connecticut (Fig. 10) because of greater effort, with increases seen in mean days fished, mean hours trawled, and mean hours per day of fishing (Table 8). In addition, an increasing pro- portion of the catch was sustained by fish taken in eastern Connecticut waters from the mouth of the Connecticut River to the Rhode Island border (Fig. 11). Commercial landings elsewhere in New Eng- land have recently decreased. Despite the declining resource, high landings in Massachusetts have been maintained by increasing effort and the num- ber and percentage of small fish landed (MDMF 1985). NMFS (1987) reported that in each of the three major stock assessment geographical ar- eas (Southern New England-Middle Atlantic, Gulf of Maine, Georges Bank) winter flounder abun- dance decreased to historical low levels in 1986, with declines in commercial landings generally re- flecting assessment survey CPUE indices. In conclusion, the fluctuations in abundance seen for Niantic River winter flounder have oc- curred concurrently with other populations in New England. Examination of long time-series of abundance data showed that winter flounder numbers have fluctuated throughout tliis century, with production of several large year-classes that resulted in periods of peak abundance. These events have been viewed by some as the result of favorable environmental conditions during periods of reproduction and early life history (.lefFries and Johnson 1974; McIIugh 1977; Jeffries and Terceiro 1985). 168 1000- 900- \ le 800 i o o o X. 700 c/i o 9 600 o 500 400 CONNECTICUT TRAWLER UXNDINGS 100 90 o 80 rn 70 T 60 ^- 50 79 80 81 82 83 84 85 86 87 YEAR FMg. 10. Comparison of Connecticut total annual winter flounder landings with trawler CPUE from 1979 through 1987. TABLE 8. Connecticut-licensed trawler commercial fishing statistics for winter flounder from 1979 through 1987 (derived from unpublished data provided by CT DEP). Year Total pounds landed Total days fished Hours trawled Average days fished Average hours trawled Average hours per day Pounds per trawl -hr Pounds per trawl -day 1979 768644 1644 7691 16.6 76.8 46 98.6 456 1980 489419 1411 5604 11.8 46.6 4.0 846 336 1981 568463 1311 5549 14.4 61.0 4.2 98.6 417 1982 401061 1024 4377 10.6 45.1 4.3 89.7 381 1983 911694 1736 10058 15.9 92.3 5.8 90.0 521 1984 944381 2184 12700 20.0 116.5 5.8 70.9 413 1985 912685 2089 12469 18.9 112.3 6.0 70.6 421 1986 657221 1185 1 11 88 20.1 119.0 5.9 57.2 340 1987 611428 1938 11993 20.6 127.6 6.2 50.2 315 Winter Flounder Studies 169 1000 900 800 700 600 o 500 400-1 CONNECTICUT TRAWLER UVNDINGS 50 45 51 o -n 40 — 1 o — 1 > 35 1 — 1 > ?=; 30 I'l 7- 25 LO 20 —i m -At _, 15 o — i 10 5 79 80 82 83 YEAR 84 85 86 87 Fig. 11. Comparison of the total annual winter flounder landings with the percentage taken in eastern Connecticut from 1979 through 1987. Reproduction The sex ratio of winter flounder larger than 20 era during the spawning season in the Niantic River ranged from 0.78 to 2.03 females for each male during the past 11 years (Table 9). The geometric mean was 1.33, but the last 2 years were the only ones in which more males than females were taken. This was unusual, based on past ratios in the river and reported ratios of 1.50 to 2.33 in favor of females by Saila (1962a, 1962b) cmd Howe and Coates (1975) for other populations in southern New England. The reduced number of females may have resulted from increased fish- ing pressure on females. They are larger than males and tend to move longer distances away from the Millstone area (NUSCO 1987), which could have increased their vulnerability to offshore commercial fisheries. TABLE 9. Female to male sex ratios of winter flounder taken during the spawning season in tlie Niantic River from 1977 through 1987. Geometric 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 mean All fish 1.03 2.23 1.37 2.66 1.42 1.16 1.52 1.07 1.37 0.92 0.78 1.32 captured Measured fish 1.26 1.95 1.21 2.03 1.61 1.50 1.52 1.07 1.37 0.92 0.78 1.33 >20 cm Female winter flounder can become sexually mature when they are age 3 or when about 20 cm in length; northerly populations mature at smaller sizes and older ages than in Southern New England (Dunn and Tyler 1969; Dunn 1970; Kennedy and Steele 1971; Beacham 1982). Re- 170 suits of a probit analysis showed that the length of 50% sexual maturation of Niantic River females during 1981-87 was 26.8 cm with a 95% confi- dence interval of 26.3 to 27.2 cm. On an aimual basis, values ranged from 25.1 cm in 1983 to 29.4 cm in 1981; most of these fish were 3 or 4 years old. Based on our observations, many males mature at even smaller sizes (10-12 cm) and younger ages (2 years) than females. Winter flounder spawning was followed by noting the weekly change in the percentage of gravid females larger than 25 cm in the Niantic River. Generally, most spawning was completed by late March or early April as relatively few gravid females remained in the river (Fig. 12). Ice in the river prevented starting population surveys earlier in January or February, so for most years approximately two-thirds of the females examined during late February or early March had spawned before sampling began. In most cases, spawning appeared to be correlated with water temperature. In relatively cold years, proportionately fewer fe- males spawned during the earlier portion of the survey, whereas in warmer years more were spent at the beginning of sampling (NUSCO 1987). The annual proportion of females larger than 26 cm was used with the Jolly index of abundance for that year to obtain a relative index of female spawners. Spawning females made up 20 to 51 % 60 1984 \ \ 50- \ \ \ 40- 1986 \ 1983 ^ \ 1985 -\ ^ 30- 20 ^-^^^^ 10- n- ^-=^=^^4^ FEBRUARY MARCH APRIL Fig. 12. Percentage of adult female winter flounder in 1983 through 1987. of the population during the past 1 1 years (Table 10). Percentages for 1977-82 were lower because they were based on abundance estimates for all fish larger than 15 cm and thus included more immature fish in the total. The value for 1980 (20%) was particularly low and was an indication that comparatively more small, immature fish were caught during that survey. spawning condition by week in the Niantic River from Annual average fecundity estimates were cal- culated using a length-fecundity relationship de- scribed in NUSCO (1987) with annual length- frequency data. Values have been relatively con- sistent with somewhat greater averages found since 1982, when the start of the surveys was advanced into February. During the past several years, most females larger than 40 cm were found Winter Flounder Studies 171 in the Niantic River early in the season and many left the estuary in March. Surveys during earlier years started after February and missed many of these large winter flounder, resulting in lower av- erages. The mean length of all females 20 cm and larger in 1980 was only 29.7 cm, in comparison to .31.4 to 32.1 cm for 1979 and 1981-85, respec- tively. This was additional evidence that fewer large females were sampled during the 1980 survey. In contrast, the means of 33.4 cm in 1986 and 32.8 cm in 1987 were particularly large, indicating that along with decreasing abundance, the female population included more larger and older speci- mens than in previous years. TABLE 10. Annual indices of female spawners and egg production for Niantic River winter flounder from 1977 through 1987. Year Relative index of spawning females^ % of population comprised by mature females Average fecundity (xlO^) Relative index of egg production 1977 5.6 31 4.8 2.69 1978 5.5 38 5.1 2.81 1979 4.8 26 5.3 2.55 1980 3.7 20 4.7 1.70 1981 10.7 37 5.3 5.66 1982 18.9 38 5.7 10.79 1983 15.2 51 5.6 8.51 1984 12.6 43 5.7 7.20 1985 10.1 47 5.9 5.92 1986 3.5 43 6.5 2.31 1987 4.0 40 6.1 2.46 From Jolly index of abundance and percentage of mature females, assuming that all females 26 cm and larger were mature. For winter flounder larger than 15 cm during 1976-82 and 20 cm thereafter. A relative index for year-to-year comparisons and not an absolute estimate of production. Relative egg production indices were deter- mined using .lolly abundance indices with the length, maturity, and fecundity data. Since the indices reflect both annual average fecundity and abundance, the value for 1980 was probably un- derestimated. The egg production index peaked in 1982 and declined about 80% since then as the increase in average fecundity was not large enough to olfset declines in abundance of adult females. However, adult abundance and absolute egg production alone were not necessarily the most important factors determining year-class strength (i.e., production of young in a particular year). This is discussed further in the following section. Stock and recruitment Although not overly long in comparison to data sets typically used in stock assessment studies, the 12 years of abundance estimates for the Niantic River winter flounder allowed for the in- vestigation of a relationship between adult stock SL7,e and the number of offspring they produced. This relationship, termed stock and recruitment, has been described in various forms (e.g., Ricker 19.54; Beverton and Holt 1957; Gushing 1973; Shepherd 1982). The parameter estimates of a stock and recruitment model may be used in other fisheries models (e.g., to predict future yields) or, more specifically for MNPS studies, in the stochastic population dynamics impact assessment 172 model. In the following discussion of the model, the recruitment index was determined as the frac- tion that age 3 (21.4-28.0 cm) fish made up of each annual Niantic River adult CPUE. These lengths represented the midpoints of the 10th and 90th percentiles of length between ages 2 and 3 and ages 3 and 4, respectively. The index of age 3 fish was multiplied by a scaling factor (1.938) reflecting their lifetime reproductive contribution (maximum life span was assumed to be 12 years). Age 3 fish were used as the best estimate of re- cruitment since a majority of them were mature and in the Nicintic River for spawning each year. This also extended the period of compensatory mortality throughout the entire period of imma- turity without restricting it to a particular larval or juvenile life stage. The index of parental stock producing the age 3 fish was the fraction of each annual CPUE made up by all adults age 3 and older (^21.4 cm) during the spawning season 3 yeais prior to the age 3 recruitment estimate. Jones (1982) noted that the largest stock abun- dances for a number of species were usually about three to six times that of the smallest ones and that the difference between them usually approx- imated the mean. Ursin (1982) reported that recruitment variation over 13 years for the North Sea plaice {Pleuronectes platessa), which is an European flatfish closely related to the winter flounder (Burton and Idler 1984), varied by a factor of five from the smallest to the largest year- class. Many other species, such as cod {Gadus morhud) and herring {Clupea harengus), also had ratios of this order. However, a few fishes (e.g., haddock, Melanogrammus aeglefmus) varied by a factor of 100 or more, suggesting the lack of a stabilizing mechanism for recruitment. For the Niantic River parental stock index, the range seen from 1976 through 1987 was 10.2 to 32.3 (differ- ence of 22.1) with a mean of 19.0. This suggested that our stock and recruitment data set probably included representative small and large stock sizes typical for this population. The two Rhode Is- land commercial fishing indices had ratios of abun- dance that were 4 and 1 1 and the URI series had a difference of 15. However, the former may have been influenced by economic and social fac- tors and the latter time-series included many ju- veniles, which, as illustrated by our data, can be quite variable in number and may not absolutely reflect true abundance. A plot of recruit versus parental indices sug- gested a dome-shaped curve with reduced recruit- ment at high levels of adult abundance. This is typical for fishes with high fecundity (Gushing 1971; Gushing and Harris 1973), although the reliability of this relationship will be ascertained as more data are collected. Because a dome was presumed for the stock-recruitment data, Ricker's (1954, 1975) model was fit to the data. His two- parameter (a and P) model did not explain much (R = 0.44) of the variability seen in recruitment (Fig. 13). For example, the 1978 and 1984 year- classes were remarkably different, although pro- duced by similar adult abundance. Numerous examples have been given where recruitment success of a species has been strongly linked with environmental factors (Gushing 1973, 1977; Sissenwine 1974, 1977, 1984; Roff 1981; Shepherd et al. 1984; Lorda and Grecco 1987). In particular, water temperature has been found to be inversely related to strong year-classes of winter flounder in Rhode Island (Jeflries and Johnson 1974; Jeffries and Terceiro 1985; Gibson 1987). Roff (1981) also noted that abnormally cold temperatures have been related to strong year-classes in several other winter-spawning flatfishes. To examine possible temperature ef- fects, yearly mean water temperatures during the winter flounder spawning and larval seasons (January- May) were calculated for single months and for various combinations of two or more months. For each month or group of months, deviations from the long-term mean were com- puted and compared to annual recruitment indices. The strongest negative correlation (r=-0.78) was found between February temperature deviations and recruitment indices (Fig. 14). Gorrelations between recruitment and March and February- March temperatures were also significant. Winter Flounder Studies 173 Q 15- R = 3.630 * P * exp(-0.093 * P) (R2 = 0 44) . 78 / /' ,•'79 /^ 7^^^^^77 / /' 76^ / / "" --. .82 / /' • ^'^ . 83 81^^-.^,^^^ 1/ 10 15 20 25 30 35 PARENT STOCK INDEX (P) Fig. 13. Slock-recruilmenl model (Iwo-paramclcr) Tor Nianlic River winter flouncler. 20 16- Q 12 . 77 78 r= -0.78 p = 0.01 79 80 76 . 82 . 81 Q\ , , ^ , , r- ^ ^ ^ . n , -2.25 -1.75 -1.25 -0.75 -0.25 0.25 0.75 ANNUAL FEBRUARY TEMPERATURE DEVIATION - 1.25 1.75 Fig. 1 4. Relationship between annual age 3 recruitment index and corresponding February water temperature deviations from a long-term mean. 174 Shepherd et al. (1984) and Sissenwine (1984) cautioned in the use of data exploration exercises not based on plausible a priori hypotheses and pointed out the risks of finding spurious correla- tions from empirical studies. Fundamentally dif- ferent empirical models may also be indistinguish- able because they account for virtually the same proportion of variability in recruitment. However, even without clear evidence for causal relation- ships, circumstantial evidence may be adequate and empirical models can serve as the basis for the creation and testing of future hypotheses (Garrod 1982; Shepherd et al. 1984). Definitive explanations do not yet exist to explain why Feb- ruary water temperatures were correlated with winter flounder recruitment. This period encom- passes much of the period of egg deposition, in- cubation, and hatching. Most likely, temperature is a surrogate for a complex set of physical and biological interactions. Winter flounder egg in- cubation time and hatching success (Scott 1929; Williams 1975; Rogers 1976) and larval growth (Laurence 1975; NUSCO 1987) were related to temperature. Temperature may also be correlated with other ecological factors affecting reproductive success. Most importantly, it probably affects the timing of the match of larval production with that of their prey and predators, thereby strongly influencing the success of a year-class (Gushing 1973). Accordingly, the difference between each an- nual February mean water temperature from the 1 1-year mean of 2.4°C was used as a third vari- able in the Ricker stock and recruitment model. The result was an improved fit (R^ = 0.78) to the observed data (Fig. 15). Estimated values of re- cruitment using the three-parameter model fol- lowed the predictions relatively closely (Fig. 16). February water temperatures helped to explain the previously mentioned diflerence between 1978 and 1984; the former year was among the coldest and the latter the warmest of the series. The 1978 year-class was also reported by Gibson (1987) to be exceptionally large in Rhode Island. This was not unexpected nor were the high correlations found among abundance indices for the Niantic River stock and others in the region. Climatic Q 15 R = 3.342 * P * exp(-0.012 * P) * exp(-0.197 * Temp) (R^ = 0.78) • = actual • ^^ y •» = predicted 15 20 25 30 PARENT STOCK INDEX (P) Fig. 1 5. Temperature-dependent stock -recruitment model (ihree-parameter) for Nianlic River winter flounder. Winter Flounder Studies 175 q: 10 79 80 81 YEAR-CU\SS Fig. 16. Nianlic River winter flounder rccruilment indices Tor the 1976-84 year-classes predicted by Ihe three-parameter mode! (line) compared to values actually observed (circles). influences on recruitment appear to be phenomena fairly pervasive and occurring over large areas and for a number of different species (Gushing 1973; Sissenwine 1984). The use of the temperature parameter (p in the model as well as the relatively large estimate of the a parameter implies that environmental and other density-independent pro- cesses are important factors in winter flounder reproduction as noted by Roff (1981) for several flatfishes. However, the density-dependent pa- rameter (P) of the model must operate to adjust year-class strength, as the relatively small varia- tions in recruitment suggest some stabilization mechanism. Given the relatively cold winters that occurred during the late 1970s (Fig. 1 7) along with moderate parental stock abundance, good to exceptional year-classes resulted in large parent stock sizes from 1981 through 1983. TTie relatively abundant adults coupled with mostly above-average tem- peratures through the present have resulted in the below-average recruitment and decreased winter flounder abundance presently seen. Recent warm winters suggest that a large increase in winter flounder will not occur in the near future. How- ever, when parental stock size and prevailing water temperatures for 1985-87 were used with the stock and recruitment relationship, moderately im- proved recruitment seems likely during 1988-90 as shown by the three predicted points (Fig. 15). Future values of age 3 CPUE may be compared to those predicted by the three-parameter model to examine its credibility. This would also address a criticism of Sissenwine (1984) in that empirical models often fail to predict post-publication events. However, as the number of data points used for the model was relatively small, the ad- dition of others in forthcoming years will likely change parameter estimates to an unknown degree and model reliability will likely improve. Finally, increasing knowledge of the reproductive process and early life history of the winter flounder may enable the formulation and testing of plausible hypotheses concerning stock and recruitment mechanisms and density-dependent mortality. 176 1.75- I 1.50 ^ 1.25 S 1.00 ° 0.75 3 0.50 ^ 0.25 ^ 0.00 g -0.25 ^ -0.50 [d -0.75 > -1.00 ^ -1.25 m -1-50 UJ U. -1.751 -2.00 -2.25 31 82 83 84 85 86 87 YEAR Fig. 17. Annual February water temperature deviations Trom the long-term mean from 1976 tlirough 1987 as determined from MNPS operating records. Larval studies Abundance and distribution Larval winter flounder abundance in 1986 and 1987 was examined in the Niantic River (stations A, B, and C combined) and Bay (stations EN and NB combined) using estimated abundance curves from the Gompertz function (Fig. 18). In the river, larvae were much more abundant in 1987 and in both years maximum densities oc- curred during late February through mid- March. The highest densities in the bay occurred later in April and the difference in abundance in the bay between the 2 years was not as great as it was in the river. This suggested that greater larval mor- tality may have occurred in 1987 compared to 1986 and that the mortality occurred early in the season while a majority of the larvae were in the river. A comparison of the abundances in 1986 and 1987 to 1983-85 was based on the a parameter from the Gompertz function, which was used as an index of abundance (Table 11). This parameter is actually an estimate of the area under the abun- dance curves presented above (Fig. 18). The Gompertz function fitted the cumulative abun- dance data well with R values exceeding 0.98. TABLE 11. Larval winter flounder abundances and 9.'5"(i confidence intervals for the Niantic River and Bay as estimated by the a parameter from the Gompertz function. Year Niantic River Niantic Bay 1983 1984 1985 1986 1987 1814 (1748-1879) 5077 (4940-5215) 11715 (11558-11871) 1677 (1629-1726) 5066 (4870-5262) " 2911 (2873-2950) 1823 (1742-1904) 1604 (1548-1660) 881 (843-920) 1279 (1234-1324) Winter Flounder Studies 177 RIVER 1987 13FEB 05MAR 25MAR UAPR 04MAY 24MAY 13JUN 03JUL DATE 300 250 200 150 100 50 BAY 15MAR 25MAR 04APR 14APR 24APR 04MAY 14K/IAY 24MAY 03JUN 13JUN 23JUN DATE Fig. 1 8. Eslimaled abundance curves (number per 500 m ) for larval winter flounder at Niantic River and Bay stations for 1986 and 1987. 178 Larvae were least numerous in 1986 in both the river and bay. The abundance of larvae in the river was similar in 1984 and 1987. Abundance in the river did not necessarily reflect that in the bay, as larvae were most abundant in the river in 1985, but were only found in moderate numbers in the bay that year. Comparisons of armual spatial abundance of the first four developmental stages were based on the cumulative weekly geometric means; because few Stage 5 larvae were collected, their abundance and distribution was not examined. At the three river stations (A, B, and C), Stage 1 abundance was lowest in 1986, except for 1983 (Fig. 19). The low abundance in 1983 was attributed, in part, to undersampling due to net extrusion (NUSCO 1987) and this was rectified in 1984 when a smaller 202-|im mesh net was used during the early portion of the larval season. Comparison among years at each river station showed a similar pattern in Stage 1 abundance with 1985 the high- est, followed by 1987, 1984, 1986, and 1983. Stage 1 larvae were rarely collected in Niantic Bay at station EN and NB, suggesting that little, if any, spawning occurred in the bay. By developmental Stage 2, larvae were more prevalent in the bay, but a majority were still collected in the river. Stage 2 larvae were least abundant in 1986 at all five stations. The order of annual abundance for Stage 2 larvae among years was generally the same for the three river stations, although 1987 dropped from second to fourth in rank. Most Stage 3 larvae were collected in the lower portion of the Niantic River (station C) and in the the bay (sta- tions EN and NB) with very few present in the upper river (station A). Abundance in 1986 was lowest at all stations, but the magnitude of the difference compared to other years was not as great. The pattern and relative abundance of Stage 3 larvae were similar among years at stations EN and NB. Stage 4 larvae were collected pri- marily in the lower river and in Niantic Bay. The low abundance ol other developmental stages in 1986 was not as apparent at Stage 4 of develop- ment. The large decline in abundance from Stage 3 to 4 in each year was probably related to less effective sampling for older larvae. By Stage 4 of development, the left eye has migrated to or past the mid-line and the larvae have become mostly demersal and thus were less susceptible to either the bongo sampler or entrainment at MNPS. The similar abundance of all stages at EN and NB, which are approximately 1 km apart, suggested a relatively uniform distribution of larvae through- out Niantic Bay. The decline in abundance of larvae as they passed through developmental stages was quite variable between years. For example. Stage 1 and 2 larvae were the most abundant in 1985 at all river stations compared to other years, but Stage 3 larvae were among the least abundant, implying that high mortality occurred during Stage 2 of development. In 1984, however. Stage 1 and 2 abundances were moderate and Stage 3 abundance at station C was the highest of the 5-year period, indicating low mortality during Stage 2. Although this variability could be attributed to the impre- cision of plankton sampling, the consistency in the relative ranking of years for Stages 1 and 2 at the three river stations and for Stage 3 at the two bay stations suggested that the precision in quan- tifying larval abundance was good. Also, the similar abundance of Stage 3 larvae at EN and NB each year implied that both techniques pro- vided comparable results, even though the sam- pling methods at these stations were different. Variability in the relative ranking of abundance from stage to stage among years was reported by Bannister et al. (1974) for egg and larval stages of the plaice, which they felt was consistent with density-dependent mortality. The lack of a pattern in the decline in abundance among years indicated that the processes that regulated larval winter flounder abundance were complex and operated at different levels from year to year. A comparison of the temporal occurrence of developmental stages was based on the date of peak abundance in the river and bay, which was estimated from the inflection point of the Gompertz function (Table 12). Because Stage 1 larvae were rarely collected in the bay, the dates of peak abundance could not be estimated for this area. The dates of peak abundance of each Winter Flounder Studies 179 10000 9000 8000 ^ 70001 Q 6000 > 5000 4000 3000 20001 1000 STAGE 1 rnn HTTl r?CT rsXN 34567 34567 34567 34567 34567 EN NB 7000 6000- ^ 5000 4000 3000 2000- 1000- i i I 1 ^ i i-i i i STAGE 2 I i M ■ 34567 34567 34567 34567 34567 Fig. 19. Cumulative density by developmental stage for larval winter flounder at each station from 1983 through 1987. 180 3000 ^ 2000 o 1000 STAGE 3 34567 34567 34567 34567 3 4 5 6 7 I NB 1 600 500 ^ 400 300 o 200 100 STAGE 4 34567 34567 34567 34567 34567 Fig. 19. Cont'd. Winter Flounder Studies 181 TABLE 12. Estimated dates of peak abundance of larval winter flounder for each developmental stage in the Niantic River and Day. Year Stage 1 Stage 2 Stage 3 Stage 4 Niantic River 1983 Mar 5 Mar 15 Apr 18 May 1 1984 Mar 7 Mar 9 Apr 26 May 19 1985 Mar 12 Mar 16 Apr 28 May 15 1986 Feb 26 Mar 7 Apr 23 May 12 1987 Mar 10 Mar 15 Niantic Bay Apr 22 May 9 1983 - Apr 7 Apr 22 May 7 1984 - Apr 9 May 2 May 23 1985 Mar 31 Apr 26 May 15 1986 - Apr 7 Apr 28 May 9 1987 - Apr 5 Apr 24 May 16 developmental stage in the river and bay were fairly consistent during the 5-year period. Stage 1 larvae generally peaked in early March, or in 1986, during late February. This corresponded with the observations on spawning adult females during the adult surveys. Based on water temper- atures of 2 to 3°C during the latter portion of February and egg incubation times reported by Buckley (1982), peak spawning probably occurred in mid-February. In the river, Stage 2 larvae peaked in mid-March, but the dates of peak abun- dance in the bay were 15 to 31 days later. As noted in NUSCO (1987), the lag in peak abun- dance of Stage 2 larvae in the bay may have been related to flushing rate, because the average re- tention time of a passive particle in the Niantic River was reported as 25 to 27 days (Moore and Marshall 1967; Kollmeyer 1972). In each year the peak abundance dates for Stage 3 and 4 larvae were very similar in both the river and bay. This similarity, along with the lag in the date of peak abundance for Stage 2 larvae, suggested that the larvae were flushed from the river primarily during that developmental stage. Predation could have affected larval abundance and there are numerous accounts that jellyfish are predators of fish larvae. Several species of hydromedusae and the scyphomedusa Aurelia aurila were found to prey upon herring larvae (Arai and Hay 1982; Moller 1984). laboratory studies with cod, plaice, and herring showed that the capture success by A. aurelia increased with mcdusal si7,e (Bailey and Batty 1984). Evidence of a causal predator-prey relationship on larvae of two European flatfishes [Pleuronectes platessa and Plalichthys flesus) by A. aurita and the ctenophore Pleurobrachia pileus was reported by van der Veer (1985). Pearcy (1962) stated that Sarsia tubulosa medusae were important predators of larval winter flounder in the Mystic River, CT, and had greatest impact on younger, less motile individuals. Crawford and Carey (1985) reported large numbers of the moon jelly {A. aurala) in Point .ludith Pond, RI and felt that they were a significant predator of larval winter flounder. The medusae of the jellyfish Cyanea sp. has been sus- pected of being an important predator of larval winter flounder in the upper portion of the Niantic River (NUSCO 1987). This hypothesized preda- tion was based on data collected at station A 182 during 1983-85 as measured volumes of jellyfish were lowest in 1985 when Stage 2 larval abundance was the highest. In addition, laboratory studies had shown that a larva which contacted a tentacle was stunned and ultimately died, even if not con- sumed. Jellyfish abundance in 1986 was similar to 1983 and that in 1987 was similar to 1985, but the numbers of Stage 2 larvae at station A in 1986 and 1987 were both low compared to 1985 (Fig. 20). However, this does not discount the potential of jellyfish predation. Abundance of Stage 2 larvae in 1987 was low at all stations and this possibly obscured observations on the effects of predation. Although the predator-prey rela- tionship of jellyfish on winter flounder larvae in the Niantic River was not as clear as once thought (NUSrO 1987), the fewer older larvae (Stages 3 and 4) in the upper portion of the river may have been related to jellyfish predation. 40 30 20 10 1984 15FEB 01 MAR 15MAR 29MAR 12APR 26APR 10MAY 24MAY DATE Fig. 20. Weekly mean volume (liters per 500 m ) of Cyanea sp. medusae collected at station A in the Niantic River from 1983 through 1987. Entrainment of larvae at MNPS The number of winter flounder larvae entrained by MNPS was related to larval densities in Niantic Bay and plant operations. Generally, larvae were entrained from late February through June with most entrainment occurring from mid-April through May. The median larval densities from entrainment collections in 1986 and 1987 were some of the lowest since 1976, but the estimated numbers entrained were among the highest due to the start-up of Unit 3 (Table 13). Although Unit 3 did not start producing commercial power until April 23, 1986, condenser cooling- water pumps in operation varied throughout the occur- rence of winter flounder larvae in Niantic Bay during 1986. The percentages of each develop- mental stage respectively entrained in 1986 and 1987 were similar, with Stage 1 representing 2% in both years; Stage 2, 21% and 29%, Stage 3, Winter Flounder Studies 183 TABLE 13. Annual median densities (number per 500 m ) of winter flounder larvae in entrainment samples during their season of occurrence and total entrainment estimates with approximate 95% confidence intervals for MNPS in 1976 through 1987. Total Year Median 95% CI Estimate (xlO ) 95% CI (xio') 1976 158.0 114-188 94.8 68-113 1077 64.1 53-87 29.3 24-40 1978 86.6 65-106 57.8 43-70 1979 90.3 70-108 36.7 28-44 1980 201.5 164-235 40.6 114-164 1981 139.2 99-183 47.4 34-62 1982 183.5 148-215 126.6 102-148 1983 244.4 158-315 171.7 111-221 1984 185.5 108-226 90.4 52-110 1985 107.1 79-153 66.0 49-94 1986 94.0 73-120 109.4 85-139 1987 88.9 65-109 126.2 93-154 61% and 62%; and Stage 4, 15% and 8%. The proportion of each developmental stage entrained was similar to previous years with Stage 3 pre- dominating (NUSCO 1987). The 1 2 years of entrainment sampling provided a long time-series of data that were examined to determine if seasonal water temperatures affected the timing of peak abundance. Seasonal water temperatures were expressed as the deviation from the 12- year mean. The date of peak abundance was highly correlated to water temperatures during March and April (Fig. 21). The warmer the water temperature, the earlier the peak that occurred, which suggested that the rate of larval development increased with increasing temperature. This was in agreement with the findings of Laurence (1975), who found that winter flounder larvae metamor- phosed 31 days earlier at 8°C than at 5°C. The date of peak abundance, estimated from the inflection point of the Gompertz function. was calculated using the p and K parameters. The K parameter was correlated to February water temperatures and as temperatures decreased, the parameter declined (Fig. 22). The K parmeter determined the shape of the abundance curve with steeply peaked curves for larger K values. To demonstrate how the this parameter affects abun- dance distribution, abundance curves were simu- lated with two different K values, but with the same a and P values (Fig. 23). For winter floun- der, February water temperatures would primarily affect spawning and egg incubation, because peak larval abundance occurs later (Table 12). A sig- nificant relationship was found between the annual K value and the age 3 recruitment indices that were used for the analysis of stock and recruitment (Fig. 24). It appeared that the shape of the larval abundance curve, as determined by February wa- ter temperature, was important in determining year-class strength, although the actual causal mechanisms were not known. 184 16MAY- • ^^- ■ 77 y 06MAY- 78 z n < • 79 • ■~-\ 81 ii: 26APR- 6 a. u. o • 80 82 ^"^^ ^^^^ 84 ■ 85 Ld Q 16APR- r=0.82 p<0.001 83 -^^^ • ^^ 85 • 76 06APR- -2.0 -1.5 -1.0 -0.5 0.0 0.5 1.0 MARCH -MAY TEMPERATURE DEVIATIONS Fig. 21. The relationship between the estimalcd annual date of peak abundance of entrained winter floundc larvae and the annual temperature deviation during March through May. < ^ 0.06- R=0.58 P=0.049 0.02 -L -1.5 -1.0 -0.5 0.0 0.5 1.0 FEBRUARY TEMPERATURE DEVIATIONS Fig. 22. T he relationship between the ic parameter of the Cjompertz function for entrained winter flounder lar\'ae and the annual February Icmperalure deviations with the fitted regression line. Winter Flounder Studies 185 5001 400 300 200 100 K =0.09 08FEB 28FEB 20MAR 09APR 29APR 19MAY 08JUN 28JUN DATE Fig. 23. A simulation illustrating the efTect of difTerent Gompcrtz k parameter values with constant a and p values on the shape of larval abundance curves. 20 16 12 R = 25.15 - 218.24 AT r 2 = 0.80 p = 0.001 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.10 K PARAMETER Fig. 24. The relationship between the k parameter of the Gompertz function for entrained winter flounder larvae and the age 3 recruitment index lagged 3 years with the fitted regression line. 186 Growth Examination of the length-frequency distribu- tion of larvae collected in 1986-87 showed a sep- aration between the first three developmental stages by predominant 0.5-mm size-classes (Fig. 25). Stage 1 larvae were primarily in the 2.5 to 3.5-mm size-classes (95%), Stage 2 were 3.0 to 4.5 mm (86%), Stage 3 were 4.5 to 7.5 mm (86%), and Stage 4 were 6.5 to 8.5 mm (92%). These predominant size-classes for each develop- mental stage were consistent with previous fmd- ings (NUSCO 1987), indicating that development and length were closely related. This allowed for the approximation of developmental stage from length-frequency data. A comparison was made of the length- frequency distribution between Niantic River and Bay in 1986-87 (Fig. 26). The pattem found was similar to that of the spatial distribution of devel- opmental stages for the same year (Fig. 19). Smaller size-classes predominated in the river, with over 60% of them 3.5 mm or smaller. In contrast, over 50% of the larvae in the bay were 5.0 nun and larger. These patterns were similar to those found in previous years (NUSCO 1987) and was further evidence that a majority of the larvae hatched in the Niantic River and then grad- ually flushed into the bay. Based on the large decline in the river from the 3.0- to the 4.0-mm size-class, it is likely that this was the size range where most of the mortality occurred. Larvae in these size-classes would have been yolk-sac larvae (Stage 1) and first-feeding Stage 2 larvae. In a bioenergetic study on winter flounder larvae, Laurence (1977) found that they had a low energy conversion efficiency at first feeding compared to later development, and that this stage of develop- ment was probably a "critical period" for mortality. The "critical period" concept was first hypothe- sized by Hjort (1926) and discussed by May (1974) for marine fishes. In many cases, the strength of a year-class was thought to have been determined by the availability of sufficient food after yolk-sac absorption was completed. How- ever, the occurrence of a "critical period" depended upon a number of environmental and species- specific factors (May 1974). The small increase in frequency of larvae from the 5.0- to 7.0-mm size-classes in the river may have been caused by the net import of these larger larvae into the river due to the behavioral retention mechanisms dis- cussed in NUSCO (1987). Previous sampling during ebb and flood tides at the mouth of the river showed a net loss of larvae smaller than 5 mm from the river, but a net import of larger larvae. This was attributed to vertical migration by larger larvae in relation to tidal stage as a retention mechanism to remain in the river. These larvae apparently swam up from the bottom dur- ing flood tides and remained near bottom during ebb tides. The decline in frequency after the 7.0-mm size-class probably was due to undersampling as larvae metamorphosed and be- came less susceptible to capture with a plankton net. The effects of temperature on larval growth was examined in the laboratory in 1986 under four temperature regimes. Larvae were reared from hatches on March 6 (treatments I and II) and April 1 (treatments III and IV). Linear re- gression was used to estimate growth rates and reasonable fits were obtained with r values of 0.85 and higher (Fig. 27). The rate of yolk ab- sorption was similar in all treatments, but growth rates differed (Table 14). Temperature compari- sons between treatments were based on the first 40 days from hatching. Growth rates were sig- nificantly lower in lowest and highest temperature regimes compared to the intermediate treatments. This limited laboratory study suggested that larval winter flounder have an optimum temperature range for growth, and as the temperature decreased or increased from the optimum, grovi^h rates de- creased. Winter Flounder Studies 187 LiJ 30- STAGE 1 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8,5 9.0 9.5 10.0 LENGTH (MM) STAGE 2 2.0 2.5 3.0 3,5 4,0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0 LENGTH (MM) I"ig. 25. Length -frequency distribution of larval winter flounder by developmental stage for all stations in the river and bay combined for 1986 and 1987. STAGE 3 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0 LENGTH (MM) STAGE 4 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0 LENGTH (MM) Fig. 25. Cont'd. Winter Flounder Studies 189 NIANTIC RIVER 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6,5 7.0 7.5 8.0 8.5 9.0 9.5 10.0 LENGTH (MM) ■ NIANTIC BAY 2.0 2.5 3,0 3.5 4.0 4.5 5,0 5,5 6.0 6.5 7.0 7.5 8.0 8.5 9,0 9.5 10.0 LENGTH (MM) Fig. 26. Length-frequency distribution of larval winter flounder for all stations combined in the Niantic River and Bay for 1986 and 1987. 190 10- TREATMENT 1 9- TEMPERATURE 5.4 8- : ' 7- \'-^^^^"^^ 6- -J:^.^^^^: 5- , ^^-^(^ • •' 4- \\^^^^ o O -^ ^ 30 20 10- Station WA 87- MAY JUNE JULY AUGUST SEPTEMBER 100 (/) 90 a: v- tj 80 S 70 _ < X Z) I- o O o Ld O CD "- < a: ^ Q. < X CD o O -, 60 50 40 30 20 10 0 Staf on LR \ r^\ \ \ 84 -/^"■■"""^^--/^v 85/^" 86 ^tb^ y MAY JUNE JULY AUGUST SEPTEMBER Fig. 32. Moving average of weekly mean CPUE of young winter flounder taken in the Nianlic River from 1983 through 1987. Winter Flouncler Studies 201 60 50 ^ 40 X h- o z -' 30 i ^ 20 UJ LU 10 986 ^KtX.^.4-4--f-p--^-l MAY JUNE JULY AUGUST SEPTEMBER 60 50 5 40 X (— CJ) z UJ -■ 30 z b 20 is: LU UJ 10- O-lr MAY JUNE JULY AUGUST SEPTEMBER Fig. 33. Weekly mean length (±2 standard errors) of young winter flounder taken in the Niantic River during 1986 and 1987. 202 TABLE 18. Monthly survival rate estimates for young winter flounder taken at two stations in the lower Niantic River from 1983 through 1987 as determined by a catch curve. Year Monthly survival rate (S) at station: LR WA 1983 1984 1985 1986 1987 0.552 0.564 0.569 0.553 0.597 0.661 0.335 (early); 0.440 (late) 0.511 Average annual survival rate 0.567 0.507 Determined from the average of the corresponding estimates of instantaneous mortality rate Z. Weekly mean lengths at LR during 1983-85 were similar by year until about late June; means were 6 to 8 mm greater thereafter during 1983 (Fig. 34). Growth in 1986 and 1987 was alike, but means were less than in previous years. Smaller differences were seen in growth among years at WA. Average size of young appeared to be greater in July during 1987, but if 95% confi- dence intervals were shown, no significant differ- ences would have been found in August and Sep- tember. The reasons for these differences among years are not known. Water temperatures ap- peared to be comparable; seasonal 19-week means ranged from 18.9''C in 1987 to I9.5°C in 1985. Life history data such as food preferences and rates of feeding were not available, but both were considered important factors in the growth of young plaice (Steele and Edwards 1970; Poxton et al. 1983). Apparent annual changes in growth may also have been caused by differentia] move- ment of larger young away from the station, which also would have increased the apparent mortality rate. However, few young were taken during the summer at trawl monitoring program stations. Thus, neither large-scale movements offstation nor differential movements by size seemed to have occurred, at least into areas sam- pled by trawl. A tentative explanation for the differences ob- served among years is density-dependent growth, especially at LR. Densities there in 1986 and 1987 were greater than during 1983-85. Benthic production and food availability at the stations may have been a limiting factor for growth. Density-dependent growth in juvenile fish has been regularly observed (Gushing and Harris 1973; Ware 1980; Poxton et al. 1983; van der Veer 1986). This may be demonstrated conclusively in forthcoming years, if greater densities are ob- served with concurrent slower rates of growth. Mortality To determine instantaneous mortality (Z) and weekly and monthly survival rates (S), catch curves were constructed using annual abundance data from LR for 1983-87 and WA for 1985-87. This method assumed that young comprised a single-age cohort which was followed from week to week during the sampling season. Catch curves and estimates of Z and S for 1983-85 were given in NUSCO (1987). The catch curves for LR data from 1986 and 1987 had relatively good fits with r^ of 0.70 and 0.89, respectively (Fig. 35). Re- markably similar values of Z were obtained over all years, resulting in monthly survival estimates ofO.552 to 0.597 (Table 18). Estimates of survival were more variable at WA (Fig. 36). The fit to the data in 1985 (NUSCO 1987) was not as good (r =0. 56) and monthly survival rate was estimated as 0.661. Winter Flounder Studies 203 80- Station WA 70- 60- 50- ^^ ->o^'' 40- .-^-=--— 05— 86 30- 20- /// MAY JUNE JULY AUGUST SEPTEMBER Station LR 80- 70- ^^ — " ^"^ ''/ ^^^^^^^ 83 ^- — ^-^ 85 60- 50- ^^/>^/ ^^'\^86 "~" 87 A^-—-:^^^ ^^^^ ^— -''' — 40- jA\-^ ^^- ""^ 30- / 20- ^ / in- ^ MAY JUNE JULY AUGUST SEPTEMBER Fig. 34. Comparison of weekly mean length of young winter flounder taken at stations LR and WA in the Niantic River from 1983 through 1987. 204 5- 1986 * - LR 4- 3- * * * * ~-^ * Z = 0.136 S (weekly) = 0.873 * *^ "^ *^ ^* 2- 1 - * S (monthly) = 0.553 r^ = 0.89 MAY JUNE JULY AUGUST SEPTEMBER 1987 - LR Z = 0.118 S (weekly) = 0.888 S (monthly) = 0.597 r2 = 0.70 MAY JUNE JULY AUGUST SEPTEMBER Fig. 35. Mortality determined by catch curve for young winter flounder taken at station LR in the Niantic River during 1986 and 1987. Winter Flounder Studies 205 1986 - WA o * \ \ * \ * o \. o * \ Early season: * O ^s,,^ Late season: ^ Z = 0,251 Z = 0.189 S (weekly) = 0.778 S (weekly) = 0.828 S (monthly) = 0.335 S (monthly) = 0.440 r2 = 0.80 r2 = 0.82 MAY JUNE JULY AUGUST SEPTEMBER ■ 1987 - WA 4- * * - *^^*\^ *\, * 3- Z = 0.154 * *\^ V* * * 2- S (weekly) = 0.857 S (monthly) = 0.511 * ^ *v * * 1 ■ r2 = 0.88 * MAY JUNE JULY AUGUST SEPTEMBER Fig. 36. Mortality determined by catch curve Tor young winter flounder taken at station WA in the Niantic River during 1986 and 1987. 206 Because of the previously mentioned increase in abundance during mid-summer of 1986, two catch curves were used, resulting in early (0.335) and late (0.440) seasonal estimates of monthly survival rate. Although the above partitioning of the data resulted in good fits, this apparent sam- pling problem made the survival estimates unre- liable. The 1987 monthly survival rate of 0.511 appeared to be based on more accurate density estimates. Overall, survival may not have been as high at WA as LR, although additional sam- pling will be necessary to confirm this. The monthly survival rates in the Niantic River were less than the value of 0.69 reported by Pearcy (1962) for the Mystic River estuary, which is the only published estimate for young winter flounder. Most monthly survival rate estimates for young plaice in British coastal embayments also were about 50% per month (Lockwood 1980; Poxton et al. 1982; Poxton and Nasir 1985). Density-dependent processes in the fu^st year of life following the larval stage are believed to occur in a number of species and can greatly affect subsequent recruitment to adult stocks (Bannister et al. 1974; Gushing 1974; Sissenwine 1984). Density-dependent mortality was not ap- parent for post-larval Niantic River winter floun- der, at least over the range of abundances seen to date. In fact, the highest survival rate estimated was in 1987, when densities were greatest. Density-dependent mortality was reported for young plaice by Bannister et al. (1974), Lockwood (1980), and van der Veer (1986). However, ex- amination of their fmdings indicated that greatest rates of mortality occurred only when extremely large year-classes of plaice were produced (three to more than five times larger than the average). The numbers of juvenile winter flounder in the Niantic River since 1983 were probably not ex- tremely large or small enough to have produced density-dependent mortality (should it exist), al- though densities were apparently different enough to have affected rates of growth. Impingement of winter flounder at MNPS Annual estimates of the number of winter flounder impinged on the traveling screens of MNPS have been made since 1972-73 (NUSCO 1987). Since 1976, the winter flounder has been the second-most abundant fish impinged at MNPS, making up 5.9% of the total. About two-thirds of the impingement occurred during the winter with relatively little in summer. Fish return sluiceways have been in place at Unit 1 since December 1983 (NUSCO 1986b, 1987) and at Unit 3 since it commenced commercial oper- ation (NUSCO 1988c). These sluiceways have considerably reduced the impact of impingement as the winter flounder is very hardy and had good ( > 85%) survival following return to Long Island Sound. The estimated impingement of winter flounder at Unit 2 was 1,212 for October 1985 through September 1986, 547 during 1986-87, and 77 from October through mid-December 1987. The 1985-86 estimate was less than 50% of the next smallest aimual estimate. As in most previous years, about two-thirds of the fish impinged in 1985-87 were less than 20 cm in length. The numbers taken at Unit 2 during the past few years have been small, due to varying plant operations; declining winter flounder abundance; and possibly the construction and operation of Unit 3, which may have altered fish availability and movements near the Unit 2 intake. The latter effect on fish impingement is discussed in the Fish Ecology sec- tion of this report. Although impingement is no longer routinely monitored at MNPS, a require- ment exists for reporting significantly large ( > 300 specimens per day) impingement events, should they occur (NUSCO 1988a). Impact assessment Approaches to impact assessment Considerable effort has been expended during the past 15 years evaluating the impact of fish mortality resulting from the operation of power plants in the United States (Van Winkle 1977; MacCaU et al. 1983). For adult fish subject to Winter Flounder Studies 207 impingement the assessment of impact is straight- forward and consists of the direct enumeration of fish losses. However, for eggs and larvae subject to entrainment, measurement of primary loss is less direct and assessment of the resulting impact at the population level (as adult fish) is much more difficult. This type of assessment involves two basic tasks: the estimation of organisms en- trained as a fraction of the annual production; and the measurement of any resulting population changes. The first task involves sampling at the plant discharge and often the use of hydrodynam- ics models to predict the spatial distribution of fish eggs and larvae in the vicinity of the plant; and the second requires measuring the size of the adult population in either absolute or relative terms. Because fish populations are complex dy- namic entities strongly influenced by the physical environment, the determination of their size and spatial distribution is not easy. It has long been known (Gushing 1977) that marine fish stocks undergo very large fluctuations in abundance, both annual and long-term, primarily induced by climatic and hydrographic factors. This natural and erratic variability creates serious problems for measuring anything but the largest changes. In addition, when knowledge of the range and fre- quency of long-term fluctuations is lacking, short- tcnn monitoring data may show a misleading and alarming population decline if observations started at the begining of a downturn. ITiis rules out the use of short-term empirical assessment meth- ods because they address impact on a single co- hort or year-class in the sense of Goodyear's (1978) adult-equivalent method. An additional problem inherent of the latter method and of most fishery stock-assessment methods is that they assume equilibrium population conditions, which implies a regularity in nature that is contrary to all evidence. All these problems point to the need for parsimonious impact assessment ap- proaches leading to the creation of a substantial data base on which to build a comprehensive fishery assessment model capable of predicting long-term effects at the population level. Prefer- ably, this model should incorporate key biological processes governing recruitment, the effect of en- vironmental factors known to affect year-class strength, and it should not require that the pop- ulation be at equilibrium. In assessing the possible impact of MNPS op- eration on the local winter flounder, NU has rec- ognized the importance of larval losses on the long-term stability of the adult population. The problems of estimating the fraction of larvae en- trained and of measuring the resulting population change were examined taking into account the geographic and hydrographic features of Niantic Bay and the spawning and nursery areas in the Niantic River. Therefore, the approach for impact assessment consisted of a combination of sampling programs and analytical methods leading to the development of a comprehensive simulation model which included hydrodynamics and popu- lation dynamics submodels. The sampling pro- grams and methods for estimating population pa- rameters, describing larval behavior, and deter- mining the stock-recruitment relationship have been discussed in previous sections. The simula- tion model components (larval dispersal and en- trainment, and population dynamics submodels) and the probabilistic risk analysis methodology for long-term impact assessments are described below. Larval dispersal and entrainment model As mentioned previously, one of the two basic tasks in assessing entrainment impacts is to esti- mate the fraction of total larval production lost to entrainment. The problem here is that, given the location of the plant intakes relative to the spawning and nursery areas in Niantic River and the prevailing tidal currents in Niantic Bay, the use of the number of entrained larvae as a direct loss to the locally spawning stock is not appropiate for impact assessment at MNPS. An early hy- drodynamics model of the area used by Saila (1976) predicted that, if larvae behaved as passive particles, most of those flushed out of the river by tidal action would progress towards Millstone Point and would continue moving in an east- southeast direction untU they left the area via the Twotree Island Channel. Saila and his coworkers estimated that only 30% of the organisms entering 208 the bay remain in the vicinity of Millstone after twenty tidal cycles (i.e., about 10 days). Their model also showed that, although larvae leaving the river were subject to entrainment losses, a large fraction of them would have been flushed from the area by tidal action in the absence of power plant entrainment. Because it seems rea- sonable to expect lower larval survival in the bay (and even more so in open waters of Long Island Sound) than in the protected nursery areas in Niantic River, absolute entrainment numbers do not represent a fair estimate of additional loss due to the operation of MNPS. In this context, the modeling work carried by Dr. Saila's research team at URI represented a first attempt to estimate the actual larval losses attributed to entrainment at MNPS. Since then, new data on the early life history of winter flounder resulting from our own studies (NUSCO 1987) suggested that vertical movements of older larvae in response to tidal and diel cycles may invalidate some of the as- sumptions made in the early URI models. In order to address this problem NU has contracted for the development of a new larval dispersal and entrainment model with the Department of Civil Engineering at the Massachusetts Institute of Technology (MIT). This new MIT model wiU use the hydrody- namics TEA (tidal circulation) and LEA (advective transport) submodels as a framework for larval dispersal and entrainment simulation. These two model components were recently used to describe the dynamics of the thermal plume at MNPS (Adams and Cosier 1987). Although TEA and LEA are nominally similar to the correspond- ing models used by Hess et al. (1975) and Saila (1976), there are important differences as well. Eirst, both TEA and LEA use irregular, triangular, grid elements, rather than regular, square, grid elements. The former configuration allows for easy grid refmements in critical regions, such as the Niantic River and the plant intake area. Sec- ondly, both submodels take advantage of newly developed computational methods which provide better accuracy and higher speed for extended and more detailed simulations. In addition, the area covered by TEA and LEA in the previous appli- cation mentioned above will be extended to in- clude the northern reaches of the Niantic River and to the south, in Niantic Bay, to include a larger portion of the local area in Long Island Sound. The larval dispersal component of the MIT model will be able to simulate continuous pro- duction of newly hatched larvae that matches the actual length of a typical spawning season, and will track larval ages in days. Larval behavior will be simulated by reducing advection (corre- sponding to larvae moving to the bottom) as a function of tide phase and time of day and ac- cording to larval age. Although four larval stages will be simulated separately, simulation results can be presented in terms of total larval population by integrating over larval stages. The model will be run to simulate various scenarios where each will be of seasonal duration involving a repeating average tide. Among contemplated simulations are comparisons between runs with and without larval "behavior", between runs with and without power plant operation, and between runs employ- ing different distributions of larval hatching in space and time. Drifting of "foreign" larvae into the simulation area through the open boundaries will also be simulated to assess the effect of larval sources other than the Niantic River. Important inputs required by the hydrodynamic components of the MIT model are the tidal boundary conditions and dispersion coefficients which influence flushing from the rivers and dis- persion away from the power plant intakes. The tidal boundary conditions will be established by comparing measured and simulated tidal currents, larval dispersion coefficients will be validated by comparing measured and simulated larval flushing rates from the Niantic River and by comparing measured and computed intake dye concentrations resulting from instantaneous dye releases in Niamtic Bay. Additional inputs related to larval dispersion will be the empirically estimated (NUSCO 1987) vertical distribution of larvae at various tides and times of the day. Daily larval mortality rates, needed to simulate naturally oc- curring larval concentrations in the river and bay Winter Elounder Studies 209 for given spawning locations and hatching rates, will be supplied by NU. Once MIT has validated its model, the computer code will be installed in our own computer system to run additional sim- ulations as new or better input data become avail- able. ITie MIT model is expected to be completed before the end of 1988. Population dynamics model A problem central to modeling the dynamics of fish populations is the difficulty of finding an adequate formulation for describing the recruit- ment process. Because egg and larval survival is more dependent on environmental factors than adult fish survival, a prominent feature of recruit- ment data for many fisheries is the large amount of variability present that cannot be attributed to changes in parental stock size. Despite this well- documented fact, past deterministic population models (Christensen et al. 1977; Saila and Lorda 1977) have included in their formulations recruit- ment equations whose parameters describe only variation in parental stock and assume populations at equilibrium. By contrast, our modeling ap- proach recognizes that the very great temporal variation in recruitment nullifies the concept of "equilibrium" conditions and this mandates stochastic models that account for this variability. This is important because, for commercially ex- ploited species like the winter flounder, the higher the fishing effort, which reduces the average life- time of the fish and their reproductive potential, the greater is the significance of year-to-year vari- ability. Our population model for the Niantic River winter flounder will use a temperature-dependent stock-recruitment relationship to generate year- classes whose size depend upon both the water temperature during larval development and the size of the spawning population. This particular representation of the recruitment process includes compensatory mortality (based on the stock- recruit model previously described) and will permit the mtroduction of realistic environmental vari- ability related to the water temperature in Febru- ary. Although the temperature variability will be simulated stochastically, this will be done using an empirical distribution derived from actual records of annual mean water temperatures in February. Because an earlier version of our stochastic population dynamics model was already described by Lxjrda et al. (1987), only the basic features of the updated version now under devel- opment are presented here. The final model will describe the dynamics of an age-structured population of winter flounder with stochastic recruitment and compensatory mortality that occurs during the first year of the life of the fish. ITie model makes no assumptions about the stability of the population or its age structure, which can vary as a result of environ- mental variability introduced via the recruitment equations. Adult fish are subject to annual natural and fishing mortalities, and mature sundvors spawn according to fecundity rates that depend on the age of the fish. Although a Leslie matrix formulation is used to carry the computations corresponding to these annual processes, its func- tion is simply one of book-keeping. This is so because the size of each year-class is determined by the stochastic recruitment, and the I^slie ma- trix is only used to update the number of fish in each age-class at the end of each year. A box- and-arrow diagram of the underlying life-cycle simulation scheme is shown in Figure 37. The impact of larval entrainment is represented in this diagram as additional density-independent mor- tality in age-class 0. This mortality can be varied annually and the actual value will be based on estimates provided by the MIT larval dispersal and entrainment model. A sample of graphical model output corresponding to a simulation that assumes a 10% larval mortahty due to entrainment is shown, for illustrative purposes only, in Figure 38. The vertical line at 35 years, represents the point in the simulation at which larval entrainment ceases and the population size begins its climb back to its initial level. This pattern of slow population decline while entrainment takes place, followed by a faster return to initial levels, is typical of populations in which compensatory mortality operates. 210 The basic inputs required by the model are: survival estimates for each larval stage, eggs, and age-class; fecundity rates for each age-class; an initial age-distribution for the adult population; estimated parameters of the recruitment model; estimates of the mean and variance of the water temperature during February in the Niantic River; and the estimated larval mortality due to plant entrainment. Except for the latter input (see dis- cussion of the larval dispersion model), we have data for most of the other model inputs. However, better estimates of egg and larval mortality are still needed. Long-term impact assessment The larval dispersal and entrainment model to- gether with the stochastic population model will allow us to simulate the long-term effect of larval entrairmient at the adult population level under a variety of scenarios chosen to describe prevailing levels of entrainment, multiple or single sources of winter flounder larvae, and various levels of fishing effort. The final output from this simu- lation scheme will consist of time-series of ex- pected population sizes suitable for estimating long-term averages and standard errors for direct comparisons, or for applying probabilistic risk analysis (PR A). In the latter case, projected pop- ulation changes are expressed in terms of the probability that a postulated change (or no change at all) will occur in a specified number of years. Although both analytical methods are acceptable for impact assessment purposes, PRA methodol- ogy has been favored in recent environmental im- pact and risk assessment studies by the U.S. En- AGEO Eggs Larvae Juveniles \. Natural Mori. Age 1 7 Age 2 J ( ADDED ] y Mort. J SIMULATED IMPACT Egg Production Age / Age Oldest Age Annual Catch (Fishing mortality) Fig. 37. Box and arrow diagram of the life-cycle simulation scheme in the population model. Transfers within the population fi.e., aging and reproduction) are indicated by thick arrows and occur once a year. Losses due to mortality are indicated by thin arrows and occur both daily (dashed) and yearly (thin solid). Winter Flounder Studies 211 u_ a Fig. 38. Simulated power plant impact on winter flounder with a projected effect of 10% mortality over 35 years of plant operation followed by the recovery of population abundance. vironmental Protection Agency. We hope to start preliminary work on this type of long-term impact assessment by the end of 1988. Wliile our data on the Niantic River winter flounder population is extensive, additional infor- mation is necessary to further refme MNPS as- sessment models. For example, the stock and recruitment relationship, larval mortality, and larval behavior are especially important for reliable population dynamics simulations. About two- thirds of the winter flounder larvae entrained each year at MNPS are in Stages 3 or 4 of development. Therefore, it remains necessary to determine how critical the loss of these larvae is to the population. To date, no relationship has been found between the absolute estimate of the number of larvae entrained and subsequent year-class strength (Fig. 39). However, the long-term effects of present levels of three-unit entrainment on the winter flounder population remain to be determined. In addition, detailed descriptions of larval move- 212 20 16- Q 12- .78 .79 . 77 .76 . 82 . 80 . 81 r = -0.42 . 84 . 83 Not significant (p=0.26) 25 50 75 100 125 150 ESTIMATED ENTRAINMENT OF LARVAE IN MILLIONS 175 Fig. 39. Relationship between the annual age 3 recruitment index and annual estimated entrainment of winter flounder larvae at MNPS for the 1976 through 1984 year-classes. ments, especially those flushed from the Niantic River, are needed. Also, little is known about the fate of larvae in Niantic Bay or the number removed from the area by tidal currents; the mag- nitude of larval metamorphosis in the bay; and whether or not post-larval young reside in the bay or are successful in moving into suitable nurs- ery areas, such as the Niantic River. Larval aging is necessary to accurately determine rates of mor- tality and grovrth. The source of entrained larvae is not known with certainty. Although evidence from the larval sampling program suggests that the Niantic River is the likely source of most entrained larvae, drift of larvae from other areas of Lx)ng Island Sound into Niantic B?.y carmot yet be ruled out. Sampling programs designed specifically to address the above concerns will be initiated in 1988 (NUSCO 1988a, 1988b). Conclusions Abundance of adult winter flounder in the study area around Millstone remained low in 1987, with numbers similar to those of 1984-86. The changes in abundance of local winter flounder during the past 12 years were similar to those taking place in other areas of Southern New Eng- land, as shown by significant correlations among the various abundance indices examined. This suggested that winter flounder populations were most likely affected by factors operating region- wide, which influenced reproduction and subse- quent recruitment and, perhaps, from similar fish- ing pressures on the adult stocks. Historical records showed that the species has repeatedly fluctuated in numbers with periods of low and high abundance. Winter Flounder Studies 213 Several important developments in our winter flounder program this past year included the es- tablishment of a stock and recruitment relation- ship and data analyses leading to increased knowl- edge of the dynamics of larval and juvenile stages. The three-parameter stock and recruitment model helped explain the variability seen in adult abun- dance and demonstrated that water temperature during spawning and early life history was as im- portant as the parental stock size in determining recruitment success. Poor recruitment was asso- ciated with wanner-than-average years and strong year-classes were produced during cold years; this has also been found for a number of other winter- spawning marine fishes. Although strong envi- ronmental (i.e., density-independent) factors were implicated as important to winter flounder repro- ductive success, density-dependent mortality must also have been a significant stabilizing mechanism, given the relatively small range within which the absolute abundance of the winter flounder varies. The use of the Gompertz function to describe larval abundance led to the fmding that the k parameter (i.e., the shape parameter determining the steepness of the Gompertz curve) described well the magnitude and breadth of larval distri- bution over time (see Figure 23). This parameter was found to be directly related to February water temperatures, with larger K values associated with warmer years and smaller values for colder years. This illustrated that in a cold year, larval abun- dance, as characterized by the Gompertz function, was slow in reaching a peak and had a more broadly based abundance curve. Conversely, in a warm year the curve peaked quickly and had a narrow base. A highly significant relationship was found between the k parameter and winter flounder recruits 3 years later (see Figure 24). This implied that in cold years somewhat fewer larvae were present at any one time, but the sea- son was longer in duration; the result was a stronger year-class. The February temperatures would have most likely affected spawning and egg incubation, resulting in protracted spawning and longer time to hatching. With larvae less con- centrated over time, effects of predation upon lar- vae may have been less and there likely would have been a better chance for larvae to encounter adequate food densities. However, exact causal mechanisms are still unknown. Results of both the larval analyses and the three-parameter stock-recruitment model showed that year-class strength was related to events in the early life history of the winter flounder, for which water temperature was an important factor (by itself or as a surrogate for other factors). .lust as the stock and recruitment relationship v/as used to describe recruitment as a function of adult stock and February temperatures, the relationship found between the k parameter and the age 3 recruitment index can be used as a second and independent indicator of future adult recruitment. The empirical relationship between k and the age 3 recruitment index (R) given on Figure 24 was used to appro xiinate recruitment levels from 1976 through 1984 (Fig. 40). For the following 3 years, recruitment from the 1985 year-class was predicted to be greater than from the 1986 and 1987 year- classes and and close to the recruitment from the 1977 and 1980 year-classes. Very similar predic- tions were made for the same 3 years using the three-parameter stock and recruitment relation- ship (see Figure 15), although somewhat smaller differences in recruitment were predicted. Results of the larval sampling program showed the greatest mortality occurring during Stage 2 of development in the Niantic River. During this stage, which is characterized by the transition from yolk-sac larvae to first feeding and limited mobility, is also when most larvae are flushed from the river and when most jellyfish predation may occur on larvae remaining in the upper river. All of this suggests that density-dependent mor- tality probably occurs during this stage of larval development. Van der Veer (1986) pointed out that, for plaice, the highest coefficients of variation for yearly abundance estimates of diflerent life stages occurred during larval development in late winter and first settlement of pelagic juveniles in spring. Much less variation was seen for post-larval young during mid-summer and for age 2 recruits. He 214 20- 16- 12- A P 76 77 78 79 80 51 82 83 YEAR 84 85 86 87 Fig. 40. Actual values (A) for the annual age 3 recruitment index (R) compared to predicted values (P) determined from the relationship between the k parameter of the Gompertz function fitted to annual cumulative cntrainment densities (see Figure 24). attributed the sharp decline in variation of abun- dance for older juveniles to a density-dependent regulatory mechanism occurring during and shortly after larval settlement. The coefficients of variation determined for indices of abundance for various life stages of Niantic river winter flounder are shown in Table 19. Although larval and juvenile data were relatively sparse, remarkably similar coefficients were found for nearly all life stages. The only exception was Stage 2 larvae, which had the largest CV, further supporting the notion that compensatory mortality probably oc- curs during this early period of winter flounder life history. The models currently under development for MNPS impact assessment will incorporate esti- mates and variability of population parameters, hydrodynamics of the waters in the Millstone area, and elements that realistically mimic larval behavior. Once finalized, the model features or inputs will be updated as soon as more complete data are available. The direction of research in the future will, in many cases, be governed by the need to obtain specific information necessary for the assessments. Sustained monitoring coupled with specific new sampling programs in 1988 and a completed impact assessment model will further lead towards the resolution and quantification of MNPS impact on the winter flounder. Winter Flounder Studies 215 TABLE 19. Coefficients of variation (CV) for abundance indices of various life stages of Niantic River winter flounder. Life stage Abundance index used CV Eggs Stage 1 larvae Stage 2 larvae Stage 3 larvae Stage 4 larvae Early juvenile l^te juvenile Age 3 recruits Adults Relative index of egg production a parameter of Gompertz function a parameter of Gompertz function a parameter of Gompertz function a parameter of Gompertz function Mean CPUE at station LR (mid-June-early July) Mean CPUE at station LR (late August-September) Age 3 recruitment index (R) Parental stock index (P) 11 63 4 63 4 114 5 57 5 45 5 51 5 55 9 47 12 43 Number of observations. Summary Indices of abundance estimates (Jolly com- posite abundance index and trawl CPUE) of the Niantic River spawning population of adult winter flounder have been made since 1976. The 1987 abundance survey had the latest start since 1980 because of ice cover in the river. The 1987 Jolly index showed a slight increase over the 12-year low abun- dance index of 1986, but remained below the levels of abundance from previous years. The 1987 median trawl CPUE also increased over 1986. This median was similar in magnitude to those for 1984 and 1985, but was only 30-60% of CPUE values for 1976-83. Median CPUE values of juvenile ( < 1 5 cm) winter flounder taken during the adult surveys were very low in 1986 and 1987. The small values implied poor reproductive success in recent years. However, factors such as dif- ferential distribution among the Niantic River sampling stations probably affected tliis index of abundance. Annual 5-mean CPUE values were computed for the first time for aU winter flounder taken by the trawl monitoring program (TMP) from 1976 through 1986. The 5-mean abun- dances did not perfectly track the Niantic River median CPUE values. The peak in 5-means persisted from 1977 through 1983 and was not as pronounced as it was in the Niantic River. The 1985 and 1986 5-means were greater than those in 1977 and 1978, whereas for the Niantic River medians the opposite was true. The TMP catches were mostly made up by fish larger than 15 cm (55-75%). However, smaller fish predominated from January through April. Catch of the latter fluctuated less outside of the Niantic River during the spawning season than inside and 5-means for 1986 and 1987 suggested that juvenile abun- dance was not as low as the Niantic River median CPUE for these fish would have in- dicated. Annual Niantic River median CPUE values and TMP 5-means were compared to other regional indices of abundance. With few ex- ceptions, most indices were significantly cor- related and thus described real trends in abun- dance that occurred throughout Southern New England. 216 6. Examination of available historical data (commercial fishing CPUE for Rhode Island and a URI trawl time-series) showed that winter flounder abundance typically fluctu- ated with sharp increases in catch most likely related to the occurrence of one or more particularly strong year-classes in succession. 7. For the first time in the past 1 1 years, more males than females were taken during the Niantic River surveys in 1986 and 1987. 8. The length of females at 50% sexual matu- ration was 26.8 cm, when fish were 3 or 4 years old. Most spawning in the Niantic River was completed by late March or early April. Spawning appeared to have been re- lated to water temperature as in relatively cold years proportionately fewer females spawned earlier during the season them in warmer years. 9. Based on the abundance indices of females and their size distribution, a yearly index of egg production was determined. This index peaked in 1982 and has declined about 80% since then. However, adult abundance and absolute egg production were not the only factors in determining year-class strength. 1 0. The 1 2 years of Niantic River winter flounder relative abundance data were used with a Ricker stock-recruitment model. Parental stock was defined as all winter flounder age 3 and older and recruits were those fish turn- ing 3 years old each spawning season. The two-parameter Ricker model did not explain much of the variability (44%) seen in annual recruitment. However, annual February wa- ter temperatures were found to be significantly and inversely correlated with recruitment in- dices. The addition of a temperature param- eter to the model resulted in a much improved fit to the observed data (R = 0.78) and helped to explain large differences seen in recruitment for similar parental stock sizes. Although the actual mechanisms affecting winter floun- der recruitment were unknown, the February water temperature appeared to have been re- lated to those factors. 1 1 . Larval winter flounder studies have been con- ducted in Niantic River and Bay since 1983 and entrainment collections have been made since 1976. In 1986 and 1987, peak abun- dance of larvae occurred first in the river and, after a lag, in the bay. Comparison of the dates of peak abundance showed that most flushing of larvae from the river to the bay occurred during Stage 2 of development. 1 2. An apparently higher mortality of larvae was found in 1987 compared to the previous year; this occurred early in the season, when the majority of the larvae were in the river. Ef- fects of predation by jellyfish was not as ap- parent in 1986 and 1987 as it had been in previous years. Examination of length- frequency distributions indicated that most mortality occurred during the 3 to 4-mm size-class, suggesting that this was a critical period for mortality. Annual total mortality rates were estimated from the difference be- tween the abundances of larvae at 3 and 7 mm in length. For 1983-87, these estimates ranged from 84.6 to 96.9%. 13. In entrainment samples. Stage 3 larvae dom- inated. As expected, total entrainment esti- mates for 1987 following the startup of MNPS Unit 3 were among the highest during the last 12 years, even though the median density (number per 500 m ) was among the lowest. Entrainment estimates were depen- dent upon plant operating conditions as well as larval densities each year. The dates of peak abundance for entrainment samples were positively correlated with March and April water temperatures. 14. From the 12 years of entrainment data, the shape of the abundance curve, as measured by the k parameter of the Gompertz function, was found to be a good predictor of subse- quent recruitment of age 3 winter flounder. The shape of the abundance curve was related Winter Flounder Studies 217 to February water temperatures, with a nar- row, high-peaked curve found during warmer years (low recruitment) and a broad, flatter curve during colder years (high recruitment). 15. Laboratory studies showed that larval growth rates were dependent upon water tempera- ture. These studies suggested that the opti- mum temperatures for growth were interme- diate (6.9-7. 5°C), with decreased growth oc- curring at lower (5.4°C) or higher (10.8°) temperatures. 1 6. Yearly growth rates were estimated using field data since 1983 at station C in the Niantic River and since 1976 for entrainment data. Estimated growth rates were consistent with laboratory estimates, again showing that growth was dependent upon water tempera- tures. 17. Post-larval young-of-the-year winter flounder have been sampled at two stations in the Niantic River since 1983. Densities at station LR in 1987 were liigher than in previous years. Smaller differences in growth were noted between stations LR and WA during 1986 and 1987 than during 1984 and 1985. Differences among years may have been due to density-dependent growth, especially at LR. Survival rates were very similar among years, regardless of densities of young. 18. The winter flounder was the second-most abundant fish impinged on the traveling screens at MNPS since 1976. Relatively few specimens were impinged at Unit 2 during the past 3 years due to declining abundance, varying plant operations, and possible reduc- tions related to the construction and opera- tion of Unit 3. The installation of fish return sluiceways at Units 1 and 3 lessened the im- pact of impingement on the winter flounder because it has good ( > 85%) survival when returned to the water. Routine impingement monitoring at Unit 2 was discontinued in December 1987 upon agreement between NU and CT DEP. 19. To predict the long-term effects of larval en- trainment, an impact assessment model for winter flounder is currently under develop- ment, which includes hydrodynamics and population dynamics submodels. The func- tion of each submodel is the estimation of the fraction of total larval production lost to entrainment at the plant and the measurement of any resulting population changes, respec- tively. A newer, more accurate and detailed hydrodynamics submodel is under develop- ment at MIT. Larval behavior will be sim- ulated to correspond more realistically to ob- servations made in the field. A stochastic age-structured population submodel will in- corporate the three-parameter stock- recruitment relationship, which includes a measure of compensatory mortality and the introduction of realistic environmental vari- ability. 20. Results from both larval analyses and the three-parameter stock-recruitment relation- ship showed that year-class strength was re- lated to events in the early life history stages, with colder winters associated with better re- productive success. Greatest winter flounder mortality took place during Stage 2 of devel- opment, during which density-dependent mortality probably occurred. References Cited Adams, E., and D. Cosier. 1987. Numerical models of coastal circulation and dispersion with application to power plant discharges. MIT Energy Laboratory Report. Dept. of Civil Engineering, Massachusetts Institute of Technology, Cambridge, MA. Arai, M.N., and D.E. Hay. 1982. Predation by medusae on Pacific herring {Clupea harengus pallasi) larvae. Can. .1. Fish. Aquat. Sci. 39:1537-1540. Bailey, K.M., and R.S. Batty. 1984. Laboratory study of predation by Aurelia aurelia on larvae of cod, flounder, plaice and herring: develop- 218 ment and vulnerability to capture. Mar. Biol. (Berl.) 83:287-291. Bannister, R.C.A., D. Harding, and S.J. Lockwood. 1974. Larval mortality and subsequent year-class strength in the plaice {Pleuronecles platessa L.). Pages 21-38 /« J.H.S. Blaxter, ed. The early life history of fish. Springer- Verlag, New York. Beacham, T.D. 1982. Biology and exploitation of winter flounder {Pseudopleuronectes americanus) in the Canadian Maritimes of the Northwest Atlantic Ocean. Can. Tech. Rep. Fish. Aquat. Sci. No. 1113. 33 pp. Begon, M. 1979. Investigating animal abundance: capture-recapture for biologists. University Park Press, Baltimore, MD. 97 pp. Beverton, R.J.H., and S.J. Hoh. 1957. On the dynamics of exploited fish populations. Fish. Invest. Lx)ndon, Ser. 2, 19:1-533. Buckland, S.T. 1980. A modified analysis of the JoUy-Seber capture-recapture model. Biometrics 36:419-435. Buckley, L.J. 1980. Changes in ribonucleic acid, deoxyribonucleic acid, and protein content dur- ing ontogenesis in winter flounder, Pseudopleuronectes americanus, and effect of starvation. Fish. Bull., U.S. 77:703-708. . 1982. Effects of temperature on growth and biochemical composition of larval winter flounder Pseudopleuronectes americanus. Mar. Ecol. Prog. Ser. 8:181-186. Burton, M.P., and D.R. Idler. 1984. The repro- ductive cycle in winter flounder, Pseudopleuronectes americanus (Walbaum). Can. J. Zool. 62:2563-2567. Carothers, A.D. 1973. The effects of unequal catchability on Jolly-Seber estimates. Biometrics 29:79-100. Christensen, S.W., D.L. DeAngeUs, and A.G. Clark. 1977. Development of a stock -progeny model for assessing power-plant effects on fish populations. Pages 196-228 in W. Van Winkle, ed. Proceedings on the conference on assessing the effects of power-plant-induced mortality on fish populations. Pergamom Press, New York. Cormack, R.M. 1968. The statistics of mark- recapture methods. Oceanogr. Mar. Biol. Ann. Rev. 6:455-506. 1979. Models for capture-recapture. 217-255 in R.M. Cormack and D.S. Statistical ecology. Vol 5. Sam- Intemational Co- MD. Pages Robson, eds pling biological populations operative Publishing Mouse, Fairland, (not seen, cited by Ilightower and Gilbert 1984). Crawford, R.E., and C.G.Carey. 1985. Retention of winter flounder larvae within a Rhode Island salt pond. Estuaries 8:217-227. Cushing, D.H. 1971. The dependence of recruit- ment on parent stock in different groups of fishes. J. Cons. int. Explor. Mer 33:340-362. . 1973. Recruitment and parent stock in fishes. University of Washington, Div. of Mar. Resources, Washington Sea Grant Publ. WSG 73-1. 197 pp. . 1974. The possible density-dependence of larval mortality and adult mortality in fishes. Pages 103-111 in i.H.S. Blaxter, ed. The early life history offish. Springer- Verlag, New York. . 1977. The problems of stock and recruit- ment. Pages 116-133 inJ.A. Gulland, ed. Fish population dynamics. John Wiley and Sons, New York. , and J.G.K. Harris. 1973. Stock and re- cruitment and the problem of density depen- dence. Rapp. P. -v. Reun. Cons. int. Explor. Mer 164:142-155. Winter Flounder Studies 219 Davies, R.G. 1971. Computer programming in quantitative biology. Academic Press, New York. Draper, N., and H. Smith. 1981. Applied re- gression analysis. John Wiley and Sons, New York. 709 pp. Dunn, R.S. 1970. Further evidence for a three- year oocyte maturation time in the winter floun- der {Pseudopleuronectes americanus). J. Fish. Res. Board Can. 27:957-960. , and A.V. Tyler. 1969. Aspects of the fish population. 113:633-641. Trans. Am. Fish. Soc. anatomy of the winter flounder ovary with hypothesis on oocyte maturation time. J. Fish. Res. Board Can. 26:1943-1947. Garrod, D.J. 1982. Stock and recruitment - again. Great Britain Ministry of Agriculture Fisheries and Food Directorate of Fisheries Re- search. Fish. Res. Tech. Rep. 68. Gibson, M.R. 1987. Preliminary assessment of winter flounder {Pseudopleuronectes americanus) stocks in Rhode Island waters. Rhode Island Div. Fish Wildl., Res. Ref. Doc. 87/7. 51 pp. Gilbert, R.O. 1973. Approximations of the bias in the Jolly-Seber capture-recapture model. Biometrics 29:501-526. Goodyear, P.C. 1978. Entrainment impact esti- mates using the equivalent adult approach. U.S. Fish Wildl. Serv., FWS/OBS-78/ July 1978. 14 pp. Hess, K.W., M.P. Sissenwine, and S.B. Saila. 1975. Simulating the impact of entrainment of winter flounder larvae. Pages 1-30 in S.B. Saila, ed. Fisheries and energy production: a sympo- sium. D.C. Heath and Co., Lexington, MA. Hjort, J. 1926. Fluctuations in the year classes of important food fishes. J. Cons. int. Explor. Mer 1:5-38. (not seen, cited by May 1974). Ilorst, J.T. 1975. The assessment of impact due to entrainment of ichthyoplankton. Pages 105-118 in S.B. Saila, ed. Fisheries and energy production: a symposium. D.C. Heath and Co., Lexington, MA. Iloude, E.D. 1987. Fish early life history dy- namics and recruitment variability. Am. Fish. Soc. Symposium 2:17-29. Howe, A.B., and P.G. Coates. 1975. Winter flounder movements, growth and mortality ofi" Massachusetts. Trans. Am. Fish. Soc. 104:13-29. Jeffries, II. P., and W.C. Johnson. 1974. Seasonal distributions of bottom fishes in the Narragansett Bay area: seven-year variations in the abundances of the winter flounder {Pseudopleuronectes americanus). J. Fish. Res. Board Can. 31:1057-1066. , and M. Terceiro. 1985. Cycle of changing abundances in the fishes of the Narragansett Bay area. Mar. Ecol. Prog. Ser. 25:239-244. , A. Keller, and S. Hale. 1986. Catch com- pilation: weekly trawl program Narragansett Bay-Rhode Island Sound 1966-1985. The Narragansett Bay Project. Finfish Rep. 1986-1. Jolly, G.M. 1965. Explicit estimates from capture-recapture data with death and immigration stochastic model. Biometrika 52:225-247. Hightower, J.E., and R.J. Gilbert. 1984. Using the Jolly-Seber model to estimate population size, mortality, and recruitment for a reservoir Jones, R. 1982. Population fluctuations and recruitment in marine populations. Phil. Trans. R. Soc, Lond. B 297:353-368. 220 Kennedy, J.S., and D.H. Steele. 1971. The win- ter flounder {Pseudopleuronectes americanus) in Long Pond, Conception Bay, Newfoundland. J. Fish Res. Board Can. 28:1153-1165. I^xjckwood, S.J. 1980. Density-dependent mor- tality in 0-group plaice {Pleuronectes platessa L.) populations. J. Cons. int. Explor. Mer 39:148-153. Klein-MacPhee, G. 1978. Synopsis of biological data for the winter flounder, Pseudopleuronectes americanus (Walbaum). NOAA Tech. Rep. NMFS Circ. 414. 43 pp. I^rda, B.C., and V.A. Crecco. 1987. Stock- recruitment relationship and compensatory mortality of American shad in the Connecticut River. Am. Fish. Soc. Symposium 1:469-482. KoUmeyer, R.C. 1972. A study of the Niantic River estuary, Niantic Connecticut. Final re- port phases I and II, physical aspects of the Niantic River estuary. Rep. No. RDCGA 18. U.S. Coast Guard Academy, New London, CT. 78 pp. Kuipers, B. 1975. On the efficiency of a two- metre beam trawl for juvenile plaice {Pleuronectes platessa). Neth. J. Sea Res. 9:69-85. , D.J. Danila, J.D. Miller, L.E. Bireley, and P.M. Jacobson. 1987. Assessing power plant impacts on fish populations at Northeast Util- ities sites: winter flounder studies at Millstone Nuclear Power Station. Pages 5-1 to 5-56 (Part 2) in Mechanisms of compensatory response of fish populations: workshop proceedings. Pre- pared by R.G. Otto and Associates and Science Applications International Corporation. EPRI Res. Rep. EA-5202. Electric Power Research Institute, Palo Alto, CA. Laurence, G.C. 1975. Laboratory growth and metabolism of the winter flounder Pseudopleuronectes americanus from hatching through metamorphosis at three temperatures. Mar. Biol. (Beri.) 32:223-229. . 1977. A bioenergetic model for the anal- ysis of feeding and survival potential of winter flounder, Pseudopleuronectes americanus, larvae during the period from hatching through meta- morphosis. Fish. Bull., U.S. 75:529-546. . 1979. Larval length- weight relations for seven species of Northwest Atlantic fishes reared in the laboratory. Fish. Bull., U.S. 76:890-895. Leim, A.H., and W.B. Scott, the Atlantic coast of Canada. Board Can. 155:1-485. 1966. Fishes of Bull. Fish. Res. Lobell, M.J. 1939. A biological t,arvey of the salt waters of \jong Island, 1938. Report on certain fishes. Winter flounder (Pseudopleuronectes americanus). Suppl. 28th Ann. Rep., N.Y. Cons. Dep., Pt. 1:63-96. MacCall, A.D., K.R. Parker, R. I^ithiser, and B. Jessee. 1983. Power plant impact assessment: a simple fishery production model approach. Fish. Bull, U.S. 81:613-619. May, R.C. 1974. Larval mortality in marine fishes and the critical period concept. Pages 3-20 in J. U.S. Blaxter, ed. The early life history of fish. Springer- Verlag, New York. McCracken, F.D. 1963. Seasonal movements of the winter flounder, Pseudopleuronectes americanus (Walbaum), on the Atlantic coast. J. Fish. Res. Board Can. 20:551-586. McIIugh, J.L. 1977. Fisheries and fishery re- sources of New York Bight. NOAA Tech. Rep. NMFS Circ. 401. 51 pp. MDMF (Massachusetts Division of Marine Fish- eries). 1985. Massachusetts marine fisheries. Assessment at mid-decade. Economic, envi- ronmental and management problems facing Massachusetts' commercial and recreational marine fisheries. Dept. Fish., Wildl. and Envir. Winter Flounder Studies 221 Law Enforcement, Exec. Off. of Envir. Affairs. 34 pp. + 10 fig. + 4 tab. Moller, n. 1984. Reduction of a larval herring population by jellyfish predator. Science (Wash., D.C.) 224:621-622. Moore, J.K., and N. Marshall. 1967. The re- tention of lameUibranch larvae in the Niantic estuary. The Veiiger 10:10-12. Nichols, J.D., B.R. Noon, S.L. Stokes, and J.E. nines. 198 1 . Remarks on the use of capture-re- capture methodology in estimating avian pop- ulation size. Studies in Avian Biol. 6:121-136. (not seen, cited by Hightower and Gilbert 1984). NMFS (National Marine Fisheries Service). 1987. Status of the fishery resources off the North- eastern United States for 1987. NOAA Tech. Mem. NMFS-F/NEC-50. U.S. Dept. of Com- merce, NOAA., Northeast Fisheries Center, Woods Mole, MA. 132 pp. NUSCO (Northeast Utilities Service Company). 1983. Millstone Nuclear Power Station Unit 3 environmental report. Operating license stage. Vol. 1-4. . 1986a. Monitoring the marine environ- ment of Long Island Sound at Millstone Nu- clear Power Station, Waterford, Connecticut. Annual report, 1985. . 1986b. The effectiveness of the Millstone Unit 1 sluiceway in returning impinged organ- isms to Long Island Sound. 18 pp. . 1987. Monitoring the marine environment of Long Island Sound at Millstone Nuclear Power Station, Waterford, Connecticut. Sum- mary of studies prior to Unit 3 operation. . 1988a. Ecological studies proposed for 1988 at Millstone Nuclear Power Station. Submitted to CT DEP. 1 1 pp. + 1 tab. + 2 fig. . 1988b. Winter flounder studies proposed for 1988 at Millstone Nuclear Power Station. Submitted to CT DEP. 21 pp. + 5 tab. -I- 8 fig. + 20 pp. app. . 1988c. The effectiveness of the Millstone Unit 3 fish return system. Submitted to CT DEP. 21 pp. OUa, B.L., R. Wicklund, and S. Wilk. 1969. Behavior of winter flounder in a natural habitat. Trans. Am. Fish. Soc. 98:717-720. Pearcy, W.G. 1962. Ecology of an estuarine population of winter flounder Pseudopleuronectes americanus (Walbaum). Bull. Bingham Oceanogr. Coll. 18(l):l-78. Pennington, M. 1986. Some statistical techniques for estimating abundance indices from trawl surveys. Fish. Bull., U.S., 84:519-525. Perlmutter, A. 1947. The blackback flounder and its fishery in New England and New York. Bull. Bingham Oceanogr. Coll. 11:1-92. Poxton, M.G., A. Eleftheriou, and A.D. Mclntyre. 1982. The population dynamics of 0-group flatfish in the Clyde Sea area. Est. Coast. Shelf Sci. 14:265-282. , A. Eleftheriou, and A.D. Mclntyre. 1983. The food and growth of 0-group flatfish on nursery grounds in the Clyde Sea area. Est. Coast. Shelf Sci. 17:319-337. , and N.A. Nasir. 1985. The distribution and population dynamics of 0-group plaice {Ple.uronecte.i platessa L.) on nursery grounds in the Firth of Forth. Est. Coast. Shelf Sci. 21:845-857. Rickcr, W.E. 1954. Stock and recruitment. J. Fish. Res. Board Can. 11:559-623. ____. 1975. Computation and interpretation of biological statistics of fish populations. Bull. Fish. Res. Board Can. 191:1-382. 222 Roff, D.A. 1981. Reproductive uncertainty and the evolution of iteroparity: why don't flatfish put aU their eggs in one basket? Can. J. Fish. Aquat. Sci. 38:968-977. Rogers, C.A. 1976. Effects of temperature and salinity on the survival of winter flounder em- bryos. Fish. Bull., U.S. 74:52-58. Limnol. Oceanogr. 6:292-298. Saila, S.B. 1961. A study of winter flounder movements. Limnol. Oceanogr. 6:292-298. . 1962a. The contribution of estuaries to offshore winter flounder fishery in Rhode Island. Proc. Gulf Caribb. Fish. Inst. 14th Annu. Sess. 1961:95-109. . 1962b. Proposed hurricane barriers related to winter flounder movements in Narragansett Bay. Trans. Am. Fish. Soc. 91:189-195. . 1976. Effects of power plant entrainment of winter flounder populations near Millstone Point, Connecticut. URI-NUSCO Rep. No. 5. 139 pp + 4 app. , and E. Lorda. 1977. Sensitivity analysis applied to a matrix model of the Hudson River striped bass population. Pages 311-332 in W. Van Winkle, ed. Proceedings on the conference on assessing the effects of power-plant-induced mortality on fish populations. Pergamom Press, New York. SAS Institute, Inc. 1985. SAS user's guide: sta- tistics. Version 5 edition. SAS Institute Inc., Cary, NC. 956 pp. Scott, W.C.M. 1929. A note on the effect of temperature and salinity on the hatching of the eggs of the winter flounder {Pseudopleuronectes americanus Walbaum). Contrib. Can. Biol., New Ser. 4:139-141. construction of sustainable yields curve. J. Cons. int. Explor. Mer 40:67-75. , J.G. Pope, and R.D. Cousens. 1984. Variations in fish stocks and hypotheses con- cerning their links with climate. Rapp. P. -v. Reun. Cons. int. Explor. Mer 185:255-267. Sissenwine, M.B. 1974. Variability in recruitment and equilibrium catch of the Southern New England yellowtail flounder fishery. J. Cons, int. Explor. Mer 36:15-26. 1977. A compartmentalized simulation model of the southern New England yellowtail flounder fishery. Fish. Bull., U.S. 73:465-482. 1984. Why do fish populations vary? Pages 59-94 in R.M. May, ed. Exploitation of marine communities. Springer- Verlag, New York. Smigielski, A.S. 1975. Hormonal-induced ovulation of the winter flounder, Pseudopleuronectes americanus. Fish. Bull., U.S. 73:431-438. Snedecor, G.W., and W.C. Cochran. 1967. Sta- tistical methods. The Iowa State University Press, Ames, lA. 593 pp. Sokal, R.R., and F.J. Rohlf. 1967. Biometry. W.H. Freeman and Company, San Francisco. 775 pp. Southwood, T.R.E. 1978. Ecological methods. Halstead Press, New York. 523 pp. Steele, J.II., and R.R.C. Edwards. 1970. The ecology of 0-group plaice and common dabs in I^ch Ewe. IV. Dynamics of the plaice and dab populations. J. Exp. Mar. Biol. 4:174-187. Ursin, E. 1982. Stability and variability in the marine ecosystem. Dana 2:51-67. Shepherd, .I.G. 1982. A versatile new stock- recruitment relationship for fisheries, and the Van Winkle, W. (ed.). 1977. Assessing the effects of power-plant-induced mortality on fish Winter Flounder Studies 223 populations. Pergamom Press, New York. 380 pp. Veer, H.W. van der. 1985. Impact of coelenterate predation on larval plaice Pleuronectes platessa and flounder Platichthys flesus stock in the western Wadden Sea. Mar. Ecol. Prog. Ser. 25:229-238. . 1986. Immigration, settlement, and density-dependent mortality of a larval and early postlarval 0-group plaice {Pleuronectes platessa) population in the western Wadden Sea. Mar. Ecol. Prog. Ser. 29:223-236. Ware, D.M. 1980. Bioenergetics of stock and recruitment. Can. J. Fish. Aquat. Sci. 37:1012-1024. White, G.C., D.R. Anderson, K.P. Burnham, and D.L. Otis. 1982. Capture-recapture and re- moval methods for sampling closed popula- tions. LA-8787-NERP, Los Alamos National laboratory, Los Alamos, NM. 235 pp. Williams, G.C. 1975. Viable embryogenesis of the winter flounder Pseudopleuronectes americanus from -1.8° to 15°C. Mar. Biol. (Berl.) 33:71-74. Erratum NUSCO (1987): The intercept for the length- weight relationship for Niantic River winter floun- der found on Table 10 in this report should have been -2.2636 instead of the value of -2.2261 that was given. 224 Contents Exposure Panel Program 227 Introduclion 227 Exposure Panel Study 228 Materials and Methods 228 Seawatcr Temperature 228 Sampling Procedure 228 Sample Processing 230 Data Analysis 230 Results 231 Seawater Temperatures 231 Fouling Species 233 Wood-boring Species 237 Interactions Between Fouling and Wood-boring ("ommunitics 239 Discussion 243 Distribution Study 246 Materials and Methods 246 Results 247 Discussion 247 Summary 250 Conclusions 250 References Cited 250 Exposure Panel Program Introduction The Exposure Panel Program was designed to determine what effect the operation of Millstone Nuclear Power Station (MNPS) has on the abun- dance and distribution of marine woodborers, and the rate of wood degradation in the marine ecosystem of Long Island Sound (LIS). Two groups of marine woodborers are common in these waters. The first group consists of small crustaceans, isopods {Limnoria spp.) and amphipods {Chelura terebrans), which make shal- low tunnels and excavations in the surfaces of submerged wood. The second group consists of the more destructive shipworms Teredo navaUs and Teredo bartschi, wliich are molluscs that pen- etrate the surface of wood as larvae (pediveligers) prior to metamorphosing into their adult, worm- like form. Woodborers are very important eco- logically because they decompose wood entering estuarine, coastal and deep ocean areas (Turner 1984). However, they also cause destruction of man-made structures. Particular attention has been paid in recent years to the response of woodborer populations exposed to heated effluents from power plants, because elevated temperatures may increase growth, reproduction, and sur\'ival of woodborers, all of which contribute to accelerated destruction of wooden structures (Naylor 1965; Turner 1973; Iloagland 1981). These effects were seen near Jersey Central Power and I ,ight Company's Oyster Creek Nuclear Generating Station, where destruc- tion of docks at nearby marinas was attributed to increased woodborer activity in the effluent (Turner 1973; Hoagland and Turner 1980; Macieolek-Blake et al. 1981). The objectives of the MNPS Exposure Panel Program are: 1. to monitor the abundance of marine woodborers at five sites in the Millstone Point area, 2. to quantify the loss of wood associated with the presence of woodborer populations in the vicinity of MNPS, 3. to monitor the dispersal of Teredo barlschi in terms of distance from the Millstone Quarry, and, 4. to monitor the abundance of prevalent fouling organisms, and to investigate their relation- ship to woodborer abundances in the Mill- stone Point area. To achieve these objectives, three separate stud- ies were conducted. The first (Exposure Panel Study) used exposure panels to monitor the abun- dance of fouling and wood-boring species, as well as the associated wood-loss. The second study (Distribution Study) used exposure panels de- ployed in close proximity to the MNPS discharge to monitor the density of Teredo navalis and Teredo barlschi relative to the distance from the thermal discharge. The third study (Timber Study) used commonly available dock building materials to quantify wood-loss. Development of these studies, and results prior to 3-unit operation (1968-1986), are summarized in NUS(X) (1987). This report represents an initial assessment of the effects of 3-unit operation on marine woodborers, based on results from the Exposure Panel Study and the Distribution Study. Results from the Timber Study will be presented in a future report, scheduled for 1989. Exposure Panel Program 227 Sites 0 y Exposure Panel Study WP, Fl, EF, BP, GN Distribution Study 100 m, 500 m, 1000 m 1 km Fig. 1. Location of exposure panel sites in the vicinity of the Millstone Nuclear Power Station (WP = White Point, FI = Fox Island, EF = FfTlucnt, BP = Black Point, GN = Giants Neck, 100 m = trawl-line at 100 m, 500 m = trawl-line at 500 m, 1000 m = trawl-line at 1000 m). Exposure Panel Study IVIaterials and Methods Seawater Temperature Seawater temperature data during each expo- sure period at several of the sites near MNPS have been summarized. Water temperatures were derived from the EDAN (Environmental Data Acquisition Networic) system, which continually records a variety of environmental parameters at 1 5-minute intervals. Ambient water temperatures were recorded by sensors in Unit 1 and Unit 2 intake bays, and effluent water temperatures by sensors in the quarry cuts. Prior to 1987, tem- peratures at White Point and Eox Island were measured continuously with thermistors and either strip or circle chart recorders. In 1987, these data were collected using solid state data loggers. Sampling Procedure The present study used sets of six replicated wood panels submerged at five sites: White Point (WP), Eox Island (EI), Black Point (BP), Giants Neck (GN), and Effluent (E,E), which was in the Millstone Quarry where panels were exposed to maximum discharge water temperatures (Eig. 1). 228 Wood 1.9 an Plexiglass 0,6 cm Fig. 2. Frame and rack assembly used for holding sixmonlh, six-replicate exposure panels at sites in the vicinity of the Millstone Nuclear Power Station. Each exposure panel was a knot-free pine board (25.4 X 8.9 X 1.9 cm) with one face covered by plexiglass. Only the uncovered wood side of each panel was processed. Two sets of six replicated panels were bolted to a stainless steel rack attached to a stainless steel frame at each site (Fig. 2). The rack and frame assemblies deployed at WP, FI, BP, and GN were suspended from docks by ropes in waters not exceeding 2 m in depth; the lower edge of the panels was maintained 0.2 m off the bottom. At EF, two rack and frame assemblies were used; the first (ES) was 1 m below the sur- face, and the second (EB) about 1 m off the bottom at high tide, at a depth of 10 m. The panels at EB were moved in May 1987 to the cantilever dock leading to the Floating Lab, a location used for exposure panels prior to 1979. At this new location, the EB panels are maintained in shallow water 0.2 m off the bottom, as at the other sites. The panels were placed at each site in February, May, August and November and collected six months later in August, November, February and May, respectively. This provided four exposure periods, each overlapping the next by three Exposure Panel Program 229 months. At the start of each exposure period, one rack of panels was removed for processing and a new rack with fresh panels was deployed. ITiroughout this report the exposure periods will be referred to using the following abbreviations: Feb-Aug, May-Nov, Aug-Feb and Nov-May. Each abbreviation refers to the month of panel deployment followed by the month of panel col- lection. Sample Processing After collection, panels were either refrigerated at 5 "C and processed immediately or frozen and processed at a later time. Primary cover, as a percentage, was estimated for each organism that occupied more than 1% of the panel surface, e.g., barnacles, biyozoans, tunicates and some algae. Beginning in 1980, cover was estimated for frecspace, mud and the dead tests of fouling spe- cies, to complete the description of total primary cover for each panel. Numerical abundance was determined for barnacles and mussels by counting the individuals on each panel. If the number of individuals per panel exceeded 100, six subsamples of 1 X 1 inch were randomly selected, three from the upper half and three from the lower half of the panel. In 1981, only the asbestos side (cur- rently replaced by plexiglass) of each exposure panel was used for determining numerical abun- dance of fouling species; therefore, numerical abundance data for barnacles and mussels were not recorded for the wood panels in 1981. The abundance of woodborers was determined after the panel had been scraped of fouling species. All individuals of the genera Limnoria and Chelura were counted when densities were less than 100 individuals per panel; otherwise, the subsampling scheme previously described for barnacles and mussels was used. Subsampling was always con- ducted evenly between the top and the bottom halves so that approximately 100 individuals were collected from each panel. After assessing the limnorid and chelurid abundances, panels were frozen and subsequently examined by means of X-ray photography (80 kV, 5 mA, for 1.2 min). The radiographs were used to count the number of shipworms. Teredo navalis and T. hartschi, and visually estimate the percentage of wood lost per panel. The percentage of wood lost was deter- mined by rating the general proportions of bright areas, caused by various densities of shipworm tubes and the dark areas caused by various degrees of wood-loss. To determine the species of shipworms collected, shipworms were randomly removed from the panels until all or at least 100 individuals were identified from each site. Shipworms smaller than 5 mm in length were classified as Teredo juveniles because their pallets were too small and undeveloped to allow accurate identifications. Data Analysis The first set of exposure panels for which re- sults are presented in this report was deployed in November 1978, and the last set was collected in August 1987. Each exposure period is represented, therefore, by six panels replicated over years, a maximum of four times during 2-unit operation, and one or two times during 3-unit operation. I'he actual numbers of panels processed for each site/exposure period combination are presented in Table 1 . Unit 3 began commercial operation on April 23, 1986, and although some intermittent operation and testing of circulation pumps oc- curred before this date, we have considered all sampling periods from Nov-May of 1979 to Nov- May of 1986 as 2-unit operation. The monitoring of all panels was suspended from November 1981 to February 1985 to investigate the life histories of two shipworms. Teredo navalis and T. hartschi in relation to seawater temperature. All averages and other summary statistics for each taxon were computed by exposure period. Temperature data were averaged by month. Per- centage of primary cover and counts of individuals were summarized as means. Data provided in histograms were summarized by year within each exposure period, while corresponding tables com- pare data collected before vs. after 3-unit opera- tion. This format represents the annual variability as well as the differences between 2-unit and 3-unit operational periods. 230 TABLE 1. The total number of panels used in comparisons for the time periods of 'before' (B = panels collected from May 1979 to May 1986) versus 'after' (A = panels collected from August 1986 to August 1987) three-unit operation at the Millstone Nuclear Power Station, nominally 6 panels/exposure period/year. E X I ' O S U R E P E R 1 O D PARAMETER SITE Aug - Feb Nov - May Feb - Aug May Nov B A B A B A B A Effluent bottom 6 6 6 6 6 12 6 6 Primary cover Effluent surface 18 6 24 6 24 12 24 6 and Black Point 3^ 5^ 6 6 0^ 12 0=^ 6 counts of Giants Neck 18 6 24 6 24 12 24 6 Woodborers Fox Island 18 6 24 6 18^ 12 18= 6 White Point 18 6 24 6 24 12 23=^ 6 Effluent bottom 6 6 6 6 6 12 6 6 Counts of Effluent surface 12 6 18 6 18 12 18 6 Barnacles Black Point 3=^ 5" 6 6 0= 12 0=^ 6 and Giants Neck 12 6 18 6 18 12 18 6 Mussels Fox Island 12 6 18 6 12= 12 12= 6 White Point 12 6 18 6 18 12 17^ 6 Some panels lost, attributed to Hurricane Gloria (Sept. 1985). One panel lost from frame and rack assembly. Panels inadvertently exposed for 3 months or 9 months excluded from analyses. Percentage of wood lost from panels in 1979 was based on visual assessments made while split- ting the panels in search of shipworms. From 1980 to 1987 wood-loss was based on percentages derived from radiographs, which correlated well (r^ = 0.98) with panel weight loss data (NUSCo 1987). Results Seawater Temperatures Each of the four exposure periods was charac- terized by a different water temperature regime (Fig. 3). The seawater temperatures during the Feb-Aug and Aug-Feb exposure periods are best described by their ranges because temperatures continually increase and decrease, respectively. Feb-Aug began when ambient water temperatures (measured at MNPS intakes) were coldest ( < 3 °C) and ended when water temperatures were warmest ( > 20 °C). Its converse, the Aug- Feb period started with warm temperatures and ended with cold. Seawater temperatures during the May- Nov and Nov-May exposure periods are more appropriately described by average values because they are composed of the warmest (av- erage = approx. 16 °C) and coldest (average = approx. 7 "C) months of the year, respectively. Panels at the effluent sites (EB, ES) were exposed to undiluted thermal effluent; since 1978, effluent temperatures have averaged 9-11 °C above ambi- ent, and maximum AT's were in the range of Exposure Panel Program 231 MAY JUN JUL AUG NOV DEC FEB MAR APR MAY Fig. 3. The average monthly seawater temperatures at the Unit 1 & 2 intakes (ambient, bottom pair of lines in each graph) and the Millstone Quarry cuts (effluent, the top pair of lines in each graph) during tfie four six-month exposure periods. The monthly averages are from the 15th of one month to the 15th of the next (— average temperatures during 3-unit operation, — average temperatures from November 1978 to May 1986, vertical bars represent the range of average monthly temperatures over this seven and one half year period of 2-unit operation). 12-15 °C. Effluent temperature regimes (i.e., annual tein- perature ranges) have not changed substantially since Unit 3 began operation; however, the volume of cooling water has almost doubled. Prior to 1986, sampling sites at WP, FI, BP, and GN were unaffected by the MNPS thermal plume; annual average temperatures at these sites were within 0.6 °C of those at the intakes. From early spring through early autumn, temperatures at the FI sampling site were up to 2 °C above ambient, but these elevated temperatures were attributed to natural warming of shallow waters in Jordan Cove by solar radiation. Similar insolation of shallow estuaries has been reported elsewhere (Dale and Gillespie 1977; Dean and Officer 1977; de Wilde and Berghuis 1979). Since Unit 3 began operation, the added volume of cooling water has increased the areal extent of the MNPS thermal plume, which on an ebbing tide raises water temperatures in Jordan Cove (see the I lydrothermal Studies section of this report). During 3-unit operation in 1987, water tempera- tures at FI and WP were close to ambient during most of the tidal cycle (9-10 h), but maximum AT's of 2-4 °C occurred for 2-3 hours per cycle (Fig. 4). Insolation in summer might raise tem- peratures an additional 2 °C. 232 19 White Point i 1 IS' 17- A ^n l/l fl 1 16- h L-^d— Sf^ i^ 15- 14-- -"v-V" ^ Day 1 Day 2 Day 3 i9T Fox Island — 1 — P— : — : — I — r Day 1 I I I Day 2 Day 3 Fig. 4. Seawater temperatures at White Point and Fox Island from October 15-17, 1987 during 3-unit operating conditions. The solid line across the bottom of each graph represents the daily mean temperatures as recorded at Unit 1 & 2 intakes. Fouling Species Percent cover. The percentage of panel surfaces covered by fouling organisms varied among sites and years and between operational periods (Fig. 5). Average values ranged from 95.5% at FI in Nov- May 1987 to < 1% at WP and GN in Aug- Feb 1986. In general, highest percentages occurred at BP and FI, and lowest at EB. The remainder of the panel surfaces was free space, or covered by the calcareous remains of dead barnacles or bryozoans (these dead remains occupied up to 40% cover; Fig. 5). Almost 300 species of plants and animals, in- cluding both motile and sessile members of the local fouling community, have been identified during the exposure panel study (NUSCO 1982). Present studies focus on the sessile assemblages because these aie more likely to influence woodborer abundances (cf. Naix and Saraswathy 1971). Of these, only 22 taxa (3 plant and 19 animal) occupied a mean of 1 % or more of the panel surface at any site during any exposure pe- riod, during 2-unit or 3-unit operation. The abun- dances of these 22 taxa are presented in Table 2. These species exhibited different spatial and temporal patterns of distribution. Some species were ubiquitous, found at all sites in both oper- ational periods, e.g., Balanus crenatus, especially abundant in the Nov-May exposure period. Exposure Panel Program 233 100 80 60 40 20 0 100' 20- Effluent Bottom LH n LlJ > 100 (J RO U Ll. 60 o 40 U 20 < 0 ^ 100 Ld O ct: Ld CL 80 60 Ld 20 < 0 Q; 100 Ld > 80 "^ 60 40 20 0 100 AUVE Effluent Surface Black Point Giants Neck Fox island White Point 0 1 6 7 AUG-FEB 7 8 8 8 8 8 9 0 15 6 7 FEB -AUG 7 8 8 8 8 9 0 15 6 MAY-NOV rig. 5. Mean % cover of sessile fouling organisms on exposure panels collected from 1979-1987. Histograms reflect the total percentage of cover, which was comprised of both living plant and animals and their dead remains (* = values < 1 %). 234 TABLE 2. Mean % cover of the most prevalent fouling organisms on exposure panels before (B) and after (A) three-unit operation at MNPS. EXPOSURE PERIOD SITE TAXON Aug - Feb Nov May Feb - Aug May Nov B A B A B A B A EB Alcyonidium spp. Anomia simplex a 1.3 17.8 t'' ... ... ... ... Balanus crenatus t ... 23.7 7.7 ... t ... ... Balanus eburnew; ... 1.5 ... ... 1.3 t 8.7 9.0 Balanus improvisus ... t ... ... t t t 1.3 Balanus juveniles ... ... ... 1.0 t 1.1 ... 1.7 Bugula spp. ... 11.3 ... 2.5 ... 2.8 ... 7.0 Me Iridium senile ... t ... ... 5.0 1.0 ... ... Mycale fibrexilis ... ... — ... ... ... ... 3.3 Mytilus edulis ... t 6.3 5.8 ... ... ... ... Serpulid tubes t 1.3 ... ... ... t t ... Tubularia crocea ... ... ... 1.2 ... ... ... ... ES Alcyonidium spp. 9.2 Balanus crenatus t ... 10.7 3.0 t ... ... t Balanus eburneus 2.7 t ... — 3.5 2.6 9.2 3.5 Balanm improvisus 22.2 t 3.4 ... 9.9 11.8 1.0 Balanus juveniles 1.0 ... 1.8 ... 1.6 10.3 t Bugula spp. t 9.0 ... 2.7 t ... ... t Halichondria spp. ... t ... ... ... 1.8 ... ... Metridium senile ... — ... t 1.5 ... ... Mycale fibrexilis ... ... ... ... ... ... ... 21.3 Mytilus edulis 1.6 1.0 17.8 16.0 ... ... ... ... Tubularia crocea 1.5 ... 3.8 t t ... t ... BP Balanus balanoides ns= 37.8 ns Balanus crenatus ... ... 6.7 3.0 ns 15.5 ns ... Balanus eburneus .- t ... ... ns ns 4.7 Balanm improvisus ... t 2.0 ... ns ns 1.3 Balanus juveniles ... — 4.0 15.2 ns ns ... Botryllus schlosseri 2.4 5.6 ... ... ns ns t Codium fragile 27.3 t ... ... ns ... ns ... Cryptosula pallasiana 1.7 19.6 ... ... ns 1.3 ns 20.8 Derbesia marina ... 12.8 ... ... ns ... ns 3.8 Halichondria spp. ... ... ... ... ns ns 2.0 Laminaria saccharina ... ... 8.8 4.5 ns ... ns ... Schizoporella errata ... 7.0 ... — ns 4.5 ns 24.5 Exposure Panel Program 235 TABLE 2. (cont'd) SITE TAXA Aug B I Feb A i X P O Nov - B S U R I May A i PE Feb - B R I O C Aug A ) May - B Nov A GN Balanus amphitrite Balanus balanoides Balanus crenatus Balanus eburneus Balanus improvisus Balanus juveniles Botryllus schlosseri Cryptosula pallasiana Derbesia marina Halichondria spp. Laminaria saccharina Schizoporella errata t t t 4.8 t t 1.0 1.0 11.4 6.3 2.3 2.7 20.3 1.7 13.9 2.5 4.9 6.5 t t t t t 1.7 1.4 4.9 1.5 3.2 0.6 3.3 1.1 t 1.0 2.4 t 5.5 FI Balanus balanoides Balanus crenatus Balanus eburneus Balanus improvisus Balanus juveniles Botryllus schlosseri Bugula spp. Cryptosula pallasiana Schizoporella errata t t t 1.1 t 6.0 t 1.3 t 56.8 35.2 12.1 79.8 15.0 14.3 t 1.0 t t t 6.0 26.1 8.7 t t t t 1.0 t 1.1 2.0 1.3 1.0 14.4 t t 3.9 29.2 WP Balanus crenatus Balanus eburneus Balanus improvisus Balanus juveniles Botryllus schlosseri Bugula spp. Cryptosula pallasiana Derbesia marina Halichondria spp. Laminaria saccharina Schizoporella errata Scypha spp. 1.2 t 12.0 2.4 4.3 3.7 10.8 t 4.6 5.8 5.8 12.5 1.2 6.6 2.9 1.2 t 1.3 1.7 1.8 t 22.1 t 1.3 t t 5.9 1.2 2.2 t 1.1 6.5 1.5 1.4 2.9 t 0.7 6.8 1.2 (— ) = taxon has never occurred on a panel (t) = trace percentage cover, < Wo of the panel's surface (ns) = exposure period not sampled Balanus improvisus was found at all sites, but was abundant only at ES before 3-unit operation. Other species found during both operational pe- riods were consistent components of the fouling communities at only effluent sites (EB, ES), e.g., Mytilus edulis, or at only ambient sites (WP, FI, BP, GN), e.g., Cryptosula pallasiana and Laminaria saccharina. Still other species had 236 more restricted distributions, e.g., Codium fragile only at BP, and Alcyonidium spp. and Tuhularia crocea only at effluent sites during 2-unit opera- tion, or Mycale fibrexilis at ES, and Balanus halanoides at ambient sites during 3-unit opera- tion. Balanus balanoides is typically an intertidal barnacle; its presence on subtidal panels is unpre- dictable. For example, Battelle researchers (1968-1978) found 65-70% cover of/?, balanoides in 1970 and 1971 at GN, and less than 2% in all other years (Battelle, unpublished data). Numerical abundance. Of the 22 species prev- alent in terms of percentage of cover, seven were assessed by counting the number of individuals per panel: Balanus amphitrite, B. balanoides, B. crenatus, B. ehurneus, B. improvisus, Balanus ju- veniles, and Mylilus edulis (Table 3); these organ- isms occur as individuals, and are most likely to influence woodborer attack. Generally, numerical abundance data support conclusions drawn from percentage cover, and demonstrate similar distri- butions, e.g., Balanus crenatus was the most abun- dant fouling organism, B. improvisus was more abundant during 2-unit operation than 3-unit op- eration, and was most abundant at ES, and B. balanoides was found only during the Feb-Aug period during 3-unit operation. In some cases, general patterns can have a single probable cause. For example, the lower abundances of juvenile Balanus in the Aug-Feb 'after' period, and the higher abundances in the Nov- May period are attributable to a slightly later set during this first year of 3-unit operation. If these juveniles were primarily B. crenatus, it would explain the gener- ally lower densities for B. crenatus during the Nov-May 'after' period (i.e., individuals, settling late, were too small to be identified in May). Wood-boring Species Numerical abundance. Most of the wood de- composition (both naturally-occurring and in man-made structures, e.g., docks and wooden lobster pots) that occurs in local waters is the result of tunneling and feeding by shipworms. Teredo spp. Therefore, one measure of the inten- sity of potential wood-loss is the numerical abun- dance of teredinids. Counts of the native species of shipworm. Teredo navalis, were highest in pan- els exposed from May to November (Fig. 6), and ranged to a maximum of 300 per panel. This exposure period encompasses the entire settlement period (roughly July-September, Graves 1928; Ilillman et al. 1985), and densities were higher than in Aug-Feb or Feb-Aug exposure periods, which included only part of the settlement season. Settlement of T. navalis did not occur during the Nov-May exposure period. Seasonal patterns of shipworm abundance are also seen in densities of Teredo juveniles (Fig. 7). At ambient water sites (GN, FI, WP), young shipworms ( < 5 mm; too small to be identified to species) were common only in the Feb-Aug period, indicating recent settlement. At the effluent sites (EB, ES), Teredo juveniles were found in each exposure period; however, these were presumably young Teredo bartschi, a non- native shipworm with an extended reproductive season. Adult T. bartschi have occurred in the MNPS effluent since 1975, and were particularly abundant (100-150 per panel) in Aug-Feb and May-Nov exposure periods (Fig. 8). Comparisons of abundance during 2 -unit and 3-unit operation, for each category of woodborers, are presented in Table 4. Of particular interest are the increased densities of T. bartschi at the effluent sites, and the increased densities of T. navalis during the May-Nov exposure periods at WP and FI (the sites most likely to be influenced by the 3-unit plume). Densities of the wood- boring crustaceans, Limnoria spp. and Chelura terebrans, although variable in time and space, show little direct relationship to operational his- tory or to levels of wood-loss. For example, densities of Limnoria and Chelura at WP decreased during 3-unit operation, when wood-loss in- creased, and densities of Limnoria at GN were higli during the Nov-May exposure period, when wood-loss was negligible. Percentage of wood-losss. The amount of wood lost from panels is closely related to the numerical abundance of shipworms. Wood-loss varied Exposure Panel Program 237 TABLE 3. Average density per panel (30.6 in ) for barnacles and mussels before (B) and after (A) three-unit operation at MNPS. Means with *'s indicate a significant difference exists between B and A (Mann-Whitney U-test, p < 0.05). EXPOSURE PERIOD FAXON SITE Aug - Eeb Nov - May Eeb Aug May - Nov B A B A B A B A EB 0 1 0 0 0.2 0.5 0 1 ES 11 0.5 0 0 1 0.4 0 0.3 Balanus amphilrite BP 1 0.2 0 0 - 1 - 1 GN 1 0.2 0 0 12 0 0 0 EI 0 0 0 0 3 0 0.1 0 WP 9 0.3 1 0 15 0.2 0.1 0 EB 18 1 148 71 0 1 0 0 ES 7 0 32 21 0 0.1 0 1 Balanus crenatus BP 0 0 280 71 - 266 - 0 GN 0.1 0 236 136 101 69 1 0 EI 3 0 435 768* 13 53 0.4 0 WP 0.1 0.0 111 170 73 272 0 0 EB 0 2 0 0 9 2* 10 10 ES 8 1 0 0 8 9 19 3 Balanus eburneus BP 0 0.6 0 0 - 8 - 6 GN 0 0 0 0 3 0.1* 5 0 EI 0 0 0 0 0.5 0.2 0.4 0.2 WP 1 0 0 0 26 0 8 0 EB 0.3 k 2 0 3 10 6 1 15 ES 239 4* 16 3 66 11* 64 13 Balanus improvisus BP 2 1 0 0 - 10 - 4 GN 1 0.5 0 0 6 1 3 0 EI 7 7 0.2 0.3 1 0.4 0 6 WP 32 r 0.04 0.2 23 0.1* 4 0 238 TABLE 3. (cont'd) EXPOSURE PERIOD TAXA SITE Aug - Feb Nov - May Feb - Aug May Nov B A B A B A B A EB 0 0 0 0 0 0 0 0 ES 0 0 0 0 0 0.1 0 0 Balanus balanoides BP 0 0 0 0 0 450 0 0 GN 0 0 0 0 0 50 0 0 FI 0 0 0 0 0 386 0 0 WP 0 0 0 0 0 4 0 0 EB 12 0.3 3 122 211 77 0.2 30 ES 56 o.r 8 14 60 41* 109 5 Balanus juveniles BP 0.6 0.2 820 2546* -- 48 -- 0 GN 96 20* 722 2237* 45 23* 19 0 FI 86 107 761 762 9 16 10 2 WP 29 2 561 2060* 308 70* 11 0 EB 0 12 91 159 0 0 0 0 ES 169 20 122 318 0 0 0 0 Myti/us edulis BP 0 0 22 3 -- 2 0 0 GN 4 0 3 0.3 22 0.1 0 0 FI I 0 7 33 0.3 1 0 0 WP 150 0.2 26 6 9 0.6 2 0 among sites, exposure periods, and years (Fig. 9). At WP, FI, BP and GN, wood-loss was greatest in the May-Nov exposure period (up to 95%) and greater at GN and WP ( > 60%) than at BP and FI (<40%). At the effluent sites (EB and ES), considerable wood-loss occurred in both Aug-Feb and May-Nov exposure periods during 3-unit operation. At all sites, wood-loss was low during the Feb-Aug exposure period; in the Nov- May period, wood-loss was very low at the effluent sites, and zero at the ambient sites. The disproportionately low levels of wood-loss (rela- tive to Teredo abundance) in Feb-Aug are ex- plained by shipworm size; they are young and small, i.e., recently settled. Similarly, high densities of Teredo juveniles in Feb-Aug have little effect on wood-loss. Interactions Between Fouling and Wood-boring Communities Analysis of relationships among components of the communities that develop on and in the exposure panels show that densities of shipworms are negatively correlated with the abundance of foulcrs. In some cases, the effect of foulers on woodborers is direct, e.g., high densities of bar- nacles on a panel reduce the amount of space available to settling Teredo larvae (Fig. 10). In Exposure Panel Program 239 AUG-FEB FEB -AUG MAY- NOV Fig. 6. Mean numerical abundance of the shipworm, Teredo navalis, in exposure panels collected during 1979-1987 (* - abundance < 5). 240 125- Effiuent aoitom 3 UNIT OPERATION ^ A^R 100- 1 1 RFFORF 75- 50- 25- m **^ A A A A 0 - 125- Effluent Surface 100- 75- 50- 25- * * *m * * * * n **^ AAA A A co 0- -J 125- Black Point 3 100- Q 50- ? 25- 1 1 - * * * * * A A '--'125-1 lI] 100- 1 75- ZD 50- Z 25- ^ n Giants Neck * * * * * A AAA AAA A A O 0 < 125-] m U] 100- Fox Island 50- 25- * * * * * * AAA A AAA A A 0 - 125- White Point 1 DO- TS- 50- 25- A * * * * r— r a* A A A * A AAA A A Q- 8 8 8 8 0 1 6 7 7 8 8 8 8 8 9 0 15 6 7 7 8 8 8 8 9 0 15 6 7 8 8 8 8 9 0 1 6 7 AUG-FEB FE B- -A UG MAY- NOV NOV-MAY Fig. 7. Mean numerical abundance of 7>rerf(7 juveniles in exposure panels collected during 1979-1987 (* = abundance < 5). Exposure Panel Program 241 TABLE 4. Average density per panel (30.6 in ) for shipworms, limnorids and chelurids before (B) and after (A) three-unit operation at MNPS. Means with *'s indicate a significant difference exists between B and A (Mann- Whitney U-test, p < 0.05). EXPOSURE PERIOD Aug ■ Feb Nov - May Feb - Aug May Nov TAXON SITE B A B A B A B A EB 2 32' 0 0 1 1 6 8 ES 8 17 0.04 0 0.5 1 7 9 Teredo navalis BP 6 8 0 0 - 0.3 - 40 GN 74 59 0 0 47 4* 210 131 FI 9 8 0 0 7 0 15 32* WP 33 37 0 0 9 11 50 223* EB 2 99' 0 3 3 20 8 118' ES 13 140* 0.1 4* 1 14* 2 72* Teredo barlschi BP 0 0 0 0 - 0 - 0 GN 0 0 0 0 0 0 0 0 FI 0 0 0 0 0 0 0 0 WP 0 0 0 0 0 0 0 0 EB 0 17 0 0 1 4 0 0.7 ES 0.4 11 0.6 0.2 1 2 6 4 Teredo juveniles BP 0 0 0 0 - 0.2 - 0 GN 0.1 0 0 0 72 2* 0.5 0 FI 0.6 0.2 0 0 13 0.1* 0.2 0.2 WP 0.1 0 0 0 24 6 0 1 EB 3 0.2 0 0.3 0.5 69* 1 0 ES 1 0.2 1 0.2 86 * 2 7 0.3 Limnoria spp. BP 3 25 4 116* - 142 - 216 GN 153 26* 209 77* 567 141 335 557 FI 26 2 0.3 0 0 0.2 0 0.3 WP 95 37 ' 29 66 520 185* 1610 319* EB 0 0 0 0 0 0 0 0 ES 0 0 0 0 0 0 0 0 Ckelura terebrans BP 0 0 0 0 - 0.1 - 1 GN 2 0,3 0 0 1 0 101 56 FI 1 0 0 0 0 0.2 0 0.3 WP 1 0.2 0 0 18 2 871 24* 242 _l < 9 LJ > < z: Q^ — LJ Li_ ^° Ld DQ 200 1 Effluent Bottom 150- 100 50 0 200 150 100 50 0- 3 UNIT OPERATION Eg^^g AFTER tffluent Surface Ik * * * 0 1 6 7 AUG-FEB 7 8 8 8 8 8 9 0 15 6 7 FEB -AUG BEFORE * * * * * * * * 7 8 8 8 8 9 0 15 6 MAY- NOV 7 8 8 8 8 9 0 1 6 7 NOV- MAY Fig. 8. Mean numerical abundance of the shipworm, Teredo harischiy in exposure panels collected during 1979-1987 {* = abundance < 5). other cases, effect is more indirect; Limnoria, even at high densities, does not occupy a substantial area of the panel, but its feeding and excavating activities alter panel surface characteristics and re- duce shipworm recruitment (Fig. 1 1). Discussion Teredo navalis densities and percentage of wood-loss in panels were higher, in the May-Nov exposure period at WP and FI during 3-unit op- eration, than the average values during 2-unit op- eration. This pattern did not occur at at our reference sites. The sampling site at BP was not established until 1985, and most panels were lost during Hurricane Gloria; the data from the re- maining panels do not provide a temporal trend. Values at GN, highest among sites during 2-unit operation, were lower than WP in 1986-1987. Because the increase at WP and FI represented only a single exposure period during 3-unit oper- ation, it carmot yet be determined whether these values are the result of natural variability in a complex system, or the result of power plant op- eration. Continued association of higher shipworm densities and percentage wood-loss at sampling stations with elevated water temperatures would indicated a possible plant impact. Warm water could influence the local woodborer community in several ways. Temper- ature tolerance studies conducted from 1982-1985 (NUSCO 1987) corroborate findings of other re- searchers, e.g., elevated water temperatures in- crease shipworm growth rates (Board 1973), as well as the fecundity and length of their breeding season (Naylor 1965). Warm water could also alter the competitive relationships between fouling species (Nair and Sarawathy 1971; Sutherland and Karison 1977), or the distribution of a haplosporidian fungal parasite, implicated as a source of shipworm mortality (Hillman et al. 1982). Naylor (1965) reported that heated effluents could encourage breeding of non-native species in areas which received warm water species. Exposure Panel Program 243 100 80 601 40 20 0 100' 80- 60' 40 GO 20 ^ n O 0 _I100 i so O 60 ^ 40 Lu 20 O 0 LjJioo CD < 80 \— -Z. 60 CL 0 Oso Q^ 60 LU > 40 < 20 0 100 80 60 40 20 0 Effluent Bottom 3 UNIT OPERATION BEFORE Effluent Surface Black Point Giants Neck Fox Island H White Point Ilf 0 1 6 7 AUG-FEB * A * * * * * * * * * * * * n^ * * * * * * * * * * 7 8 8 8 8 8 9 0 15 6 7 7 8 8 8 8 9 0 15 6 MAY-NOV 7 8 8 8 8 9 0 1 6 7 FEB-AUG MAY-NOV NOV-MAY Fig. 9. iVIean wood-loss from exposure panels collected during 1979-1987 (* = value <2%). 244 5.0 4.5 1 '^ 4 0 in ^ 3 5- o 00 3.0 - 1 2.5- Q_ S 2-0 Ld CD 15 1.0 0.5 Sites: l-ox Island and Black Point Exposure Period: Feb— Aug r 2 = 0.77 Fig. in. The barnacles on (a = o.nooi 9 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 PERCENTAGE OF BARNACLES (LOG SCALE) linear regression of the density of shipworms in exposure panels versus the total percentage of panels at Fox Island and Black Point during the February to August exposure period and r - 0.88). Sites: White Point and Giants Neci< Exposure Period: May— Nov r 2 = 0.23 NUMBER OF LIMNORIA (LOG SCALE) Fig. II. The linear regression of the density of shipworms in exposure panels versus the density oflimnorid isopods at White Point and GianU Neck during the May to November exposure period (a = 0.0001 and r = 0.53). Exposure Panel Program 245 At present, the application of these mechanisms to our sample sites at WP and FI is premature; one year of 3-unit operation is insufficient to de- termine whether thermal incursions to these sites, of short duration, are great enough to influence the Teredo navalis populations. We have not identified a non-native shipworm at WP or FI, even though Teredo bartschi has been found in the MNPS effluent quarry since 1975. In contrast, the woodborer populations at effluent sites were clearly affected by 3-unit oper- ating conditions; panels at EB and ES have shown higher Teredo bartschi densities and greater wood- loss in May-Nov, Aug-Feb, and Feb-Aug since Unit 3 began operation. However, effluent water temperatures have not increased during 3-unit op- eration; average AT's have remained within 2-unit operation ranges. Therefore, the operational effect must be related to a factor other than temperature. A non-thermal effect resulting from start-up of Unit 3 is increased flow and turbulence in the Millstone Quarry. Teredo larvae are relatively dense, and usually settle near the bottom (Graves 1928; Scheltema and Truitt 1956; Turner 1966; Nair and Saraswathy 1971). The increased tur- bulence may be suspending the larvae in the water column long enough to allow them to attack pan- els hung from the floating lab. We have observed changes in the densities of Teredo navalis populations since Unit 3 began operation, and have postulated mechanisms through which these changes might have occurred. However, owing to the limited 3-unit operation database, and the high degree of natural variability typical of biological systems, at present we do not have sufficient information to assess whether the observed increases in shipworm density and wood- loss at WP and FI are related to MNPS operation. Further monitoring will increase our understand- ing of the system, and allow us to make that assessment. Distribution Study Materials and Methods The distribution of shipworms in the MNPS effluent mixing zone was assessed, using their abundance in panels placed 100, 500, and 1000 m from the Millstone Quarry cuts (Fig. 1). At each location, five panels were attached to each of three modified lobster pots, deployed on the bottom in trawl lines (Fig. 12). Panels were first set out in May 1985. In November 1985, three panels from each pot were collected, and replaced with fresh panels (the remaining two panels in each pot served as a source community for overwintering shipworms and larvae). In May 1986, all panels were collected, and replaced with new panels. Pased on the severe wood-loss seen in panels collected in November 1985, sample design was modified to provide a 5-month expo- sure period (May-Oct), including the time of max- imum settlement and growth, and a 7-month ex- posure period (Oct-May), comprising months of low infestation. The trawl-line placed at 100 m in May 1986 proved impossible to maintain; peo- ple fishing in the area of the discharge frequently moved the pots or cut the buoy lines. Therefore, data from 100 m during the May-Oct 1986 expo- sure period are excluded from this report. Begin- ning in October 1986, pots at 100 m were unbuoyed, deployed and collected individually by divers. Six of the nine panels collected from each dis- tance in October (or November of 1985) were each cut into six segments (Fig. 12). A different section of each panel was examined; all shipworms found were removed and identified. Woodborer abundance was determined as a composite total for each distance. Shipworm attack was minimal in panels col- lected in May; to quantify infestation, all panels were examined by means of X-ray photography, using the techniques described in the previous section. If shipworms were present, they were removed and identified. 246 I Rock Weight Exposure Panels Q_ EXPOSURE PANEL Rock Weight / 1 2 (J 4 5 6 / } / Fig. 12. Diagram of an exposure panel trawl-line used to sample the distribution ofshipworms in relation to the effluent discharge point at the Millstone Nuclear Power Station (A. trawl-line of five lobster pots with the locations of the 15 pine panels; B. pine panel showing the sections for subsampling). Results In the four exposure periods sampled to date, 2712 shipworms have been removed and identi- fied, excluding data fi-om the 100 m panels in May-Oct 1986. Virtually all were from the May- Oct/Nov exposure period; four Teredo navalis were found in panels collected in May 1 987 (Table 5). Most of the shipworms have been identified as T. nmalis. Fourteen individuals (0.5%) were T. bartschi, all from 100 m panels in Nov 1985. A trend of decreased Teredo navalis abundance with increased distance from the quarry cuts has been observed in the past two years (Table 5; Fig. 13). Our data from 500 m (654 shipworms) and 1000 m (525 shipworms) support conclusions drawn last year (NUSCO 1987). Temperature data, collected at 100, 500, and 1000 m sites in March 1988 (3-unit operation), are presented in Figure 14. Preliminary data show that water tem- peratures at the 100 m site varied with tidal stage; on a flooding tide, ambient temperatures occurred for 3-4 hours per tidal cycle, and elevated tem- peratures (8 °C AT), occurred for the remaining 8-9 hours per cycle. A temperature gradient was not seen between 500 and 1000 m, at least during early spring. Temperatures at both sites were similar to those at the intakes (i.e., ambient) and showed the same 1 °C tidal fluctuation. Discussion Teredo bartschi have not been found in the effluent mixing zone since Unit 3 began operation. However, our samplirg gear at 100 m was removed Exposure Panel Program 247 TABLE 5. Distribution of shipworms in relation to the effluent discharge point at the Millstone Nu- clear Power Station. TO T A , NUMBER PER PANEL Distance Species May-Nov 1985 Nov- May 1986 May-Oct 1986 Oct- May 1987 100 m Teredo navalis 601 0 a 3 Teredo bartschi 14 0 — 0 500 m Teredo navalis 526 0 654 1 Teredo bartschi 0 0 0 0 1000 m Teredo na\>aUs 388 0 525 0 Teredo bartschi 0 0 0 0 'Data omitted because the pots had been moved from the sample lo'^ation. 100 300 400 500 600 700 800 DISTANCE FROM QUARRY CUTS (METERS) 900 1000 Fig. 13. The density of shipworms, Teredo navalis, in panels placed 100, 500 and 1000 m from the eflluent discharge at the Millstone Nuclear Power Station. 248 6- 4- a C ■ LJ Z) 4- t— < LU Q. 1000 meters 500 meters Ld 100 meters L2- Ld < e- h- ^ AM /% Af) >^ \J M V hhtiC y H ^ A y TIME IN DAYS Fig. 14. Seawaler temperatures on the bottom (5-1 1 m) at 100, 500 and 1000 m during 3-unit operation at the Millstone Nuclear Power Station. Temperature data from 100 m were collected from March 3 to March 12, 1988, while those from 500 and 1000 m were collected from March 16 to March 25, 1988. from its designated site during the May-Oct 1986 exposure period. Based on mformation collected during 2-unit operation, this was the distance/ period most likely to support a population of the non-native shipworm; in November 1985, we identified 14 T. bartschi in panels at 100 m. Therefore, the relative distribution of T. bartschi at 100 m during 3-unit operation will not be known until the October 1987 pemels are pro- cessed. However, it is known that under 3-unit operating conditions, they have not extended their distribution to 500 or 1000 tn, even though T. navalis settlement was high. The continued trend of decreasing T. navalis abundance with increasing distance from the quarry appears to be a response to an environ- mental gradient. Among the possible variables are water depth, water flow, and water temperature (surface and bottom). Such a gradient would not have to persist year-round, as long as it existed during the period of shipworm settlement (June through October). Exposure Panel Program 249 Regardless of other environmental factors that might be influencing shipworm abundance, the thermal incursions at 100 m provided an extended settlement season for Teredo navalis. Enhanced settlement and growth were demonstrated for shipworms grown in a mixture of Effluent and Jordan Cove water (NUSCO, unpublished data). Imai et al. (1950) noted that boring activity of larvae does not begin until water temperature reaches 14 °C. This occurs in early June at am- bient sites, but given an 8 °C AT, it could occur in May at 100 m from the cuts. Therefore, in- creased densities of T. navalis near MNPS dis- charges would be expected, because seawater tem- peratures would enhance settlement, grow^th and survival. Summary 1. The fouling community on exposure panels has vshown no clear response to 3-unit oper- ation. The assemblages continue to be di- verse, and the abundance and distribution of the component species remain patchy. Throughout the study, there has been a neg- ative correlation between fouling cover and shipworm recruitment. 2. There has been an increase in density of shipworms. Teredo na\'alis, and an increase in the amount of wood lost at the WP and EI sites during the first May-Nov exposure period of Unit 3 operation. At a reference site, GN, shipworm density decreased. 3. Higher densities of shipworms and increased wood-loss have occurred in panels in the un- diluted effluent (EB and ES) during 3-unit operation, resulting from increased attack by Teredo bartschi, a non-native shipworm. These increases are attributed to the oppor- tunistic life liistory of T. bartschi, but altered water circulation patterns in the effluent quarry may also expose the panels to more larvae. 4. Panels placed at 100, 500, and 1000 m from the quarry cuts continued to show increased recruitment of T. navalis at panels closer to the discharge. Teredo bartschi, which had been found in 100 m panels during 2-unit operation, was not sampled during initial 3-unit operation, because the May-Oct 100 m panels were dragged off-station by fisher- men. Conclusions Since Unit 3 began operation, increased shipworm abundance and increased wood-loss were observed at sites in the MNPS effluent, and during one exposure period, at White Point and Fox Island, which were potentially exposed to the 3-unit thermal plume. Further monitoring will be required to determine whether these changes are related to 3-unit operation, or are expressions of natural variability. References Cited Board, P.A. 1973. The effects of temperature and other factors on the tunnelling of Lyrodus pedicellatus and Teredo navalis. Pages 797-805 iji Proc. 3rd Int. Congr. on Marine Corrosion and Fouling, Natl. Bur. Std., Gaitherburg, Maryland, U.S.A. Dale, H.M., and T. Gillespie. 1977. Diurnal fluc- tuations of temperature near the bottom of shallow water bodies as affected by solar radi- ation, bottom color and water circulation. Hydrobiologia 55:87-92. Dean, D., and C.B. Officer. 1977. Development document in support of alternative effluent limitations pursuant to section 316(a) of the Federal Water Pollution Control Act for Maine Yankee Nuclear Generating Station. Maine Yankee Atomic Power Company, Wiscasset. de Wilde, P.A.W.J., and E.M. Berghuis. 1979. Cyclic temperature fluctuations in a tidal mud- flat. Pages 435-441 in E. Naylor and R.G. HartnoU ed. Cyclic phenomena in marine plants 250 and animals. Pergamon Press Oxford, New York. Grave, B.H. 1928. Natural history of shipworm, Teredo navalis, at Woods Hole Massachusetts. Biol. Bull. Woods Hole 55:260-282. Hillman, R.E., C.I. Belmore, and R.A. McGrath. 1985. Annual report for the period December 1, 1983 to November 30, 1984 on study of woodborer populations in relation to the Oyster Creek Generating Station. As submitted to GPU Nuclear Corporation. Battelle, New Eng- land Marine Research Laboratory Report. 148 pp. Hillman, R.E., N.J. Maciolek, J.I. Lahey, and C.I. Belmore. 1982. Effects of a haplosporidian parasite, Haplosporidium sp., on species of the molluscan woodborer Teredo in Bamegat Bay, New Jersey. J. Inverteb. Pathol. 40:307-319. Hoagland, K.E. 1981. Life history characteristics and physiological tolerances of Teredo bartschi, a shipworm introduced into two temperate zone nuclear power plant effluents, [n 3rd Int. Waste Heat Meetings Proc. 14 pp. , and R.D. Turner. 1980. Range extensions of teredinids (Shipworms) and polychaeta in the vicinity of a temperate-zone nuclear gener- ating station. Mar. Biol. 58:55-64. Imai, T., M. Hatanaka and R. Sato. 1950. Breed- ing of marine timber-borer, Teredo navalis L., in tanks and its use for anti-boring tests. Tohoku J. Agricult. Res. 1:199-208. Maciolek-Blake, N.J., R.E. Hillman, P.I. Feder and C.I. Belmore. 1981. Annual report for the period December 1, 1979 to November 30, 1980 on study of woodborer populations in relation to the Oyster Creek Generating Station to Jersey Central Power and Light Company. Battelle-William F. Clapp laboratories. Report No. 15040. 15 pp. Nair, N.B. and M. Saraswathy. 1971. The biology of wood-boring teredinid molluscs. Adv. Mar. Biol. 9:335-509. Naylor, E. 1965. Effects of heated effluents upon marine and estuarine organisms. Adv. Mar. Biol. 3:63-103. NUSCO (Northeast Utilities Service Company). 1982. Monitoring the marine environment of Long Island Sound at Millstone Nuclear Power Station, Waterford, Connecticut. Annual re- port, 1981. . 1987. Monitoring the marine envirorunent of Lx)ng Island Sound at Millstone Nuclear Power Station, Waterford, Connecticut. An- nual report, 1986. Schehema, R.S. and R.V. Truitt. 1956. The shipworm Teredo navalis in Maryland coastal waters. Ecology 37:841-843. Sutherland, J.P., and R.H. Karlson. 1977. Devel- opment of stability of the fouling community at Beaufort, North Carolina. Ecol. Monogr. 47:425-446. Turner, R.D. 1966. A survey and illustrated cat- alogue of the Teredinidae. Cambridge Mus. Comp. Zool. Harvard Univ. 265 pp. . 1973. In the path of a warm saline effluent. Am. Malacol. Union Bull. 39:36-41. . 1984. An overview of research on marine borers: past progress and future direction. Pages 3-16 in J.D. Costlow and R.C. Tipper, ed. Marine biodeterioration: an interdisci- plinary study. Proceedings of the symposium on marine biodeterioration. Uniformed Services University of Health Sciences, 20-23 April 1981. Library of Congress Cat. no. 81-85468. Naval Institute Press, Annapolis, Maryland. Exposure Panel Program 251 Contents Fish Ecology Studies 255 Introduction 255 Material and Methods 256 Trawl program 256 Seine program 256 Impingement program 256 Ichthyoplankton program 257 Data analyses 258 Results and Discussion 260 Trawl monitoring 260 Seine monitoring 261 Impingement monitoring 261 Ichthyoplankton 266 American sand lance 267 Anchovies 268 Sticklebacks 272 Atlantic tomcod 273 Silversides 275 Grubby 277 Taulog 278 Cunner 283 Conclusions 288 Summary 289 References Cited 290 Appendices to Fish Fx-ology Studies 295 Fish Ecology Studies Introduction Fish are an important marine resource in Con- necticut and millions of dollars in annual revenues are generated by the fishing industry in Long Is- land Sound (LIS) (Sampson 1981; Blake and Smith 1984). Commercially and recreationally important fishes are abundant in the area around Millstone Nuclear Power Station (MNPS) along with important forage species that contribute to ecosystem productivity. Some vSpecies inhabit the area seasonally for feeding, spawning, or nursery activities while others are year-round residents. The operation of MNPS could affect fish as- semblages in the area by increasing mortality rates and altering spatial distribution. Adult and juve- nile fish may be removed from populations by impingement on the intake screens. Fish eggs and larvae may be removed through entrainment with the condenser cooling water. The effects of uicreased mortality rates on the abundance of these populations can differ depending upon the size, life span, and age structure of the affected population and on the existence of compensatory mechanisms. Spatial distributions of local fish populations may change in response to alterations in the thermal or chemical regime of the effluent or modifications to the physical habitat. Water temperature increases can attract or exclude fish from areas affected by the thermal plume of MNPS. Physical alterations caused by bottom scouring or dredging could also affect the spatial distribution of fish. To determine if the operation of MNPS has impacted the local fish assemblages, monitoring studies have been established. The objectives of these monitoring programs are: 1. Describe the occurrence and abundance of fish in the Millstone area. Identify spatial and temporal patterns of fish assemblages and establish the extent and di- rection of natural changes in these assem- blages. Evaluate whether observed changes are the result of MNPS operation, and if so, the significance of these changes, with particular emphasis on the period since Unit 3 began operating. To meet these objectives, four sampling pro- grams have been established to collect data on the available life history stages of those fishes susceptible to impact. These programs are the demersal trawl; shore-zone seine; ichthyoplank- ton, including entrainment sampling; and impingement. In this report, the life history and population characteristics of potentially impacted species are presented and evaluated to determine if MNPS has had any detrimental effects on them. Although Unit 3 did not start producing com- mercial power until April 23, 1986, variable num- bers of condenser cooling water pumps were op- erated starting in November 1985. For species potentially impacted by entrainment or impingement, all of 1986 will be considered as a three-unit operational period. The Fish Ecology section of the two-unit sum- mary report (NUSCO 1987b) emphasized time- series analyses to describe the natural fluctuations of potentially impacted species. These time-series models represent a baseline that can be used with intervention analysis (Madenjian et al. 1986) to assess three-unit operations. But at this time, with less than 2 years of data since the start-up of Unit 3, there are insufficient data to apply intervention analyses. Therefore, in this report more conventional indices of abundance will be used for assessment. Fish Ecology Studies 255 Material and Methods Trawl program Data used for this report are from the period of January 1976 through May 1987. A reporting year included data collected from June of one year through May of the following year; thus, the report y^ar 1986-87 included data from June 1986 througli May 1987. A complete history offish ecology programs was presented in a two-unit summary report (NUSCO 1987b) using a calender year (January through December). Prior to the summary report, the reporting period was October through September. This report period was not based on biological considerations, but on the timing of report requirements to regulatory agen- cies. Many of our analyses used the seasonal period of occurrence of a species (i.e., the period when 95% of the cumulative abundance was ob- served) and in some cases the seasonal abundance transcended the arbitrary reporting periods of Oc- tober througli September or the calender year. Considering the seasonal occurrence of our abun- dant species, June was the best trauvsitional period. By the end of May the early life history stages of the winter-spawning species were no longer sus- ceptible to entrainment and summer spawners were not yet abundant. Because of occasional overlap in the occurrence of a species during tliis May-June transitional period, species-specific analyses are based on the period of occurrence of each species and not absolutely constrained to June 1 as the starting point. If a life history stage occurred during the June- December period, only the 1986 season of three-unit operation is given in this report, whereas, if it occurred in the January-May period, the 1986 and 1987 seasons of three-unit operation are reported. The materials and methods presented are for 1985-86 and 1986-87 reporting periods, except for impingement monitoring, which was discontinued at Unit 2 on December 11, 1987. Impingement data are sum- marized through 1987 using a calender reporting year. Demersal fishes were collected using a 9.1-m otter trawl with a 0.6-cm codend liner. Triplicate tows were made biweekly at six stations: Niantic River (NR), Jordan Cove (JC), Twotree (TT), Bartlett Reef (BR), Intake (IN) and Niantic Bay (NB) (Fig. 1). A standard tow covered 0.69 km and this distance was measured using radar. The total length of up to 50 randomly selected indi- viduals of each species per station was measured to the nearest millimeter. Catch was expressed as the number per tow. Data are reported from June 1976 through May 1987. Seine program Shore-zone tishes were sampled using a 9. 1 x 1.2-m knotless nylon seine net of 0.6-cm mesh. Triplicate 30-m tows were made parallel to the shoreline at White Point (WP), Jordan Cove (JC) and Ciiants Neck (GN), monthly from November througli March and biweekly April through Oc- tober (Fig. 1). Collections were made during the period of 2 hours before to 1 hour after high tide and all three stations were sampled the same day. Fish in each haul were identified to the lowest possible taxon, counted, and the total length of up to 50 randomly selected individuals of each species in each replicate was measured to the nearest millimeter. Catch was expressed as the number per haul. Data are reported from June 1976 through May 1987. Impingement program Fish impinged on the intake screens at Unit 2 were washed at least once ever 8 hours into a 1.5 X 0.8 x 1.75-m perforated collection basket. Impingement sampling consisted of sorting spec- imens from all material washed from the screens during a 24-hour period. All were identified to the lowest possible taxon, counted, and the total length of up to 50 randomly selected specimens of each species was measured to the nearest mil- limeter. Catch was calculated as number impinged per 24-hour period. Sampling effort was stratified 256 Fig. 1. Location of trawl, seine, and ichlhyoplankton sampling stations. by month, with 8 samples collected in January, 15 in February, 14 in March, 5 in April, 4 per month during May through November, and 10 in December. Data are reported from January 1976 through December 11, 1987. Ichthyoplankton program Weekly samples of entrained ichthyoplankton (fish eggs and larvae) were collected during 3 day and 3 nights from June through September, on 1 day and 1 night from October through February, and during 4 days and 4 nights from March through May. Sampling alternated weekly be- tween the discharges of Units 1 and 2 (station EN) when plant operations pennitted. A 1.0 x 3.6-m conical plankton net with 333-nm mesh was deployed with a gantry system. Four General Oceanic flowmeters (Model 2030) were positioned in the mouth of the net to account for horizontal and vertical flow variations. Sample volume (about 400 m ) was determined by averaging the four volume estimates from the flowmeters. larvae were collected at station NB located in mid-Niantic Bay (Fig. 1). Weekly 2 day and 2 night samples were taken from June through Au- gust and April through May, and biweekly 1 day and 1 night sample taken from September through March. Paired 0.61 x 3.3-m conical plankton nets, mounted on a bongo frame, were used to take stepwise oblique tows. Sampling duration was 5 minutes each at surface, mid, and bottom depths. Sample volumes were measured using one General Oceanics flowmeter in each net and approximately 300 m of seawater were filtered Fish Ecology Studies 257 for each sample. Net mesh size was 333 [im, except for a period from mid- February through March, when 202-|im mesh nets were used to reduce the extrusion of yolk-sac winter flounder larvae (see Winter Flounder Studies section). Plankton samples were split using a NOAA- Boume splitter (Botelho and Donnelly 1978) and sorted for ichthyoplankton using dissecting mi- croscopes. Successive splits were completely sorted until at least 50 larvae and 50 eggs (for samples processed for eggs) were found, or until one-half of the sample was examined. Samples sorted for larvae included all those from FN col- lected during .January through May and .July through December, one day and one night sample collected per week during .June, and all NB sam- ples. Three day and three night FN samples collected in April through September were sorted for fish eggs. Fish eggs and larvae were identified to the lowest practical taxon. Gunner and tautog eggs were differentiated from a weekly composite sample of their eggs using the criterion of bimodality of egg diameters (Williams 1967). Ichthyoplankton density was expressed as number per 500 m" . Included in this report are ichthyoplankton data through May 1987, starting with egg collections at EN in May 1979; larval collections at EN in January 1976 and at NB in .January 1979. Data analyses To assess impacts it was necessary to identify potentially affected species, document their spatial distribution, and describe the natural temporal fluctuations of their life history stages collected near Millstone. The selection of potentially af- fected species was based on their prevalence in entrainment or impingement samples or their abundance in the shore-zone area of .Jordan Cove, an area which may be impacted by the thermal plume. Indices to describe temporal and spatial abundance for all life history stages of fishes must be selected based on the knowledge of the under- laying assumptions of each index. Failure of the data to conform to these assumptions may reduce the precision of the estimates or, worse, provide invalid results. Since fisheries data typically have numerous zero values and follows a lognormal distribution, the 5-mean (Pennington 1983, 1986) was used as the index of abundance of various life history stages of selected species. A detailed description and evaluation of this statistic is pro- vided in a separate section (see Delta Distribution section). The S-mean was used as an index of abundance for juveniles and adults collected in the trawl and seine programs and for larvae that were not consistently collected during their sea- sonal occurrence. For species that were collected seasonally, the data used to calculate the 5-mean were restricted to the period of occurrence to re- duce the number of zero values. An alternative index of abundance, used for ichthyoplankton that were collected consistently during their sea- sonal occurrence, was the a parameter from the Cjompcrtz function (Draper and Smith 1981). Typically, the distribution of ichthyoplankton abundance over time is skewed, with a rapid in- crease to a maximum followed by a slower decline. This skewed density distribution results in a sigmoid-shaped cumulative distribution and the time of peak abundance is the time at which the inflection point occurs in the cumulative distribu- tion. The Gompertz function was chosen to de- scribe the cumulative distribution data because the inflection point of this function is not con- strained to the central point of the sigmoid curve. 'I'he form of the Gompertz function was: C,= a(exp[-pe"'''l) where Q = cumulative density at time t a = total or asymptotic cumulative density P = location parameter K = shape parameter t = time in days The origin of the time scale was arbitrary and for our data was set to the time of the year that the respective developmental stage generally starts to appear in ichthyoplankton samples. Least- 258 squares estimates of these parameters and their asymptotic 95 "/o confidence intervals were ob- tained by fitting the Gompertz function to the cumulative abundance data (based on the weekly geometric means) using nonlinear regression meth- ods (SAS Institute Inc. 1985). The a parameter was used as an index to compare annual abun- dances and the time of peak abundance was es- timated as the date t; corresponding to the inflec- tion point of the function defined by its parameters P and K as: (log,P) '/ = The presence of compensatory mortality during the early life history stages would help mitigate the loss of entrained eggs and larvae. When abun- dance estimates of both eggs and larvae were available, density-dependent mortality was inves- tigated with the following relationship (Rickcr 1975): lo8.(|) a-H pE where L = larval abundance estimate E = egg abundance estimate a = intercept P = index of density-dependent mortality If the slope (P) is positive the density-dependent mortality is depensatory and if negative it is com- pensatory. Armual entrainment estimates were calculated for dominant ichthyoplankton species entrained. These estimates were obtained by multiplying the median density at EN during the period when 95% of the annual cumulative abundance oc- curred times the total volume of water passed through MNPS during the same period. A nonparametric method (Snedecor and Cochran 1967) was used to construct 95% confidence in- tervals around each median density and corre- sponding entrainment estimate. Monthly impingement estimates were based on the extrapolation of actual counts using a volu- metric ratio. The daily cooling water volume was calculated based on 15-minute flow rates from 0800 to 0745, the time corresponding to the actual impingement period. Within each month, an es- timate for every day not sampled was calculated by multiplying the average impingement density (number of fish per m' of cooling water) based on the days sampled in that month times the volume of cooling water on each day not sampled. All of these daily estimates were then added to the sum of the actual sample counts to arrive at the monthly totals for each species. Annual impingement estimates were calculated by sum- ming the monthly estimates. As stated previously, seine sampling effort was stratified by season and impingement sampling was stratified by month. Therefore, whenever appropriate, the length-frequency data were weighted to account for unequal effort during the year. Because seine sampling effort from April through October was twice that during the re- mainder of the year, data collected from November through March were weighted by a factor of two. I'or impingement collections, monthly weight fac- tors of 4 (January), 2 (February and March), 6 (April), 7 (May through November), and 3 (De- cember) were used to standardize the effort. Data on the annual abundance of fishes in LIS and adjacent areas were examined to determine if observed changes in the Millstone area were lo- calized or evident over a larger area. Trawl and ichthyoplankton data were available from moni- toring studies at the Shoreham Nuclear Power Station (SNPS) and summarized for 1977-82 ((Jcomct Tech. 1983) and 1983-1986 (EA Eng., Sci., and Tech. 1987). SNPS is located on the southern shore of LIS almost directly south of New Haven, CT. The available trawl data were converted to annual catch-per-unit-effort for day collections. The ichthyoplankton data were sum- marized in the reports for 1977-82 as the annual sum of the mean densities ( 1 000 m ) per sampling trip and for 1983-86 as the annual sum of the monthly mean densities. Because the 1977-82 Fish Ecology Studies 259 period contained some months when two sam- pling trips were made, a direct comparison of 1977-82 to 1983-86 data cannot be made, but the information should be sufficient to determine long-term trends. For the potentially impacted species in the Millstone area, sufficient data were available from the SNPS data base for compari- sons with egg abundance of anchovies, cunner, and tautog; larval abundance of sand lance, an- chovies, cunner, and tautog; and trawl catches of anchovies, cunner, and tautog. Additional trawl data were available from the National Marine Fisheries Service (NMFS) ground trawl survey (Grosslein 1974; Azarovitz 1981). Data were ob- tained from selected strata off southern Ix)ng Is- land, NY; Rhode Island; and southwestern Mas- sachusetts. These data were provided by NMFS as the annual 5-mean for both spring and fall surveys. Because of limited catches of most po- tentially affected species, only sand lance data were sufficient for comparisons. Results and Discussion Over 100 fish taxa from ichthyoplankton, impingement, trawl, and seine samples have been collected in the Millstone area from 1976 through May 1987 (Appendix I). The most common were American sand lance {Ammodytes americanus), winter flounder {Pseudopleuronecles americanus), anchovies (Anchoa milchilli and A. hepsetus), sticklebacks {Gasteroslem aculeatus and G. wheatlandi), silversides [Menidia menidia and M. betylUnd), Atlantic tomcod {Microgadus tomcod), grubby {Myoxocephalus aenaeus), skates (Raja erinacea, R. ocellata, and R. eg/anteria), scup {Stenolomus chrytsnps), windowpane (Sc.ophthalmus aquosus), tautog {Tauloga onitis), and cunner {Tautogolahrus adspersus). These taxa were typical of fish assemblages found in LIS (Greeley 1938; Warfel and Merriman 1944; Wheatland 1956; Richards 1959; Pearcy and Rich- ards 1962; McMugh 1972; Saila and Pratt 1973; Geomet Tech. 1983). Important recreational and commercial fishes, such as, bluefish {Pomalomus sallatrix) and striped bass {Morone saxatilh), that occurred in the Millstone area were not susceptible tp our sampling gear. However, they were rare in the entrainment and impingement collections, so the potential impact of MNPS on these species is minimal. The following is a summary of the fishes collected in each sampling program to show which taxa and life history stages predominated in the Millstone area. Trawl monitoring In the trawl program, over 90 taxa of juvenile and adult fishes were taken at six stations in the Millstone area during the past 1 1 years (Appen- dices II and III). The demersal fishes collected in the trawl monitoring program were similar to those found in Narragansctt Bay (Oviatt and Nixon 1973). Since 1976, six fish taxa comprised over 80% of the trawl catch and winter flounder accounted for over 40% of the total. The winter flounder was caught throughout the year in the Millstone area and due to its commercial and recreational importance is discussed in detail in a separate section (see Winter F'lounder Studies sec- tion). The second most abundant species (15%) was the scup, which was found from .June through October. Most scup were juveniles and over 40% of them were caught at NB. Anchovies, which accounted for over 8% of the trawl catch, were also found primarily at NB from August through October and were primarily young-of-the-year. Our demersal trawl does not uniformly sample anchovies, most likely because of their location in the water column, small size, and patchy dis- tribution. Annual catches of anchovies were vari- able and the highest catch occurred in 1985. Windowpane and skates, both resident taxa, to- gether accounted for an additional 13% of the catch. Both were most often found at the deeper water stations (TT and BR). Silversides were the sixth most abundant taxon caught by trawl. They were the dominant shore-zone taxon in the Mill- stone area and were caught in trawls during the the winter, primarily from October through Feb- ruary. Annual 5-mean catches (all stations combined) were calculated to examine year to year variation for the six dominant taxa in the trawl monitoring program (Table I). For taxa that occur seasonally. 260 TABLE 1. The seasonal 5-mean (CPUE) of the abundant fish taxa caught in trawls during each report period (June- May). Taxon 16-n 77-78 78-79 79-80 80-81 81-82 82-83 83-84 84-85 85-86 86-87 Psevdoplmronecle.'! amerkanus 16.6 13.5 16.7 26.8 32.6 24.1 41.8 27.7 29.5 22.0 19.8 Stenotomus chrysops 10.6 19.8 13.3 18.5 17.0 20.4 27.5 26.6 22.3 13.6 30.6 Anchoa spp. 11.1 3.3 39.3 0.1 0.1 4.0 0.2 0.4 0.7 113.8 57.38 Scophthalmus aquosus 2.9 2.4 1.8 2.9 3.5 2.9 6.7 5.0 4.4 4.7 3.8 Raja spp. 1.4 1.2 0.8 0.8 2.0 1.4 6.1 5.3 3.1 8.5 4.5 Menidia spp. 16.2 9.7 2.8 6.2 6.5 1.8 1.5 2.1 0.5 1.9 17.8 Data seasonally restricted to June-October for Stenotomus chrysops, to August-October for Anchoa spp. , to October-February for Menidia spp. , and remaining taxa year-round (June-May) data for this analysis were limited to their season of occurrence: June through October for scup, August through October for anchovies and Oc- tober through February for silversides. The 5-mean catch for all of the dominant species has fluctuated and during three-unit operation the catches were within or above historical ranges with some of the highest abundances for scup, anchovies, and silversides. Seine monitoring Approximately 40 different taxa have been caught by seine during the past 1 1 years of mon- itoring (Appendices IV and V). Silversides dom- inated the shore-zone catches and accounted for over 80% of the total. About 80% of the total seine catch was collected at JC. This station is a productive nursery area, and hundreds of juvenile silversides are routinely caught at this site during the summer months (June through September). Because silversides dominated all the annual catches, total catches were largely a function of silverside catches. Total catches for all shore-zone taxa were higliest during the 1976-77 and 1977-78 report periods and were dominated by juvenile silversides at JC (NUSCO 1987b). Because the silversides dominated the shore-zone area of Jor- dan Cove, which may be thermally impacted by the condenser cooling water discharge, 't was se- lected as a potentially impacted taxon and is dis- cussed in further detail later. Impingement monitoring Impingement has been monitored at Millstone Unit 2 since it began operating in September 1975. The objective of the monitoring program was to quantify total annual species-speciftc loss due to impingement. Because impingement losses have been well-documented and measures to mit- igate impingement losses have been investigated, a request was vSubmitted to the Connecticut DEP in July 1987 to discontinue impingement moni- toring at Unit 2 (NUSCO 1987a). The DEP accepted our request and sampling was discontin- ued on December 11, 1987. Annual impingement estimates were calculated from January 1, 1976 through December 11, 1987. Over 100 fish and invertebrate taxa were impinged during the past 12 years (Tables 2 and 3). Sand lance accounted for over 60% of the total because an estimated 480,000 were impinged during the week of July 18, 1984. Impingement of sand lance during that week accounted for over 98% of the 12-year total for that species. The sand lance is a schooling species (Leim and Scott 1966) and a large school encountered the intake struc- ture. Excluding sand lance, six fish taxa dominated the collections: winter flounder, anchovies, grubby, silversides, sticklebacks, and Atlantic tomcod (Table 4). These taxa were selected as potentially impacted species and discussed in fur- ther detail later, except for winter flounder (see Winter Flounder Studies section). Fish Ecology Studies 261 TABLE 2. Annual impingement estimates for Tish taxa impinged at MNPS Unit 2 from January 1, 1976 through December 11, 1987. Taxon 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 Total Ammodyies spp. 46 20 239 59 168 223 81 161485411 73 42 8 486531 Pseudopleiironectes americanus 2783 4604 3184 10077 3576 6207 2735 6213 2542 2769 1049 624 46363 Myoxocephalus aenaeus 1027 1299 3980 1721 9167 3262 2671 7023 2359 4621 1427 647 39204 Anchoa spp. 848 177 774 2508 4073 3722 4085 12726 4200 342 38 12 33505 Metxidia spp. 853 503 2292 3319 8676 2769 800 3759 1042 1484 511 136 26144 Microgadus tomcod 45 86 1956 768 1213 2809 10302 2264 4938 1130 8 206 25725 Casterostetts wheatlandi 580 1090 11691 702 21 1799 1822 17705 Gasterosteus spp. 1533 3609 3480 2241 6710 17573 Gasterosleus aculeatus 5883 1188 4638 1055 859 921 572 15116 Syngnathus fuscus 559 260 875 425 766 1417 557 3503 1467 460 858 748 11895 Feprilus triacanlhus 114 122 233 1091 781 1416 2414 465 1455 1337 1406 946 11780 Tautogolahrus adspersus 357 598 1399 1656 751 883 1787 694 1188 466 57 642 10478 Merluccius bilinearis 586 304 282 361 1659 277 422 998 133 106 44 41 5213 Scophthalmus aquoyus 434 188 243 503 570 367 354 1241 569 174 292 224 5159 Morone americana 136 260 458 230 489 491 1340 476 375 48 19 2 4324 Tautoga onitls 464 98 731 578 106 434 397 168 664 122 96 241 4099 Cyclop terus lumpus 26 265 209 248 689 329 9 499 120 1010 128 204 3736 Raja spp. 292 165 133 170 231 464 274 626 275 99 285 268 3282 Osmenis mordax 282 204 390 62 101 283 184 897 71 99 105 60 2738 Paralkhthys dentatus 230 17 59 7 27 261 88 72 646 29 377 469 2282 Brevoortia tyrannus 177 50 62 135 154 101 247 3M 167 242 55 56 1757 Prionotua spp. 364 112 63 49 61 242 147 86 49 72 142 71 1458 Cynoscion regatis 26 568 70 4 7 466 40 38 34 14 8 54 1329 A losa pseudoharengus 48 274 26 92 lis 128 203 192 79 59 32 21 1272 Alosa aestivalis 86 125 88 140 94 121 63 234 91 51 42 2 1137 Sphoeroides maculatus 165 4 17 49 12 80 126 166 174 86 81 44 1004 Liparis spp. 6 208 86 11 25 371 19 155 39 66 0 5 991 Anguilla rostrata 60 25 56 84 73 66 207 104 60 48 10 59 852 Stenotomus chrysops 114 76 19 87 35 78 115 95 53 105 23 6 806 Opsanus tau 69 27 77 96 49 123 55 67 98 28 23 75 787 Pholis gunnellus 48 28 39 86 28 88 42 121 49 12 24 35 600 Pomatomus saltatrix 23 44 9 47 27 81 40 108 110 46 34 6 575 Pollachius virens 5 6 2 2 71 55 41 41 253 0 37 0 513 Trinectes maculatus 20 6 14 53 35 21 75 29 194 21 21 12 501 Fundulus spp. 12 16 91 14 75 99 33 13 20 8 0 0 381 Clupea harengus 33 114 0 5 2 35 9 16 12 12 5 98 341 Urophycis chuss 3 17 0 80 19 42 26 41 71 13 0 6 318 licmitripterus americanus 9 2 5 2 25 64 51 94 22 6 12 0 292 Urophycis regia 5 0 0 3 14 12 19 187 7 0 13 0 260 Caranx hippos 5 0 2 4 2 47 91 9 21 34 8 0 223 Gadus morhua 0 0 0 0 0 10 24 2 142 7 16 0 201 Ophidion marginatum 16 0 0 10 0 0 7 4 50 6 17 49 159 Urophycis spp. 0 29 124 0 2 2 0 0 0 0 0 0 157 Centropristis striata 22 2 0 0 7 0 2 6 74 0 28 0 141 Mugil cephalus 3 4 10 4 18 5 7 17 39 4 19 0 130 Sphyraena borealis 12 4 10 0 0 25 63 0 12 0 0 0 126 Leiostomus xanthurus 12 2 83 0 0 0 2 16 0 0 8 0 123 Monacanthus hispidus 5 0 34 45 8 4 4 7 3 9 0 0 119 Etropus microstomus 2 0 0 0 0 5 4 41 20 10 22 14 118 Apeltes quadraais 2 4 2 2 31 45 12 3 2 0 6 2 111 Alosa sapidissima 12 1 0 2 33 16 6 17 16 0 0 0 103 Melanogrammus aeglefmus 0 0 4 88 3 0 0 0 0 0 0 0 95 Selene selapinnis 30 0 0 0 0 0 2 2 20 34 7 0 95 Urophycis tenuis 0 1 4 0 4 17 13 2 45 0 0 0 86 Pungilius pungitius 0 0 2 0 0 4 19 17 5 10 19 2 78 Paralichthys oblongus 16 0 4 7 4 16 2 4 0 6 0 15 74 Selene vomer 0 22 0 0 5 0 2 0 14 22 0 7 72 Scomber scombrus 0 4 0 4 0 12 46 2 0 0 0 0 68 Alulerus schoepft 1 36 4 0 3 0 9 0 6 0 0 0 59 Caranx crysos 0 9 7 0 0 5 14 0 24 0 0 0 59 262 TABLE 2. Conlinued. 'I'axon 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 Total Alectis ciliaris 1 7 2 0 0 28 2 0 6 7 0 0 53 Mugil curema 0 0 0 0 0 10 0 3 27 8 4 0 52 Mustelis canis 2 6 18 2 0 6 4 0 0 8 6 0 52 Fistularia tabacaria 3 4 5 0 10 2 5 0 0 6 0 0 35 Lophius americanus 0 0 0 0 0 0 2 0 0 7 0 24 33 Squalus acanthias 2 0 0 0 6 0 0 21 0 0 2 2 33 Myoxocephalus spp. 1 31 0 0 0 0 0 0 0 0 0 0 32 Morone saxalilis 0 0 3 2 0 0 13 5 0 7 0 0 30 Hippocampus erectus 2 0 0 0 0 0 0 4 0 3 13 7 29 Ulvaria mbbifurcala 2 1 3 2 0 7 2 0 6 4 0 0 27 Conger oceanicus 13 4 0 0 0 2 5 2 0 0 0 0 26 Myoxocephalus 0 0 0 5 5 2 4 8 0 0 0 0 24 octodecemspinosus Dactylopterus votitans 0 2 0 0 0 0 4 0 0 7 0 10 23 Myoxocephalus scorpius 0 16 6 0 0 0 0 0 0 0 0 0 22 Ophidiidae 17 0 0 0 0 0 5 0 0 0 0 0 22 Trachurus lathami 0 0 0 16 0 4 0 0 0 0 0 0 20 Chaetodon ocellatus 0 0 2 0 0 2 7 0 0 7 0 0 18 Gadidae 0 0 5 0 0 0 13 0 0 0 0 0 18 Decaplerus macarellus 0 0 0 0 0 0 0 0 15 0 0 0 15 Enchelyopus cimbrlus 0 0 0 0 0 0 0 0 8 0 0 6 14 Pristigenys alta 0 0 0 0 0 0 2 0 12 0 0 0 14 Cypnnodon variegalus 0 0 4 0 4 2 2 0 0 0 0 0 12 Chilomycterus schoe.pji 0 0 0 0 0 0 2 0 8 0 0 0 10 Macrozoarces americanus 0 0 0 2 0 0 0 2 0 6 0 0 10 Etrumeus teres 0 2 0 5 0 0 2 0 0 0 0 0 9 Priacanthus arenatus 0 0 0 0 0 0 2 0 6 0 0 0 8 Clupeidae 0 0 0 0 0 0 0 0 3 4 0 0 7 Menticirrhus saxatilis 0 0 2 3 0 0 2 0 0 0 0 0 7 Alosa spp. 0 0 0 0 0 0 0 0 6 0 0 0 6 Hippocampus spp. 0 0 0 0 0 0 0 0 0 6 0 0 6 Priacanthus cruentatus 0 0 0 0 0 2 0 4 0 0 0 0 6 Selar crumenopthalmus 3 0 0 0 0 2 0 0 0 0 0 0 5 Seriola zonata 0 5 0 0 0 0 0 0 0 0 0 0 5 Aulostomus maculatus 3 0 0 0 0 0 0 0 0 0 0 0 3 Monocanthus spp. 0 0 0 0 0 0 0 0 3 0 0 0 3 Alosa mediocris 0 0 0 0 0 0 2 0 0 0 0 0 2 Decapterus punctatus 0 0 0 0 0 2 0 0 0 0 0 0 2 Ictalurus catus 2 0 0 0 0 0 0 0 0 0 0 0 2 Rhinoptera honasus 0 0 0 0 0 2 0 0 0 0 0 0 2 Salmo trutta 0 0 0 2 0 0 0 0 0 0 0 0 2 Total 12077 14677 21981 27268 40822 34636 32745 60410 : 511387 16360 10199 8560 791122 Many researchers have found that impingement rates were directly influenced by cooling-water flow (ConEd and PASNY 1977; Lawler, Matusky and Skelly Engineers 1980, 1987). The estimated total number offish impinged each year by MNPS was compared to annual cooling-water volume (Fig. 2). Annual impingement at Unit 2 from 1976 to 1982 appeared to be related to cooling- water flow. In 1983, cooling water flow was low while the number offish impinged was high; Unit 2 was at full power during the winter of 1983, but was shutdown from June through November. Except for anchovies, the more abundant fishes were usually impinged in greater numbers during the winter; this accounted for the differences noted between catch and flow in 1983. In the fall of 1983, there was a noticeable reduction in daily impingement at Unit 2 after the summer removal of a cofferdam surrounding the Unit 3 intake. The cofferdam existed when Unit 2 began oper- ating and it provided a reef-like habitat in the vicinity of the Unit 2 intake that may have at- tracted fish. Analyses of daily monitoring data revealed a significant (p<0.01) reduction in the average number of organisms impinged at Unit 2 after the removal of the cofferdam (NUSCO Fish Ecology Studies 263 TABl-E 3. Annual impingement estimates for invertebrate taxa impinged at MNPS Unit 2 from January 1 , 1976 through December 11, 1987. Taxon 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 Total Ovalipes ocellatus 1402 2289 1831 4708 14224 13054 23194 4876 4118 2838 1552 1343 75429 Loligo pealei 5095 1257 2430 10763 11083 7895 6846 2779 14748 3298 1912 1491 69597 Cancer irroratijs 457 389 442 438 560 3919 5179 4036 6680 4456 6264 2365 35185 Carcinus maenus 425 408 368 408 670 2339 1532 4311 5647 2727 2960 1571 23366 CalUnectes sapidus 564 193 499 945 927 964 1120 468 1020 733 621 988 9042 Homarus americanus 663 308 261 402 392 1043 1032 458 1167 505 549 825 7605 Libinia spp. 1244 205 93 91 124 516 475 887 1484 407 689 283 6498 Neopanope texana 0 13 16 13 18 104 362 288 1496 244 373 186 3113 Limulus polyphcmus 281 11 17 54 46 152 86 10 164 0 41 275 1137 Pagurus spp. 227 37 8 4 88 38 14 11 12 7 83 54 583 Squilta empusa 52 38 4 2 193 23 20 25 3 3 0 65 428 Cancer borenlls 12 4 5 77 9 10 11 12 2 0 0 0 142 Argnpecten irradians 0 0 0 0 0 2 2 37 38 0 0 0 79 Penaeus aztecus 0 0 0 0 2 2 45 0 0 7 7 0 63 Upogebia affinis 0 0 0 0 0 0 14 39 0 0 0 0 53 lllex illecebrosus 0 5 6 0 0 9 0 0 6 0 0 0 26 CalUnectes similis 0 0 0 0 0 0 14 0 0 0 0 0 14 Lunalia heros 0 0 0 0 2 0 2 0 0 0 0 0 4 Aptysia wilcoxi 0 0 0 0 2 0 0 0 0 0 0 0 2 Rana plpens 0 0 0 0 0 0 0 2 0 0 0 0 2 Hexapanopeus anguslifrons 1 0 0 0 0 0 0 0 0 0 0 0 1 Total 10423 5157 5980 17905 28340 30070 39948 18239 36585 15225 15051 9446 232369 60 40 20 FISH IMPINGED (*) COOLING WATER FLOW (+) * + * + 120 (- 110 100 90 80 70 76 77 78 79 80 31 82 83 84 85 86 87 YEAR Fig. 2. Annual total impingement estimates and total condenser cooling water flows at MNPS Unit 2. The large impingement of sand lance in 1984 was excluded from that annual total. 264 TABLE 4. Percent contribution of tlie dominant laxa collected in entrainment samples at Units 1 and 2 and estimated impingement at Unit 2. lintramment Taxon Impingement Anchoa spp. Pseudopleuronectes americanus Ammodytes spp. Myoxocephalus aenaeus Pholis gunnellus Brevoorlia tyrannus Tautogolabrus adspersus Tautoga onitis Syngnathus fuscus Liparis spp. Vivaria subbifurcata Scophthalmus aquosus Pepnlus triacanihus Enchelyopus cimbrius Gobiidae Prionotus spp. Myoxocephalus oclodecemspinosus Slenolomus chrysops Cynoscion regalis Scomber scombrus Anguilla rostrata Paralkhthys oblongus Menidia spp. Clupea harengus Clupeidae Urophycis spp. Sphoeroides maculatus Gadus morhua Paralkhthys dentatus Merluccius bilinearis Etropus microstomus Mlcrogadus tomcod Trinectes maculatus Gasterosteus aculeatus Alosa spp. Osmerus mordax Hemltripterus americanus Pollachius virens Fundulus spp. Gasterosteus wheallandi Cyclopterus lumpus Alosa aestivalis Alosa pseudoharengus Caranx hippos l.abridae Melanogrammus aeglejinus M or one americana MoTone saxa tills Opsanus tau Pomatomus saltatrix Raja spp. 62.7 10.5 14.5 10.3 0.0 16.0 9.1 0.0 0.3 3.9 0.0 11.4 1.9 0.0 0.2 1.7 0.0 0.5 1.7 53.4 3.6 1.7 29.3 1.6 1.0 0.0 3.2 0.9 0.0 0.3 0.8 0.0 0.0 0.7 1.6 1.7 0.7 0.0 3.1 0.6 0.5 0.0 0.3 0.0 0.0 0.3 2.0 0.3 0.2 0.0 0.0 0.2 1.0 0.2 0.2 0.0 0.4 0.1 0.0 0.0 0.1 0.0 0.3 0.1 0.0 0.0 0.1 0.1 10.4 0.1 0.0 0.1 0.1 0.0 0.0 0.1 0.4 0.3 0.0 0.0 0.4 0.0 0.0 0.1 0.0 0.0 0.5 0.0 0.0 3.0 0.0 0.0 0.1 0.0 0.0 6.5 0.0 0.0 0.1 0.0 0.0 9.6 0.0 0.8 0.0 0.0 0.0 0.8 0.0 0.0 0.2 0.0 0.0 0.3 0.0 0.0 0.3 0.0 0.0 3.9 0.0 0.0 1.1 0.0 0.0 1.0 0.0 0.1 0.4 0.0 0.0 0.1 0.0 0.3 0.0 0.0 0.0 0.1 0.0 0.0 1.3 0.0 0.0 0.1 0.0 0.0 0.3 0.0 0.0 0.3 0.0 0.0 0.9 Large impingement event in July 1 984 excluded. 1987a). Be^nning in 1984, the number of fish impinged continued to decline as cooling-water flow remained high. Certainly, the removal of the cofferdam contributed to the reduction of impingement at Unit 2, but impingement levels were expected to remain constant instead of con- tinuing to decline. Change in water circulation patterns caused by the start-up of Unit 3 may have contributed to the continued decline in impingement at Unit 2. Fish Ecology Studies 265 TABLE 5. The seasonal 5-mean density (no. per 500 m ) of abundant fish egg taxa collected at EN by report period (June- May)^ Taxon 79-80 80-8! 81-82 82-83 83-84 84-85 85-86 86 Taulogolabnis adspersus Tautoga onitis Anchoa spp. 'Data seasonally restricted to May 23-July 22 for Tautogolabrus adspenrus, to May 24- August 19 for Tautoga onitis, and to June 16-Ajgust 4 for Anchoa spp. . 6001 8298 5132 5519 7114 5745 7560 2941 1378 2891 2628 2268 2107 2165 3264 2731 1445 1242 1090 786 2275 5009 148 930 TABLE 6. The seasonal' 5-mean density (nc 1. per 500 m ) of abundant Tish larval taxa collected at EN by report period (June -May). Taxon 15-16 76-77 77-78 78-79 79-80 80-81 81-82 82-83 83-84 84-85 85-86 86-87 Anchoa spp. 1156 932 482 2235 2542 5701 845 14.34 306 1114 1270 Pseudopleuronectes americanus 170 104 137 114 287 129 237 301 212 166 83 108 Ammodytes americanus 22 96 238 119 no 137 22 28 17 10 2 12 Myoxocephalus aenaeus 16 41 21 36 38 107 72 68 50 68 34 30 Data seasonal restricted to July-September for Anchoa spp. , to March-June for Pseudopleuronectes americanus, to December- April for Ammodytes americanus , and to February-May for Myoxocephalus aenaeus. Althougli impingement has decreased at Unit 2, potential impacts could still occur there which would go undetected because of the elimination of routine monitoring. Therefore, a procedure was established so that operations personnel will examine and record all large impingement events (over 300 fish m a 24-hour period). Mitigative measures to reduce impingement impacts from MNPS have been incorporated and include the addition of a fish-return sluiceway at Unit 1 and continued modification of the fish-return system at Unit 3 to increase its efficiency. Ichthyoplankton Ichthyoplankton studies at MNPS have pro- vided entrainment estimates, seasonal density in- dices, and species composition. Over 50 taxa of ichthyoplankton have been collected (Appendix I). The additional mortality due to entrainment could affect local fish populations since natural ichthyoplankton mortality rates are one of the most important controlling factors of adult fish stock abundance (Gushing and Harris 1973; Ban- nister et al. 1974; Gushing 1974; May 1974; DeAngelis et al. 1977). S-mean densities were calculated for the most abundant fish egg and larval taxa entrained (Tables 5 and 6). Gunner eggs and anchovy larvae have consistently been the most abundant taxa col- lected. For all dominant larvae collected at EN, there appeared to be a general pattern of high abundance in the late 1970s or early 1980s fol- lowed by a decline to present levels that were similar to the mid 1970s. Gunner, tautog, and anchovies have accounted for over 93% of eggs collected at EN; and anchovies, winter flounder, sand lance, and grubby have accounted for over 85% of the larvae collected at EN (Table 4). These taxa were selected as potentially impacted and discussed in greater detail following this sec- tion, except for winter flounder (see Winter Floun- der Studies section). ITie dominant ichthyoplankton taxa collected at EN were compared to taxa collected in the trawl and seine programs. Except for winter floun- der, few adults of abundant ichthyoplankton taxa were collected by trawls or seines. Adult ancho- vies, cunner, and tautog apparently were not very susceptible to our demersal trawl or shore-zone seine sampling. Adult abundance, or stock re- productive capacity, for these three taxa was prob- ably best measured by seasonal egg abundance. The early life history stages of scup, windowpane, skates, and silversides were rare in ichthyoplankton 266 collections. This indicated that their spawning and early life history strategies reduced their suscepti- bility of being entrained. American sand lance ITie American sand lance is found from the Arctic to Cape Hatteras (Bigelow and Schroder 1953). They are primarily pelagic plankton feeders (Richards 1982). Individuals form large schools and are found over sandy bottoms from near shore to the edge of the continental shelf (Richards 1963; I^im and Scott 1966). Sand lance mature in 1 to 2 years and spawn between December and March (Westin et al. 1979). The life span of the sand lance has been reported as 5 to 9 years (Westin etal. 1979; Grosslein and Azarovitz 1982). The sand lance was collected in all fish ecology programs, but primarily during its larval stage in the winter and spring. It has generally contributed less than 1 % to armual total impingement (Table 4), except for the previously discussed one large impingement occurrence in 1984. Few were col- lected in the trawl and seine samples, possibly because juveniles and adults burrow into the sand (Leim and Scott 1966), thus avoiding these gears. Eggs were rarely collected because they are demersal and adhesive (Frizsche 1978). The sand lance was the third most abundant larval species at EN and NB. Aimual entrainment estimates, based on median densities, ranged from 2.8 to 66.7 million (Table 7). TABLE7. Annual entrainment estimates and 95% conndence intervals for American sand lance larvae entrained at MNPS. Enlrainmen^ estimate (xlO ) Year 95% CI 1976 18.6 14.7-23.5 1977 66.7 56.6-77.5 1978 36.4 21.9-48.6 1979 62.4 52.4-75.6 1980 66.6 57.6-83.8 1981 ';7.4 50.0-67.4 1982 12.4 10.4-15.6 1983 19.6 14.0-27.1 1984 13.1 10.4-15.7 1985 5.8 5.5-8.60 1986 2.8 0.0-5.00 1987 19.8 13.1-24.8 Annual larval abundance and temporal occur- rences were compared based on parameters from the Gompertz function. The function fitted the data well; all R^ values exceeded 0.94. Generally, larvae were collected from December through May, but the estimated dates of peak abundance (as determined by the inflection point) were quite variable both from year to year and between EN and NB within a year (Table 8). The a parameter was used as an index of annual abundance and it declined considerably since the early 1980s (Fig. 3). Based on the 95% confidence intervals, abun- dances at EN since 1982 have been significantly lower than the 1978 to 1981 period, but were similar to abundance in 1976. TABLF 8. Estimated date of peal< abundance for American sand lance based on the inflection point of the Gompertz function for stations EN and NB. Year EN NB 1976 Mar 22 . 1977 Feb 3 - 1978 Jan 31 - 1979 Apr 7 Mar 5 1980 Mar 23 Mar 12 1981 Mar 21 Mar 9 1982 Apr 2 Feb 21 1983 Apr 4 Mar 7 1984 Feb 29 Jan 16 1985 Feb 15 Jan 28 1986 Mar 31 Mar 29 1987 Apr 30 Apr 15 To determine if the decline of larvae in the Millstone area was localized, data on sand lance were examined from other areas (Table 9). Their abundance in the NMFS spring trawl survey data showed a similar decrease to levels reported in the mid 1970s. Larval densities reported for the waters off the SNPS were variable and declined from the early 1980s but not as evident as seen in the vicinity of MNPS. Similar to the MNPS program, few sand lance were taken in SNPS trawls, but there was an apparent decline in the annual CPUE. From these comparisons, it was apparent that decreasing sand lance abundance occurred regionally. Apparently, adult sand lance abundance has greatly fluctuated along the North- east Atlantic coast during the last 20 years. Meyer Fish Ecology Studies 267 6- 5- 3 - 2 - OH —I 1 1 1 1 r- 76 77 78 79 80 81 83 84 85 86 87 YEAR TABLE 9. Annual abundance indices of American sand lance, expressed as an annual sum of means, and trawl catch as annual CPUE at Shoreham Nuclear Power Sta- tion (SNPS), and 5-mean catch of the National Marine Fisheries Service (NMFS) spring trawl survey at selected stations. Fig. 3. American sand lance larval abundance estimates and 95% confidence intervals based on the a parameter from the Gompertz function for stations EN and NB. et al. (1979) reported that the average trawl catch in the spring for an area north of Cape Cod during the 1967-75 period was near 0, increased to 50 in 1976, and exceeded 10,000 in 1977. It appears that the dechne of larvae in the Millstone area resulted from the regional decrease of adult stocks, which are returning to levels similar of the early to mid 1970s. Anchovies Two anchovy species, the bay anchovy and the striped anchovy, have been collected in the Mill- stone area. Based on the proportion of eggs col- lected for each species, the bay anchovy was by far the most common in the area (more than 95%). In addition, nearly all specimens collected by trawl and in impingement collections were the bay anchovy. Due to the preponderance of the bay anchovy and the difficulty in separating the two species in larval and juvenile stages, the re- Sums for 1977-82 based on mean density per sam- • • j- • -ii r iU„ ^ , „„„u„.„, , , ■ ,r- , T- u ,oo-,v jr ,oo, <,<: u J maming discussion Will focus On thc bay anchovy, pling trip (Geomet Tech. 1983) and for 1983-86 based .,,.-,/- • i , on monthly mean density (EA Eng., Sci., and Tech. With the relatively few stnped anchovy eggs m- 1987). eluded in egg abundance estimates. Year Sum" SNSP NMFS larvae trawls trawls 1975 0.1 1976 123.0 1977 18463 0.11 3.8 1978 9446 1.03 249.2 1979 51545 0.61 20.0 1980 18925 0.06 145.9 1981 25989 0.66 45.9 1982 6278 0.71 37.1 1983 10165 0.68 26.3 1984 21753 0.16 7.7 1985 3284 0.00 6.5 1986 10474 0.16 0.9 1987 - 7.9 268 The bay anchovy is perhaps the most abundant fish along the Atlantic Coast and is usually the dominant ichthyoplankton species in estuaries within its range (McHugh 1977; L^ak and Houde 1987). Its range extends from Cape Cod to Mex- ico, with occurrences as far north as Maine (Hildebrand 1943; Bigelow and Schroeder 1953). They are commonly found inshore during the warmer months and move offshore in the winter, but are seldom found in water deeper than 25 m (Grosslein and Azarovitz 1982). Hildebrand (1943) believed that each section of the coast had discrete anchovy populations and movements were inshore and offshore. In LIS, spawning takes place at depths of less than 20 m during .lune through September (Richards 1959). Eggs are pelagic and at 27°C hatch in about 24 hours (Kuntz 1914). Since water temperatures in LIS near Millstone rarely exceed 22°C, incubation probably takes longer here. Development is rapid and individuals may mature within 2.5 months of hatching in Delaware Bay, and maximum life span is probably not more than 2 or 3 years (Stevenson 1958). Various anchovy life history stages were very abundant in some programs, but rarely collected in others. Their period of occurrence in each program has been consistent among years (NUSCO 1987b). Adults were primarily found in impingement collections from May through June, which corresponded to the spring inshore spawning migration. The estimated numbers im- pinged at Unit 2 have declined dramatically since 1984, particularly in 1986 and 1987 (Table 2). This was probably related to the previously dis- cussed overall decline in Unit 2 impingement for all species. Adult anchovies were rare in the demersal trawl program. Vouglitois et al. (1987) reported large numbers of adults collected in Bamegat Bay, NJ, but their data were from semiballoon otter trawl collections as opposed to our flat otter trawl, which may have accounted for this difference. .luvenile anchovies, resulting from the summer spawning were susceptible to our trawling and were captured during August through October, primarily at NB (Appendix III). Even though anchovies ranked third in trawl catch, they were collected infrequently and in large numbers with over 70% of them collected in only 13 of the over 5,000 tows. This infrequent collection of individuals greatly limited the use- fulness of trawl data as an index of abundance. At EN, anchovy larvae ranked first among all species and was the third most abundant egg taxon (Table 4). Since over 50% of the eggs collected annually occurred during a 2- to 3-week period, spawning was during a short period of time. The annual date of peak abundance, esti- mated from the Gompertz function inflection point, ranged from late June to mid- July (Table 10). The date of the larval peaks did not appear TABLE 10. Estimated date of peak of abundance anchovy larvae at EN and NB and eggs at EN based on the inflection point of the Gompertz function. Eggs at Larvae at Year EN EN NB 1976 . Jul 22 1977 - Jul 23 - 1978 - Aug 16 - 1979 Jul 13 Jul 21 Jul 18 1980 Jul 18 Jul 20 Jul 19 1981 Jul 12 Jul 21 Jul 21 1982 Jul 7 Jul 15 Jul 11 1983 Jul 28 Jul 18 Jul 12 1984 Jun 28 Jul 18 Jul 12 1985 Jul 16 Jul 17 Jul 18 1986 Jul 4 Aug 13 Aug 9 to be related to the date of peak abundance for eggs, but the dates for larvae at EN and NB were similar. This similarity suggested that the factors affecting the timing of maximum abundance were probably the same throughout the Millstone area. The causes of the later larval peaks in 1978 and 1986 were not known. Any potential impact of the operation of MNPS on the anchovy popula- tion would probably be due to entrainment with annual estimates ranging from 16.0 to 807.7 mil- lion for eggs and from 1.5 to 1,284 million for larvae (Table 11). The estimated number of eggs entrained increased in 1986 relative to the previous year, because egg densities were low in 1985 and the increased cooling water demands of Unit 3 in 1986. However, this increase was not apparent Fish Ecology Studies 269 TABLE U. Annual entrainment estimates and 95% confidence intervals for anchovy eggs and larvae at MNPS. Eggs Larvae Entrainment estimate (XIO ) 95% CI Entrainment estimate (XIO^) 95% CI 448.4 334.2- 576.4 162.5 119.0- 248.4 160.0 111.6- 236.9 600.9 432.6- 766.5 558.1 480.9- 707.1 1284.1 1061.5-1531.9 299.7 229.6- 396.3 485.6 346.3- 678.2 91.3 59.3- 159.4 454.9 375.6- 750.3 238.5 71.5- 398.7 1976 - 1977 - 1978 - 1979 464.1 1980 183.1 1981 369.3 1982 213.6 1983 503.5 1984' 807.7 1985 16.0 1986 347.9 366.0- 540.0 47.2- 250.4 285.2- 462.1 148.1- 277.0 348.2- 700.2 388.3-1249.9 0.0- 53.70 142.7- 533.0 Revised larval estimates due to error in previous calculations (NUSCO 1987b). for larval entrainment because Unit 3 was shut- down from July 25 to August 17, 1986 during the period of peak larval abundance; the total entrain- ment estimate was a function of both abundance and volume of cooling water. Examination of anchovy egg and larval annual abundances, using the a parameter from the Gompertz function as an index, showed large fluctuations for both developmental stages (Fig. 4). The Gompertz function fitted the cumulative data well (all R values exceeded 0.94). During three-unit operations, egg abundance was low 30 25 20 15 - 10 - LARVAE 1_ARVAE AT EN 76 77 78 79 81 82 YEAR 83 84 85 Fig. 4. Anchovy egg and larval abundance estimates and 95"/'o confidence intervals based on the a parameter from the Gompertz function for stations EN and NB. 270 compared to 1983 and 1984, but similar to the period of 1980-82. Larval abundance at EN dur- ing 1986 was similar to or higher than most other years, except for 1981. ITie fluctuations in Icirval abundance at NB were similar to EN; the 1984 abundance was low at both stations. No long term-trends in abundance were apparent for eggs or larvae. Patterns of annual abundance of anchovy eggs and larvae in the vicinity of S NFS were not sim- ilar to the Millstone area (Table 12). The low abundances of eggs at Millstone in 1985 and 1986 were not found at SNPS, which had high annual egg abundance during this period. Further, the peak that occurred in 1984 at Millstone was not evident at SNPS. During the early 1980s, larval abundance at Millstone was the highest, but at SNPS it was the lowest. These dissimilarities sug- gest that the spatial distribution of anchovies in LIS may differ from year to year. Trawl catches of anchovies at SNPS were variable, but a decline has occurred since the early 1980s, and similar to our trawl data, a majority of the catch was taken in late summer and early fall, probably of young- of-the-year. TABLE 12. Annual abundance indices of anchovy eggs and larvae, both expressed as an annual sum of means; and trawl catch as annual CPUE at Shoreham Nuclear Power Station. Sum Sum Trawl Year eggs larvae CPUE 1977 126369 38849 1.02 1978 36687 4475 0.09 1979 100589 6129 0.74 1980 32388 934 10.82 1981 3587 1245 1.08 1982 6756 839 3.75 1983 173171 16638 0.17 1984 49565 6261 0.02 1985 109688 15440 0.00 1986 107613 9728 0.56 Sums for 1977-82 based on mean density per sam- pling trip (Geomet Tech. 1983) and for 1983-86 based on monthly mean density (EA Eng., Sci., and Tech. 1987). A comparison of the annual abundance of eggs and larvae at EN indicated two apparent discrep- ancies. First, the index of armual abundance (a) for eggs was generally lower than for larvae, and secondly, there was no relationship between the annual egg and larval abundances. The apparent low abundance of eggs relative to larvae may be related to the length of time that each develop- mental stage was available for capture; the incu- bation period for eggs was only a few days and larvae were available for capture over a period of weeks. In order to make a direct comparison of abundance of eggs to larvae, the abundance index would have to be weighted according to the av- erage developmental time for each respective stage. Another possible explantion for low egg abun- dance, compared to larval abundance, was that Niantic Bay was not a primary spawning area for anchovies and that larvae were transported by tidal currents to the bay from more preferred spawning grounds. In either of the above two cases, there should be some relationship between the abundance of eggs (a measure of spawning stock size) and the resulting larvae, unless there were annual fluctu- ations in the mortality rates of eggs and larvae. In some years there appeared to be an inverse relationship between egg and larval abundances. Vouglitois et al. (1987) reported a similar pattern for the bay anchovy in Bamegat Bay, NJ during a 2-year period. This would suggest that com- pensatory mortality occurred during the early life history stages. Density-dependent mortality was examined by comparing annual a values for eggs (E) to the 8-mean density of larvae (L) collected during August at EN. Compensatory mortality was evident, as the slope (P) was negative and significantly different from zero (Fig. 5). This relationship was most evident in 1981 which had relatively low egg abundance and the greatest larval abundance, and in 1984, with the greatest egg and the lowest larval abundance. Compensa- tory mortality can be caused by several factors, including starvation due to competition for prey and increased predatory pressure. Houde (1977, 1978a, 1978b) found in laboratory studies that food prey availability affected the survival of the Fish Ecology Studies 271 6000 9000 EGG ABUNDANCE INDEX I'ig. 5. I jnear relationship between tfie larval anchovy recruitment index and the egg abundance index, suggesting density-dependent mortality in the early life history stages. bay anchovy larvae. However, Leak and Iloude (1987) reported that in a field study the highest anchovy mortality was during egg and yolk-sac stages and estimated that mortality due to preda- tion was two to three times higher than that at- tributed to star/ation. Cannibalism by adults may be a source of predation and because adult abundance is directly related to egg abundance, this would be a self-regulating mechanism. Causal mechanisms for the compensatory mortality of the early life history stages of anchovies in the Millstone area are not known, but because com- pensation was occurring at the same time as en- trainment through MNPS, this would help miti- gate the impact on the adult anchovy population. Sticklebacks The threespine stickleback and the blackspotted stickleback are small, nearshore fishes. The threespine stickleback is distributed throughout the north polar regions and as far south as Chesapeake Bay in the Western North Atlantic; the blackspotted stickleback is found only in the Western North Atlantic from Newfoundland to LIS (Perlmutter 1963). Threespine and blackspotted sticklebacks are very similar in appearance and are not easily dis- tinguished (Bigelow and Schroeder 1953). Because of this similarity, the blackspotted stickleback was not identified in MNPS collections until October 1981 (NUSCO 1982). Although Fitzgerald and Wlioriskey (1985) found no size overlap between these two species in Canada, the length frequency of individuals collected at MNPS overlapped at 30 to 55 mm (Fig. 6). Thus, length frequency could not be used to separate the species in earlier years and the data for the two species were com- bined. Sticklebacks were collected in all sampling pro- grams, but were only abund£int in impingement samples from fall through spring. During the 272 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0 THREE-SPINED 1 1 1 1 1 1 1 1 — 30 35 40 45 50 55 60 65 70 LENGTH 75 Fig. 6. Length -frequency distribution by 1-mm incremenLs of threespine and biackspotted sticklebacks impinged at MNPS Unit 2 from 1982 through 1987. spring, both species move into salt marshes and tidal rivers to spawn (Worgan and Fitzgerald 1981) and both adults and young-of-the-year re- main in the spawning areas until late summer (Fitzgerald 1983). These spawning habits proba- bly accounted for the low catch of these species during the summer. The number of sticklebacks impinged annually at Unit 2 ranged from 16,329 in 1983 to 880 in 1985 (Table 2). Approximately 32% of all sticklebacks impinged there were taken in 1983 and there has been a decrease in impingement since the mid- 1 980s to levels similar to the mid and late 1970s. The impact of impingement on these two species at MNPS has been mitigated with the addition of sluiceways at Units 1 and 3 and high survival estimates for returned fish of over 90% and 70%, respectively (NUSCO 1988). Atlantic tomcod The Atlantic tomcod is the most abundant member of the cod family collected in the MNPS monitoring programs. It ranges along the Atlantic coast of North America from Newfoundland to Virginia (Bigelow and Schroeder 1953). The Atlantic tomcod was caught in all sampling programs, but was only abundant in impingement and trawl samples. Eggs are adhesive and found attached to the substrate and larvae tend to remain in brackish water spawning areas (Howe 1971), habitats not sampled by the monitoring programs. Tomcod reach sexual maturity at about 130 mm (Howe 1971) and over 98% of those impinged were adults larger than 130 mm (Fig. 7). About 90% of those impinged were taken in the fall and winter, during the time of their spawning migration (Howe 1971; Klauda et al. 1987). Over 40% of the total impingement at Unit 2 since 1976 oc- curred in 1982 (Table 2). A marked decrease Fish Ecology Studies 273 SEASON LENGTH SPR 1 NG-SUMMER 60 80 1 00 1 20 140 1 60 FALL-WI NTER 60 80 100 1 20 140 1 60 200 400 FREQUENCY 2000 4000 FREQUENCY I MP I NGEMENT Fig. 7. Spring-summer and fall-winter seasonal length-frequency distribution by 20-mm size-classes of Atlantic tomcod in trawl and MNPS Unit 2 impingement collections from 1976 through 1987. occurred in the number impinged since three-unit operations. In trawls, over half of the tomcod were caught at the nearshore stations (NR and JC). Young-of-the-year dominated the trawl catches in the spring and summer; adults were caught mostly in the fall and winter (Fig. 7). Trawl catches peaked from April through June, except at NR, where most were caught during their winter spawning season. The annual abundances of tomcod in trawls were analyzed using the 5-mean as an index of relative abundance; only data from catches at NR and JC were sufficient to determine annual trends (Table 13). Armual 5-mean catches were low and during 2 report periods (1980-81 and 1985-86), none were found at JC. The Atlantic tomcod is a short-lived species and becomes sexually mature during the fall of its first year (Klauda et al. 1987). Marine fishes that are short-lived usually exhibit large year-to-year variations in abundance because population size is determined in a single spawning TABLE 13. The annual S-mean catch (CPUE) and 95% confidence interval of Atlantic lomcod caught by trawls at stations JC and NR for each report period (June-May). Year JC NR 1976-77 0.3 ± 0.3 0.1 ± 0.2 1977-78 0.5 ± 0.7 0.2 ± 0.3 1978-79 0.2 ± 0.4 0.3 ± 0.2 1979-80 0.2 ± 0.4 0.2 ± 0.2 1980-81 0 2 ± 1.8 1981-82 0.04 ± 0.08 3 ± 6.9 1982-83 0.08 ± 0.1 1 ± 0.7 1983-84 0.3 ± 0.4 0.3 ± 0.2 1984-85 0.2 ± 0.1 0.3 ± 0.2 1985-86 0 0.7 ± 0.5 1986-87 0.04 ± 0.08 0.07 ± 0.1 season and conditions during that season often control spawning success. Tliis is evident in the distribution of annual catches at JC. In 1980-81, no tomcod were collected at this station, but 3 274 years later (1983-84) the 5-mean catch was one of the highest recorded. The 1986-87 catches were low at both stations but will probably re- bound when conditions are better suited for re- productive success. Silversides Two species of silversides dominate the shore zone along the Connecticut coast, the Atlantic silverside and the inland silverside. Both species are sympatric along the Atlantic coast, wdth the Atlantic silverside ranging from the Gulf of St. Lawrence to the Chesapeake Bay and the inland silverside ranging from Cape Cod to South Car- olina (Johnson 1975). Both species spawn as yearlings and have a life cycle that ranges from 1 to 2 years. Both are omnivorous, feeding on copepods, mysid shrimp, fish eggs, and young squid. They are important as forage food for larger fish species (Bigelow and Schroeder 1953). Silversides collected in MNPS programs were not always identified to species. When identified, the Atlantic silverside was the most abundant (over 90%). However, to determine long-term trends the two species were analyzed as a single taxon. Silversides dominated seine collections, accounting for over 80% of the total catch (Ap- pendix IV). They were also among the top-ranked taxa collected in the trawl and impingement pro- grams (Tables 1 and 4). Silversides were not abundant in plankton samples because their eggs are adhesive (Bigelow and Schroeder 1953) and larvae and juvemles stay close to shore (Bayliff 1950). Seasonal patterns of abundance occurred in the seine, trawl, and impingement collections. They were found in seine collections in the spring, sum- mer, and fall and in trawl and impingement col- lections during the winter. This pattern suggested that silversides in the MNPS area overwintered in waters close to shore. Offshore winter migra- tions of silversides have been reported in other studies (BayliPf 1950; Bigelow and Schroeder 1953; Conover 1979; Conover and Murawski 1982). Conover (1979) found that silversides in Massa- chusetts migrated about 50 km offshore in winter to depths of 100 m and only 1% of the population returned inshore the following spring. In the MNPS area, the rate of return (mean number per haul in the fall compared to the spring) was cal- culated starting in 1984, the first year sampling was done at least once a month and biweekly from April through October. The average return rate at the three stations ranged from about 9% in 1985 to about 40% in 1986. Silversides in LIS may have a different overwintering strategy than fish that overwinter in the Gulf of Maine. Winter water temperatures in LIS do not approach the rC lethal limit for silversides (Hoff aind Westman 1966), but average 2.5-4°C. Thus, silversides in LIS may not have to move as far offshore to find tolerable winter water temperatures. Because of the additional thermal effluent in Jordan Cove, a potential exists for changes in the distribution of silversides inhabiting the shore- zone at JC and WP. Sampling at these two sites and at control station GN was doubled in 1984 from monthly to biweekly (April through Octo- ber) to increase our ability to detect any changes. Annual 5-mean catches were calculated for each of the three seine sites (Table 14). The 1986 TABLE 14. Seasonal 5-mean catch (CPUE) and 95% confidence interval of silversides by seine. 1976" 1977 1251 ± 2094 151 ± 569 62 ± 168 1978 26 ±20 32 ± 64 24 ± 35 1979 41 ±55 26 ± 27 15 ± 12 1980 479 ± 930 104 ±96 55 ± 70 1981 114 ± 107 81 ±72 31 ± 40 1982 108 ± 102 48 ± 112 192 ± 567 1983 580 ± 989 42 ± 59 112 ±158 1984 35 ± 35 11 ± 9 3 ± 1 1985 23 ± 14 18 ± 11 6 ± 4 1986 200 ± 404 58 ±48 16 ± 10 Season used for calculating 5-mean was June-November at all stations Not enough data available in 1976 to calculate 5-mean Fish Ecology Studies 275 SEASON LENGTH Wl NTER 20 40 60 80 100 120 SPR 1 NG 20 O 1000 2000 FREQUENCY SE 1 NE 100 200 300 FREQUENCY 1 O 20 FREQUENCY I MP I NGEMENT Fig. 8. Seasonal length-frequency distribution by 20-mm size-classes of silversides in seine, trawl, and MNPS Unit 2 impingement collections during 1986 and 1987. catches were within the range of historic data. The length-frequency distribution of the 1986 catch was compared to the long-term frequency, and the size composition found in 1986 was sim- ilar to previous data (NUSCO 1987b; Fig. 8). Seasonal patterns of occurrence in 1986-87 were also similar to historic patterns. The estimated 136 silversides impinged in 1987 was a historic low. The pattern of decreased impingement was previously discussed and was not specific to silversides, but was probably due to physical changes near the MNPS intakes. The 5-mean catch in trawls, an index of winter adult abundance, for the 1986-87 report period was similar to previous catches (Table 15), indicating that the decrease in impingement was not due to a change in adult abundance. TABLE 15. Seasonal 5-mean catch (CPUE) and 95% confidence interval of silversides by trawls at selected stations by report period (June-May). Report period IN JC NB NR 1976-77 15 ± 16 13 ± 20 6 ± 8 77 ± 283 1977-78 29 ± 92 6 ± 612 18 ± 25 10 ± 21 1978-79 60 * 105 9 ± 8 8 ± 7 2 ± 1 1979-80 42 ± 276 6 ± 16.8 0.7 ± 1.6 4 ± 6 1980-81 8 ± 17.4 4 ± 4.6 19 ± 41.5 3 ± 3.8 1981-82 6 ± 9.2 0.7 ± 0.4 5 ± 6.4 6 ± 8 1982-83 2 ± 3.5 1 ± 2.4 0.5 ± 12.5 12 ± 4.7 1983-84 2 ± 4.2 4 ± 1.3 4 ± 0.6 1 ± 6.3 1984-85 2 ± 6.3 5 ± 11.3 1 ± 1.4 0.5 ± 0.9 1985-86 7 ± 8.2 6 ± 7.6 2 ± 1.4 3 ± 5.9 1986-87 5 ± 3.1 8 ± 6.9 4 ± 2.9 110 ± 222 Season used for calculating S-mean was November-February at IN, NB and NR and October-January at JC. 276 Grubby The grubby is found in coastal waters, com- monly in eelgrass habitats, along the Atlantic coast of North America from the Gulf of St. Lawrence to New Jersey (Bigelow and Schroeder 1953). They spawn throughout the winter (Lund and Marcy 1975) and Richards (1959) reported finding larvae in shallower areas of US from Feb- ruary to April. The grubby tolerates a wide range of temperatures and salinities (Bigelow and Schroeder 1953). The grubby is a resident in the Millstone area and both larvae and adults were collected in the monitoring programs. However, eggs were rarely taken because they are demersal and adhesive (Lund and Marcy 1975). Juveniles and adults were rarely collected in the shore zone by seine. Grubby larvae accounted for about 4% of the total larval collection in entrainment samples (Ta- ble 4). Annual entraiimient estimates have ranged from 8.9 to 50.0 million and the estimates during three-unit operations were similar to those in pre- vious years during the 1980s, even with the addi- tion of Unit 3 (Table 16). Larvae were collected at both EN and NB, primarily in February through May and the dates of peak abundance varied from mid March to early April (Table 17). The a parameters calculated from the Gompertz TABLE 16. Annual entrainment estimates and 95% confi- dence intervals for grubby larvae at MNPS. TABLE 17. Estimated date or peak abundance for grubby larvae based on the inflection point of the Gompertz function for stations EN and NB. Entrainment Year estimate (xlO ) 95% CI 1976 12.0 10.M5.2 1977 30.2 24.9-33.7 1978 8.9 7.3-10.8 1979 19.8 17.9-22.2 1980 30.2 25.8-37.1 1981 45.0 33.4-51.7 1982 ^6.4 41.8-57.2 1983 50.0 41.6-63.5 1984 35.8 29.0-42.7 1985 36.5 28.9-42.5 1986 47.1 38.0-61.0 1987 45.5 32.0-58.4 Year EN NB 1976 Mar 23 1977 Apr 7 1978 Mar 20 - 1979 Mar 27 Mar 21 1980 Mar 23 Mar 22 1981 Apr 4 Apr 4 1982 Mar 28 Apr 1 1983 Mar 14 Mar 27 1984 Mar 11 Mar 12 1985 Mar 24 Apr 5 1986 Mar 17 Mar 22 1987 Mar 15 Mar 14 equation (all R values exceeded 0.98) were used as indices of annual larval abundance (Fig. 9). larvae were most abundant from 1981 to 1983 at EN and the abundance during 1984 to 1987 decreased to a level similar to the late 1970s. Annual abundance at NB was consistently lower than at EN. Larval abundances during three-unit operation were within the range of historical data at EN and NB. In impingement collections, the grubby was the third most abundant taxon, accounting for over 1 1 % of the total (Table 4). Most were adults (60 to 120 mm) taken during their winter spawning season (December through April). As was found for other species, there was a recent decline in impingement at Unit 2 with the estimated number in 1987 the lowest recorded (Table 2). Almost 80% of the grubby taken by trawl were found at the nearshore stations (IN, JC, and NR) (Appendix III). They were collected throughout the year at JC and NR, and primarily during their spawning season at IN. At these stations, 5-mean catches during three-unit operations were within the historical range (Table 18). Highest annual catches at IN and NR occurred in the early 1980s and corresponded to high larval abundance at EN during the same time period. The annual mean length at IN and JC has remained fairly consistent, but at NR mean length decreased during the 1980s (Fig. 10). This smaller mean length at NR Fish Ecology Studies 277 12- 10- k 8- \ T 6 - V'\ 4- / h--^-^. iU EN NB n - 76 77 78 79 80 n 82 YEAR 83 84 85 86 87 Fig. 9. Grubby larval abundance estimates and 95% confidence intervals based on the re parameter from the Gompertz function for stations EN and NB. was concurrent with larger catches and suggested higher recruitment of young. TABLE 18. Seasonal 5-mean catch (CPUE) and 95% con- fidence intervals of grubby caught by trawl at selected stations during each report period (June-May). Report period IN JC NR 1976-77 1 ± 0.4 1 ± 1.1 1 ± 0.8 1977-78 2 ± 0.9 2 ± 1.1 0.5 ± 0.4 1978-79 1 ± 0.5 2 ± 1.3 1 ± 0,5 1979-80 2 ± 0.9 1 ± 0.3 5 -h 3.0 1980-81 2 ± 1.3 1 ± 0.4 2 ;t 1.7 1981-82 5 ± 2.7 0.7 i 0.4 9 ± 8.2 1982-83 5 ± 3.1 0.7 ± 0.3 16 lb 9.5 1983-84 2 ± 0.6 2 ± 0.6 5 ± 2.8 1984-85 2 ± 3.3 0.8 ± 0.6 7 ± 4.1 1985-86 2 ± 0.7 0.6 ± 0.3 4 ± 2 1986-87 0.9 ± 0.3 0.7 1: 0.5 8 ± 6.4 Season used for calculating the 5-niean was December- June at IN and year-round at JC and NR (June-May). Except at NR, no long-term changes in mean length or abundance indices occurred for the grubby, even though it was a dominant taxon in both entrainment and impingement. The esti- mated number impinged has declined since 1985, but there has been no decline in trawl catch, in- dicating this was probably a result of physical changes (removal of the cofferdam) near the Unit 2 intake. Impingement impacts for the grubby have been reduced with the addition of sluiceways at MNPS. The grubby is a hardy species and had high sluiceway survival of 74% at Unit 1 and 97% at Unit 3 (NUSCO 1988). Tautog The tautog is found from New Brunswick to South Carolina, but is m.ost common from Cape Cod to the Delaware Capes (Cooper 1965). Its distribution is limited primarily to inshore regions with individual populations being highly localized (Cooper 1966). They are commonly found in 278 no- 100- 90- 80- ^ V . ^ ^ \ 70- 1 ^^"^^^--^' 'N RO ■ 76-77 78-79 80-81 82-83 84-85 86-87 YEAR 110 100- -§■ 90 S 80 76—77 7S— 79 80—81 82-83 84-85 86-87 YEAR 110 100 90 80- 70- 60 76—77 78-79 80-81 82-83 84—85 86-87 YEAR Fig. 10. The annual mean length and 95% confidence interval of grubby taken by trawl at stations IN, NR, and JC. Fish Ecology Studies 279 waters less than 20 m around rocky areas, ledges, mussel beds, breakwaters, and other similar nearshore habitats from early May until late Oc- tober (Bigelow and Schroedcr 1953; Wheatland 1956; Cooper 1965). Juveniles are also found in eelgrass beds and among macroaigae in coves and channels (Tracy 1910; Briggs and O'Conner 1971). Both juveniles and adults have a home site where they remain inactive and under cover at night; during the day larger fish move to other locations to feed, but juveniles remain close to their home sites (OUa et al. 1974). During winter, adults move to deeper water (about 25 to 55 m) and remain inactive while juveniles stay inshore to overwinter in a torpid state (Cooper 1965; Olla et al. 1974). Tautog males become sexually ma- ture at age 2-3 and females at age 3-4 and the maximum reported age for males is 34 years and females is 22 years (Chenoweth 1963; Cooper 1965). Spawning occurs from mid-May until mid- August in LIS (Wheatland 1956; Chenoweth 1963). The eggs are pelagic, hatch in 42-45 hours at 22°(", and are concentrated in the upper 5 m of the water column (Williams 1967; Fritzsche 1978). Young become benthic and move inshore after metamorphosis, which is completed by 10 mm (Fritzsche 1978). TABLE 19. Total number of tautog caught by trawl at each station during each report period (June-May)./ Report period JC NR NB TT BR IN 1976-77 71 39 20 26 10 63 1977-78 106 16 35 29 27 70 1978-79 59 30 43 15 30 86 1979-80 57 45 32 26 42 68 1980-81 22 25 27 10 15 47 1981-82 20 129 23 6 23 27 1982-83 37 90 23 26 13 50 1983-84 18 16 23 15 27 41 1984-85 15 11 18 9 20 46 1985-86 31 22 7 11 16 47 1986-87 57 no 2 4 19 23 Total 493 533 253 177 242 568 evident at the other stations. Some of the largest annual fluctuaaons have occurred at NR. Mean length has remained fairly constant at all stations, except at NR (Fig. 11). The decline in mean length at NR was concurrent with increased abun- dance, particularly evident in the 1981-82, 1982-83, and 1986-87 report periods, indicating greater re- cruitment of young. Similar to the Millstone area, no trends were apparent in trawl CPUE at SNPS (Table 20). The tautog was collected in all sampling pro- grams, but only in high abundance as eggs from May through August. Due to its habitat prefer- ence, it was raiely collected in the shore-zone seine program (Appendix IV). Even though tautog prefer rocky shores, such as those adjacent to the MNPS intakes, they were not impinged in large numbers and contributed less than 2% to the total impingement estimate; annual estimates at Unit 2 ranged from 96 in 1986 to 731 in 1978 (Table 2). Annual 5-mean abundance indices could not be calculated from trawl data because individuals were collected infrequently. Because trawl sampling effort was nearly the same each year, the total catch from each station and report period (June- May) was used to examine spatial and temporal distributions (Table 19). Tautog were caught primarily at nearshore stations (JC, IN, and NR). The lowest catches at NB have occurred in recent years, but this decline was not TABLE 20. Annual abundance indices of tautog eggs and larvae, both expressed as an annual sum of means, and trawl catch as annual CPUE at Shoreham Nuclear Power Station. Sum Sum " Trawl Year eggs larvae CPUE 1977 17727 424 0.76 1978 5930 197 0.33 1979 11337 113 0.61 1980 1711 50 0.63 1981 4062 12 0.40 1982 3239 0.28 1983 3428 99 0.34 1984 4415 b 0.19 1985 6003 168 0.32 1986 11562 166 0.29 Sums for 1977-82 based on mean density per sam- pling trip (Gcomet Tech. 1983) and for 1983-86 based on monthly mean density (EA Rng., Sci., and Tech. 1987). Abundance not reported (Geomet Tech 1983) appar- ently due to low densities. 280 y 200 y 200- 77 78 79 80 81 82 83 84 85 86 YEAR 76 77 78 79 80 81 82 83 84 85 86 YEAR 400 300 200 y 200 77 78 79 80 81 82 83 84 85 86 YEAR 76 77 78 79 80 81 82 83 84 85 YEAR 9 200 OT— I- 77 78 79 80 81 82 83 84 85 86 YEW 76 77 78 79 80 81 82 83 84 85 86 YEAR Fig. 11. The annual mean length and 95% confidence interval of tautog taken by trawl at each station. Fish Ecology Studies 281 Probably the greatest potential impact to the tautog from three-unit operations is from the en- trainment of eggs. The entrainment estimate for 1986 was at least twice that of previous years (Table 21). The Gompertz function was fitted to the annual cumulative egg densities at EN (all R values exceeded 0.98) and the a parameter was used to compare abundances. Annual egg abun- TABLE 21. Annual entrainment estimates and 95% confi- dence intervals for tautog eggs at MNPS. Entrainment Year estimate (xlO ) 95% CI 1979 645.8 508.5- 809.7 1980 992.1 836.5-1158.2 1981 1385.3 1201.5-1655.4 1982 1443.4 1181.1-1579.0 1983 953.7 718.7-1275.2 1984 1211.9 915.9-1543.2 1985° 1435.9 1037.2-2416.1 1986 3163.7 2597.5-4021.1 Revised estimate due to error in previous calculations (NUSCO 1987b). dance fluctuated from a low in 1979 to relatively high abundances in 1985 and 1986 (Fig. 12). The time of peak abundance (estimated from the in- flection point of Gompertz function) generally occurred mid to late June (Table 22). TABLE 22. Estimated date of peak abundance for tautog eggs based on the inflection point of the Gompertz func- tion for station EN. Year EN 1979 1980 1981 1982 1983 1984 1985 1986 Jun 23 Jun 22 Jun 21 Jun 28 Jun 21 Jun 23 Jun 17 Jun 12 Tautog larvae did not appear consistently in ichthyoplankton samples during their period of occurrence and when present, occurred in low densities. Due to the numerous zero values for sample densities, larval entrainment estimates, 32 24 20 16 79 82 83 84 85 YEAR Fig. 12. Tautog egg abundance e,stimates and 95% confidence intervals based on the a parameter of tfie Gompertz function for station EN. 282 based on medians, could not be calculated. The seasonal 5-mean densities were used as an abun- dance index (Table 23). Larval densities have been consistently higher at NB than at EN, but similar annual trends were evident at both stations; abundance increased from the late 1970s to early 1980s and then declined to levels comparable to those in the mid-1970s. This decline was not evident in annual egg abundances. TABLE 23. The seasonal 5-mean density (no. per 500 m ) of tautog larvae collected at stations EN and NB by year. Year EN NB 1976 37.3 ± 16.0 1977 36.3 ± 15.3 1978 1.2 ± 0.6 1979 11.6 ± 4.8 50.8 ± 22.6 1980 46.9 ± 17.8 91.5 ± 50.0 1981 82.9 ± 36.0 92.5 ± 62.0 1982 44.4 ± 21.9 111.8 ± 60.0 1983 34.1 ± 21.1 124.2 ±129.5 1984 3.1 ± 2.1 29.2 ± 21.1 1985 18.2 ± 12.7 44.2 ± 15.0 1986 3.2 ± 1.8 12.2 ± 6.7 Data seasonally restricted to June through August. Comparison of annual egg to larval abundances, based on the 8-mean (see Table 5 for egg 5-mean), indicated that egg to larval survival was low; gen- erally the larval index was less than 2% of of the corresponding egg index. Similar low annual larval to egg ratios were evident for collections at SNPS (Table 20). Further discussion of this low survival for the wrasses (tautog and cunner) is provided in the following section on cunner. This apparent low natural egg survival would reduce the potential impact of the large numbers of eggs entrained by three-unit operations. A comparison of annual larval and egg abundances did not reveal a density-dependent relationship, as was found for anchovies. Because the tautog takes 2 to 4 years to reach maturity, the possible impact of entrainment by three-unit operation on the adult stock size (best measured by annual egg abun- dance) will not be evident for several years. Cunner The cunner is a coastal marine fish that prefers rocky habitats (Bigelow and Schroeder 1953; Serchuk 1972; Olla et al. 1975, 1979; Dew 1976; Pottle and Green 1979). It ranges from northern Newfoundland to the mouth of the Chesapeake Bay (Leim and Scott 1966). Most cunner have limited home ranges (less than 4 km) and probably stay within several meters of their nighttime shel- ter. Adults generally display highly localized abundance in areas they inhabit and their numbers greatly decrease only a short distance from cover (Gleason and Recksiek, in preparation). They are active only during the day and activity declines in cold weather as individuals become dormant at temperatures below 8°C and lie torpid among and under rocks (Green and Farwell 1971; Green 1975; Dew 1976; Olla et al. 1979). Individuals mature in 1 to 2 years and the maximum reported age is 10 years (.lohansen 1925; Dew 1976). Cunner spawn primarily in June through August and the pelagic eggs hatch in 2-6 days depending on water temperature (Williams 1967; Dew 1976; Fritzsche 1978). Metamorphosis of larvae is com- plete by 10 mm and juveniles move to the bottom (Miller 1958). All life stages of the cunner were collected in the Millstone area. Eggs and larvae were found in ichthyoplankton collections, primarily from June through August. Juveniles and adults were caught at aU six trawl stations and in greatest abundance during the spring through fall. Cunner were rarely collected by seine (Appendix IV). As cunner prefer the rocky habitats that surround MNPS, the species was among the top dozen of those impinged (Table 2). Annual impingement estimates at Unit 2 decreased from a high of 1,787 in 1983 to a low of 57 in 1987. This decrease followed the general decline in total impingement of all species. Cunner eggs predominated in entrainment col- lections and were abundant during May through July; the date of peak abundance consistently oc- curred during the first half of June (Table 24). Fish Ecology Studies 283 TABLE 24. Estimated date of peak abundance for cunner eggs based on the inflection point of the Gompertz function for station EN. Year EN 1979 1980 1981 1982 1983 1984 1985 1986 Jun 12 Jun 14 Jun 8 Jun 18 Jun 14 Jun 9 Jun 7 Jun 5 The a parameter from the Gompertz function was used to estimate annual egg abundance and all R^ values exceeded 0.98. Annual egg abun- dances peaked during 1985 and declined to a his- torical low in 1986 (Fig. 13). This low abundance was reflected in the annual entrainment estimates with the 1986 estimate among the smallest, even with the additional cooling water demands of Unit 3 (Table 25). Similar to tautog, cunner larval TABLE 25. Annual entrainment estimates and 95% confi- dence intervals for cunner eggs at MNPS. Entrainment Year estimate (xlO ) 95% CI 1979 1674.7 1341.6-1964.3 1980 2031.8 1654.5-2971.6 1981 1610.5 1335.3-2145.5 1982 2103.0 1693.5-2903.9 1983 2589.3 1763.3-3087.9 1984 2153.6 1563.9-2595.4 1985' 2216.2 1415.7-3083.5 1986 1812.0 1420.1-2603.1 Revised estimate due to error in previous calculations (NUSCO 1987b). entrainment estimates based on median densities could not be calculated because larvae were in- consistently found in samples collected during their period of occurrence. The seasonal 5-mean density of larvae in 1986 was one of the lowest at both EN and NB (Table 26). Abundances 60 50 40 30 20- 10 79 80 32 83 YEAR 85 Fig. 13. Cunner egg abundance estimale.s and 95% confidence intervals ba.sed on the a parameter of the Gompertz function for station EN. 284 TABLE 26. The seasonal 5-mean density (no. per 500 m ) of cunner larvae collected at stations EN and NB by year. vived to the larval stage. Further, survival in re- cent years was even less than usual and probably 1976 29.4 ± 14.1 TABLE 27. Ratio 0 if annual 5-mean densities of cunner 1977 58.1 ± 28.1 — and taulog larvae to eggs collected at station EN. 1979 13.6 ± 58.7 ± 4.8 19.6 94.7 ± 42.2 148.1 ± 86.5 Year Cunner Tautog 1980 1979 0.0023 0.0084 1981 77.4 ± 36.3 98.5 ± 72.0 1980 0.0071 0.0162 1982 31.8 ± 13.7 153.0 ± 74.0 1981 0.0151 0.0315 1983 49.9 ± 26.1 207.9 ±266.0 1982 0.0058 0.0196 1984 1985 3.7 ± 12.6 ± 2.6 8.9 37.6 ± 28.6 30.3 ± 11.5 1983 0.0070 0.0162 1986 3.0 ± 1.6 8.4 ± 4.8 1984 1985 1986 0.0006 0.0017 0.0010 0.0014 Data seasonally restricted to June through August. 0.0012 EN, suggesting that larval densities may be lower near the MNPS intaJkes compared to other areas. Larval abundance began declining in 1984 and has not returned to historic levels. However, a similarly large decrease in larval abundance, which occurred from 1977 to 1978, was followed by an increase to relatively high densities, in the early 1980s. Low larval abundance in 1984 may have been due to predation or other factors operating concurrently on the entire summer ichthyoplankton assemblage, because larval den- sities of tautog and anchovies were low that year aswell(NUSCO 1987b). A comparison of cunner and tautog larval abundance showed a remarkably similar pattern at both EN and NB (Tables 23 and 26). At EN, lowest densities occurred in 1978, 1984, and 1986, and the greatest densities occurred in 1981. At NB, the greatest abundances were in 1983 and lowest in 1986. However , these similarities between the two species were not ev- ident in their egg abundance estimates. Cunner egg survival was reported as low (about 5%) by Williams et al. (1973), based on the ex- amination of embryonic development in field- collected eggs; this survival estimate did not take into account the possible additional loss due to predation. An index of egg survival wat, estimated by the ratio of the annual 6-mean for larvae to eggs (see Table 5 for egg 5-mean) at EN for cunner, and also tautog for comparison (Table 27). These low ratios indicated that few eggs sur- produced fewer juvenile recruits from 1984 through 1986. Egg densities seem unrelated to the corresponding larval densities and no com- pensatory relationship was found. Whatever fac- tors affected survival of cunner larvae also affected tautog larvae; ranks of the larval to egg ratios were nearly the same for the two species and they were significantly correlated (Spearman rank cor- relation, r = 0.83, p<0.01). Cunner were taken only frequently enough at three (FN, JC, and NB) of the six trawl stations to use the 5-mean for describing abundances (Ta- ble 28). At aU three stations there was a decline TABLE 28. Seasonal 5-mean catch (CPUE) and and 95% confidence intervals of cunner caught by trawl at selected stations. Year IN JC NB 1976 22 ± 19 4 ± 2.0 1 ± 0.7 1977 30 ± 23 3 ± 1.0 1 ± 0.6 1978 6 ± 3.7 3 ± 1.4 0.7 ± 0.3 1979 29 :t 23 9 ± 5.0 2 ± 1.0 1980 23 ± 16 6 ± 2.0 3 ± 1.2 1981 12 ± 10 5 ± 2.2 3 ± 0.9 1982 5 ± 3,0 4 ± 2.0 2 ± 0.9 1983 3 ± 1.3 4 ± 2.0 1 ± 0.6 1984 2 ± 1.0 2 ± 1.0 0.4 ± 0.2 1985 1 ± 0.6 1 ± 0.5 0.4 ± 0.7 1986 0.1 ± 0.2 0.5 ± 0.4 0.08 ± 0.1 Season used for calculating 5-mean was May-August at IN, May-September at JC and April-November at NB. Pish Ecology Studies 285 in abundance, particularly at IN. Because trawl sampling effort was about the same each year, total catch was also used as an index of relative abundance for comparison among the six stations (Table 29). Decreases of juveniles and adults were TABLE 19. Total number of cunner caught by trawl for each station during each report period (June-May). Report period JC NR NB TT BR IN 1976-77 97 14 37 43 15 632 1977-78 78 4 39 34 54 666 1978-79 90 7 40 12 24 227 1979-80 232 11 55 25 54 1022 1980-81 191 7 89 42 15 596 1981-82 263 91 77 23 44 342 1982-83 209 58 77 24 36 207 1983-84 120 60 36 12 58 76 1984-85 73 16 23 20 48 68 1985-86 23 15 1 9 44 27 1986-87 28 38 16 5 51 9 Total 1404 321 490 249 443 3878 apparent at the four stations closest to MNPS (IN, NB, JC, and TT), but at BR numbers re- mained stable over the period and they fluctuated at NR. Most (57%) cunner were taken at IN, where numbers began to decrease in the 1980-81 report period and decreased further in 1983-84. Simultaneous decreases occurred at JC, TT, and NB, 2 years after the decline at IN began. Al- though causes for these declines may never be known with certainty, physical alterations of the habitat in the MNPS intake area near IN occurred just before the decreases at that station and could have accounted for the observed change. In March 1975, a bottom fish boom was installed at Unit 1 intake to reduce impingement. It was unsuccessful and was removed in the spring of 1981 just before the initial decline at IN. The Unit 3 cofferdam was constructed in March of 1976 and was removed in the summer of 1983 just before the second decline. Both of these structures provided ideal habitats for cunner. Un- fortunately, trawl monitoring was not done at IN prior to the installation of these structures and it could not be determined if these structures actually caused an increase in the cunner population around the MNPS intake. During the latest report period of 1986-87, the mean length of cunner caught by trawl decreased markedly at all stations, except at BR (Fig. 14), indicating a reduction in the abundance of older fish. To determine the age structure of the local population at the three trawl stations where they were most abundant (JC, NB, and IN), cunner were assigned to age-classes using an age-length key (Serchuk 1972). From 1976-77 through 1980-81, ages I, II, and III were dominant (Table 30). The low percentage of age 0 (young- of-the-year) during this time period indicated that the trawl may not have sampled smaller individ- uals as effectively as larger ones. In 1981-82, age 0 became the dominant age-class. The estimated number of age 0 fish was significantly correlated to the 5-mean density of larvae at EN during the same time period (Spearman rank correlation, r = 0.71, p = 0.015). Although sample size was small, over 80% of the cunner taken in 1986-87 at the three stations near MNPS were age 0. These findings showed that older cunner were less abundant near MNPS. This predominance of younger fish and low abundance during the 1986-87 report period probably accounted for the decease in eggs collected at EN during the same period. TABLE 30. Percentage of cunner by estimated age-class caught by trawl at JC, NB and IN combined during each report period (June-May). Percentage by age- class Report Total period 0 I II Ill IV V number 1976-77 4 23 29 26 13 5 698 1977-78 8 20 22 23 17 10 516 1978-79 6 20 25 25 13 11 357 1979-80 4 36 31 17 8 4 927 1980-81 11 37 33 16 7 6 779 1981-82 32 18 21 17 6 6 617 1982-83 24 18 21 20 8 9 477 1983-84 25 22 20 16 7 7 231 1984-85 21 27 12 16 9 14 164 1985-86 12 29 20 16 10 14 51 1986-87 83 7 0 2 4 4 53 Length-age key from Serchuck (1972) Data collected at SNPS were examined to de- termine if trends in cunner abundance were similar 286 275 250- 225 200 175 150 125 100 75 50 25 76 77 78 79 80 81 82 83 84 85 86 76 77 78 79 80 81 82 ^ YEAR 83 84 85 86 275 250 225- 200- 175 150 125 100 75 50 25 75 77 78 79 80 81 82 83 YEAR 275 250 225 200 1751 150 125 100 75- 50- 25 275 250 2251 200 175 150 125 100 75 501 25 76 77 78 79 80 81 82 83 84 85 YEW? 76 77 78 79 80 81 82 83 84 85 YEAR Fig. 14. The annual mean length and 95% confidence interval ofcunner taken by trawl at each station. Fish Ecology Studies 287 to the Millstone area (Table 31). Larval to egg ratios were also low indicating poor egg survival. The egg abundance index at SNPS in 1986 was the lowest compared to other years, but the esti- mated annual abundance fluctuated much more than at Millstone. I^arval abundance did not decline at SNPS in recent years as it has here, but similar to Millstone, some of the lowest catches by trawl at SNPS occurred in 1985 and 1986. TABLE 31. Annual abundance indices of cunner eggs and larvae, both expressed as an annual sum of means, and trawl catch as annual CPUE at Shoreham Nuclear Power Station. Sum Sum Trawl Year eggs larvae CPUE 1977 8119 1064 0.07 1978 4706 307 0.14 1979 14225 308 0.51 1980 3848 42 0.68 1981 3587 43 0.25 1982 2367 b 0.09 1983 9221 151 0.27 1984 9724 64 0.52 1985 2429 156 0.08 1986 1454 487 0.07 Sums for 1977-82 based on mean density per sam- pling trip (Geomet Tech. 1983) and for 1983-86 based on monthly mean density (EA Eng., Sci., and Tech. 1987). Abundance not reported (Geomet Tech. 1983) appar- ently due to low densities. In summary, the abundance of all life history stages of cunner at stations close to MNPS has recently declined. The decrease in impingement can, in part, he attributed to an overall decline in impingement at Unit 2 since the start-up of Unit 3, but also may be related to an apparent decrease in the cunner population near MNPS. Some of the decline of juveniles and adults was likely the result of several changes in physical habitat near IN trawl station. Concurrent with the decrease in trawl catch was a reduction in mean length. If changes in juvenile and adult abundance were related to entrainment losses, then juvenile re- cruitment would decrease arid cause the mean length to increase. The cause of the change in the cunner population is not known, and moni- toring of its abundance will continue to determine if the decrease was a result of natural long-term fluctuations or the operation of MNPS. Conclusions The life history stages of fishes collected in the fish ecology programs were examined to determine which species were most susceptible to potential impact due to the operation of MNPS, with par- ticular emphasis on the period of three-unit op- erations, lliere was a significant decrease in the total impingement at Unit 2 starting with the removal of the Unit 3 cofferdam in the summer of 1983 and a continuing decline through 1987, possibly due to a change in water circulation pat- terns with the start-up of Unit 3. Because the objectives of the impingement program were com- pleted. Unit 2 impingement monitoring was dis- continued on December 11, 1987. Eiglit taxa were selected for detailed examina- tion: American sand lance, anchovies, sticklebacks, Atlantic tomcod, sUversides, grubby, tautog, and cunner. There was no apparent change in the distribution or abundance of the silversides in Jordan Cove related to the increased thermal plume with three-unit operations. As ex- pected, increased cooling water demands of three- unit operations increased annual entrainment es- timates for most of the abundant ichthyoplankton taxa. The sand lance and cunner were the only taxa that showed a decline concurrent with three- unit operations. The decrease in sand lance larvae has been occurring throughout the 1980s and was attributed to a regional decline in adults. The cause for the cunner decline was not ascertained, but monitoring will continue to determine if the decrease was due to natural long-term fluctuations or the operation of MNPS. 288 Summary 1. The operation of MNPS could affect fish assemblages in the Millstone area in several ways. Juveniles and adults could be lost due to impingement on the intake screens. The mortality rates of early history stages could be increased by entraining eggs and larvae through the condenser cooling water system. The local distributions could be altered by the thermal plume. This report emphasizes the comparison of data collected during two- unit operations to those collected since the start-up of Unit 3. 2. Impingement monitoring at Unit 2 was dis- continued on December 11, 1987 because losses have been well-documented and all feasible mitigative measures have been inves- tigated. There was a significant decline in total impingement in recent years, which was attributed to physical changes near the Unit 2 intake and possible changes in water circu- lation patterns because of the operation of Unit 3. Losses due to impingement by MNPS were reduced with the installation of fish return sluiceways at Units 1 and 3. 3. Over 100 fish taxa have been collected in the monitoring programs since 1976. These pro- grams were demersal trawl, shore-zone seine, impingement, and ichthyoplankton. Eight taxa were selected for detailed examination due to their prevalence in entrainment or impingement collections or their abundance in the shore-zone area of Jordan Cove, an area which may be impacted by the thermal plume. 4. The American sand lance was primarily col- lected as larvae and was a dominant entrained taxon. A decline in larval abundance since the early 1980s was attributed to a regional decrease in adult stock size. 5. Several life history stages of anchovies were very abundant in some sampling programs. Adults were present in impingement collec- tions, juveniles were caught by trawl, and eggs and larvae were abundant in entraiimient samples. The numbers impinged have de- clined in recent years. Comparison of annual egg and larval abundance indices suggested compensatory mortality during the early life history stages, which could help mitigate losses due to entrainment. 6. Sticklebacks and Atlantic tomcod were pri- marily found in impingement collections. The impact of MNPS impingement on sticklebacks has been reduced due to high ( > 70%) survival of individuals returned by sluiceways at Units 1 and 3. There was a marked decrease in Atlantic tomcod numbers impinged at Unit 2 since the start-up of Unit 3. 7. Silversides dominated the shore-zone area of Jordan Cove and adults were abundant in winter trawl and impingement collections. There was a recent decline in the number impinged at Unit 2, but this was not evident in the number caught by trawl. There were no apparent changes in length-frequency dis- tribution or seasonal abundance in Jordan Cove related to the three-unit thermal plume. 8. Grubby larvae were present in ichthyoplank- ton collections and juveniles and adults were present in trawl and impingement collections. Larval abundance has declined in recent years to levels similar to the late 1970s. Numbers impinged at Unit 2 have decreased and those returned by the Units 1 and 3 sluiceways had high survival (>74%). Except for station NR, there has been no long-term changes in the mean length or abundance indices of adults collected by trawl. 9. The tautog is an important recreational fish in the Millstone area and the greatest potential impact of MNPS on it is through the en- trainment of eggs. Egg abundance, the best index of adult stock size, has increased in recent years, but larval densities declined to levels similar to the late 1970s. There was Fish Ecology Studies 289 an apparent poor egg to larval survival in all years examined, with the lowest during recent years. Because the tautog takes 2 to 4 years to reach maturity, the possible impact of en- trainment by three-unit operations on adult stock size will not be evident for several years. 10. The abundances of all life history stages of cunner collected near MNPS have recently declined. Decreased impingement was attrib- uted, in part, to the overall decline in total impingement at Unit 2. Decreases in the trawl catch were evident at stations closest to MNPS. Part of the decreasing trend in juvenile and adult abundance at station IN could have been caused by physical changes to the habitat, but reasons for the decline at other stations are not known. In 1986, the abundance of eggs and larvae were among the lowest found since 1976. There was an apparent low egg to larval survival for all years examined, and, similar to tautog, poor- est survival occurred in recent years. In ad- dition, there were similar fluctuations in the annual abundance of cunner and tautog lar- vae, suggesting that factors affecting the sur- vival of early life history stages were similar for both species. Further monitoring will continue to determine if the apparent decrease in the cunner population was a result of nat- ural long-term fluctuations or from the op- eration of MNPS. References Cited Azarovitz, T.R. 1981. A brief historical review of the Woods Hole laboratory trawl survey time scries. Can. Spec. Publ. Fish. Aquat. Sci. 58:62-67. Bannister, R.C.A., D. Harding, and S.J. Lockwood. 1974. Larval mortality and subsequent year-class strength in the plaice {Pleuronectes platessa L.). Pages 21-38 in J. M.S. Blaxter, ed. The early life history of fish. Springer- Verlag, New York. Baylifl", W.H., Jr. 1950. The life history of the silverside, Menidia menidia (Linnaeus). Md. Board Natur. Resour. Publ. 90:1-27. Bigelow, H.B.,andW.C. Schroeder. 1953. Fishes of the Gulf of Maine. U.S. Fish Wildl. Serv. BuU. 53:1-577. Blake, M.M., and E.M. Smith. 1984. A marine resouces management plan for the State of Connecticut. Connecticut Dept. Envir. Prot., Mar. Fish. 244 pp. Bothclho, V.M., and G.T. Donnelly. 1978. A statistical analysis of the performance of the Bourne plankton splitter, based on test obser- vations. NMFS unpub. ms. Briggs, P.T., and J.S. O'Conner. 1971. Com- parison of shore-zone fishes over natural vege- tated and sand-filled bottoms in Great South Bay. N.Y. Fish Game J. 18:15-41. Chenoweth, S.B. 1963. Spawning and fecundity of the tautog, Tautoga onitis (Linnaeus). M.S. Thesis, Univ. of Rhode Island, Narragansett, RI. 60 pp. Conover, D.O. 1979. Density, growth, produc- tion and fecundity of the Atlantic silverside, Menidia menidia (Linnaeus), in a central New England Estuary. M.S. Thesis, Univ. of Mas- sachusetts, Amherst, MA. 60 pp. , and S. A. Murawski. 1982. Offshore winter migration of the Atlantic silverside. Fish. Bull., U.S. 80:145-150. Consolidated Edison Co. of New York, Inc., and Power Authority of the State of New York (Con Ed and PASNY). 1977. Near-field efi"ects of once-through cooling system operation on Hudson River biota. Consolidated Edison Co., New York, NY. Section 10. 34 pp. Cooper, R.A. 1965. Life history of the tautog, Tautoga onitis (Linnaeus). Ph.D. Thesis, Univ. of Rhode Island, Narragansett, RI. 153 pp. 290 . 1966. Migration and population estima- tion of the tautog, Tautoga onitis (Linnaeus) from Rhode Island. Trans. Am. Fish. Soc. 95:239-247. Gushing, D.H. 1974. The possible density- dependence of larval mortality and adult mor- tality in fishes. Pages 103-111 wJ.H.S. Blaxter, ed. The early life history of fish. Springer- Verlag, New York. , and J.G.K. Harris. 1973. Stock and re- cruitment and the problem of density depen- dence. Rapp. P.-v. Cons. int. Explor. Mer 164:142-155. DeAngelis, D.L., S.W. Christensen, and A.G. Clark. 1977. Response of a fish population model to young-of-the-year mortality. Oak Ridge National Laboratory Publ. No. 1065. Dew, C.B. 1976. A contribution of the life history of the cunner, Tautogolabrus adspersus, in Fishers Island Sound, Connecticut. Chesapeake Sci. 14:101-113. Draper, N., and H. Smith. 1981. Applied regres- sion analysis. John Wiley and Sons, New York. 709 pp. EA Engineering, Science, and Technology, Inc. 1987. Preoperational aquatic ecology study Shoreham Nuclear Power Station, Unit 1. 1983-1986. (Draft) Prepared for Long Island Lighting Company, HicksviUe, NY. Fitzgerald, G..I. 1983. The reproduction ecology and behavior of three sympatric sticklebacks (Gasterosteidae) in a salt marsh. Biol. Behavior 8:67-79. , and E.G. Whoriskey. 1985. Effects of interspecific interactions upon male reproduc- tive success in two sympatric sticklebacks, Gasterosleus aculeatus and G. wheat landi. Biol. Behavior 9:112-126. Fritzsche, R.A. 1978. Development of fishes of the Mid- Atlantic Bight. An atlas of egg, larval and juvenile stages. Vol V. Chaetodontidae through Ophidiidae. Power Plant Project, Off. Biol. Serv., U.S. Fish Wildl. Serv., U.S. Dept. of the Interior, FWS/OBS-78/12. 340 pp. Geomet Technologies, Inc. 1983. Preoperational aquatic ecology study Shoreham Nuclear Power Station, Unit 1. 1982. Prepared for Long Island Lighting Company, Hicksville, NY. Gleason, T., and C. Recksiek. In preparation. Synopsis of biological data for the cunner Tautogolabrus adspersus (Walbaum). Univ. of Rhode Island, Contrib 2420 of the Rhode Is- land Experimental Station. Greeley, J. R. 1938. Fishes and habitat conditions of the shore zone based upon July and August seining investigations. Section II. Pages 72-91 in A biolo^cal survey of the saltwaters of Ix)ng Island, Pt. II. N.Y. Conserv. Dept. Green, J.M. 1975. Restricted movements and homing of the cuimer, Tautogolabrus adspersus (Walbaum) (Pisces; Labridae). Can. J. Zool. 53:1427-1431. , and M. Farwell. 1971. Winter habits of the cunner, Tautogolabrus adspersus (Walbaum), in Newfoundland. Can. J. Zool. 49:1497-1499. Grosslein, M.D. 1974. Bottom trawl survey methods of the Northeast Fisheries Center, Woods Hole, Massachusetts, U.S.A. Int. Comm. Northw. Atl. Fish. Res. Doc. No. 74/96. Ser. No. 3332 (mimeo). 27 pp. , and T.R. Azarovitz. 1982. Fish distribu- tion. Mesa New York Bight Atlas Monograph 15. New York Sea Grant Institute, Albany, NY. 182 pp. Hildebrand, S.F. 1943. A review of the American anchovies (Family Engraulidae). Bull. Bingham Oceanogr. Coll. 8:1-165. Fish Ecology Studies 291 Hoff, J.G., and J.R. Westman. 1966. The tem- perature tolerance of three species of marine fishes. J. Mar. Res. 24:131-140. Houde, E.D. 1977. Food concentration and stocking density effects on survival and growth of laboratory-reared larvae of bay anchovy Anchoa mitchilli, and lined sole Achirus lineatus. Mar. Biol. (Berl.) 43:333-.341. . 1978a. Critical food concentrations for larvae of three species of subtropical marine fishes. BuU. Mar. Sci. 28:395-411. ecosystem level effects related to intake of cool- ing water. Prepared for Electric Power Research Institute, Palo Alto, CA. 320 pp. . 1987. Brayton Point Station Unit No. 4. Angled screen intake biological evaluation program. Vol 1. Prepared for New England Power Company, Westborough, MA. Leak, J.C, and E.D. Houde. 1987. Cohort growth and survival of bay anchovy, Anchoa mitchilli, larvae in Biscayne Bay, Florida. Mar. Ecol. 37:109-122. . 1978b. Simulated food patches and sur- vival of larval bay anchovy, Anchoa mitchilli, and sea bream, Archosargus rhomboidalis. Fish. Bull., U.S., 76:483-487. Howe, A.B. 1971. Biological investigation of Atlantic tomcod, Microgadus tomcod, in the Weweantic River Estuary, Massachusetts, 1967. M.S. Thesis, Univ. of Massachusetts, Amherst, MA. 82 pp. Johansen, F. 1925. Natural history of the curiner. Contrib. Canadian Biol. 2:423-468. Johnson, M.S. 1975. Biochemical systematics of the atherinid genus Menidia. Copeia 1975:662-691. Klauda, R.J., R.E. Moos, and R.E. Schmidt. 1987. Life history of Atlantic tomcod, Microgadus tomcod, in the Hudson River Es- tuary, with emphasis on spatio-temporal distri- bution and movements. Pages 219-254 in C.L. Smith, ed. Fisheries research in the Hudson River. State University of New York Press, Albany, NY. 407 pp. Kuntz, A. 1914. The embryology and larval development of Bairdiella chrysura and Anchovia mitchilli. U.S. Bur. Fish., Bull. (1913)33:1-19. Lawler, Matusky and Skelly Engineers. 1980. Methodology for assessing population and Leim, A.H., and W.B. Scott. 1966. Fishes of the Atlantic coast of Canada. BuU. Fish. Res. Board Can. 155. 485 pp. Lund, W.A., and B.C. Marcy, Jr. 1975. Early development of the grubby, Myoxocephalus aenaeus (MitchiU). Biol. Bull. 149:373-383. Madenjian, C.P., D.J. Jude, and F.J. Tesar. 1986. Intervention analysis of power plant impact on fish populations. Can. J. Fish. Aquat. Sci. 43:819-829. May, R.C. 1974. Larval mortality in marine fishes and the critical period concept. Pages 3-20 in J.H.S. Blaxter, ed. The early life history of fish. Springer- Verlag, New York. McHugh, J. L. 1972. Marine fisheries of New York State. Fish. Bull., U.S. 70:585-610. . 1977. Fisheries and fishery resources of New York Bight. NOAA Tech. Rep. NMFS Circ. 401. 51 pp. Meyer, T.L., R.A, Cooper, and R.W. Langton. 1979. Relative abundance, behavior, and food habits of the American sand lance, Ammodytes americanus, from the Gulf of Maine. Fish. Bull., U.S. 77:243-253. Miller, D. 1958. A key to some of the more common larval fishes of the Gulf of Maine. Woods Hole Lab. M.S. Rep. 58-1. 56 pp. 292 Northeast Utilities Service Company (NUSCO). 1982. Monitoring the Marine Environment of Long Island Sound at Millstone Nuclear Power Station, Waterford, Coimecticut. Annual re- port, 1981. . 1987a. Justification for discontinuing impingement monitoring at Millstone Unit 2. Submitted to CT DEP. 1 1 pp. . 1987b. Monitoring the marine environ- ment of Long Island Sound at Millstone Nu- clear Power Station, Waterford, Connecticut. Summary of studies prior to Unit 3 operation. . 1988. The effectiveness of the Millstone Unit 3 fish return system. Submitted to the CT DEP. 21 pp. Olla, B.L., A.J. Bejda, and A.D. Martin. 1974. Daily activity, movements, feeding, and seasonal occurrence in the tautog, Tautoga onitis. Fish. Bull., U.S. 72:27-35. . 1975. Activity, movements, and feeding behavior of the cuimer, Tautogolabrus adspersus, and comparison of food habits with young tautog, Tautoga onitis, of Long Island, New York. Fish. BuU., U.S. 73:895-900. . 1979. Seasonal dispersal and habitat se- lection of cunner, Tautogolabrus adspersus, and young tautog, Tautoga onitis, of Long Island, New York. Fish. BuU., U.S. 77:255-262. Oviatt, C.A., and S.W. Nixon. 1973. The demersal fish of Narragansett Bay: an analysis of community structure, distribution and abun- dance. Est. Coast. Mar. Sci. 1:361-378. Pearcy, W.G., and S.W. Richards. 1962. Distri- bution and ecology of fishes of the Mystic River estuary, Connecticut. Ecology 43:248-259. Pennington, M. 1983. Efficient estimators of abundance, for fish plankton surveys. Biometrics 39:281-286. . 1986. Some statistical techniques for es- timating abundance indices from trawl surveys. Fish. BuU., U.S. 84:519-525. Perlmutter, A. 1963. Observations on fishes of the genus Gasterosteus in the waters of Long Island, New York. Copeia 1963:168-173. Pottle, R.A., and J.M. Green. 1979. Field ob- servations on the reproductive behavior of the cunner, Tautogolabrus adspersus (Walbaum), in Newfoundland. Can. J. Zool. 57:247-256. Richards, S.W. 1959. Pelagic fish eggs and larvae of Long Island Sound. BuU. Bingham Oceanogr. CoU. 17:95-124. . 1963. The demersal fish population of Long Island Sound. I. Species composition and relative abundance in two locjdities, 1956-1957. BuU. Bingham Oceanogr. CoU. 18:5-31. . 1982. Aspects of the biology of Ammodytes americanus from the St. I^wrence River to Chesapeake Bay, 1972-75, including a comparison of the Long Island Sound postlarvae with Ammodytes dubius. J. Northw. Atl. Fish. Sci. 3:93-104. Ricker W.E. 1975. Competition and interpreta- tion of biological statistics of fish populations. BuU. Fish. Res. Board Can. 191:1-382. SaUa, S.B., and S.D. Pratt. 1973. Mid Atlantic Bight fisheries Pages 6.1-6.125 in Coastal and offshore environmental inventory. Univ. of Rhode Island Mar. Pub. Ser. Sampson, R. 1981. Connecticut marine recrea- tional fisheries survey 1979-1980. Conn. Dept. Envir. Prot., Mar. Fish. 49 pp. SAS Institute Inc. 1985. SAS user's guide: sta- tistics. Version 5 edition. SAS Institute Inc. Cary, NC. 956 pp. Serchuk, F.M. 1972. The ecology of the cunner, Tautogolabrus adspersuf (Walbaum) (Pisces: Fish Ecology Studies 293 Labridae), in the Weweantic River Estuary, Wareham, Massachusetts. M.S. Thesis, Univ. of Massachusetts, Amherst, MA. 1 1 1 pp. Snedecor, G.W., and W.C. Cochran. 1967. Sta- tistical methods. Iowa State Univ. Press, Ames, lA. 593 pp. Stevenson, R.A. 1958. The biology of the an- chovies Anchoa mitchilli and Anchoa hepsetus in Delaware Bay. M.S. Thesis. Univ. of Del- aware, Newark, DE. 56 pp. Tracy, H.C. 1910. Annotated list of the fishes known to inhabit the waters of Rhode Island. R.I. Ann. Rep. Comm. Inland Fish. 40:35-176. Vouglitois, J..I., K.W. Able, R.J. Kurtz, and K.A. Tighe. 1987. Life history and population dynamics of the bay anchovy in New Jersey. Trans. Am. Fish. Soc. 116:141-153. Warfel, lI.E., and D. Merriman. 1944. Studies on the marine resources of southern New Eng- land. I. An analysis of the fish population of the shore zone. Bull. Bingham Oceanogr. Coll. 9:1-53. Westin, D.T., K.J. Abemethy, L.E. Meller, and B.A. Rogers. 1979. Some aspects of biology of the American sand lance, Ammodytes americanus. Trans. Am. Fish. Soc. 108:328-331. Wheatland, S.B. 1956. Oceanography of Lx)ng Island Sound. 1952-1954. II. Pelagic fish eggs and larvae. Bull. Bingham Oceanogr. Coll. 15:234-314. Williams, G.C. 1967. Identification and seasonal size changes of eggs of the labrid fishes, Tautogolabrus adspersus and Tautoga onitis, of Lx5ng Island Sound. Copeia 1967:452-453. , D.C. Williams, and R.J. Miller. 1973. Mortality rates of planktonic eggs of the curmer, Tautogolahnis adspersus (Walbaum), in Long Island Sound. Pages 181-195 in A. Pacheco, ed. Proceeding of a workshop on egg, larval and juvenile stages of fish in Atlantic Coast estuaries. National Marine Fisheries Service, Middle Atlantic Coastal Fisheries Center. Tech. Publ. No. 1. Worgan, J.P., and G.J. Fitzgerald. 1981. Habitat segregation in a salt marsh among adult sticklebacks. Env. Biol. Fish. 6:105-109. 294 Appendices to Fish Ecology Studies Fish Ecology Studies 295 APPRNDIX I. l.ist of fishes collected in the Fish Ecology sampling programs. Scientific name Common name Trawl Impingement Icthyopjankton A cipenxer oxyrhynchiis A lectis ciliaris Alosa aestivalis Alosa mediocris Alosa psmidoharengus Alosa sapidissima Alosa spp. Alulcnis schoepfi Ammodyles americanus Anchna hepsetus Anchoa mitchilli Anguilla rostrata Apeltes quadracus Aulostnmus macula tus Bairdiella chrysoura Bothidae Brevonrlia lyrannus Brosme brosme Caranx crysns Caranx hippos Centropristis striata Chaetodon ocellatus Chilomyctems schoepfi Clupeidae Clupea harengus Conger oceanicus Cyclopterus lumpus Cynoscion regalis Cyprinodon variegatus Dactylopterus volitans Dasyatis centroura Dncapterus macarellus F.nchclyopus cimbrius Etropus microstomus Eucinostomus lefroyi Fistularia tabacaria Fundulus diaphanus Fundulus heteroclitus Fundulus luciae Fundulus majalis Atlantic sturgeon African pompano blueback herring hickory shad alewife American shad alosid orange filefish American sand lance striped anchovy bay anchovy American eel fourspine stickleback trumpelfish silver perch left-eye flounder Atlantic menhaden cusk blue runner crcvalle jack black sea bass spotfin butterflyfish striped burrfish herrings Atlantic herring conger eel lumpfish weakfish sheepshead minnow fiying gurnard roughtail sling-ray mackerel scad fourbeard rockling smallmouth flounder mottled mojarra bluespotted cornctfish banded killifish mummichog spotfin killifish striped killifish 296 APPENDIX 1 continued. Scientific name Common name Seine Impingement Icthyoplankton Gadidae Gadus morhua Gasterosteus aculeatus Gasterosteus wheatlandi Gobiidae Gobiosoma ginsburgi Hemitripierus americanus Hippocampus erectus Labridae Lactophrys spp. LMostomus xanthunis Liparis spp. Lophius americanus Lucania parva lAimpenus lumpretaeformis Macrozoarces americanus Melanogrammus aeglefinus Menticirrhus saxalilis Menidia beryllina Menidia menidia Merluccius bilinearis Microgadus tomcod Monacanthus hispidus Monocanthus spp. Morone americana Morone saxatilis Mugil cephalus Mugil aircma Mullus auratus Muslelis canis Myliobatis freminvillei Myoxocephalus aenaeus Myoxoceph alus oclodecemspinosus Myoxocephalus spp. Ophidjidae Ophidian marginatum Ophidian welshi Opsanus lau Osmerus mordax codfishes Atlantic cod threespine stickleback blackspotted stickleback gobies seaboard goby sea raven lined seahorse wrasses boxfish spot seasnail goosefish rainwater killifish snakehlenny ocean pout haddock northern kingfish inland silverside Atlantic silverside silver hake Atlantic tomcod planehead filefish filefish white perch striped bass striped mullet white mullet red goatfish smooth dogfish bullnose ray grubby longhorn sculpin sculpin cusk-eels striped cusk-eel crested cusk-eel oyster toadfish rainbow smelt Fish Ecology Studies 297 APPENDIX I continued. Scientific name Common name Impingement Iclhyoplankton Paralichlhys dentatus Paralichthys oblongus Peprilus triacanthus Petromyon marimis Pholis gunneilus Pollachius virens Pomatomus saltatrix Priacanlhus arenaius Priacanthus cruenlatus Prlsligenys alta Prionotiis carolinus Prionotiis evolans Pseudopliiuronectes ame.ricanus Pungllius pungiiius Raja eglanteria Raja erinacea Raja ocellata Salmo trulta Sciaenidae Scophthalmus aquonis Scomber scombrus Scyliorhinus retifer Selar crumenopthalmus Selene, setapinnis Selene vomer Seriola zonata Synodus foelena Sphyraena borealis Sphoeroides maculatiis Squabn acanthias Stenotomus chrysops Slrongylura marina Syngnathus fuscus Tautogolabriis adspersus Tautoga onitis Trachinotus fakatus Trachurus lathami TTrachirxocephalus myops summer flounder fourspot flounder butterfish sea lamprey rock gunnell pollock bliiensh bigeye glasseye snapper short bigeye northern scarobin striped searobin winter flounder ninespine stickleback clearnose skate little skate winter skate brown trout drums windowpane Atlantic mackerel chain dogfish bigeye scad Atlantic moonfish lookdown banded rudderfish inshore lizardfish northern sennet northern puffer spiny dogfish soup Atlantic needlefish northern pipefish cunner tautog permit rough scad snakefish 298 APPENDIX I continued. Scienlific name Common name Trawl Seine Impingement Ictliyoplankton Tr'mectes maculatus Ulvaria subbifurcata Upeneus parvus Urophycis chuss Urophycis tenuis Urophycis spp. hogchoker radiated shanny dwarf goatfish red hake while hake hake Fish Ecology Studies 299 APPENDIX II. Total trawl catch offish taxa and number of samples collected by report period (June-May). Taxon 16-11 77-78 78-79 79-80 80-81 81-82 82-83 83-84 84-85 85-86 86-87 Total Number of samples 468 468 468 468 468 468 473 477 468 468 468 5162 Pseudopleuionectes americanua 7415 6045 7236 11442 13296 10749 19201 12560 13260 9849 9321 120374 Stenotomus chrysops 1918 4040 2556 4094 3844 3403 4896 5268 4206 2640 5205 42070 Anchoa spp. 979 580 2226 16 109 578 38 109 157 10003 8038 22833 Scophlhalmus aquosus 1480 1296 875 1508 2016 1518 3517 2475 2199 2483 1655 21022 Raja spp. 661 579 362 402 954 696 2797 2493 1583 3801 2207 16535 Menidia spp. 2152 1647 1463 1340 882 501 518 583 322 519 3438 13365 Gadidae 112 326 230 211 3296 1424 476 481 562 630 168 7916 Tautogolabrus adspersus 8.-?8 875 400 1399 940 840 611 362 248 119 147 6779 Myoxocp.phalus aenaeus 266 636 297 342 632 870 996 672 477 341 727 6256 Prionotus spp. 338 322 138 313 405 661 1059 422 371 395 436 4860 Paralichlhys dentatus 286 141 92 75 122 240 250 269 1937 28! 653 4346 Merlucclus bilinearis 425 163 69 134 558 220 382 147 100 175 197 2570 Urophycis spp. 99 87 103 69 163 313 615 286 251 272 286 2544 Ga.slerosteus aculeatus 30 12 47 77 206 103 63 218 1102 116 354 2328 Tanloga onitis 229 283 263 270 146 228 239 140 119 134 215 2266 Hemitripterus americanus 34 48 39 148 278 410 557 377 125 41 45 2102 Pholis gunnellus 85 106 99 65 251 273 302 145 127 151 186 1790 Syngnathus fuscus 43 54 49 88 151 264 232 202 254 196 207 1740 Osmenis mordax 111 286 90 5 123 63 89 26 227 391 257 1668 Peprilus triacantlms 37 44 407 174 44 69 182 244 19 135 132 1487 Apeltes quadracus 10 6 24 27 194 765 76 11 112 130 107 1462 Etropus micros tomus 43 7 0 3 31 91 94 56 85 218 640 1268 Centropmtis striata 33 9 3 4 10 63 23 38 30 80 412 705 Myoxocephalus 11 10 97 40 30 145 172 51 20 13 12 601 octodecemspinosus Alosa pseudoharengus 11 272 13 17 4 15 5 26 4 16 208 591 Parallchthys oblongus 31 7 21 11 51 32 138 34 81 66 72 544 Ammodytes americanus 5 59 128 36 117 14 19 11 19 6 11 425 Opsamis tau 98 21 7 18 31 35 25 23 24 32 56 370 AnguiUa roslrata 19 16 8 5 10 37 29 24 22 34 28 232 Cyctoptems tumpus 19 1! 28 58 11 0 14 1 29 1 1 173 liparis spp. 9 27 10 10 18 33 15 16 11 3 18 170 Cynosciorx regalis 9 21 4 2 2 45 7 0 1 5 36 132 Alosa sapidissima 33 6 1 5 40 12 0 29 0 0 1 127 Clupeidae 2 1 0 0 0 n 0 0 0 110 0 113 Clupea harengus 1 9 13 0 0 1 0 2 9 63 10 108 Sphoeroides maculatus 16 10 1 0 9 14 16 15 7 7 3 98 300 APPENDIX II continued. Taxon lb-11 77-78 78-79 79-80 80-81 81-82 82-83 83-84 84-85 85-86 86-87 Total Mustelis canis 2 5 45 11 1 5 4 6 0 2 2 83 Brevoortia tyrannus 1 14 11 1 I 1 0 1 0 34 10 74 Alosa aestivalis .? 11 8 12 4 1 1 17 5 2 4 68 Monacanthus hispidus 3 6 8 4 0 n 8 1 8 9 2 49 Limanda ferniginea 7 5 5 2 3 15 6 0 4 0 0 47 Morone americana 8 17 3 5 8 2 1 0 0 0 0 44 ■ Macrozoarces americanus 5 7 9 2 2 2 2 2 3 1 0 35 Hippocampus spp. 0 0 0 0 0 0 0 1 4 7 20 32 Gobiidae i 0 0 0 4 0 0 3 9 7 2 28 Fistiilaria tabacaria 2 3 0 0 3 0 1 0 8 1 2 20 L^ioslomus xanthurus 5 6 0 0 0 0 2 0 0 3 1 17 Pungitius pungitius 0 0 0 0 1 2 0 0 5 1 5 14 Gasterosteidae 0 0 0 13 0 0 0 0 0 0 0 13 Pomatomus saltatrix 1 1 0 2 1 2 3 3 0 0 0 13 Alutems schoepfi 0 2 2 1 1 0 0 1 1 2 2 12 Dactylopterus volitans 3 0 0 0 0 1 3 I 0 1 3 12 Fundulus spp. 0 0 0 0 0 5 2 0 0 2 1 10 Mentidnhus saxatilis 0 1 0 1 0 3 1 0 0 0 4 10 Synodus foetens 0 1 4 0 0 3 1 0 0 0 0 9 Ophidian marginatum 0 0 0 0 0 0 0 0 1 2 4 7 Priacanlhus cruentatus 0 0 0 0 0 1 0 2 3 I 0 7 Trinectes maculatus 3 1 0 0 0 0 0 0 0 1 2 7 Gasterosteus wheatlandi 0 0 0 0 0 1 1 1 0 1 2 6 Lophius americanus 2 0 0 0 I 0 1 1 0 0 I 6 Morone saxatilis 0 0 2 1 0 1 1 0 0 1 0 6 Ulvaria subbifurcata 0 2 0 0 1 1 0 0 0 1 1 6 Caranx crysos 0 0 0 0 1 0 1 0 1 2 0 5 Conger oceanicus 1 0 0 0 1 0 0 0 2 0 1 5 Pristigenys alta 0 0 0 0 0 1 0 0 2 1 1 5 Mullus auratus 0 0 1 0 0 0 2 0 0 0 1 4 Selene vomer 1 2 0 n 0 0 0 0 0 0 1 4 Sphyraena borealis 0 0 0 0 0 0 0 I 1 2 0 4 Trachurus lathami 0 0 0 4 0 n 0 0 0 0 0 4 Chaetodon ocellatus 0 0 0 0 1 0 0 1 0 0 1 3 Lactophrys spp. 0 0 0 0 0 n 0 0 3 0 0 3 Mugil cephalus 0 0 0 0 0 0 1 0 0 2 0 3 Priacanthus arenatus 0 0 0 0 0 n 0 0 2 1 0 3 Alosa mediocris 1 0 0 0 1 0 0 0 0 0 0 2 Fish Ecology Studies 301 APPENDIX 11 continued. Taxon 16-11 77-78 78-79 79-80 80-81 81-82 82-83 83-84 84-85 85-86 86-87 Total Caranx hippos 0 0 0 0 0 0 1 0 0 1 0 2 Decapterus macarellus 0 0 0 0 0 0 0 0 2 0 0 2 Enchelyopus cimbrius 0 0 0 0 0 I 0 0 0 0 1 2 Scomber scombrus 0 1 0 1 0 0 0 0 0 0 0 2 Squalua acanthias 0 0 0 0 0 0 1 0 1 0 0 2 Aclpenser oxyrhynchus 0 0 0 I 0 0 0 0 0 0 0 Aulostomus maculalus 1 0 0 0 0 0 0 0 0 0 0 Bairdiella chrysoura 0 0 0 0 0 0 0 1 0 0 0 Bothus oceltatus 0 0 0 0 0 0 1 0 0 0 0 Da sy a lis centroura 0 0 0 0 0 0 0 0 1 0 0 Melanogrammus aeglefmus 0 0 0 0 0 0 0 1 0 0 0 Monocanthus spp. 0 0 0 0 0 0 0 0 n 0 1 Myliobatis freminvillei 0 0 0 0 0 0 1 0 0 0 0 Myoxocephalus spp. 0 0 0 0 0 0 0 0 0 0 1 Ophidiidae 0 0 0 0 0 0 0 0 0 0 1 Salmo trutta 0 0 0 0 1 n 0 0 0 0 0 Scyliorhinus retifer 1 0 0 0 0 0 0 0 0 0 0 Selar crumenopthalmus 0 0 0 0 0 0 0 0 0 1 0 Selene selapinnis 0 0 0 0 0 0 0 0 1 0 0 Trachinocephalus myops 0 0 0 0 0 0 0 0 1 0 0 lyachinotus falcatus 0 0 0 0 0 0 0 0 1 0 0 Upeneus parvus 0 0 0 0 0 0 0 0 0 1 0 Total 17941 18147 17497 22469 29010 2477.-? 37699 27860 28169 33546 35566 292677 302 APPENDIX III. Total trawl catch of fish and the number of samples collected by station. Taxon JC NR NB TT BR IN Total Number of samples 861 861 861 860 858 861 5162 Pseudopleuronectes americanus 11591 38093 17033 19956 14873 18828 Stenotomus chrysops 3015 239 17398 6543 3739 11136 Anchoa spp. 1004 249 17914 295 16 3355 Scophthalmus aquosus 1199 1742 1888 2828 10931 2434 Raja spp. 191 9 2734 4433 6643 1919 Menidia spp. 312\ 3950 1438 656 184 3416 Gadidae Mil 719 2632 998 242 1548 Tautogolabnis adspersus 1404 321 490 249 443 3872 Myoxocephalus aenaeus 919 2614 354 394 666 1309 Prionotux spp. 82 444 374 964 2439 557 Paralkhthys dentatus 649 830 529 1642 167 529 Merluccius bilinearis 142 3 376 316 1224 509 Urophycis spp. 306 28 247 237 1377 349 Gasterosteus aculeatus 1599 700 9 6 6 8 Tautoga onitis 493 533 253 177 242 568 Hemitripterus americanus 442 82 401 294 484 399 Phoiis gunnetlus 958 208 217 109 19 279 Syngnathus fuscus 518 827 112 65 74 144 Osmems mordax 1082 237 111 71 59 108 Peprilus triacanlhus 24 3 202 409 790 59 Apeltes quadracus 133 1324 1 1 1 2 Etropus microstomus 94 9 282 127 558 198 Centroprlstis striata 66 147 33 30 26 403 Myoxocephalus octodecemspinosus 3 0 20 52 511 15 Alosa pseudoharengus 7 63 19 12 250 240 Paralichthys oblongus 0 2 59 6 453 24 Ammodytes americanus 19 94 5 29 269 9 Opsanus tau 6 353 0 0 0 11 Anguilla rostra la 35 173 0 16 2 6 Cyclopterus lumpus 107 4 11 6 2 43 Liparis spp. 19 1 34 28 65 23 Cynoscion regalis 20 0 24 10 59 19 Alosa sapidissima 8 17 51 9 20 22 Clupeidae 0 1 0 1 0 111 Ctupea harengus 63 4 13 9 15 4 Sphoeroides maculatus 11 57 9 2 12 7 120374 42070 22833 21022 16535 13365 7916 6779 6256 4860 4346 2570 2544 2328 2266 2102 1790 1740 1668 1487 1462 1268 705 601 591 544 425 370 232 173 170 132 127 113 108 Fish Ecology Studies 303 APPENDIX III continued. Taxon Mustelis cams Brevoorlia tyraimus Alosa aestivalis Monacanthus hispidus Limanda ferruginea Morone americana Macrozoarces americanus Hippocampus spp. Gobiidae Fistularia tabacaria Leiostomus xanthurus Pungitius pungitius Gasterosteidae Pomatomus saltatrix Aluterus schoepfi Dactyloptcrus volitans Fundulus spp. Menticirrhus saxalilis Synodus foetens Ophidian marginatum Priacanthus cruentalus Trinectes macutalus Gasterosteus wheatlandi Lophius americanus Morone saxatilis Ulvaria subbifurcata Caranx crysos Conger oceanicus Pristigenys alta Mullus auratus Selene vomer Sphyraena borealis Trachurus lathami Chaetodon ocellalus L/ictophrys spp. Mugil cephalus Priacanthus arenatus Alosa mediocris 4 1 39 3 32 4 0 58 12 1 0 3 1 20 14 7 14 12 16 1 7 .S 11 9 0 0 0 4 43 0 6 It 4 1 5 17 0 0 n 1 33 1 12 13 1 2 1 3 2 26 0 0 0 0 16 I 0 0 0 3 2 0 8 0 4 3 10 3 0 0 0 1 2 11 2 3 2 0 5 1 6 n 0 2 1 i 1 8 0 n n 3 1 9 0 0 0 0 0 2 3 3 0 2 0 3 0 2 4 0 0 0 1 2 4 0 1 0 1 2 0 3 5 1 0 0 0 1 6 0 0 1 1 4 0 0 6 0 n 0 0 3 0 0 1 1 I 0 0 2 0 1 2 1 1 1 n 2 0 2 0 0 1 1 1 1 0 0 0 0 3 1 0 2 0 0 1 4 0 0 n 0 0 1 0 3 0 n 0 2 1 0 0 0 0 2 1 0 n 0 0 1 I 1 n 0 0 0 1 0 0 0 2 I 0 0 0 1 0 304 APPENDIX III continued, Taxon Caranx hippos Decapterus macarellus Enchelyopus cimbrlus Scomber scombrus Squalus acanthias Acipenser oxyrhynchus Aulostomus maculatus Bairdiella chrysoura Bothus ocellatus Dasyatis cenlroura Melanogrammus aeglejinus Monocanthus spp. Myliobatis freminvillei Myoxocephalus spp. Ophidiidae Salmo trutta Scyliorhinus retifer Selar crumenopthalmus Selene setapinnls Trachinocephalus myops Trachinotus falcatus Upeneus parvus Total 0 0 0 0 0 2 1 0 1 0 0 0 0 0 0 0 2 0 0 0 1 n 0 1 0 0 0 0 2 0 0 0 1 0 0 0 1 0 0 0 0 0 0 0 0 0 1 0 0 1 0 0 0 0 t 0 0 0 0 0 1 0 0 0 0 0 1 0 0 0 0 0 0 0 1 0 0 0 0 1 0 0 0 0 0 0 0 n 1 0 0 1 0 0 0 0 0 1 0 0 0 0 0 0 0 1 0 0 0 0 1 0 0 0 1 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 32432 54266 65380 41019 47034 52546 Fish Ecology Studies 305 APPENDIX IV. Total seine calch of fish taxa and number of samples collected by report period (June-May). Taxon 16-11 77-78 78-79 79-80 80-81 81-82 82-83 83-84 84-85 85-86 86-87 Total Number of samples 66 72 72 72 72 72 98 ilo~ 174 156 156 1130 Menidia spp. 40619 18194 1335 1062 7996 3186 5413 9807 1538 1375 5441 95966 Fundulus spp. 1695 1199 815 659 952 613 915 1081 1463 906 111 10409 Applies quadracus 464 603 258 266 49 94 89 1827 167 106 297 4220 Cyprlnodo. i variegatus 48 673 39 30 10 352 146 50 29 28 2 1407 Ammodytes americanus 6 -■520 16 51 10 318 82 30 21 0 7 1061 Pungilius pungitius 5 1 28 2 5 2 10 32! 8 11 8 401 Gasterosteu.i aculeatus 9 154 27 5 3 2 5 53 6 6 19 289 Syngnathus fuscus 9 3 9 108 6 8 21 12 35 30 33 274 Pomatomiis saltatrix ! 0 1 6 0 2 135 4 19 35 12 215 MugU cephalus 0 4 3 23 41 I 4 4 1 0 38 119 Pseudoplei-ironectes americanus 4 6 4 1 6 5 2 3 17 40 18 106 Gadidae 2 0 9 2 20 16 11 8 1! n 8 98 A losa pseudoharengus 0 0 0 0 0 0 0 1 93 0 0 94 Gasterosteus wheatlandi 8 6 6 19 12 9 60 Brevoortia tyrannus 0 0 17 0 4 0 7 1 0 8 6 43 Anguilla rostra ta 10 5 12 3 2 0 1 1 0 0 3 37 Clupea harengus 0 0 0 0 0 0 2 0 0 0 30 32 Myoxocephalus aenaeus 3 2 1 2 0 0 3 1 3 3 3 21 Anchoa spp. 0 0 0 0 2 0 7 2 1 0 0 12 Gasterosteus spp. 12 12 Mugil curema 0 0 0 0 0 0 0 1 9 0 0 10 Alosa aestivalis 2 6 0 0 0 0 0 0 0 0 0 8 Sphoeroides maculattis 0 0 0 1 0 0 I 0 0 3 3 8 Lucania parva 1 2 0 0 0 0 0 2 0 1 0 6 Tautogolahnis adspersus 0 0 2 0 0 0 3 0 1 0 0 6 Tautoga onitis 0 0 0 0 0 0 4 0 0 0 0 4 jyachinotiis falcalus 0 0 1 0 3 0 0 0 0 0 0 4 Caranx hippos 0 0 1 0 0 1 0 0 0 1 0 3 A losa sapidissima 1 0 0 0 0 0 0 0 0 1 0 2 Mentidrrhus saxatilis 1 0 1 0 0 0 0 0 0 0 0 2 Osmerus mordax 0 0 0 0 0 0 0 0 0 2 0 2 Peprilus Iriacanlhus 0 0 0 0 0 0 1 0 1 0 0 2 Pholis gwmellus 0 0 0 0 0 0 0 0 0 1 1 2 Strongylura marina 0 0 0 0 0 1 1 0 0 0 0 2 Clupeidae 1 0 0 0 0 0 0 0 0 0 0 Cynoscion regalis 0 0 0 0 0 0 0 0 1 0 0 Prionotus spp. 0 0 0 0 0 0 0 0 0 1 0 Scopbthalnms aquosus 0 0 0 0 0 0 0 0 0 1 0 Urophyds spp. 0 0 0 0 0 0 0 0 1 0 0 Total 42881 21372 2579 2221 9109 4609 6869 3215 3444 2582 6061 114942 306 APPENDIX V. Total seine catch of fish taxa and number of samples collected by station. Taxon JC WP GN Total Number of samples 345 389 396 1130 Menidia spp. Fundulus spp. Apeltes quadracus Cyprinodon varlegatus Ammodytes americanus Pungitius pungitius Gasterosteus aculeatus Syngnathus fuscus Pomatomus saltatrix Mugil cephalus Pseudopleuronectes americanus Gadidae Alosa pseudoharengus Gasterosteus wheatlandi Brevoorlia tyrannus Anguilla rostra ta Clupea hare.ngus Myoxocephalus aenaeus Anchoa spp. Gasterosteus spp. Mugil curema Alosa aestivalis Sphoeroides macutatus Lucania parva Tautogolabrus adspersus Tautoga onitis llachinotus falcatus Caranx hippos Alosa sapidissima Menticirrhus saxatilis Osmerus mordax Peprilus triaranthus Pholis gunnellus Strongylura marina Clupeidae Cynoscion regalis Prionotus spp. Scophthalmus aquosus Urophycis spp. lotal 75911 11439 8616 7807 1513 1089 4199 6 15 626 758 23 2 198 861 330 67 4 241 21 27 48 39 187 141 12 62 55 40 24 25 7 74 63 29 6 5 89 0 25 13 22 2 6 35 31 2 4 32 0 0 6 8 7 11 1 1 0 11 10 0 0 1 5 2 0 1 7 4 0 2 5 I 0 4 0 0 2 2 0 2 0 1 0 1 0 0 2 1 0 0 2 0 1 I 0 0 2 2 0 0 0 1 0 1 0 0 0 1 0 0 0 1 0 1 0 89592 14262 11088 95966 10409 4220 1407 1061 401 289 274 215 119 106 98 94 60 43 37 32 21 12 12 10 Fish Ecology Studies .•^07 Contents The Usage and Estimation of DELTA Means 311 Statement of the Problem 311 Ihc Delta I>istribution 311 Comparison of the 5-mean to other statistics 313 Application to NUEL's Monitoring Data 319 References Cited 319 The Usage and Estimation of DELTA Means statement of the Problem The average number of marine organisms caught per tow in trawl surveys, or observed in a sample in other monitoring work involving dif- ferent sampling schemes, is often used as an index of a species abundance. When large areas are sampled and the target species occupies only a part of the total area, the occurrence of samples with no organisms (i.e., zero observations) is un- avoidable. Even in situations where a species is known to be present in the entire area, the oc- currence of zero observations in varying propor- tions is still common. The frequency of zero data is particularly high for mobile organisms such as fish, and for plankton, which generally exhibit a high degree of spatial variability or "patchiness". The presence of zero observations in monitor- ing data complicates the data analyses on two accounts. First, zero data are diRicult to interpret because they may arise from natural patchiness, low population density, undetected sampling gear problems, and other reasons singly or combined; and second, the presence of zero observations in- creases both the coefficient of variability (i.e., the variance-to-mean ratio) and the skewness of the data. Because both higli variability and high skewness contribute to non-normality in data, es- timation methods based on normal theory do not apply. Since logaritlunic transformations to cor- rect skewness are not effective when many zero observations are present, the usual approach is to use order statistics such as the sample median and nonparametric variance estimators of unknown power. In extreme cases where over 50% of the data are zeros, the median cannot be used because it would always be zero regardless of obvious differences among samples. This study addresses the estimation problems described above and suggests the use of the ?>-mean as a more desirable statistic for describing the relative abundance of marine organisms when large numbers of zero observations are present in the data. The performance of the ?>-mean relative to three other statistics commonly used to estimate population abundance is investigated through nu- merical simulation. The results of this simulation also serve to illustrate how the four statistics are affected by the presence of zeros in lognormally distributed data. The Delta Distribution The delta distribution first introduced by Aitchison (1955) and later described by Aitchison and Brown (1969), is a generalized form of the lognormal model in which some of the observa- tions may be zeros and the nonzero values follow the lognormal distribution. The latter has two parameters (\i) and (a^) which are the mean and variance of the log-transformed observations (Hastings and Peacock 1975). The delta distri- bution has the same two parameters ( \i and a ) of the undcriying lognormal model, plus a third parameter (5) which is the proportion of zeros in the data. Thus, the lognormal distribution is a particular case of the delta distribution in which the parameter (5) is zero (i.e., when the data do not contain zero observations). Because the abundance of living organisms is the result of an inherently multiplicative process (i.e., the number of female parents times a fecun- dity rate, times a survival rate), random samples of naturally occurring organisms tend to follow the lognormal distribution (Demetrius 1971). In the case of marine organisms, fish in particular, it has been shown that recruitment variability is generally well described by the lognormal distri- bution (Ilennemuth et al. 1980; Peterman 1981; Ililbom 1985). Therefore, the delta distribution appears particularly well suited to describe the DELTA Means 311 variability of marine organisms when the data contain zero observations. The typical lognormal distribution is asymmet- ric, with a long tail on the right-hand side, and with a population mean that lies to the right of the middle point. Because the distance between the middle point and the mean increases with the variability of the data, the geometric mean or its estimator the sample median are often used to describe central tendency in lognormal data. However, there is no known estimator for the variance of those statistics when the lognormal data contain zeros. The basis for applying the delta distribution to describe the abundance of marine organisms is that, for approximately lognormal data with many zeros, the best estima- tors of the population mean and its variance are the mean of the delta distribution {h-meati) and its variance {h-variance). Like the sample mean (i.e., the average or arithmetic mean of the sam- ple), the h-mean estimates the population mean rather than the mid-point of the data distribution. Recent applications of the delta distribution to describe the variability of ichthyoplankton and fish in the MARMAP program (National Marine Fisheries Service) have been reported by Pennington (1983, 1986). ^^mCF) = 1 + m + m\2^?)(m + 1) {m - \)'y^ m\y.){m + \){m + 3) The constant y in these series is computed as (2) (3) where .!■ is the sample variance of the log- transformed m nonzero observations. The num- ber of terms needed in the series (Eq. 2) to achieve reasonable accuracy was found to be six to ten, depending on the number of significant digits in the logarithmic mean [x). The unbiased estimator of the 5-mean variance was also derived by Aitchison (1955). However, for large n and a proportion of zeros (5) appre- ciably less than 1.0, Owens and DeRouen (1980) found that the approximate (asymptotic) variance of h-mean given by Aitchison and Brown (1969) was accurate enough and much easier to compute. 1 his simplified estimator is computed as: Estimation of the 8-mean and its variance: The minimum variance unbiased estimator of the h-mean was derived by Aitchison (1955). Its computation in practice is rather involved and it generally requires the use of a computer program to evaluate a series iteratively until enough accu- racy is reached. The estimate of the h-mean for a given data collection is computed as: h-mean = exp(jc) G^{y) , (1) where m is the number of nonzero values in the data, n is the total number of observations in the data, X is the arithmetic mean of the log- transformed nonzero observations, and Cj„,{y) is a Bessel function that is evaluated as the scries: h-var - ^exp(2Jc + .r^)|s{l - 8) + ^(I - 5)(2.v^ + j'')}(4) for a proportion of zeros 5 = 1 (5) and where m, n, x, and s are as previously de- fined for equations 1 through 3. In applications with actual data it is often of interest to construct confidence intervals for the h-mean estimates. Not knowing the exact statis- tical distribution of the h-mean as estimated by r.q. 1, it may be reasonable to assume asymptotic normality for this estimator in order to form ap- proximate confidence intervals. Owen and DeRouen (1980) investigated this possibility in a 312 simulation study and concluded that the estimate of 5-variance obtained by Eq. 4 provided quite accurate coverage, in both 95% and 99% confi- dence intervals, for samples with more than 15 observations. Therefore, an approximate 95% confidence interval for large samples (n > 1 00) can be constructed as: 95%C/ = 5-mean ± 1 .96^/ 8-variance . (6) For smaller sample sizes ( 15 < n < 100 ) the corresponding two-tail t-value (a = 0.05) replaces the value of 1.96 above. Comparison of the h-mean to Other Statistics A numerical simulation was conducted to com- pare the 5-mean to other statistics often used to describe the abundance of marine organisms at NUEL and elsewhere. The statistics chosen for this study where the sample mean, the sample median and the geometric mean. The properties and common usage of these three statistics are briefly discussed first. The sample mean: The arithmetic mean of a sample or "sample mean" is the unbiased estimator of the true population mean under normality, but it is biased in the case of non-normal data (spe- cially with small samples). The actual estimator of the population mean has specific forms other than a simple arithmetic mean for each known statistical distribution (e.g., Eq. 1 is the form of the unbiased estimator of the delta-distribution mean). However, the sample mean is generally an acceptable estimator of the population mean for symmetric distributions not far from normal when the sample is large. Tor lognormal distri- butions with high variance (i.e., long tails), the sample mean is a poor estimator of the population mean and the standard error of the sample mean underestimates the true variance of th^ lognormal mean (Stuart and Ord 1987). The reason for this is that both lognormal mean and variance increase exponentially with the variance (a^) of the log- transformed data. As a^ approaches zero the distribution becomes symmetric and the sample mean, lognormal mean and median coincide (Hastings and Peacock 1975). The sample median: The median of a sample is the value corresponding to the mid-point of the ranked observations in the sample. Unlike the sample mean, the sample median always es- timates the mid-point of a distribution. If the distribution is symmetrical its mean and its median coincide (e.g., the normal distribution). For nor- mally distributed random samples, however, the sample mean is a more accurate or "efficient" estimator of the tme mean than the median be- cause the standard error is larger for the median than for the mean (Snedecor and Cochran 1980). For this reason, and also because of the superior statistical properties of the mean, the latter is pre- ferred over the median with symmetrical distribu- tions not far from normal. On the other hand, with data whose distribution is highly skewed, the median should be the preferred statistic because it conforms with the concept of an "average" bet- ter than the mean. An additional advantage of the median is that it is not affected by extreme values in the sample (i.e., outliers), whereas the sample mean can be greatly affected, more so in the case of small samples. A final consideration regarding the median is that, for data far from normal, nonparamctric confidence intervals and tests to compare medians are readily available. The performance of the median and its confidence interval with lognormal data that contain many zeros had not been reported prior to this study. Tlic geometric mean: The chief application of the geometric mean lies in lognormally distributed data for which the sample mean is a biased esti- mator of the lognormal mean and the median has only nonparamctric (very conservative) estimators of its variance. Unlike the sample mean, the geometric mean is also an estimator of the middle point of a lognormal distribution because it co- incides with the median. In practice, a simple logarithmic transformation of the lognormal data allows the estimation of the logarithmic sample mean and variance which then can be used to estimate the geometric mean and its asymmetric DELTA Means 313 confidence inlerval in the original scale. When there are zero observations, however, the data must be rescaled prior to log-transformation (by adding some arbitrary positive value to each ob- servation) and this causes the geometric mean and its confidence interval to become biased. Al- though no studies have been reported investigating the effect of zero observations on the geometric mean, it seems reasonable to assume that the bias will depend on the choice of the constant added to each observation and that it will increase with the proportion of zeros in the data. Numerical simulation: The basic premise in this simulation was that, for data with an approx- unate delta-distribution, the h-mean mean and the sample median were the only correct statistics to estimate the population mean and the mid-point of the distribution, respectively. On the basis of distribution theory it was already known that the sample mean and its variance were biased estima- tors of the lognormal population mean and its variance, and that the geometric mean was a bi- ased estimator of the middle point of lognormal Table 1. Typical sample size, proportion of zero observations, and variability for data collections from NUEL programs where 5-means have been used. data with many zeros. Therefore, the purpose of the simulation was to describe the relationships among the four statistics (i.e., their relative loca- tions) and to investigate how their magnitudes and the width of their 95% confidence intervals (95%C/) were affected by the presence of zeros in the data and by different amounts of variability. The data base used in the numerical simulation consisted of three data sets of 100 normal random numbers with identical mean (3c = 2.00) and in- creasing variances {s = 0.50, 1.00, and 2.25) so that the CV's would be 25, 50, and 75 %. These data were converted into lognormally distributed data by exponentiation of each observation. Therefore, the three lognormal data sets had iden- tical geometric mean, GM = exp(2) = 7.39, and increasing variabilities: CV = 25%, 50%, and 75% in the logarithmic scale. These CV values roughly corresponded to typical (low, moderate, and high) variabilities encountered in samples of marine organisms in various monitoring programs at NIJCL (Table 1). Monitoring Program Sample Size (n) Proportion of Zeros (5) Variability (CV)' I'ish trawl surveys Ichthyoplankton surveys Lx)bster larvae entrainment Rocky Shore (% cover) ^220 = 0.55 = 65% (108 to 480) (0.2 to 0.9) (5 to 300%) S200 ^0.35 ^55% (44 to 460) (0.10 to 0.68) (35 to 85%) = 70 ^0.55 S26% (57 to 86) (0.49 to 0.60) (24 to 28%) ^30 = 0.35 = 70% (13 to 91) (0.15 to 0.90) (20 to 190%) Coefficient of variability for log-transformed data. 314 Finally, the lognormal data were converted into "delta-distributed" data by adding an equal number of zeros to each data set and increasing that num- ber in each of the 15 simulations conducted. The zeros added in these simulation were 5, 10, 15, 20, . . , up to 75, resulting in data sets with 105, 110, 115, , 175 observations and 4.8 to 43 % of zeros. The h-mean, the sample mean, the median, and the geometric mean with their respective stan- dard errors were computed separately for each of the three lognormal data sets prior to adding any zeros, and then recomputed after adding zeros in each of the 15 simulations. The 95%C/'s for each statistic were also computed and normalized as the percentage of the statistic value represented by the total width of its confidence interval. The purpose of this normalization was to facilitate the comparison of 95%C/'s among simulations and across data sets with different variability. Direct comparison of 95"/oC/'s between statistics, how- ever, can be misleading because the standard er- rors of the sample and geometric means underes- timate the true variance. As a result, the 95%C/'s for the sample mean and the geometric mean reflect a coverage which is less than the nominal 95%, whereas the 95%C/ for the 8-mean reflects an accurate coverage and the 95% C/ for the me- dian is very conservative (i.e., its coverage could be closer to 99% than to 95%). The 95%C/ of the median was estimated using the two order statistics given by Snedecor and Cochran's (1980) nonparametric formula: n+ 1 1.96 7« (V) where /; is the sample size ( 1 00 to 175 observations in this simulation study). Simulation results: The simulation results are presented in Figure la- lb for low variability data, in Figure 2a-2b for moderate variability, and in Figure 3a-3b for high variability. The magnitudes of the four statistics for each simulation are shown in Figures la, 2a, and 3a. Although the biased sample mean and the unbiased 5- mean did not coincide, they tracked each other remarkably well over the entire range of zeros and variability sim- ulated. Also the median and geometric mean tracked each other well for moderate and high variance data, but the geometric mean was badly affected by the increasing number of zeros when variability was low. Except for highly variable data, the geometric mean was always smaller than the median (very much so at low variance). This indicates that, in the presence of zeros, the geo- metric mean is not a reliable estimator of the middle point of the data which is always accurately described by the median. The two estimators of the population mean (i.e., the sample mean and the 8- mean), were always located to the right of the middle point of the data (as expected) and the separation increased with the variability in the data. Except for the geometric mean in the case of low variability data, the magnitude of the four statistics declined at similar rates in response to the addition of zeros to the lognormal data. The relative widths of the 95 %C/ for each statistic are shown in Figures lb, 2b, and 3b. The 95%C/ of the h-mean (the only CI known to be accurate for "delta" data) increased with the data variability, but its relative witdth was almost un- affected by the number of zeros in the data. Al- though the 95%C/ of the sample mean was also quite insensitive to the number of zeros present, its coverage was not accurate because the sample mean standard error underestimates the true vari- ance. Finally, the 95%C/'s of the geometric mean and median were erratic and had very different widths, especially with moderate and high vari- ability. Except at low variability, both 95%C/'s were greatly affected by the number of zeros in the data. In summary, the 5-mean and its 95%C/ be- haved predictably and consistently over the ranges of data variability and proportion of zeros simu- lated. The 8-mean increased in magnitude with the data variability and decreased as the proportion of zeros increased. The decrease caused by the presence of zeros was more pronounced when the data had high variability. The relative width of DELTA Means 315 INITIAL DATA: lOO nonzero Log-Normal obs. (Geom. Me C.V. = 255! for Log— transformed data 7.39) (a) 30 35 40 45 50 ZEROS ADDED TO THE INITIAL DA", A INITIAL DATA: 100 nonzero Log-Normal obs. tGeom. Mean = 7.39) C.V. = 25Jr Tor Log— transformed data (b) 20 25 30 35 40 45 50 ! ZEROS ADDED TO THE INITIAL DATA Fig. 1 Results of simulaling an increasing proportion of zeros in the low-variability data set: a) estimates of the four statistics; b) widths of the 95%C/'s normalized as a percentage of the estimated statistics. 316 INITIAL DATA; 100 nonzero Log-Normal obs. (Geom. Mean = 7.39) C.V. = 505! for Log— transformed data (a) ' 25 30 35 40 45 50 55 60 B5 70 75 ZEROb ADDED TO TmE INITIAL DATA S 70 INITIAL DATA: 100 nonzero Log-Normal obs. (Geom. Mean = 7.39) C.V. = 50?; T:?r Loq— transformed data 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 ZEROS ADDED TO THE INITIAL DATA Fig. 2 Results of simulating an increasing proportion of zeros in the moderate-variability data set: a) estimates of the four statistics; b) widths of the 95%C/'s normalized as a percentage of the estimated statistics. DELTA Means 317 INITIAL DATA; 100 nonzero Log-Normal obs, (.Geom. Mean = 7.39) CV. = 75)t for Log— transformed data (a) ii-__j*' n 5 10 15 20 25 30 35 40 45 50 55 60 55 ZEROS ADDED TO THE INITIAL DATA INITIAL DATA; lOO nonzero Log-Normal obs. ^Geom. Mean = 7.39) / G.V. = 75x tor Log— transformed data / / / (b) / / \ ,v / i^pla MEDIAN s/ / / Z' - X ^-i^ ^ GEOMETRIC moan 0 5 10 15 20 25 30 35 40 45 50 55 fcO 65 70 75 ZEROS ADDED TO THE INITIAL DATA Fig. 3 Results of simulating an increasing proportion of zeros in the high-variability data set: a) estimates of the four statistics; b) width of the 95%C/'s normalized as a percentage of the estimated statistics. 318 the 95 %C/ increased noticeably with the data variability, but it was almost unaffected by the proportion of zeros in the data regardless of vari- ability. This is a very desirable property of the 5-mean because, in actual applications, the pres- ence of zeros in the data should only affect the estimate of the mean without unduly inflating its standard error. The 5-mean was not a good es- timator of the middle point of the data. The only correct estimator of central tendency, the median, cannot be recommended for delta-type data be- cause its 95%C/ was unreliable except for low variability data with less than 30% of zeros (i.e., less than 55 zeros in Fig. lb). Although for highly variable data the geometric mean tracked the median quite well, its 95%C/ was also unre- liable. Application to NUEL's Monitoring Data The 5-mean should be the preferred statistic to estimate the population mean when the data con- tains many zeros (e.g., data from NUEL's pro- grams listed in Table 1). Its usage, however, should be restricted to cases where the nonzero observations are approximately lognormal and the sample size is 15 or larger. A simple approach to testing for lognormality is to log-transform the data and then test for normality using the proce- dure UNIVARIATE (SAS 1985). The most de- sirable features of the S-mean mean are an accurate standard error for delta-distributed data and 95%C/'s whose relative width (i.e., scaled by the mean) is almost unaffected by the proportion of zeros in the data. The weaknesses of the h-mean are that it does not coincide with the middle point of the data and that, unlike the median, it is not resistant to outliers. The median, however, was shown to have a 95%C/ which was unreliable except for low variability data with less than 30% of zeros. When the data contain only a few zeros (less than 10%), either the median or the geometric mean should perform reasonably well with low to moderate variability (CV < 50% for log- transformed data). In actual applications to monitoring data on species that occur seasonally, the cut-off points in the data series must be chosen consistently to insure comparability among years. The following procedure is suggested: 1) for each species, start and end the data series on the dates of the first and last occurrence of that species each year and ignore any zero data collected before and after; and 2) compute the cumulative distribution of the data series (i.e., by adding observations se- quentially) and trim the two tails of the distribu- tion by 2.5%, so that only the central observations adding up to 95% of the total sum are retained. It should be noted that this procedure will result in annual data series that will generally vary in length and starting and ending dates from year to year. This poses no problem for estimating cor- rect annual h-means and insures that the data series are a consistent and fair representation of the annual occurrence of each species. Finally, the lengthy computation of the h-mean and its standard error should be carried out using a computer program. Such a program was written for the simulation study described above and it is available at NU's computer system. The ac- curacy of this program was tested by reproducing the results of Peimington (1983). References Cited Aitchison, J. 1955. On the distribution of a pos- itive random variable having a discrete proba- bility mass at the origin. J. Amer. Stat. Assoc. 50:901-908. Aitchison, J., and J.A.C. Brown. 1969. The lognormal distribution. Cambridge University Press, New York. 156 pp. Demetrius, L. 1971. Multiplicative processes. Math. Biosci. 12:261-272. Hastings, N.A.J. , and J.B. Peacock. 1975. Sta- tistical distributions: a handbook for students and practitioners. John Wiley & Sons., New York. 130 pp. DELTA Means 319 Hcnncmuth, R.C., J.E. Palmer, and B.E. Brown. 1980. A statistical description of recruitment in eigliteen selected fish stocks. J. Northwest Atl. Fish. 1:101-111. Hilbom, R. 1985. Simplified calculation of op- timum spawning stock size from Ricker's stock-recruitment curve. Can. J. Fish. Aquat. Sci. 42:18.13-1834. Owen, W.J., and T.A. DeRouen. 1980. Estima- tion of the mean for lognormal data containing zeros and left-censored values, with applications to the measurement of worker exposure to air contaminants. Biometrics, 36:707-719. Pennington, M. 1983. Efficient estimators of abundance, for fish plankton surveys. Biometrics, 39:281-286. Pennington, M. 1986. Some statistical techniques for estimating abundance indices from trawl surveys. Fish. BuU., U.S. 84:519-525. Peterman, R.M. 1981. Form of random variation in salmon smolt-to-adult relations and its in- fluence on production estimates. Can. .1. Fish. Aquat. Sci. 38:1113-1119. SAS 1985. The univariate procedure. Pages 1181-1191, in SAS user's guide: basics, version 5 edition. SAS Institute Inc., Cary, North Car- olina. Snedecor, G.W., and W.G. Cochran. 1980. Sta- tistical methods (7th edition). The Iowa State University Press, Ames, Iowa. 584 pp. Stuart, A., and J.K. Ord. 1987. Kendall's ad- vanced theory of statistics (5th edition). Volume 1: Distribution theory. Oxford University Press, New York. 604 pp. 320 Contents Ilydrothcrmal Studies 323 Introduction 323 Review of past studies 323 1965 - 1976 324 1977 Hydrotheimal survey 326 1978 - present 328 Materials and Methods 328 Dye injection 328 Survey vessels 329 In situ current, temperature and salinity monitoring 329 Presentation of isothermal data 329 Results 330 Fnvironmental conditions 330 Station operating conditions 331 Temperature and dye plume mapping 332 Ixiw slack 332 Maximum flood 334 High slack 334 Maximum ebb 334 Supplemental temperature data collection 334 Comparison of results to predictions 343 Summar>' 346 References Cited 346 Appendix to llydrothermnl Studios 349 Altarhmrnt to Ilydrothcrmal Stiidics 355 Hydrothermal Studies Introduction During 1987, Northeast Utilities Service Com- pany (NUSCO), on behalf of Northeast Nuclear Energy Company (NNECO), completed a survey of the extent and configuration of the thermal plume generated from the combined output of all three nuclear power plants at the Millstone Nu- clear Power Station (MNPS). This survey was completed to meet two needs. The first was to determine whether the NPDES Permit (CT 0003263) condition restricting the plume size had been met. The Permit states that the permitee shall operate all facilities in such a manner as not to raise the average temper- ature of the receiving waters more than 4°F or increase the normal temperature of the receiv- ing waters above 83°F. For purposes of this condition, cognizance will be given to reason- able time and distance to allow mixing of effluent and receiving waters, but the boundary of the mixing zone shall not exceed a radius of 8,000 feet from the discharge outlet at the quarry cut. The second was a commitment made by NUSCO on behalf of NNECO in a letter dated 23 July 1 986, to the Connecticut Department of Environ- mental Protection (CT DEP) (NUSCO 1986). In this letter NUSCO agreed to verify thermal plume predictions and to map the extent of the 3-unit thermal plume during varying tidal conditions, and to undertake ad- ditional thermal plume studies in 1987. The objective would be to implement the field stud- ies when all three units were operating .umul- taneously at or near 100% power. The work completed during 1987 included water temperature and dye concentration surveys during the period 21-27 August 1987 to determine the water temperature rise due to the plant discharge and resulted in plots of isotherms of plume con- figurations during four tidal regimes. The specific objectives of the field studies were 1 . Provide dye concentration measurements that would distinguish temperature rise in Long Island Sound due to heated circulating water discharge from heating due to solar radiation and other natural sources. 2. Provide information on the effect of three- unit circulating water discharge on the hydrographic and thermal characteristics of the receiving waters off Millstone Point dur- ing four tidal phases: max ebb, slack after ebb, max flood and slack after flood The thermal plume survey was a cooperative effort involving Ocean Surveys, Inc. (OSI), and NUSCO staff. OSI provided instrumentation for positioning, dye delivery, flourometry and tem- perature recording and expertise for thermal con- tour mapping. NUSCO provided necessary boat and staff support. The survey data were supple- mented by continuously recorded temperature data collected during October 1987 at strategic locations by NUSCO staff. This document sum- marizes past hydrographic studies, presents the methodology and results of those done during 1987 and compares the results of these studies to three-unit thermal plume predictions. Review of past studies Since 1966, NUSCO has retained several differ- ent investigators to perform hydrographic and hydrothermal surveys in the vicinity of Millstone Point (Table 1). Synopses of the results foUow. Hydrothermal Studies 323 TABLE 1. Chronology of MNPS hydrographic and hydrolhermal studies. Year Study 1965 Stone & Webster Engineering Corporation. Thermal eflects analysis for a 1,200 MW plant at the Millstone Site. Units 1 and 2 Environmental Report, Docket Nos. 50-245 and 50-336, Appendix B, Section III-A. 1965 Essex Marine Laboratory. Current velocity, temperature and salinity measurement in the Millstone Point area. 1965 U. S. Coast and Geodetic Survey. Study on tidal current data. Units 1 and 2 Environmen- tal Report, Docket Nos. 50-245 and 50-336, Appendix B, Section III-M. 1966 Bechtel Corporation. Diffusion patterns of the circulating water discharge effluent for the Millstone Project. Units 1 and 2 Environmental Report, Docket Nos. 50-245 and 50-336, Appendix B, Section IH-B. 1966 Pritchard-Carpenter Consultants. Continuous discharge tracer study, Twotrcc Island Chan- nel and Niantic Bay, Long Island Sound. Units 1 and 2 Environmental Report, Docket Nos. 50-245 and 50-336, Appendix B, SecUon III-C. 1968 - 69 Raytheon Corporation. Millstone survey, study on salinity - temperature profiles. Units 1 and 2 Environmental Report, Docket Nos. 50-245 and 50-336, Appendix B, Sections II-D and E. 1970 Pritchard-Carpenter Associates. Tracer study of the circulating water system, Millstone Point Unit 1 and 2 Environmental Report, Docket Nos. 5U-245 and 50-336, Appendix B, Section III-F. 1971 VAST, Inc. Unit 1 June 1971 Millstone Point temperature survey. Units 1 and 2 Environ- mental Report, Docket Nos. 50-245 and 50-336, Appendix B, Section III-G. 1971 VAST, Inc. Dye difTusion survey in Twotrcc Island Channel UniUs I and 2 Environmental Report, Docket Nos. 50-245 and 50-336, Appendix B, Section III-J. 1972 VAST, Inc. Study of an offshore thermal difl'user outfall by dye simulation. Units 1 and 2 Environmental Report, Docket Nos. 50-245 and 50-336, Appendix B, Section III-K. 1972 VAST, Inc. Unit 1 thermal survey. Units I and 2 Environmental Report, Docket Nos. 50-245 and 50-336, Appendix B, Section MM. 1973 - 74 Braincon Corporation. Hydrographic surveys. August-September 1973 and February 1974. 1977 Environmental Devices Corporation. Postoperational UniLs 1 and 2, preoperational Unit 3 hydrothermal survey of the Millstone Nuclear Power Station. Millstone Nuclear Power Station, Unit 3, Environmental Report, Operating License State, Appendix D. 1977 Texas Instruments, Inc. Airborne thermal infrared survey. Millstone Point Nuclear Station. Millstone Nuclear Power Station, Unit 3, Environmental Report, Operating License State, Appendix D. 1965 - 1976 All analysis of the thermal effect of a proposed nuclear power plant, with a capacity of 1,200 MWe and a design circulating water flow of 1560 cfs (44.2 m /s) and a 20°F (1 TC) rise, was made in May 1965 by Stone & Webster Engineering Corporation. It was assumed that the circulating water would be discharged from the southern end of the quarry into Long Island Sound. The pur- pose of this analysis was to provide a basis for predicting thermal effects that would result from the operation of Millstone Unit 1. In 1965, Essex Marine Laboratory obtained field data to serve as a basis for the initial calcu- lations of thermal distribution patterns. Their observations established the existence of strong tidal currents in Twotree Island Channel (see Fig. 1 for orientation) and provided information on velocity, temperature, and salinity profiles in the area of the proposed circulating water discharge for Unit 1. These data were supplemented by tidal current measurements made during the same period by the U.S. Coast and Geodetic Survey (now the National Ocean Survey). These mea- surements were made at the following locations: 324 Twotree Island Channel, 0.2 mi (0.3 km) south of Bartlett Reef, and Niantic River railroad bridge. Continuous measurements were made for approx- imately four days. Measurements in Twotree Is- land Channel and Bartlett Reef consisted of cur- rent speed and direction measurements at three different depths. A study of the diffusion patterns of the circu- lating water discharge effluent was made by Bechtel Corporation (1966). The objective of this study was to determine the flow and diffusion patterns of the condenser discharge with the aid of a small scale hydraulic model. The hydraulic characteristics of the model were correlated with local field data furnished by the Essex Marine laboratory and the U.S. Coast and Geodetic Sur- vey. The model study used tracers to indicate the general flow patterns of the tidal current in the vicinity of Millstone Point. Although the variations of the water temperature in the effluent were not predicted from this technique, the study provided local flow patterns in the Millstone area. In May of 1966 Pritchard-Carpenter (1967) con- ducted Rhodamine B dye tracer tests prior to the operation of Unit 1 to determine the dilution off Millstone Point that resulted from the natural action of tides and wind. To simulate the action of the Unit 1 circulating water system, which was not functional at the time, Rhodamine B dye was released through 1/8-in (3.2-mm) holes in a 5-ft (1.5-m) pipe located 450 ft (137 m) offshore of the proposed discharge point. The results pro- vided estimates of the rates of mixing and dilution of chemical discharges, and cooling of the heated water. However, the thermal predictions did not reflect induced entrainment due to discharge mo- mentum nor did they reflect the effect that strat- ification would have on the discharge of warm water. Thus, the predictions were considered conservative for design purposes. During August l968 and March 1969 "aytheon Marine Research Laboratory (1968, 1969) made temperature and salinity measurements at three points in the greater Millstone bight to obtain vertical temperature and salinity distributions. These points represented the general locations of the ends of the Unit 1 thermal plume during ebb and flood tide as predicted by Pritchard-Carpenter (1967). The purpose of these surveys was to obtain reference information on typical summer and winter temperature-salinity characteristics. During February 1970 several tests were con- ducted (Pritchard-Carpenter 1970) with the Unit 1 circulating water system operating at full flow, but at ambient water temperature. A 30-percent Rhodamine B solution was injected into the in- take at 10 ml/min. The purpose was to evaluate the combined effect of momentum entrainment resulting from the velocity of the discharge and the natural mixing characteristics produced by tidal currents. Because there was no heat rejected to the cooling water, the effects of stratification were not determined. In June 1971, VAST, Inc. (1971a) conducted a temperature survey of three-dimensional thermal distribution patterns as they actually occurred during full operation of Unit 1. The study used a surface transect method to reduce the measure- ment time required for each of four surveys; one during maximum flood, maximum ebb, slack after flood and slack after ebb. These surveys indicated the position of the thermal plume and provided the information necessary for the strategic collec- tion of temperature at depth profiles. The result- ing profiles indicated the three-dimensional struc- ture of the thermal plume and, in deeper water, determined the depth of the thermal plume. However, no compensation was made for the onshore warming temperature gradient character- istic of the spring and summer months. Thus, the areas enclosed by contours representing water temperatures of 1.5°F and 4.0°F (0.83°C and 2.2°C) above ambient (AT) were overestimated. In November 1971 and March 1972, VAST, Inc. (1972b) conducted temperature surveys to define and map the thermal plume from Unit 1 independent of thermal inputs from natural sources. Dye was injected at a constant rate into the Unit 1 discharge while the unit was at full load, and all four circulating water pumps and Hydrothemial Studies 325 two service water pumps were in operation. Dye concentrations were measured continuously in the quarry and from boats moving along prescribed transects. The dye concentration and temperature data were digitized at predetermined grid points and converted to the equivalent temperature rise to obtain synoptic patterns of the thermal plume. These results were in closer agreement with pre- dictions than the June 1971 study (NUSCO 1979). The differences between the prediction of May 1966 and those of February 1970 in the near field were attributed to the effects of momentum en- trainment which had not been compensated for in the 1966 study. VAST, Inc. conducted two other dye studies (1971a, 1972c). Each was designed to determine the natural flushing or renewal rate of Long Island Sound water in the vicinity of Millstone Point due to tidal exchange. The surveys were required for use in the conceptual design of a multiport diffuser system that might be located in Twotree Island Channel or west of Twotree Island for cooling water discharge. During both surveys Rhodamine B dye was injected from boats in these two areas and dye concentrations were sam- pled continuously from boats moving along pre- scribed transects. In 1972, Stone and Webster Engineering Cor- poration (SWEC) completed a modeling effort that resulted in preliminary thermal plume pre- dictions presented in the Environmental Reports for Unit 2 and Unit 3 (MFC 1972a, 1972b). Hydrograpliic surveys were conducted in the Niantic Bay area of Long Island Sound during August and September 1973 and February 1974 (Braincon 1975). Tide levels, tidal current speed and direction, and wind speed and direction were measured. These data were analyzed and the tide levels were used as input for the SWEC two- dimensional tidal circulation model; the current data were used to verify this model. 1977 Hydrothermal survey During 1977, as required by the Nuclear Reg- ulatory Commission, a hydrothermal field survey of the two-unit operational plume from MNPS, was conducted (ENDECo 1978; TI 1978). The objectives of this survey were to determine the three-dimensional temperature characteristics of the thermal plume resulting from the discharge of condenser cooling water from Units 1 and 2 and to verify the original 1972 predictions of the two- unit plume. The details of these studies are best described in contractor reports (ENDECo 1978; TI 1978) but are summarized briefly. The dye study provided detailed information on dilution rates and three-dimensional temperature distributions (ENDECo 1978). Rhodamine WT dye was introduced into the Units 1 and 2 circu- lating water discharges during full plant load con- ditions to achieve 16 ppb at the quarry cut (ENDECo 1978). During a two-day period (one complete tidal cycle each day) dye concentration and temperature were monitored continuously at each intake, the quarry cut and from boats that followed six predetermined transects. It took the boats about one hour to complete all six transects; boat position was determined from four transpon- ders. In the near field (an area defined as 500 ft (152 m) from the center line of the discharge plume out to 1000 ft (305 m)), dye concentration and temperature were measured vertically (5-ft (1.5-m) at intervals down to 30 ft (9 m)) and horizontally 1.5 ft (0.5 m) below the surface. These data were used to map the plume over a complete tidal cycle. Temperature isotherms and dye concentration isopleths were drawn manually on a map. Thermal infrared scanning provided a synoptic 'picture' of the surface thermal plume at specific tidal stages. The 8- to 14-micrometer portion of the electromagnetic spectrum was scanned coinci- dent with the dye study (TI 1978). Temperature reference sources within the field-of-view provided ground truth calibration. The recorded airborne thermal infrared data were presented both quali- tatively and quantitatively. 326 Comparisons of the temperature and dye plumes When compared to two-unit predictions, however, and the infrared survey indicated that the shape it was clear that the distance to the 1.5°F (0.83°C) and extent of the temperature and dye plumes for isotherm exceeded predictions for the max ebb some tidal phases were in general agreement with tide stage. each other (NUSCO 1979) (Tables 2 and 3). TABLE 2. Comparison of predicted and actual surface areas encompased by selected AT isotherms pro- duced during two-unit operation. Area in acres encompassed by each isotherm Tide stage 8.0°F (4.4°C) 6.0°F (3.3°C) 4.0°F (2.2°C) 1.5°F (0.8°C) Max Flood Predicted for 2 units' 47.1 172.0 Hydrothermal survey, 7/29/77 Infrared survey, 7/29/77 0.8 6.4 1.0 12.4 7.1 140.0 91. r 666.0^ Slack after flood (high slack) Predicted for 2 units 48.2 223.0 Hydrothermal survey, 7/29/77 Infrared survey, 7/29/77 1.8 2.6 7.8 12.1 32.5 34.5 58.1 50.6 Max Ebb Predicted for 2 units 33.1 159.0 Hydrothermal survey, 7/29/77 Infrared survey, 7/29/77 7.1 10.1 10.6 65.1 27.3 171.0 147.0^ 290.0 Slack after ebb (low slack) Predicted for 2 units 56.0 273.0 Hydrothermal survey, 7/29/77 Infrared survey, 7/29/77 9.0 11.6 66.1 78.5 186.0, 243.0^ 331.0, 410.0^ 1 from NUSCO 1979 2 plume areas are limited due to lack of data TABLE 3. Comparison of maximum predicted and actual distances to selected AT isotherms produced dur- ing two -unit operation. Distances in feet from quarry cut to each isotherm Tide stage 8.0°F (4.4°C) 6.0°F (3.3°C) 4.0°F (2.2°C) - 1.5°F (0.8°C) Max Flood Predicted for 2 units Hydrothermal survey, 7/29/77 SWEC predicted, 3 units (extreme) Slack after flood (high slack) Predicted for 2 units Hydrothermal survey, 7/29/77 SWEC predicted, 3 units (extreme) Max Ebb Predicted for 2 units Hydrothermal survey, 7/29/77 SWEC predicted, 3 units (extreme) Slack after ebb (low slack) Predicted for 2 units Hydrothermal survey, 7/29/77 SWEC predicted, 3 units (extreme) S&A predicted 3 units 1 from NUSCO 1979 2 distance estimated from infrared survey on 7/29/77 Hydrothermal Studies 327 205 230 1,900 2,300 1,730 2,286 4,800 4,000 9,700' 520 900 2,300 2,200 1,500 3,143 5,000 2,200 4,000 780 2,100 1,120 3,900 2,500 1,500 7,428 4,900 5,500 10,900 1,150 1,800 2,300 4,800 5,000 2,500 3,600^ 5,428 8,000 5,400 4,800^ 11,500 12,000 1978 - present The results of the 1977 survey indicated that the two-unit thermal plume approached, and ex- ceeded, the 4000-ft limit specified by the NPDES Permit for Units 1 and 2, for short periods as the tide turned during the slack after ebb tidal phase (Stoltzenbach and Adams 1979). Because of these findings and recognizing that the original plume predictions were based on non-transient models, it became apparent to NUSCO that improved modeling was necessary. During 1978 and 1979, Liang and Tsai (1979) analyzed the data and calibrated an updated math- ematical model to predict the thermal plume re- sulting from the combined discharge of condenser cooling water from all three units. Also, Stohzenbach and Adams (1979) developed a near field - intermediate field model and coupled it to the existing transient far field model. This model replicated the essential surface and subsurface fea- tures of the MNPS thermal plume as observed during two-unit operation and provided reliable estimates of the surface and subsurface induced temperature rise distribution down to a AT of 1.5°F (0.83°C). Conservative estimates of the maximum extent (in terms of distance from the discharge point) of 6°F (3.3°C), 4''F (2.2''C) and 1 .5°F (0.83°C) induced temperature increases were 5,000 ft (1,524 m), 8,000 ft (2,438 m) and 12,000 (3,658 m) respectively (Stoltzenbach and Adams 1979). The NPDES Permit now limits the 4°F (2.2°C) temperature rise to 8,000 ft (2,438 m). Several factors were considered when selecting the methodology and dates for the actual plume mapping in 1987. After discussions with advisors (J. Tietjen, N. Marshall, S. Saila, W. Pearcy) and regulators, NUSCO determined that the configu- ration of the thermal plume during late summer, when ambient water temperatures were near max- imum, would represent worst case conditions. Mapping would have to include a dye survey so that natural daily warming could be separated from the heat load added by the Station. Also because NUSCO wanted to determine the maxi- mum extent of the three-unit thermal plume, the survey ^yould have to be completed when all MNPS units were operating at near-maximum capacity. Because of scheduled refuel outages (Unit 1, July and August 1987 and Unit 3 to begin 31 October 1987), a window of appropriate conditions would occur in 1987 only during late August. After that, water temperature would have cooled considerably from its normal summer max- imum. Materials and Methods As mentioned previously, Ocean Surveys, Inc. (OSI), Old Saybrook, Connecticut, provided in- strumentation and technical support to complete the actual dye survey. Preliminary activities began 21 August 1987 and data collection began 23 Au- gust 1987. Actual mapping of the thermal plume during four major tidal phases (low slack, max flood, high slack and max ebb) took place on 26 August 1987. The survey data were supplemented with continuous in-situ temperature records taken subsequent to the actual plume mapping. Details of instrumentation and data aquisition techniques used during the mapping survey are described in the OSI report, which is the Attachment to this section. A brief summary is given below along with a description of methodology used by NUSCO to collect in situ supplemental temper- ature during October 1987. Dye injection Dye plume mapping was accomplished using Rhodamine WT flourescent tracer dye. Dye was injected onto the water surface immediately down- stream of the Unit 3 discharge. Mixing occurred here and where the waters from Units 1, 2 and 3 met in the middle of the quarry. Dye concentra- tion and water temperature were monitored con- tinuously at one quarry cut using a calibrated Turner Model 1 1 1 flourometer and a Yellow Springs Instrument Company Series 700 thermistor, respectively. To insure that the dye injection rate remained constant, NUSCO staff checked the flow of dye, weighed the dye supply reservoir and checked the flourometer every two hours. Background levels of flourescence were 328 recorded from 1700 hours, 21 August to 1400 hours, 22 August. Dye injection began at 1500 hours on 23 August 1987, at a rate of 5 Ib/h (1.85 kg/h). This was increased to 7 Ib/h (2.6 kg/h) 26 hours later so that a dye concentration of 2.2 ppb by weight was achieved at the quarry cut. Dye concentrations at the cut were first observed about 3 hours after dye injection began and stabilized there about 10 hours after changes occurred at the injection point. Dye injection was suspended at 1750 hours on 26 August at the conclusion of the intensive mapping period (0657 to 1605 hours, 26 August 1987); flourescence at the quarry cut continued to be monitored until 1200 hours on 27 August 1987. Survey vessels Two NUSCO vessels, 'Northeast I' and 'North- east ir, were equipped with instrumentation ap- propriate for use during the intensive mapping phase of the study. Both vessels had Turner Model 10 flourometers equipped with thermistors. 'Northeast I' was rigged to allow surficial dye concentration and temperature measurements to be taken along 41 predetermined transects during the four tidal mapping sessions. The transects were oriented nominally perpendicular to the axis of the thermal plume and along well defined plume boundaries. 'Northeast If was rigged to collect depth profiles of temperature and dye con- centration at 27 specific locations to characterize the vertical mixing of the thermal plume. Survey vessels were positioned using either a computerized system, developed by OSI, that per- mits an accuracy of ± 50 ft ( 1 5 m) or a manual system employing a hand-held VHF range finder. WTiere the computerized system could be used, the person at the helm could locate the vessels' position with respect to the intended survey trackline on a video monitor. The desired trackline could be followed by making course corrections indicated on the display. Where site conditions did not permit the use of the computerized system, the survey vessel was controlled along a bore sight representing the intended survey transect by the transit operator using the VHF range finder. In all cases, actual position data were automatically recorded; tracklines are presented in the Appendix to this section. In situ current, temperature and salinity monitoring Continuous measurements of current speed and direction, and water temperature and salinity, were obtained from an Endeco Type 174 in situ recording device positioned at stations east and southeast of Millstone Point (Attachment Fig. 1). Stations JCE and JCW were occupied on 24 and 25 August and TTIC and MP were occupied on the day of mapping, 26 August. Between 8 and 29 October 1987, NUSCO staff deployed two solid-state temperature recorders (TempMentors), which continuously logged water temperatures at two depths, at seven stations (Fig. 1). These data loggers provided additional infor- mation on the variability of the position and ex- tent of the thermal plume between tidal stages and under a variety of meterological and opera- tional conditions. The date and duration of each deployment are listed in Table 4. The loggers were returned to the lab and data was processed using an IBM PC. TABLE 4. Deployments of the TempMentor continu- ously recording thermistors during 1987. At each lo- cation a thermistor was set 1 m below the surface and 1 m above the bottom. Date Location* Duration 10/8/87 1 108 h 10/13/87 2 75 h 10/16/87 3 76 h 10/21/87 4 48 h 10/23/87 ' 5 51 h 10/26/87 6 51 h 10/29/87 7 96 h * see rig. 1 Presentation of isothermal data Because both dye concentration and water tem- perature are considered conservative, the percent drop in the temperature of the plume with respect to the total temperature difference between the discharge water and the receiving body of water Hydrothemial Studies 329 r^\ \' / \ \^ j\ \ III 1 1 1 5000 ft { f MNPS ^ J Jordan Hi } Niantic Bay ^ atX Cove \ 3 ? \ quarry ■ 3^ a ^ \White \ Point Seaside ^/ I %/ 4 Twotree ^q^ Island ''e/ Black / Point 1 ,.0- ^% Fig. 1. The location of MNPS. TempMentor deployment locations are designated by the numerals "V through '7". is equal to the percent dilution of the dye in the plume with respect to the dye concentration of the discharge water at the quarry cut. Dye con- centrations (ppb by weight) were converted to degrees Farenheit above ambient temperature (AT) and plotted on a trackline map for each of the four surveys. Because heat loss to the atmo- sphere was neglected, the indicated temperature increases were higher than actual, particularly in the far field. These data were contoured at the 1.5, 4, 6 and 8°F (0.83, 2.2, 3.3 and AA'C) AT levels. Data from the vertical profiles were also developed; these were plotted as AT versus depth. Results The results of 1987 hydrothermal studies com- pleted between 20 August and 15 November 1987 follow. The results include those from both the OSI-supported dye study and the NUSCO- deployed temperature data loggers. Environmental conditions Meterological conditions in the vicinity of MNPS during August through October reflect the seasonal transition from summer to autumn. Based on historical data acquired during eleven years from the environmental data aquisition net- work (EDAN), air temperature decreases from an average of about 22° C in early August to about lOX by the end of October (Fig. 2). Water temperatures decrease from an average of 20°C in early August to 14°C in late October (Fig. 2). Winds tend to be liglit (2 - 2 m/s) in early August, and increase in intensity to 3 - 6 m/s in October (Fig. 2). Winds tend to be out of the north in August, and from the west northwest in October (Fig. 3). Although, conditions during this period in 1987 were fairly typical of historical conditions (Figs. 2, 3), a storm occurred 21-22 August 1987, three days before the dye mapping. Wind speed varied between 2 and 7 m/s during the period the 330 AUG01 AUG20 SEP09 SEP28 OCTIS N0V07 aevan-yaar mean (1976-86) 1987 values; », o-doys of survey and probe deployment 21- 20- 19- IB- 17- 16- 15- AUGl AUG20 SEP09 SEP28 0CT18 N0V07 Deven-year mean (1976-86) 1987 values: «, o-doys of survey ond probe deployment 0 ■4 • *o dP ' Oo°! 0 *> • ° AUG01 AUG20 SEP09 SEP28 0CT18 N0V07 Deven-year mean (1976-86) 1987 values: », o-days of survey and probe deployment Fig. 2. Air and water temperature, and wind speed recorded by the environmental data acquistion network (EDAN), August through October, 1976 - 1986. Am- bient water temperatures (°C) are averag.is of those measured at 15-minute intervals at the MNPS Units 1 and 2 intakes. Air temperature and wind speed are averages ormeasurments taken at 15-minute intervals, 10 m high on the meterological tower. TempMentors were deployed and water tempertures were below the 11 -year mean. Dye concentration mapping was conducted un- der nearly ideal weather conditions (26 August). Seas were calm (less than 0.3 m), reducing thermal plume mixing due to wave action to a minimum. The light winds, averaging 2 m/s (Figs. 2, 3), assured that the dispersal of the thermal plume was due predominantly to tidal influences. Con- ditions were good enough to allow surface expres- sion of the plume boundaries during maximum flood, high slack and maximum ebb stages of the tide. A further benefit of the good conditions was the development of the 1.5°F (0.83°C) isotherm. Greater wind and wave action would have dimin- ished the lateral extent of this isotherm. Thus, error-free equipment operation combined with excellent site conditions allowed for the clear def- inition of this isotherm. Station operating conditions All three MNPS units normally operate at a constant base load. Unit 1 resumed power pro- duction on 19 August 1987 subsequent to the completion of its refuel outage. Power production reached 100 % capacity on 21 August 1988 (Fig. 4). During the background monitoring period, the dye build-up period, and low slack and max flood mapping periods (0900 on 26 August 1987) all three units operated at full capacity and pumped a total of 4,364 cfs ( 124 m /s) of cooling water (Fig. 4). Between 0900 and 1400 hours on 26 August, Unit 1 power production was reduced. Concurrently, Unit 1 cooling water usage was reduced from 989 cfs (28 m"/s) to 521 cfs (15 m' /s) and the effluent water temperature dropped about rC (Fig. 5). The Station, however, was still using 94% of its maximum cooling water at 1200 hours, during the high slack mapping session, and nearly 90% of its maximum at 1430 hours, the start of the maximum ebb tide mapping ses- sion. Considering the nearly ideal survey condi- tions, these reductions in discharge volume should not have greatly altered the patterns and magni- tudes of the plume from those presented; adverse wind and sea conditions could be expected to Hydrothermal Studies 331 Mid draction Auguat 1 TJorST' Fig. 3. Frequency of l5-min observations of wind direction as measured al the 30-m level on the melerological tower, August through October. have a much greater impact on the thermal plume. 0) turns' I 125 .S 100 o. 75 8 50 o 25 AUG87 AUG87 SEP87 SEP87 0CT87 N0V87 • Date Fig. 4. Station cooling water usage ( — , million m /sec- ond) and power production (--, MWe) August through October 1987. Temperature and dye plume mapping As expected the the MNPS 3-unit thermal plume moved and shifted in the direction of the prevailing currents (Attachment Figs. 10 - 13). Under the survey conditions on 26 August 1987, the discharge water cooled to 4°F (2.2°C) above ambient within 3,750 ft (1,143 m) of the quarry cut on all tidal stages. At low slack, the plume pooled in an area from southwest to east of Mill- stone Point and extended toward Twotree Island. During maximum flood the plume was swept by the tide along Millstone Point, westward into Niantic Bay. At high slack the plume again pooled in an area southeast of Millstone Point but did not extend as far as Twotree Island. Dur- ing maximum ebb the plume extended into and followed Twotree Island Channel. Low slack During the low slack survey (0657 - 081 1 hours), the plume was fairly well distributed about the quarry cuts. The 4''F (2. 2''C) isotherm extended out to 3,500 ft (1,067 m) off the cuts; it also extented at the north into the eastern part of .lordan Cove to 3,760 ft (1,146 m. Attachment Fig. 10). The 6 and ST (3.3 and 4.4°C) isotherms defined the edges of the discharge jet. The 4°F (2.2°C) isotherm extended to the bottom out to 500 ft (152 m) from the quarry cuts (LSI and 332 High slack 500 600 700 800 900 1000 1100 1200 1300 1400 1500 1600 1700 1800 Time 5000 4500 5" 4000 •J" 3500 ig 3000 § 2500 o 2000 ■^ 1500 j? 1000 500 0 5000 4500 4000 3500 3000 2500 2000 1500 1000 500 500 600 700 800 900 1000 1100 1200 1300 1400 1500 1600 1700 1800 Time 500 600 700 800 900 1000 1100 1200 1300 1400 1500 1600 1700 1800 Time Low slack Max. flood survey survey High slack Max. ebb surv,ey survey 500 600 700 800 900 1000 1100 1200 1300 1400 1500 1600 1700 1800 Time Fig. 5. EfTluent and intake water temperatures ("C), total station cooling water flow (cfs), and wind speed (m/s) and direction, taken at 15-minute intervals on the day of dye mapping (26 August 1987). Hydrothermal Studies 333 Attacliment Fig. 14). At 2,000 ft (610 m) from the cuts, the base of the plume (as indicated by the 1.5"F (0.83°C) isotherm) was 10 to 15 ft (3 to 5 m) below the surface (LS3 on Attachment Fig. 14). Maximum flood During the maximum flood (0848 - 1019 hours), strong currents carried the thermal plume to the west (Attachment Fig. 11). Mixing diluted the plume so that the 4°F (2.2°C) isotherm extended a maximum of 2,000 ft (610 m) to the west in a narrow tongue; the 1.5°F (0.83''C) isotherm was observed much farther west where it appeared to begin separating into multiple branches. A well defined edge to the plume was defmed by a change in surface water texture and a line of foam on the surface; dye concentration readings dropped to background levels as this line was crossed. The 4''F (2.2°C) isotherm extended to the bottom in the vicinity of the cuts, but was confined to the upper 10 ft (3 m) 1,000 ft (305 m) southwest of the cut (MFl and 3 on Attachment Fig. 15). The 1.5°F (0.83°C) isotherm was less than 5 ft (1.5 m) from the surface 4,000 ft (1 ,220 m) west south- west of the cuts (MF5 on Attachment Fig. 15). High slack During the high slack survey (1213 - 1308 hours), the thermal plume was rather evenly dis- tributed about the cuts and the center was shorter and broader than during previous tidal phases (Attachment Fig. 12). This difference was prob- ably due to the difference in tide elevations. Dur- ing low slack the water elevation between the cut and Long Island Sound is greatest, producing higher discharge currents there than during high slack; the greater currents would tend to carry the plume jet further offshore. The 4''F (2.2°C) iso- therm extended 2,100 ft (640 m) offshore while the 1.5''F (0.83°C) isotherm was approximately 3,000 ft (914 m) oft" the cuts. About 125 ft (38 m) offshore (MSI on Attachment Fig. 16), the base of the plume was 20 ft (6 m) deep and was within the top 10 ft (3 m) beyond 1,000 ft (305 m) from the cuts. Maximum ebb During the maximum ebb survey (1434 - 1605 hours), the thermal plume was carried eastward into Twotree Island Channel (Attachment Fig. 13). The 4^ (2.2°C) isotherm extended out 2,500 ft (762 m) to the southeast and 3,300 ft (1,006 m) to the northeast mto Jordan Cove. The 6°F (3.3°C) isotherm closely followed the 4°F (2.2''C) isotherm. The 1.5°F (0.83°C) isotherm was mapped as a narrow finger extending 12,500 ft (3,810 m) toward the southeast through Twotree Island Channel. In Jordan Cove, the 4 and 1.5 °F (2.2 and 0.83°C) isotherms were limited to the eastern part; no dye was detected in the northern or western parts. The plume (4°F (2.2°C) iso- therm) extended down 17, 13, and 4 ft (5, 4 and 1.2 m) at 125, 500 and 1,000 ft (38, 152 and 305 m), respectively, from the cuts (MEl, 2 and 3 on Attachment Fig. 17). As mentioned earlier. Unit 1 reduced power, heat load and circulating water volume usage during the morning of 26 August. Because the Station was still operating at nearly 90% of full capacity, the survey was completed after steady state had been reached and due to the ideal conditions, the distributions of these iso- therms are believed to be representative of con- ditions during maximum ebb. Supplemental temperature data collection The TempMentor data loggers were deployed at a range of distances from the quarry cuts during 8 October through 2 November 1987. Deploy- ment #3 was closest, about 1 ,500 ft (457 m) north east of the cuts; deployment #4 was fartherst, about 7,500 ft (2,287 m) south east of the cuts. Deployments #5, #2, #7, #1 and ^6 were set at intermediate and increasing distances from the cuts (see Fig. 1). The temperatures recorded continuously during the seven deployments provided location-specific water temperature histories (Fig. 6), which re- flected both Station operation and environmental conditions (Fig. 7). During deployment #3, 334 Dep I oymen t #=1 31 30 29 28 27 26 25 24 23- 22 21- 20- 19- 2i 18- § 17- I 16- . 15- 14- 13" 12" 11 ' o (U cn (D - 1 ■ ! 1 t 1 i !■ 1 1 1 i ! ! I t 1 ] 1 1 t 1 1 I I I 1 1 ! ! 1 1 ! i 1 1 1 ! ! 1 1 t ■i 1 s^ 1 i i i ! 1 1 i 1 i 1 1 1 1 1 1 I ! 1 ■1 I 111 I 1 I ! 1 i 1 1 ! I I I I 1 1 1 1 A ! 1 1 1 1 1 1 ! ' 1 1 ' 1 1 1 1 ! ' 1 ^ 1 ■ 1 .... ,, 1 1— \ 1 ■ ■ y 12 24 36 48 60 72 84 Time from start (hours) 96 Fig. 6. Surface ( — ) and bottom (^— ") water temperatures recorded from TempMentor deployments. The upper and lower lines (+ +) are the efTIuent and intake water temperatures respectively; times of high (T) and low (i) slack are also indicated. Hydrothermal Studies 335 Dep I oymen t # = 2. O CO «* *VAis^ »% 1 1 1 1 1 1 , I. +■ 1 1 ! ! 1 ^26- I 1 1 1 1 ' ' ! 1 <^ ^^ ' 1 1 1 1 1 1 ■■ 1 I •■ , CO 25; 1 1 I ! 1 1 1 1 ' ! ' ! 1 I 1 [ 1 ! '• 1 , 1 ■ ' Qj 24- 1 ! ! 1 1 1 1 ! ! 1 1 ' ! S^23- I I I 1 I 1 ' 1 1 1 TD ' ! 1 1 ! ' 1 ^22- 1 1 1 1 1 ' ' ! I j,,^,^lllll|llllllll%WHtHHj))HHWtiM#'H'W- 12 18 24 30 36 Time from start (hours) Fig. 6, continued Hydrothermal Studies 339 Deployment #=6 O cn (U (U cn a i_ CD a. £ -l-l s_ N-'J * • ^ _J^^^fy^^»m. 'i y\ - Ct^'"' ' 'x^Ji,^^^ 1 4 ! 1 1^: 1 1 1 I 1 ! 1 ! 1 I II 1 1 12; 1 1 1 ' 1 I'll ' 'III 11'- 1 ! ' 1 ' 1 ' 1 ' ! ' 1 1 1 1 1 1 1 1 1 1 1 12 18 24 30 36 Time from start (hours) 42 48 Fig. 6, continued 340 Deployment #=7 O (D O) CD "a 31 ■ 30- 29" 28- 27- 26- 25- 24- 23- 22- 21 - 20- 19- 18- 17- 16- 15- 14- 13- 12- 11 ' '! i T i -f it )- ! ' 1 ! ! ' I 1 1 1 1 I 1 1 1 ' 1 ' 1 ! I : 1 1 1 ' ' ! ! ! i 1 1 ! I 1 1 1 1 1 1 1 1 I 1 1 1 ! 1 ' ] ' t -1 " 1 1 1 1 1 1 1 1 ! 1 ' 1 ' ' ; I ! Ill' I 1 I 1 1 1 1 I - : • I ' ■ ' t 1 1 II ! ' it! :Vi/ : : : : .-: MY ! II 1 . I II 1 1 '1 1 : II 1 . 1 ! 1 11 1 I ! . '■ 1 1 11 1 1 ' S*^ 1 till ' 1 ill: 1 1 '111 1 1 1 1 1 I ♦ llilTl i ! 1 . ''^ 1 11 II 1 1 I 1 I 1 1 1 1 1 1 1 1 1 1 1 1 1 ' 1 1 I 1 1 ' 1 I 1 1 1 1 1 III' 1 ' 1 '1 1 II 1 1 1 ' 1 ' ' ! ! 1 ■ ; 1 1 1 " ■ 1 ! II 11 1 II 11 ] 1 ! II 1 ■ '1 I 1 1 1 1 1 1 1 1 I 1 1 1 ' 1 ' 1 ' 1 ' 1 1 I 1 1 : I ' 1 ! 1 1 1 ' 1 12 24 36 48 60 72 Time from start (hours) 84 96 Fig. 6, continued Hydro thermal Studies 341 Dep 1 oymen t #=1 1 • M 1 W /i' ! M ' i 'i i! I \ ^Yi ! il |\i: I 45 g 360 o. 315 I •270 s 225 £ 180? 135; 0 12 24 36 46 60 72 84 96 Time from start (hours) Dep I oyment #=3 .■V «' 1'', .3011 f'l'liuiiU 'y-:A 2-10 li i ilVtSMiltil; ^ 360 a 315| 270?: o 225^ 180? 135 5 0 12 24 36 48 60 72 84 96 Time from start (hours) Deployment #=5 30 1^1 1 ^^ ;'( ■ i ' ; S';0 „ t- J ,"'* : i i , 1 -■'. 1 , -i. •>- __,-<-•' 1 ij-10 i 1 < •D ^v^ i ! ! ^vYv*,, ^ i \ i TT^^ 0 v*^^ ■ 0 6 12 18 24 30 36 42 48 Hme from start (hours) ! !' ! : I' ij' i 1 1 I ! Dep I oyment #=2 H 45 g 360 a 315 I 270 s: 225 !_ 180 « 135 3 o 90" 0 12 24 36 48 60 72 84 96 Time from start (hours) Dep I oymen t #=4 30 jMli 1 : N.20 J l/|v- 1 ' i i 1 Ij-lO c 0 0 ! 45 g 360 a 315 I 270 2: o 225 _^ 180 « 135 g 90 I «r 0! 6 12 18 24 30 36 42 48 Time from start (hours) Deployment #=6 0 12 24 36 48 60 72 84 Time from start (hours) Fig. 7. Hourly observations of wind speed and direction during periods the TempMentor devices were recording water temperatures (8 October through 3 November 1987). Also indicated are times of high (t) and low (i) slack. 342 the plume, as defined by at least a 4°F (2.2°C) temperature rise in water temperature, encoun- tered both the surface and bottom thermistors (Fig. 6), usually at the time of max ebb and low slack. Deviations from this general pattern (e.g. the period 54-66 hours from start) were probably the result of a shift in wind direction (Fig. 7), however neither the surface nor bottom thermistor recorded temperature increases larger than 6°F (3.3°C). At the other extreme, during deployment #4, the data loggers experienced no temperature increases of 4°F (2.2°C) (Fig. 6), although they did record 1.5°F excursions. Surface water tem- peratures recorded during deployment #6 in- creased almost 4''F (2.2°C) just after the time of high slack, presumably just as the tide turned, and again just before low slack, if the winds were southerly (Fig. 7). Both surface and bottom thermistors experienced increased water tempera- tures just after the time of high slack during de- ployment #1. Also during this deployment, sur- face water temperature excursions were usually 4°F (2.2°C) or greater; bottom water temperatures usually increased less than 4°F (2.2°C). The tem- perature records from deployments #2 and #7 were nearly identical. During both of these de- ployments, only the surface probe experienced brief temperature increases of 4°F (2.2°C) or greater. These excursions were typically bimodal. The first increase occurred at or just after the time of high slack and may have represented warmer water from Jordan Cove passing the probe. The second peak occurred just before low slack, presumbably at or just after the time of maximum ebb and probably represented the MNPS effluent. During deployment #5, warmer water encountered the data logger for longer pe- riods of time than when they were set at locations farther away; temperature excursions rarely ex- ceeded 4°F (2.2''C). During all deployments, slight descrepancies from the general patterns just described were, very likely, the result of shifting wind patterns (see Fig. 7). Comparison of results to predictions The configuration and extent of the thermal plume as measured by dye concentrations during each tidal stage surveyed on 26 August 1987 were similar to those predicted (Fig. 8). In general, the area encompassed by the 4°F (2.2°C) isotherm was within 10 acres of the predicted area deter- mined from average conditions and was always less than the area predicted under extreme condi- tions (Table 5). Further, the distances to the 4°F (2.2°C) isotherm also tended to be similar to the average three-unit predictions and less than the extreme predictions (Table 6). Because the ther- mal plume predictions were meant to be conser- vative and because the conditions during the sur- vey were ideal, the measured plume should have been smaller than predicted. The measured and predicted plumes were most discrepant during the low slack survey (Fig. 8, Tables 5, 6). The actual start of this survey was about 15 minutes after the time of low slack rather than about an hour before low slack as was intended (Fig. 5). Thus the plume portrayed as 'low slack' was probably influenced by increas- ing flooding currents. However, as indicated pre- viously, the low slack configuration was still within the areas and distances predicted. The data from the seven temperature probe de- ployments also supported the conclusion that, al- though the actual three-unit plume is highly dy- namic, it generally conformed to the predicted configuration. For example, temperature records from deployments #5 and #3 indicated the absence of the plume during high slack, which was the predicted situation. Temperatures recorded during deployments #1, #2, #4, #6 and #7 indicated that the plume (4°F above ambient), reached White Point for short periods of time during max ebb and low slack. Again, this situation was predicted, although it was not observed during the dye sur- vey. Hydrothermal Studies 343 / \" 0 5000 M > Niontic Bay Lb r 'coir^ Black / ■■■•A ■^ Sooaide 1 i-f J i /■ ' ^ J \" / ) r^"^! 0 5000(1 / Hay.'"' \point Ssasidi _^ Black ] Twotree "^ /;, b'F .^, r ~T J Y / ) T^M 0 5000 11 ) Nion l,e Boy I—,* r" ^^\ 1 / ■1 5-F. ^. ■■■■"\ Twolree \ Poin Seoside ^ Block \ Island Pq|nlJ --X J \ ) r'^'C-,\ 0 5000 ft Nionlic Soy i _^B / ^°^* ^ ( YPoint Sea aide 1 5-F ^ Block / Twotra\ Island ^ v_ p5 ~ Fig. 8. Locations of selected three-unit thermal plume isotherms. The lefl column contains predicted locations; the right column contains the locations determined from the 26 August dye survey. The rows, from top to bottom, contain isotherms at low slack, maximum flood, high slack and maximum ebb. 344 TABLE 5. Comparison of predicted and actual surface areas encompased by selected AT isotherms pro- duced during three-unit operation. Tide stage Area in acres encompassed by each isotherm 8.0°F (4.4°C) 6.0°F (3.3°C) 4.0°F (2.2°C) 1.5°F (0.8°C) Low slack Predicted for 3 units Predicted for 3 units (extreme) Hydrothermal survey, 8/26/87 Max Flood Predicted for 3 units Predicted for 3 units (extreme) Hydrothermal survey, 8/26/87 High slack Predicted for 3 units Predicted for 3 units (extreme) Hydrothermal survey, 8/26/8 Max Ebb Predicted for 3 units Predicted for 3 units (extreme) Hydrothermal survey, 8/26/87 6.8 11.5 40.8 31.9 248.r 583.0 241.0 1,049.8- 572.8 6.9 15.3 15.4 39.1 59.0 26.4 1,252.2- 252.1 10.6 32.8 49.3 166.0 59.0 353.9 152.2 27.5 23.8 62.9 93.6 535.0 102.2 1,046.4- 443.1 1 determined from average conditions, NUSCO 198.'? 2 reported in NUSCO 1983 3 limit readied TABLE 6. Comparison of maximum predicted and actual distances to selected AT isotherms produced dur- ing three-unit operation. Distances in feet from quarry cut to each isotherm 8.0°F (4.4°C) 6.0°F (3.3°C) 4.0°F (2.2°C) 1.5° Tide stage F (0.8"C) Low slack Predicted for 3 units Predicted for 3 units (extreme) Hydrothermal survey, 8/26/87 Max Flood Predicted for 3 units Predicted for 3 units (extreme) Hydrothermal survey, 8/26/87 High slack Predicted for 3 units Predicted for 3 units (extrem.e) Hydrothermal survey, 8/26/87 Max Ebb Predicted for 3 units Predicted for 3 units (extreme) Hydrothermal survey, 8/26/87 1,078 1,580 1,963 2,275 5,968 5,428 3.760 1 0,973' 4,800 1,150 1,155 1,550 1,925 2,286 2,000 8,278 6,800 ,975 1,540 2,233 3,143 2,160 4,620 3,250 1,980 1,887 2,420 3,465 7,428 3,300 12,204- 12,500 1 determined from average conditions, NUSCO 1983 2 reported in NUSCO 1983 3 limit reached 4 4°F isotherm extends to 6,000 ft as determined from temperature recorded during deployment A feature of the plume that was apparent from the continuous temperature records, but was not readily apparent from either the dye survey results or the plume predictions was the highly dynamic nature of the plume. Even though the plume reached White Point, it influenced water temper- atures there for brief periods only and elevated temperatures did not appear to be sustained there. Further, based on both the survey and supple- mental temperature data collected thus far, water temperatures greater than 4°F above ambient do not appear at distances greater than 6,000 ft ( 1 829 Hydrothermal Studies 345 m) from the cuts. Even in Jordan Cove, the area most likely to be affected by the thermal effluent, there were respites from elevated temperatures. As temperature records are obtained from thermistors deployed at strategic times and places, our understanding of the highly dynamic thermal plume will be enhanced. Summary 1 . The configuration and extent of the thermal plume produced during three-unit operation, as measured by dye concentration, generally matched predictions during all four tidal re- gimes. 2. Based on dye concentrations at Unit 1, very little recirculation of discharge water occurs. 3. Water temperatures recorded continuously at selected locations were also generally what was expected, based on predictions. 4. The plume is highly dynamic and those re- gions influenced by increased water tempeatures generally experienced a respite from wann water for at least several hours during a tidal cycle. 5. Based on both the survey and supplemental temperature data, at no time during the pe- riod studied did the 4°F (2.2°C) isotherm appear to extend past the 8,000-ft (2439 m) limit imposed by the NPDES permit. References Cited Liang, H.C., and Tsai, C.E. 1979. Far-field thermal plume prediction for Units 1, 2 and 3, Millstone Nuclear Power Station. NERM-49, Stone and Webster Engineering Corp., Boston, MA. Millstone Point Company (MPC). 1972a. Mill- stone Nuclear Power Station Unit 2 Environmental Report, operating license stage. MPC. 1972b. Millstone Nuclear Power Station Unit 3 Environmental Report, construction permit stage. Vol. 1 and 2. NUSCO. 1979. Millstone Point Units 1 and 2 Hydrothermal Survey Report, July 25 - August 2, 1977. Submitted to NRC. . 1983. Millstone Nuclear Power Station Unit 3 Environmental Report. Operating Li- cense Stage. Vol 1-4. . 1986. Proposed ecological studies Mill- stone Nuclear Power Station, 1987. Letter to Connecticut Department of Envirormiental Protection. OSI (Ocean Surveys, Inc.). 1987. Final report, hydrothermal survey Millstone Nuclear Power Station, Waterford, CT. Ocean Surveys, Inc., Old Saybrook, CT. Pritchard-Carpenter, Consultants. 1967. Contin- uous discharge tracer study Twotree Island Channel and Niantic Bay, lx>ng Island Sound. Report to Millstone Point Company. Braincon Corporation. 1975. Hydrographic Sur- veys, August-September 1973 and February 1974. Report to Northeast Utilities Service Company. ENDECo (Environmental Devices Corporation). 1977. Postoperational Units 1 and 2, preoperational Unit 3 hydrothermal survey of the Millstone Nuclear Power Station. Report to Northeast Utilities Service Company. . 1970. Tracer study of the circulating water system, Millstone Point Unit number one. Report to Millstone Point Company. Raytheon Marine Research Laboratory. 1968. Millstone survey, August 29, 1968. Report to Millstone Point Company. . 1969. Millstone survey, March 20, 1969. Report to Millstone Point Company. 346 Stoltzenbach, K. and E. Adams. 1979. Thermal plume modeling at the Millstone Nuclear Power Station. Report to Northeast Utilities Service Company. TI (Texas Instruments, Inc.). 1977. Airborne thermal infrared survey, Millstone Point Nu- clear Power Station. Report to Northeast Util- ities Service Company VAST, Inc. 1971. June 1971 Millstone Point temperature survey. Report to Millstone Point Company. . 1972a. Dye diffusion survey. Millstone Point, Connecticut, September - November 1971. Report to Millstone Point Company. . 1972b. Thermal survey and dye study Millstone Point, Connecticut, September - November 1971. Report to Millstone Point Company. . 1972c. Study of an offshore thermal diffuser outfall by dye simulation (Phase III - 1978 unit). Report to Millstone Point Com- pany. . 1972d. Thermal survey. Millstone Point Company, Unit no. 1, March- April 1972. Re- port to Millstone Point Company. Hydrothermal Studies 347 348 Appendix to Hydrothermal Studies Appendix to Hydrothermal Studies 349 350 Appendix to Hydrothermal Studies 351 352 Appendix to Hydrothcrnial Studies 353 354 Attachment to Hydrothermal Studies Attachment to Hydrothermal Studies 355 / / X ./ / / / FINAL REPORT HYDROTHERMAL SURVEY MILLSTONE NUCLEAR POWER STATIul WATERFORD, CONNECTICUT Submitted To: Northeast Utilities Environmental Lab PO Box 128 Waterford, CT 06385 Submitted Qy Ocean Surveys , Inc . 91 Sheffield Street Old Saybrook, CT 05475 30 September 1987 TABLE OF CONTiiNTS Page 1.0 IIMTRODUCTIOH 1 2.0 PROCEDURES Al^lD EQUIPMENT . ' 2 2.1 Hurizontal Control 2 2.2 Vertical Control 2 2.3 Navigation 3 2.4 In Situ Current, Temperature, and Salinity Data 4 2.5 Dye Tracer Study 5 2.5.1 Dye Injection 5 2.5.2 Fluorescence Monitoring 6 3.0 DATA PROCESSING AND PRESENTATION 8 3.1 Survey Trackline Reconstruction 8 3.2 Tide Level Data 8 3.3 In Situ Current, Temperature, and Salinity Data 8 3.4 Dye Concentration Data 9 3.5 Background and Recirculation 9 3.6 Isothermal Data 10 4.0 DISCUSSION OF DATA 10 4.1 Background Fluorescence 10 4.2 Current, Temperature and Salinity Data . . 11 4.3 Isothermal Data 13 4.3. 1 Low Slack 14 4.3.2 Maximum Flood 14 4. 3.3 High Slack 15 4.3.4 Maximum Ebb 16 4.4 Recirculation Data 17 APPENDICES I II Equipment Specifications Ta bul ar Data FINAL REPORT HYDROTHERMAL SURVEY iILLSTONE NUCLEAR PO^JER STATION WATERFORD, CONNECTICUT 1.0 INTRODUCTION During the period 21-25 August 1987, Ocean Surveys, Inc. (OSI) performed hydrographic and dye tracer surveys for Northeast Utilities Environmental Laboratory (NUEL) at the Millstone Nuclear Power Station, Waterford, Connecticut (Figure 1). This work was undertaken to provide NUEL with contoured delta-T isotherm maps of the generating station's thermal plume during four phases of a complete tidal cycle. NUEL assisted OSI in executing this project by supplying personnel, equipment and facilities for equipment monitoring and offshore surveying. Rhodamine WT dye was injected into the Unit 3 discharge waters. Dye concentration and water temperature data were collected along 41 transects and 27 vertical profile stations in and around the thermal plume during the times of low slack, maximum flood, high slack, and maximum ebb tides. These data were analyzed to determine the positions of the 1.5 ,4,6 and 8 F delta-T isotherms. The contoured isotherms are presented by tidal phase in Figures 10 through 13 of this report . Dye concentration and temperacure data were additionally collected at the Unit 1 discharge well to monitor for the recirculation of power station cooling water and are presented in Figure 9 of this report. 1 - 6- ft :o- .^ S i? Q O U. cc u < S S ;S oo - 1 ^ -. o «= 2 / f / •:^ -^^ /• >- /'• "^ r vA- 05 ViM^-' ^. 'J '•"?;*• y I -v. \ *^ ' A ^ o •\ ^ Co Ok ^ ^ Ci ^ 01 > liJ > C£ D z < UJ u 0 r- 00 >- Ul , , ^ :^ ^ 3 o < < < - 10 _J >- C\J ■^ 1 -3 > 1 a: 3 CM >- oi m O O o si- CM Ul a: =) LJ — < ii. l/J bi. Background fluorescence, tide, salinity, temperature, and current speed and direction data were also collected prior to and during the dye concentration survey to support the development of the delta-T isotherm data. These data are presented in graphic form in Figures 4 through 8 and in tabular form in Appendix II. A small circulation study employing free-drifting drogues at multiple depths was also conducted and the results are being held for later processing. 2.0 PROCEDURES AND EQUIPMENT 2 . 1 Horizontal Control Horizontal control points easily discernible on 1:24,000 (1 " = 2,000") scale USGS Quadrangle maps were used as control stations for navigation equipment during the survey. The estimated accuracy of determining horizontal control point coordinates in this manner is j^5 0 ft in vessel locations relative to the Connecticut state grid system. 2. 2 Vertical Control OSI established a temporary tidal benchmark (TbM) at NUEL's dock facilities by correlating recorded tide levels with predicted tide elevations for Millstone Point (NOAA Tide Tables). The elevation of OSI's TBM is 5.2 ft Mean Low Water (MLW) . A Stevens Type F recording tide gauge was installed at the NUEL dock facilities and referenced to OSI's TBM. Tidal elevation data was continuously recorded during all survey operations. A specification for the Stevens Type F recording tide gauge is included in Appendix I. 2.3 Navigation Survey vessel position data were acquired using OSI's "Haretrack" trackline control system in conjunction with a Cubic DM-40A "Auto tape" electronic positioning system. OSI's "Maretrack" system consists of an Apple computer with video display and left/right indicator interfaced with the "Autotape" system. The "Autotape" system is comprised of three components: two shore-based range responders and an interrogator unit installed aboard tiie survey vessel. Range measurements to each responder are acquired from the phase comparison of microwave reference signals transmitted and received by the responders and the interrogator. The two ranges are automatically displayed in meters by the onboard interrogator and are updated at a one-second rate and input to the "Maretrack" system. The "Haretrack" program computes vessel location, distance to beginning of line, distance to end of line, vessel speed and cross-track (distance off intended survey trackline) error. Additionally, "Maretrack" indicates required course corrections in real-time through a left/right indicator. In operation, the helmsman is able to locate the vessel's position with respect to the survey trackline on the video monitor. Once on line, he can steer the desired line following the course corrections indicated on the left/right display. All positional data are automatically recorded on a paper tape printer for later processing. Where site conditions did not permit the use of the "Maretrack" system, an alternate positioning method was employed consisting of one range of the "Autotape" system and a theodolite established on NUEL's dock. The survey vessel was controlled along a bore sight representing the intended survey transect by the transit operator using a hand-held VHF - 3 - radio. Distance along the line was measured using the "Autotape" positioning system. Alternately, the helmsman steered a predetermined course while the transit operator turned angles from a known backsight. As above, all positional data were automatically recorded on paper tape with angles logged into the field notes. Specification sheets for the "Autotape" system, the "Maretrack" system, and the theodolite are provided in Append ixl. 2 . 4 In Situ Current, Temperature, and Salinity Monitoring Continuous measurements of current speed and direction, water temperature, and salinity were obtained by installing an Endeco Type 174 in situ recording current meter on each of two su rf ace- f 0 1 1 owi ng moorings deployed at stations east and southeast of Millstone Point (Figure 1). The meters were positioned 5 feet below the surface on a suspended taut line which was itself moored to the bottom (Figure 2). Two stations (JCE and JCIJ) were occupied on 24 and 25 August 1987 and two stations (TTIC and HP) were occupied on 26 August. Data was recorded in digital form on magnetic tape for later process i ng . The Endeco type 174 instrument has been designed to eliminate the considerable effects of surface waves. The ducted iinpellor cancels the action of wave induced orbital velocities while a flexible tether decouples the instrument from mooring line motion. A specification sheet for the Endeco 174 current meter is included in Appendix I. - 4 BUOY W 1/4 IN. STAINLESS WIRE' 50 LB LEAD ENDECO TYPE 1 74 CURRENT METER CURRENT METER MOORING DESIGN figure: no. SCALE N.TS DATE 29 SEPT 87 BY RJ.L. OCEAM SURVEYB,irV*C.r^^ \OBl OLD SAYBBOOK. CONNECTICUT 2 . 5 Dye Tracer Study Dye dilution studies are based on the principle that the downstream dilution of a conservative substance is directly proportional to the mixing characteristics of the receiving water body. Temperature is assumed to be conservative in this study in that the temperature of the thermal plume is reduced only through mixing with Long Island Sound water. Dilution of Rhodamine WT dye injected into the thermal plume therefore is assumed to directly mimic the temperature reduction of the plume. 2.5.1 Dye Injection Rhodamine WT dye is a fluorescent, biodegradable tracer that is extremely soluble in water and detectable in very small concentrations (less than 0.05 parts per billion). The dye was supplied as a 20 percent solution by Crompton and Knowles Corporation, Gibralter, Pennsylvania. The specific gravity of the individual lot of Rhodamine WT which OSI used at Mi 1 1 stone was 1.112 at 8 0° F. Dye was injected into the circulation water discharge well of Unit 3 utilizing a Fluid Metering, Inc. laboratory pump. The 20% solution of dye was pumped onto the surface of the discharge water. Mixing occured where the Unit 3 water entered the quarry and again where the waters from Units 1, 2, and 3 meet in the middle of the quarry. Dye injection began at 1500 hours on 23 August initially at a rate of 5 pounds per hour and was increased to 7 pounds per hour at approximately 1700 hours on 24 August 1987. Dye injection rates were monitored by weighing the dye supply reservoir at approximately one-hour intervals to an accuracy of j-0.01 pounds. A specification sheet for the Fluid Metering, Inc. pump is included in Appendix I. Vertical profiles of dye concentration were taken at three points across each of the two quarry cuts on 25 August to insure that complete mixing of the dye was occurring within the quarry. Dye concentration readings varied less than 0.05 ppb by weight indicating complete mixing. 2.5.2 Fluorescence Monitoring Dye concentrations were monitored using Turner Designs Model 10 and Turner Associates Model 111 fluorometers. Both fluoro meters provide a relative measure of the quantity of light emitted from a fluorescent solution. In principle, a lamp within the fluoro meter emits light which is filtered and allowed to strike the sample as it flow: continuously past the light source. Any dye present in the solution will fluoresce. The emitted light spectrum is passed through a secondary filter to a sensor, compared to the source light and the relative quantity of light is indicated by the fluoro meter readout. The fluorescence of dye varies with sample temperature, therefore, the water temperature in the sampling line was monitored with a Yellow Springs Instrument Company Series 700 thermistor to enable data processors to correct recorded dye concentrations for solution temperature. Both dye concentration and temperature were continuously recorded during all dye concentration surveys. Turner Model 111 fluorometers equipped with thermistors and interfaced with Soltec two-pen strip chart recorders were installed at both the quarry cut and at the Unit 1 discharge well. The data from the instrument at the quarry cut was used to determine the concentration of dye in the discliarge water for later processing of the thermal plume dye concentration data. Background fluorescence was also recorded from 1700 hours, 21 August to 1400 hours, 22 August at the quarry cut to characterize fluctuations in the natural fluorescence of Long Island Sound waters. The instrument at the Unit 1 discharge well was used to monitor for the recirculation of cooling water. Turner Model 10 fluorometers equipped with thermistors and interfaced with Soltec two-pen strip chart recorders were installed on the horizontal transect and the vertical profile boat . Surficial dye concentrations and water temperatures were recorded along 41 transects during the four tidal mapping sessions. These transects were oriented nominally perpendicular to the axis of the thermal plume and along well defined plume boundaries. Dye concentrations and water temperature were also measured at 27 vertical profile stations to characterize the vertical mixing of the thermal plume. Quarry cut dye concentrations were monitored from 1500 hours, 23 August until 1200 hours, 27 August including both the periods of dye buildup and flushing within the quarry. Buildup took 10 hours before the quarry cut concentrations stabilized at 2.2 ppb by weight. Similarly, it required 10 hours for concentration readings to stabilize after dye inject ionwasstopped. Pre- and post-survey calibrations of the fluorometers were conducted using standard solutions prepared with site water and dye drawn from the lot used for the study. These solutions were prepared employing Class A glassware which meets or exceeds National Bureau of Standards requirements. Water samples were also taken during the survey and used in the lab to verify instrument responses. Specification sheets - 7 - for the Turner Model 10 and 111 f 1 uorometers , Yellow Springs Instrument Co. Series 700 thermistor, and the Soltec recorders are included in Appendix I. 3.0 DATA PROCESSING AND PRESENTATION 3. 1 Survey Trackline Reco ns t ruction Survey tracklines vie re reconstructed from the "Autotape" ranyes and transit angles logged at each position "fix." These values, together with the grid coordinates of the responder and theodolite locations, were input into OSI's DEC PDP 11/44 computer system which calculated the X and Y coordinates for each recorded position. During calculation of the vessel positions, geometric consideration of responder elevations, interrogator antenna height, X and Y corrections for sensor layback and offset (relative to the "Autotape" antenna), and range calibration data were also input to yield the most precise computation possible. 3.2 Tide Level Data Half-hour water elevations taken from the continuous tide level chart recordings collected during the survey were referenced to the MLW datum and are presented graphically in Figure 3 and in tabular form in Appendix II. 3.3 In Situ Current, Temperature, and Salinity Data The recorded current, temperature, and salinity data were translated employing an Endeco Type 250 data translator and input into OSI's in-house computer system for processing. Each data set was then corrected for the individual instruiTient calibration and computer plotted as time series plots. Data from the 24 and 25 August deployments at Station TIDE ELEVATIONS MILLSTONE POINT -0,5 ri I I I I I I I I I I I I I I I I I I I I I 12 24 36 43 60 72 TIME FROM START (1000, 23 AUG., 1987) 84 96 FIGURE NO. SCALE NA OAT^ /29/87 BY JAH OCEAIM SUFIVEYS, IIMC. OLD SAYBROOK. CONNECTICUT JCE are presented on Figure 4 and Station JCW on Figure 5. Data collected on 26 August at Stations TTIC and MP are presented on Figures 6 and 7, respectively. These data are also presented in tabular form in Appendix II. 3.4 Dye Concentration Data Dye concentration data recorded on strip chart must be converted from fluorescence readings to dye concentrations in parts per billion (ppb) by weight. This was accomplished by first correcting the fluorescence data to a standard temperature according to the equation: where, CONC TRUE dye concentration corrected for sample temperature CONC PP = recorded fluorometer output = recorded sample temperature = standard temperature; in this case, T3 = 68°F Using calibration data for each fluorometer and the specific gravity of the dye lot, the "equivalent" dye concentration was calculated in ppb by weight. 3 . 5 Background and Recirculation Dye concentration data from the background fluorescence instrument at the quarry cut and the Unit 1 recirculation fluorometer were processed as above and are presented as time series plots in Figures 8 and 9. 9 - 35 -| 33 - 1= 31 - Q- 27 25 24 AUG 87 1200 25 1200 26 a: 25 -| 23 - < ^ 21 H Od o L^^ 19 H UJ 17 H 15 ..y^ ju^ 24 AUG 87 1200 25 1200 26 1200 27 1200 27 30 24 H LU ^ 18 ^^ 6 H 0 24 AUG 87 1200 25 1200 26 1200 27 360-] -z. o _ 270- I— o c) 180- uj ^ a: 9 -^ 90 0 K 24 AUG 87 1200 25 1200 26 1200 27 SURVEY DATE 24-25 AUG. 87 JORDAN COVE, NIANTIC STATION - JCE Figure No. 4 OCEAN SURVEYS, INC. /== < CO 35 -| 33 - 31 - 29 27 25 24 AUG 87 1200 LU CH 25 -| 23 - < ^ 21 H ^ ^ 19 H Q_ I— 15 24 AUG 87 1200 25 1200 26 1200 27 25 1200 26 1200 27 Q '- 30 -1 24 18 H CO -- ' 6 0 24 AUG 87 1200 25 1200 26 200 27 360-, o _ 270 o o 1 80- LU LU or: S -- 90 H Q 24 AUG 87 1200 '^ 25 1200 26 200 27 SURVEY DATE 24-25 AUG. 87 JORDAN COVE.NIANTIC STATION-JCW Figure No. 5 OCEAN SURVEYS, INC. fC: 35 1 >- 33 - ^ ;= 31 - ::e:29 - < OD 27 - O^ - 26 AUG 87 '200 27 1 200 1 28 1 200 29 LU 25 n ^ 23 - 1— < ^ 21 - a: o S 17 - 1 «; ^r 26 AUG 87 1200 27 1 200 1 28 1 200 29 90 n 72 - Q -- LU 5^ 54 - ^5^36 - D r 18 - 1 ^ u -| 26 AUG 87 1200 27 1200 1 28 1 200 29 360-, O _ 270- — t- 1— o o 1 80- LU^ en 9 -- 90 - Q ^^ L 0 - 26 AUG 87 1200 27 1 200 1 28 1 200 29 SURVEY DATE TWOTREE ISLAND CHANNEL STATION-TTIC Figure No. 6 NIANTIC 26 AUG. 8 7 OCEAN SURVEYS, INC. 91 SHEFFIELD ST , OLD SAYBROOK. CT 06475 TEL (2031 386-4631 TLX 5106013995 m 35 33 H P 31 a. 27 25 26 AUG 87 1200 27 200 28 1200 29 25 n 23 - < ^ 21 H en o Ll) 17 -I 15 26 AUG 87 1200 27 200 28 1200 29 90 -| 72 - LU ^ 54 - '^ " 36 - 18 26 AUG 87 1200 27 1200 28 1200 29 360-| o _ 270- o o 1 80- uj "ij Ql 9 -- 90-1 Q f*!'^-^ 26 AUG 87 1200 27 1200 28 1200 29 SURVEY DATE M ILLSTONE POINT , Nl ANTIC 26 AUG. 87 STATION- MP Figure No. 7 background vs tide quhkry cut — FLUOR. - - TIDE ELEV. 0.16 — 0.12 0.08 0.04 -0.04 2.5.- 1.5 4 8 12 16 TU'lE FROM START (1600, 21 AUG., 1987) FIGURE NO. SCALE NA DATE 9/29/87 BY JAH OCEAN 8UFIVEY8, IMC. OLD SAYBROOK. CONNECTICUT (X lE-3) 144 TIDE VS DYE CONCEHTSATI OH UNIT i DISCHARGE DYE cone. TII/E ELEy. TTI|II^\|III|II||||||||||| ||||| || ||||||| ||||||||l I I ||||||| ||_ 1 1 II 1 1 lil 1.9 1.4 o.s 0.4 -0.1 0 6 12 18 24 30 36 42 4a 54 60 66 72 73 84 yC- TIME FROM START (1100, 23 AUG., 198?) FIGURE NO. SCALE NA DATE 9/29/87 BY JAH OCEAN SUFIVEYB, INC. OUD SAVBROOK. CONNECTICUT 3.6 Isothermal Data Since both dye concentration and water temperature Are held to be convervat i ve , the percent drop in the temperature of the plume with respect to the total temperature difference between the discharge water and the receiving body of water is equal to the percent dilution of the dye in the plume with respect to the dye concentration of the discharge water at the quarry cut. Dye concentrations in ppb by weight were converted to degrees Farenheit above ambient temperature and plotted on a trackline map for each survey. These data were contoured at the 1.5 ,4,5 and 8 F delta-T levels and are presented on Figures 10 through 13. Data from the vertical profiles at increasing distances from the quarry cut along the axis of the plume were also developed and plotted as delta-T versus depth. These profiles are presented on Figures 14 through 17. Additional vertical profiles not associated with the axis of the plume are also presented on Figuresl4throughl7. 4.0 DISCUSSION OF DATA 4.1 Background Fluorescence Prior to initiating dye injection, the natural or background fluorescence of the circulation water was recorded at the quarry cut on Millstone Nuclear Power Station. Tide elevations were also measured from 1700 hours, 21 August to 1400 hours, 22 August, 1987 and are presented together with the background fluorescence data on Figure 8. The background fluorescence remains near zero for the first 14 hours of the study. From 0700, 22 August the background fluorescence is seen to rise, coinciding with d storm event which occurred that day. The increase in fluorescence may represent the presence of naturally fluorescent material introduced into - 10 - the water by a combination of sea state and meteorological conditions. Background fluorescence had returned to pre-storni levels by the time dye injection began on 23 Au gu St . No correlation is found between tidal phase and background fluorescence as has been observed on similar surveys conducted by OSI . 4.2 Current, Temperature, and Salinity Data In situ recording current tneters were installed at Stations JCE and JCW (Figure 1) for approximately 10 hours each on 24 and 25 August. These data were collected to characterize the tidal currents and associated temperature and salinity at the entrance to Jordan Cove to the east of Millstone Point ( Fi gures 4 and 5 ) . As indicated on Figure 4, a well defined NNE-SSW trending ti dally driven current is developed between Fox Island and High Rock. Similarly, Figure 5 shows an ENE-SU tidal current developed between High Rock and Flat Rock. Flood tidal currents to the NNE and ENE have a greater speed than the SSW and SU ebb currents. Water temperatures on the flood tide tend to be lower than the ebb tide, representing the difference between Long Island Sound/Block Island Sound water and water which has been heated in relatively shallow Niantic Bay and Jordan Cove. Salinities remain relatively constant indicating no significant influx of fresh water runoff or excessive evaporation. To augment the dye concentration survey, recording current meters were deployed at Station TTIC in Twotree Island Channel off White Point and at Station MP directly off Millstone Point on 26 August. Millstone Point data (Figure 11 7) indicates a well defined E-W tidal current system with westerly flood tidal currents having a yreater peak speed than the easterly ebb currents. Temperature and salinity at Station MP are relatively constant due to the station's location in Lony Island Sound. Short duration thermal peaks appear to be associated witli periods of slack currents. These peaks probably represent the influence of the thermal plume from liillstone Point as it reaches further offshore during times of tidal current reversal. StationMP does map inside the 1.5 isotherm during high slack tide (Figure 12) and inside the 4 isotherm during low slack tide (Figure 10). Data from the Twotree Island Channel station (TTIC) (Figure 5) show the tidal currents are WNW-ESE in direction. Peak ebb current speeds to the ESE Are greater than the flood current, possibly due to the flow being restricted between the mainland to the nortii and Bartlett's reef to the south. The temperature of the ebb currents Are approximately 2-,4°F greater than the flood currents. This difference is too great to be due only to thermal plume water being carried down the channel. Thermal mapping discussed later in this report indicates that only 1.5-2°F can be attributed to the thermal plume. The additional temperature difference is probably attributable to cooler waters from Block Island Sound brought in on the flood tide. Mixing with warmer Long Island Sound water would serve to raise water temperatures for the subsequent ebo tide. The recorded times of low slack tide and maximum flood currents for Twotree Island Channel are within a half-hour of predicted times but actual times of high slack tide and maximum ebb currents are up to 1.75 hours later than 12 - predicted. The recorded times of current reversal and maximum currents at Station MP are approximately 1 hour later than at Station TTIC. 4.3 Isothermal Data Dye concentration data were collected along 41 tracklines and 27 vertical profiles during the 4 major tidal phases on 25 August. These data were converted to degrees Fahrenheit above ambient water temperatures, contoured at 1.5 , 4 , 5 , and 8 delta-T isotherms and are presented on Figures 10 through 13 and as vertical profiles on Figures 14 through 17. The term "delta-T isotherm" will be abbreviated to simply "isotherm" for the remainder of this report. The dye concentration mapping was conducted under nearly ideal weather conditions. Seas were calm (less than 1 ft), reducing thermal plume mixing due to wave action to a minimum. The light winds, averaging 2 meters per second, assures that the dispersal of the thermal plume was due predominantly to tidal influences. Conditions were good enough to allow surface expressions of the plume boundaries to be observed during the maximum flood, high slack, and maximum ebb stages of the tiae. A further benefit of the good survey conditions is the development of the 1.5° isotherm. Greater wind and wave action would have diminished the lateral extent of this isotherm. As it was, error-free equipment operation combined with excellent site conditions to allowed for clear definition of the 1.5 isotherm. - 13 4.3.1 Low Slack The lort slack isotherm map shown on Figure 10 indicates the thermal plume is fairly well distributed about the discharge point at the quarry cut on Millstone Point. The 4° isotherm extends up to 3500' directly off the quarry cut and continues around to the north into the eastern portion of Jordan Cove. The 6 and 8° isotherms define the edges of the jet itself. The vertical profiles (Figure 14) along the axis of the plume show that the 4° isotherm extends to the bottom up to 500' from the quarry cut (Stations LSI and LS2). At 2,000' feet from the quarry cut (Station LS3), the base of the thermal plume is 10'-15' below the surface as indicated by the 1.5° isotherm. A similar trend is observed on Profiles LS4 and LS5 3500' off the quarry cut. The vertical profile located in the western entrance to Jordan Cove (LS6) indicates a relatively high delta-T at the surface which is not supported by horizontal transect data. This station was sampled at the start of the low slack tide mapping session and probably represents a remnant of the previous maximum ebb tide. Similarly, the 1.5° delta-T observed throughout the water column at Station LS7 also represents the influence of the preceeding ebb tide. Since horizontal transect data collected just offshore from this station recorded delta-T's of less than 1.5°, Station LS7 was probably located in an isolated remnant of ebb tide waters trapped against the shore. 4.3.2 lax i mum Flood During the maximum flood tidal stage (Figure 11) strong currents carry the thermal plume to the west. Mixing dilutes the plume so that the 4 isotherm extends a maximum of 2,000' - 14 DEPTH vs DELTA T LOU SLhCK tide -3,3 r I (\ I I I I I I I I I I I 1 — I — I — I — I — I — L 10 12 delta T DEPTH vs DELTA T LOU SLACK TIDE FIGURE NO. 14 SCALE NA DATE 9/29/87 BY JAH OCEAIM SURVEYS, INC. OLD SAYBROOK. CONNECTICUT to the west in a narrow tongue. The 1.5 isotherm is observed much farther to the west where it appears to begin separating into multiple branches. A well defined eastern edge to the plume as defined by a change in surface water texture and a line of foam on the surface was observed in the field. Dye concentration readings dropped to background levels as this line was crossed. Vertical profiles (Figure 15) through the plume show the 4° isotherm extends to the bottom at Station MFl but is at less than 10' deep further out in the plume at Stations HFS and 4. At Station MF5 south of White Rock, the base of the plume is only 5' deep as indicated by the 1.5° isotherm. Station MF2 was located to the east of the visible plume edge and recorded background levels only. Station MF5 located to the west of Millstone Point indicates a trace of the plume at the surface. This was not seen on the horizontal mapping data and is believed to represent the effects of a minor eddy on the northern edge of the plume. 4. 3. 3 High Slack During high slack tide (Figure 12) the thermal plume is relatively evenly distributed about the discharge point and the center of the plume is shorter and broader than seen on previous tidal phases. This difference between the core of the high slack plume and the longer, narrower core of the low slack plume is probably due to the change in tide elevations. During low slack tide the elevation difference between the quarry cut and Long Island Sound is greatest, producing higher discharge currents at the cut than during high slack tide. The greater currents would tend to carry the plume jet further offshore. Lower current speeds during high slack tide would allow the plume jet to spread out more quickly, producing the shorter, broader pattern observed. The 4 15 - DEPTH vs DELTA T fttXlMUM FLOOD TIDE 12 14 DELTA T DEPTH vs DELTA T MAXIMUM FLOOD TIDE DELTA T FIGURE NO. 15 SCALE NA DATE 9/29/87 BY JAH OCEAN SURVEYS, INC. OLD SAYBROOK. CONNECTICUT DEPTH vs DELTA T HIGH SLACK TIDE J I I I I I I I I I I I I I I I I I I I M I I I I M I I M I I I I I I I I I I I l_ MSlI -30 ^" "I 'I ''''''■''''''''■' 'I ' 'I' I ' I I '' I I I I I I ' I I I Ol234567tl9 lELTA T DEPTH vs DELTft. T HICH SLACK TIDE FIOURC NO. 16 SCALE NA DATE iiimL BY JAH OCEAN SURVEYS, INC. OLD SAVKROOiC. CONNECTICUT isotherm extends up to 2,100' offshore while the 1.5 isotherm is approximately 3,000' off the cut. An extension of the 1.5 isotherm to the west is probably a remnant of the previous flood tide. Vertical profiles (Figure 16) indicate the base of the plume close to shore is approximately 20' deep and decreases to about 10' beyond 1,000 feet from shore. Station HS5 in Jordan Cove shows the base of the plume to be quite shallow; less than 5' deep. Station HS6 located off the Unit 1 intake shows no major influence from the thermal plume. A small amount of dye was detected throughout the water column indicating minor recirculation, but the uniform distribution indicates complete vertical mixiny of the recirculated water. 4 . 3'. 4 laximum Ebb During maximum ebb tide (Figure 13) the thermal plume is carried to the east toward Twotree Island Channel. The 4 isotherm extends up to 2,500' to the SE and extends northeast into Jordan Cove. The 6° isotherm closely follows the 4° isotherm while the 8° isotherm shows greater areal extent than observed on previous tidal phases. The 1.5° isotherm was mapped as a narrow finger extending up to 12,500' to the SE through Twotree Island Channel. The 4 and 1.5 isotherms in Jordan Cove are limited to the eastern portion of the cove. No dye was detected in the northern or western parts of the cove. Vertical profiles through the plume (Figure 17) show it to have a depth of 15 to 20' within 500' offshore (Profiles MEl and ME2). Beyond 1,000' from shore the plume as defined by the 4° isotherm extends down to approximately 5' (Profiles ME3 and ME4). - 16 - DEPTH vs DELTA T HAXIMUH EBB TIDE -5 -iO — D E P -15 T H -20 DEPTH vs DELTA T MAXIMUM EBB TIDE -30 2 4 DELTA T _ 1 ' 1 1 ' ' . ' 1 1 1 1 _ ~" /mE7 ME6 - — r^. — — f^- — 1 •— ~ 1 1 1 1 1 . . 1 1 1 1 .1.. - FIGURE NO. 17 SCALE NA DATE 9/29/87 Bf J AH OCEAIM 8UR\/EY8. IIMC. OLD SAYBROOK. CONNECTICUT At approximately 1000 hours on 26 August, Unit 1 experienced problems forcing a reactor shutdown and a partial shutdown of the circulating system. The total circulating system was operating at 94% capacity at 1200 hours resulting in a 6% reduction in the volume of water discharged at the quarry cut during the high slack tide mapping session. Discharge volume hdQ dropped to 89% of capacity by the beginning of the maximum ebb tide mapping session at 1430. Considering the nearly ideal survey conditions, these reductions in discharge should not greatly alter the patterns and magnitudes of the thermal plume as they are presented. Adverse wind and sea conditions could be expected to have a much greater impact on the thermal plume. 4.4 Re circulation Data Dye concentration monitoring =", t the Unit 1 discharge was begun at 1300 on 23 August to determine whether discharge water was recirculating back through the system. Only background levels were recorded for the first 26 hours (Figure 9) after which dye concentrations rose to approximately 0.12 ppb where they remained for the remainder of the survey. Dye concentrations at the quarry cut reached a relatively stable level 3 hours after dye injection began at 1500 on 23 August-so that it took between 21 and 24 hours for dye to be sensed at the Unit 1 discharge. Dye concentrations recorded at Unit 1 equate to a temperature increase of less than 1°F in the waters being recirculated through Unit 1. This may be an over-estimation of the actual level of thermal recirculation since radiant cooling of the water during the 21-24 hour transit time would reduce the temperature of the water without reducing the dye concentration. Mechanisms for temperature exchange should be considered with recirculation times of this length. 17 - APPENDIX I EQUIPMENT SPECIFICATIONS (NOT INCLUDED: FOR THIS INFORMATION CONTACT THE NU ENVIRONMENTAL LAB) APPENDIX II TABULAR DATA TIDE DATA MILLSTONE POINT 23 August 1987 Time Tide (Local ) (ft) 1100 2.50 1130 2.40 1200 2.27 1230 2.18 1300 2.05 1330 1.81 1400 1.50 1430 1.20 1500 1.01 1530 0.88 1600 0.79 1530 0.76 1700 0.88 Time Tide (Local ) (ft) 1730 1.00 1800 ■ 1.25 1830 1.44 1900 1.70 1930 1.95 2000 2.22 2030 2.43 2100 2.62 2130 2.77 2200 2.90 2230 2.83 2300 2.72 2330 2.59 24 August 1987 Time Tide (Local ) (ft) 0 2.35 30 2.14 100 1.83 130 1.55 200 1.20 230 0.87 300 0.60 330 0.35 400 0.16 430 0.06 500 -0.03 530 0.07 600 0.25 630 0.54 700 0.78 730 1.10 800 1.37 830 1.69 900 1.99 930 2.23 1000 2.40 1030 2.57 1100 2.60 1130 2.60 Time Tide ( Local ) (ft) 1200 2.50 1230 2.24 1300 2.08 1330 1.88 1400 1.62 1430 1.33 1500 1.09 1530 0.90 1600 0.70 1630 0.63 1700 0.62 1730 0.62 1800 0.85 1830 1.05 1900 1.25 1930 1.49 2000 1.76 2030 2.01 2100 2.25 2130 2.48 2200 2.64 2230 2.75 2300 2.77 233U 2.64 TIDE DATA MILLSTONE POINT 25 August 1987 Time Tide (Local ) (ft) 0 2.51 30 2.30 100 2.10 130 1.85 200 1.57 230 1.29 300 0.99 330 0.70 400 0.42 430 0.21 500 0.05 530 0.00 600 0.19 630 0.42 700 0.70 730 0.98 800 1.25 830 1.58 900 1.95 930 2.20 1000 2.43 1030 2.55 1100 2.60 1130 2.56 Time Tide (Local ) (ft) 1200 2.40 1230 2.21 1300 2.03 1330 1.85 1400 1.60 1430 1.30 1500 1.00 1530 0.72 1600 0.55 1630 0.39 1700 0.26 1730 0.24 1800 0.34 1830 0.53 1900 0.78 1930 0.98 2000 1.25 2030 1.53 2100 1.80 2130 2.05 2200 2.28 2230 2.45 2300 2.59 2330 2.60 25 August 1987 Ti me Tide (Local ) (ft) 0 2.54 30 2.38 100 2.15 130 1.90 200 1.59 230 1.41 300 1.08 330 0.80 400 0.56 430 0.35 500 0.15 530 0.05 600 0.15 630 0.26 Time Tide (Local ) (ft) 1200 2.78 1230 2.66 1300 2.27 1330 2.27 1400 2.05 1430 1.80 1500 1.50 1530 1.26 1600 0.94 1630 0.68 1700 0.46 1730 0.26 1800 0.17 1830 0.23 26 August 1987 Time Tide {Local ) (ft) 700 0.50 730 0.78 800 1.09 830 1.45 900 1.71 930 1.99 1000 2.30 1030 2.50 1100 2.70 1130 2.78 Continued) Time Tide (Local ) (ft) 1900 0.40 1930 0.64 2000 0.86 2030 1.11 2100 1.38 2130 1.63 2200 1.95 2230 2.20 2300 2.40 2330 2.60 CURRENT, TEMPFRATl.IRE AND SALINITY DATA MILLSTONE THERMAL STUDY JORDAN COVE EABT 24 AUG 1987 CURRENT CURRENT WATER TIME fJPEED SPEED DIRECTION TEMPERATURE SALINITY ( EDT ) (KTS) (CM/S) (DF:G. TRUE) (F) (C) (PPT) 1030 0.08 4.0 171 66.3 19.0 30.2 104 4 0,06 3.0 159 66.0 18,9 30,2 10?3 0.04 1 .9 145 65.7 18.7 30.2 1112 . 0.04 2.3 90 65.5 18,6 30.2 1126 0.09 4.6 106 65.8 18.8 30.3 1140 0.12 6.0 124 66.0 IB. 9 30.3 1154 0.12 6.3 125 65.7 13.7 30.3 J208 0.11 5.8 123 65.4 18,6 30.3 1222 0. 11 5 .6 119 65.2 18.4 30.3 1236 0.11 5.4 114 65.1 18.4 30.3 1250 0.11 .J . / 99 65.2 18.4 30.3 1304 0 .24 12.1 84 65.7 18.7 30.3 1318 0.40 20.4 74 65.8 18.8 30.3 1332 0.29 14.9 77 65.4 18.6 30.3 1346 0.19 9.7 77 65.5 18.6 30.3 14 00 0.26 13.5 81 65.9 18.8 30,3 1414 0.29 14.7 74 66.2 19.0 30.3 1428 0.18 9.4 87 66.6 19.2 30.4 1442 0.30 1 5 . 6 86 66.5 19.2 30.3 14 56 0.32 16.4 88 67.3 19.6 30.3 1510 0,32 16.3 82 67.9 20.0 30.3 1524 0.30 15.4 86 67.8 19.9 30.3 1538 0.22 11.3 74 66.7 19.3 30.2 3 5 52 0.17 8.8 75 66.8 19.3 30,3 1606 0 . 20 10.5 71 67.7 19.8 30.3 1620 0.34 17.3 69 67.1 19.5 30,2 1634 0.25 12.7 59 66.4 19.1 30.2 1640 0.20 10.2 42 66.3 19,1 30.3 1702 0.14 7.2 29 66.2 19.0 30.3 1716 0.14 7.3 32 66.4 19.1 30.2 1730 0. 15 7.6 24 66.3 19.1 30.2 1744 0.13 6.8 8 66.1 19.0 30.2 1758 0.13 6.6 356 66.3 1. 9 . 0 30.2 1812 0.12 6.4 336 66.3 19.1 30.2 1326 0.15 7.7 327 66.3 19.1 30.2 1840 0.19 10.0 318 66.5 19.2 30.2 1854 0.22 11.5 314 67.0 19.4 30,2 1908 0.18 9.3 286 68.0 20.0 30,3 1922 0. 19 9.9 290 67.3 19.6 30.2 CURRENT i> TFMPERATURE AMD SALINITY TiATA MILLS rONE THERMAL STUDY JORDAN COVE WEST 24 ALIG 1987 CURRENT CURRENT WATER TIHE SPEED SPEED DIRECTION TEMPERATURE SAL INITY ( EDT ) (KTS) (CM/S) (DEC. TRUE) (E) (C)