Summary
The freshwater crustacean Mysis relicta (M. relicta) is an important component of the Cayuga Lake food web. This small pelagic crustacean occurs in the lake's deep cold water (the hypolimnion) and is present in southern Cayuga Lake in the region of the proposed Lake Source Cooling (LSC) intake. The Draft Environmental Impact Statement (DEIS) gives special consideration to this species due to its importance in the food web and its potential vulnerability being entrained by (drawn into) the LSC intake. The LSC research team completed a program of field investigations to assess the animal's life history and distribution within Cayuga Lake in order to evaluate the potential impact of the LSC project. Additional field investigations were conducted to assess the effectiveness of a lighted intake at reducing entrainment of M. relicta.
Based on the results of these studies, we are able to draw the following conclusions:
· M. relicta are present in southern Cayuga Lake at the depth of the proposed LSC intake.
· Distribution of the zooplankton is vertically layered and patchy, not uniform, throughout the hypolimnion. However, M. relicta are not found in higher densities in southern Cayuga Lake in the region of the proposed LSC intake, as compared to their distribution lakewide.
· M. relicta will avoid artificial light at low levels. Installation of a light on the LSC intake would create a region surrounding the intake that M. relicta would actively avoid (an exclusion zone). A low power, eight-watt bulb on the intake will create an exclusion zone large enough to enable M. relicta to avoid the flow field induced by the LSC intake and of low enough intensity not to attract other species.
These results have led us to conclude that the LSC intake can be designed to minimize the potential entrainment of M. relicta. With proper mitigation, we anticipate no adverse impact on the population of this organism. Devices will be installed to monitor any entrainment of this species so that action can be taken to correct potential problems.
2.3.4.1 Introduction.
Mysis relicta is one of three species of freshwater crustaceans in the family Mysidacea, also known as the opossum shrimp (Figure 2.3.4-1). This species is circumpolar, occurring in deep oligotrophic and mesotrophic lakes in the northern regions of North America, as well as the northern regions of Eurasia (Pennak 1989).
Mysis relicta is an important component of the Cayuga Lake food web. The species is known to be an important food source for juvenile lake trout (Salvelinus namaycush), alewife (Alosa pseudoharengus), and smelt (Osmerus mordax) in Cayuga Lake (Brownell 1970). M. relicta is an opportunistic omnivore that feeds on detritus, benthic invertebrates, phytoplankton and zooplankton (Lasenby, Northcote, and Furst 1986). M. relicta probably compete with lake trout, alewife, and smelt for zooplankton (Johannson 1992).
A Cornell University masters thesis (Brownell 1970) is the source of much of the information that exists about Cayuga Lake M. relicta. These hypolimnetic crustaceans migrate vertically through the water column over each 24-hour day. During daylight hours, the animals are found at or near the bottom of lakes. In very deep waters, M. relicta are suspended in the lower regions of the hypolimnion during the day (Brownell 1970; Robertson, Powers, and Anderson 1968; Holmquist 1959). As daylight fades, the animals begin to ascend to shallower depths to feed. M. relicta migrations of up to 120 meters (m) (394 feet [ft]) per night have been recorded in Cayuga Lake near the New York State Electric and Gas Milliken Station (Brownell 1970). This vertical migration can occur at a rapid pace, with the animals reaching speeds of 24 to 48 meters per hour (m/hour) (79 to 175 feet per hour [ft/hour]) (Brownell 1970).
This daily migration behavior enables M. relicta to forage high in the water column during the night, when less light is present, making the animals less visible to predators. During the daytime, when light penetrates further into the water column, M. relicta avoid visual predators by traveling to the deeper, darker waters. In addition to offering protection from predators, the lower temperatures of the deep water may minimize M. relicta's metabolic costs. The benthos also provides an additional food source.
The diurnal migration is a manifestation of the animal's behavioral adaptation to avoid light at very low levels. Rudstam et al. (1989) demonstrated that a similar species (M. mixta) avoids light intensities greater than 10-4 lux and temperatures above 17°C (63°F). M. relicta display a similar avoidance (Teraguchi, Hasler, and Beeton 1975; Beeton and Bowers 1982). M. relicta have also been shown to retreat to deeper depths in the presence of an artificial light source (Teraguchi, Hasler, and Beeton 1975). This behavioral adaptation keeps M. relicta out of areas where visual predators would find them easy prey.
The actual time of migration is dependent on seasonal patterns of illumination (Brownell 1970). The upper limit of M. relicta migration may vary due to changes in light intensities, temperature, prey abundance, and predator abundance. M. relicta avoid warm temperatures and sharp gradients in temperature associated with the thermocline, and are therefore more likely to be found higher in the water column at night during destratified conditions than during stratified conditions (Brownell 1970). Juveniles migrate to shallower depths than adults, and gravid females tend to stay closer to the bottom (Brownell 1970; Moen and Langeland 1989).
Brownell (1970) found that the breeding season for M. relicta in Cayuga Lake occurs primarily from October through May. The young are released from the female's brood pouch two to three months later at a length of 3 to 4 millimeters (mm), and grow at a rate of 1 to 1.5 mm per month. This suggests that small M. relicta are released into the water column between December and August. Brownell's findings also provide some evidence of a release of young in September, since many individuals 5 and 6 mm in length were present in an October 24 sample. Brownell concluded that the Cayuga Lake population consists of one cohort with an overlapping breeding cycle that spans several months.
The average length at sexual maturity for the species is 12 to 14 mm for males and 14 to 16 mm for females. Males die shortly after mating. Individuals in the Cayuga Lake population of M. relicta can reach sexual maturity in 13 to 16 months, and females have a gestation time of 2 to 3 months before releasing young. The M. relicta life span in Cayuga Lake was reported to be 16 to 24 months (Brownell 1970).
2.3.4.2 Objectives and Approach of the LSC Field Investigations.
Lake Source Cooling (LSC) field investigations were designed to provide answers to questions relative to the proposed LSC system design: How much time (over the day and over the year) do M. relicta spend in close proximity to the LSC intake, and consequently vulnerable to entrainment? Is there a mechanism for keeping the hypolimnetic crustacean away from the intake? How large an exclusion zone would be required to keep Mysis relicta from being entrained by the hydrodynamic flow field induced by the LSC intake?
Dr. Lars Rudstam, of the Cornell Biological Field Station, and Mr. Gideon Gal, a Cornell University graduate student, provided technical leadership throughout the three years of field work in support of the LSC field program for M. relicta. Dr. Rudstam directs a research program on the Lake Ontario M. relicta population as well. Their advice and knowledge were essential throughout the experimental design, data collection, data analysis, and report phases of the project.
Specific activities conducted to determine potential impacts of LSC on M. relicta included the following:
· Evaluation of collected samples of Mysis relicta at different seasons and depths to evaluate the seasonal reproduction pattern, recruitment, growth, and life-span of the organism. These life history data were used in the evaluation of the potential impact of LSC-induced mortality on the population.
· Measurement of the lakewide distribution and abundance of M. relicta over different seasons to determine in what proportion they occur in the lake's southern basin, the area near the LSC intake.
· Monitoring of the vertical distribution and abundance of M. relicta in the water column to determine when, in what concentrations, and for how long the animals are likely to be present at the depth and location of the proposed LSC intake.
· Testing of the effect of an artificial light source on vertical migration and distribution of M. relicta to evaluate whether their light avoidance behavior could be exploited to minimize their entrainment by the LSC system.
· Examination of the projections of the hydrodynamic flow field that would be created by the LSC intake and comparison of the induced water velocity (as a function of distance from the intake) to literature values of the swimming speed of M. relicta. From this evaluation, we estimated the exclusion zone needed to protect M. relicta from entrainment.
LSC researchers carried out this series of experiments during the years 1994, 1995, and 1996. Details of the experimental protocols, equipment, and data analyses are included in Appendix C-12. Data gathered in 1994 were utilized to develop methodologies for the 1995 and 1996 field seasons. Both hydroacoustic surveys and vertical net tows were used to gather data regarding the distribution and abundance of M. relicta and their response to artificial light sources. Biological assessments to provide life history information were made on animals collected in the net tows. The field investigations are summarized in Table 2.3.4-1. Data on life history information can be found in Appendix C-2.
2.3.4.2.1 Hydroacoustic Surveys.
The hydroacoustic surveys were conducted with the use of a high-frequency echosounder, which generates sound pulses (pings) that are directed into the water column. When the pulse hits a target, a portion of the incident sound is reflected (backscattered) back to the instrument. These surveys were used to detect dense groups (or "patches") of M. relicta, which occur as suspended layers within the water column.The hydroacoustic data are converted to volume backscattering strength (the proportion of the incident sound that is reflected back to the transducer by targets within one cubic meter of water), denoted as sv. This quantity is expressed in decibels (dB) and denoted by capital SV (SV = 10 log sv). The sv is related to the density of targets (organisms) by: sv = r×sbs, where r is the density (animals/m3) and sbs is the average backscattering cross section (the ratio of incident sound intensity reflected back to the transducer from a single target). This is also expressed in dB and denoted as TS (target strength, TS = 10 log sbs ). If sv and sbs are known, it is possible to calculate the density of organisms in the water column.
2.3.4.2.2 Verification of Hydroacoustic Data with Net Data.
Net tows were often employed in conjunction with the hydroacoustic surveys to confirm the identity and relative magnitude of these acoustic targets. These net samples were also used to compare volume backscattering (Sv), as measured by the hydroacoustic equipment to measured density (net data) of M. relicta. The comparisons of Sv to M. relicta density, as measured by the net surveys, and Sv to M. relicta biomass, again measured by net surveys, are graphically displayed in Figure 2.3.4-2 and Figure 2.3.4-3.Both the hydroacoustic and the net sampling data are subject to measurement error. The hydroacoustic signal has an intrinsic limit of detection, so organisms must be a minimum size to be detected by the equipment. For the 420 kilohertz (kHz) unit, we determined that the minimum detection limit was on the order of 5 mm. M. relicta below this size were not detected in the water column unless they were concentrated in a narrow band. Once M. relicta are about 5 mm long, however, they are reliably detected using the 420 kHz echosounders. The hydroacoustic signal does not discriminate between organisms of approximately the same size. For example, large zooplankton such as Leptodora sp. were included in the signal, but not enumerated in the net data.
Overall, errors in the net samples tend to bias the result low. The mesh size of the nets (0.5 mm) would efficiently capture even the smallest M. relicta. However, net avoidance, especially of the closing net, is potentially significant. The LSC team attempted to minimize the potential for M. relicta to avoid or escape the net by drawing it through the water column at a uniform rate of 0.3 meters per second (m/sec), or 1 foot per second (ft/sec).
Three types of light experiments were completed: (1) shallow water (1995); (2) benthic (1995); and (3) long-term benthic (1996). The shallow water light experiment, which was conducted at 25 m depth, was designed to determine if light bulbs of increasing wattage would have an increasing zone of influence on the distribution of M. relicta during the night, when the animals tend to form well-defined layers in the water column. We predicted that M. relicta would avoid a light source from above, as tested in the shallow water experiment, since it simulates reactions to natural light conditions to which the species is exposed each day. In contrast, the benthic experiment (at 65 m [213 ft] depth) was designed to test the animals' response to a light source from below in deep water. The long-term benthic experiment, occurring at a depth of 60 m (197 ft), was designed to determine if the mitigating effect of light was reduced by long-term exposure.
2.3.4.3 Findings of the LSC Field Investigations.
2.3.4.3.1 Life History of M. relicta in Cayuga Lake.
Net samples collected during the 1995 and 1996 field program were analyzed for life history information. Adults and late-instar immatures (subadults) were identified based on the development of sexual structures as described in Reynolds and DeGraeve (1972). Males were either counted as adults (exopod on the fourth pleopod fully developed) or subadult (fourth pleopod enlarged, but exopod not fully developed) (Figure 2.3.4-1). Females were counted as gravid (fully developed oostegites and brood pouches, with egg or embryos), marsupial (enlarged oostegites and well-developed brood pouches, but no eggs or embryos), adult (with oostegites well developed and overlapping, but brood pouch not expanded or only slightly enlarged), or subadult (oostegites visible but not fully developed, not meeting at midline).Lengths of individuals and the number in each breeding stage were observed for the entire sample (if the sample size was less than or equal to 100) or a randomly selected subsample of at least 100 individuals (if the sample size was greater than 100). After counting, the samples were returned to their original containers and the preservative was replaced.
The review of the 1995 and 1996 net tow data revealed that the Cayuga Lake population of M. relicta is composed of two overlapping generations or cohorts. Each cohort has a generation time of 18 to 24 months. The animals are released as 3 to 4 mm juveniles between October and April, and reach sexual maturity in July to October of the following year. Thus, at any one time, there are three cohorts that overlap in size: the winter cohort, the yearling (age one), and the adults (Figure 2.3.4-4). Some females may also spawn a second time in the summer. This is consistent with Brownell's (1970) description of young present in the water column from October through May. Young were present in the water column during the majority of the year (1995 and 1996), with strong peaks in October and May samples.
Data from the 1995 and 1996 samples are consistent with Brownell's reported summer growth rates of 1 to 1.5 mm/month or 0.033 to 0.050 mm/day. Winter growth rate is approximately one-third to one-half of the summer value, thus retarding growth and sexual maturity to maintain an 18- to 24-month cycle. Length at sexual maturity was found to be consistent with the 12- to 14-mm length for males and the 14- to 16-mm length for females reported by Brownell (1970).
The life history information gathered through the 1995 and 1996 study is generally consistent with results reported in Brownell (1970). The major difference is in timing of reproduction and the number of generations. Brownell (1970) reported the release of only one generation of young, which occurred from October to May. Analysis of the LSC team's 1995 and 1996 data suggest two overlapping generations: one in which juveniles were released in February to March, and one in which juveniles were released in August to October. These generations can be seen as peaks in the size distribution displayed on Figure 2.3.4-4.
2.3.4.3.2 Lakewide Distribution and Abundance of Cayuga Lake M. relicta.
The concern that entrainment in southern Cayuga Lake could have a disproportionate effect on the lake's overall population led the LSC team to quantify the lakewide distribution of M. relicta. Lakewide surveys were conducted in three field seasons (spring, summer, and fall) to determine the distribution of M. relicta and to compare the density of the species in the southern basin (proposed intake location) with the density in the rest of the lake. Our conclusions about distribution are based on three surveys of the entire lake: two hydroacoustic surveys in 1995 (August and September) and a net survey in spring 1996 (during unstratified conditions).Results of the three lakewide surveys show that M. relicta are distributed throughout all suitable habitats in Cayuga Lake. Due to the patchiness in distribution, variance between samples is high. The animals were not more concentrated in the southern basin than they were in the rest of the lake during any of the three lakewide surveys conducted in support of LSC.
The lakewide distribution surveys were conducted during different field seasons in order to evaluate whether annual migration due to spawning or predator avoidance, or population depletion by predation, might alter the distribution of M. relicta. The LSC team considered that the organism might tend to be concentrated in southern Cayuga Lake due to higher availability of food, based on high allochthonous inputs of particulate matter originating at the major southern tributaries. We also considered that the M. relicta population might be higher in southern Cayuga Lake if the animals migrate to shallower water for reproduction. Finally, we considered that the animals might be found in proportionally lower concentrations in the southern lake basin during the spring as a means of avoiding predatory fish. The survey results provide three snapshots of the distribution of M. relicta in Cayuga Lake. The interaction of the abiotic and biotic factors that determine the distribution cannot be discerned from the limited data available. However, we are able to draw conclusions regarding the relative numbers of M. relicta in southern Cayuga Lake as compared to lakewide.
The methods used in the lakewide surveys are discussed in detail in Appendix C-2. Cayuga Lake was divided into five approximately equal strata at five minute latitude intervals. Four of the five strata contained deep water habitats utilized by M. relicta (the fifth stratum [north end of the lake] was too shallow). Using stratified random sampling techniques, three east-to-west transects were located within each stratum. Transects were joined by diagonal transects from the end of each east-to-west transect to the beginning of the next transect to the north (Figure 2.3.4-5). In order to determine if M. relicta occurs in higher densities in the region of the proposed intake than in the rest of the lake, densities measured in strata 2, 3, and 4 were combined and compared to density measured in stratum 1. Stratum 1, in southern Cayuga Lake, includes the region of the proposed LSC intake.
Figures 2.3.4-6 and 2.3.4-7 are actual echograms depicting the relative difference in density between transect 2A (just north of Myers Point) and transect 3C (north of King Ferry). These echograms represent entire transects. Net tows taken in the area around transect 2A averaged 59 individual M. relicta captured per net tow. Net tows taken around transect 3C averaged 150 individuals per net tow. The net tow data are therefore consistent with the hydroacoustic displays; higher density as measured by the net tows are associated with stronger hydroacoustic signals.
2.3.4.3.2.1 1995 Lakewide Survey Results.
Hydroacoustic lakewide surveys were performed in August and September 1995. The average volume of backscattering energy for each transect is presented in Figure 2.3.4-8. These values were averaged for each stratum and subjected to a Kruskal-Wallis nonparametric ANOVA by ranks test to determine if the average volume of backscattering energy was significantly different in a comparison of Strata 1, 2, 3, and 4.The Kruskal-Wallis test concluded that at least one stratum was different from at least one other in the August 1995 survey (H = 8.38 > x2(0.05.3) 7.8). No significant difference was found in the September 1995 survey (H = 3.78 < x2(0.05.3) 7.8). A stepwise comparison of summed ranks of the four strata results of the August 1995 survey (1 vs. 2, 1 vs. 3, 1 vs. 4, 2 vs. 3, 2 vs. 4, 3 vs. 4), however, did not reveal any significant difference between any of the four strata at an experimental error rate of 0.15. Table 2.3.4-2 presents these data, the critical values, and the stepwise comparison.
The hydroacoustic data, therefore, revealed that the density of the population of M. relicta in the southern basin of Cayuga Lake is not significantly different from densities in the rest of the lake.
Net count data were compiled for a comparison of density between the southern basin and the rest of the lake. A nonparametric test (Mann-Whitney) was performed on the 1995 data, to test for significant difference between the southern stratum (stratum 1) and the combined strata (2, 3, 4). The nonparametric test was selected because the data were not normally distributed. In both the August and September 1995 surveys, there was no significant difference between the strata. The density of M. relicta in the southern basin was not significantly different from the density of M. relicta in the rest of Cayuga Lake during these surveys (Mann-Whitney U, August U = 18 < U0.05(2)3,9 = 25; September U = 14 < U0.05(2)3,7 = 20). The relationship between stratum 1 and the combined strata 2, 3, and 4 is graphed in Figure 2.3.4-9.
2.3.4.3.2.2 1996 Lakewide Survey Results.
Net samples taken during the spring 1996 survey were also used in our analysis of the lakewide distribution of M. relicta. Results of the spring 1996 survey demonstrated that M. relicta was not present in higher concentrations in southern Cayuga Lake as compared to the lake as a whole. The southern basin had a mean of 74 M. relicta/m2 (std. dev. = 67, n = 54), and the remainder of the lake had a mean of 86 M. relicta/m2 (std. dev. = 64, n = 58). These means are graphically displayed in Figure 2.3.4-10. In this case, the sample sizes were large, and a student's t-test was used to test the hypothesis of equal density in stratum 1 as compared to strata 2, 3, and 4. This hypothesis was accepted (t = 0.98 < t0.05(2)12 = 1.982). As a final test, the data were stratified and reanalyzed, deleting samples taken in water shallower than 20 m (66 ft) (the observed threshold depth for capturing M. relicta). Using the stratified data set, southern Cayuga Lake (transect 1) had a mean of 103 M. relicta/m2 in spring 1996 (std. dev. = 57, n = 36), and the remainder of the lake had a mean of 93 M. relicta/m2 (std. dev. = 61, n = 54). These data are graphically displayed in Figure 2.3.4-11. Again, the means are not statistically different (student's t, t = 0.77 < t0.05(2)90 = 1.987). The results of each lakewide survey and the associated statistical analysis are summarized in Table 2.3.4-3.2.3.4.3.3 Vertical Distribution of M. relicta in the Water Column.
2.3.4.3.3.1 Daytime Distribution of M. relicta.
The LSC team used hydroacoustic and net surveys to identify the daytime distribution of M. relicta in the region of the proposed intake. This information was gathered to determine whether M. relicta tended to be on the lake sediments or suspended in the water column during daylight hours at the depth of the proposed intake. Surveys were conducted under different light conditions in order to assess the impact of incident light on depth distribution.Transects began south of the proposed intake location, proceeded in a northerly direction toward the proposed intake location, and continued north up the centerline of the lake until M. relicta were detected off the bottom. The survey dates are presented in Table 2.3.4-4.
In order to estimate the extent to which light level at the water surface corresponded to the depth of the M. relicta layer, we measured the light extinction coefficient through the epilimnion using a LiCor meter and projected light intensity at the upper limit of the M. relicta layer based on Beer's Law (Table 2.3.4-4). Note that the data do not demonstrate a consistent light level at which M. relicta are present. Since shallow waters contain more algae and particulate material than deeper waters, light extinction coefficients were calculated in 10 m intervals. The light level at the depth of the M. relicta layer was extrapolated from the deepest calculated light extinction coefficient after calculating the attenuation of light through each 10 m layer.
2.3.4.3.3.1.1 Light Extinction Through the Water Column.
Wavelength- and depth-specific light extinction coefficients were measured by Upstate Freshwater Institute on June 20, 1996, when daytime distribution and diurnal migration sampling events were also scheduled. The day was cloudy, with relatively constant surface illumination. The wavelength-specific light extinction coefficients were obtained using an underwater spectroradiometer.The wavelength-specific light measurements were separated into two depth categories for analysis: the surface layer (0.5 to 4.5 m [1.6 to 14.8 ft]) and a subsurface (1 to 23 m [3.3 to 75.5 ft]) layer. The thermal profile measured on this date is plotted in Figure 2.3.4-12. Note that the surface layer approximately corresponded to the weakly stratified conditions observed on this date; a thin layer of warmer water had penetrated through the top 4 m (13.1 ft) of the water column. Wavelength-specific light extinction coefficients are summarized in Table 2.3.4-5. The spectral distribution of irradiance for the two layers is plotted in Figure 2.3.4-13. A total of 98 percent of the surface irradiance was attenuated by 4.5 m (14.8 ft) depth. Wavelengths between 500 and 600 nanometers (nm) were found to have the lowest attenuation rate for the spectrum. These wavelengths penetrate deeper into the water column than other wavelengths. The findings are consistent with the optical characteristics of Cayuga Lake.
2.3.4.3.3.1.2 Spectral Sensitivity of Mysis relicta.
Spectral sensitivity of the M. relicta eye is known to vary from one isolated population to another (Lindstrom and Nilsson 1988). These investigators examined the correlation between the wavelength of light to which each population was most sensitive, and the wavelength of light best transmitted by each water body. Recently, Dr. Ellis Loew, of the New York State College of Veterinary Medicine, and Dr. Lars Rudstam measured the spectral sensitivity of photopigments of specimens of Cayuga Lake M. relicta using a microspectrometer. They found the pigment to have a peak sensitivity to the 520 nm wavelength of light (Rudstam and Loew, unpublished data 1996).The M. relicta eye is therefore sensitive to the wavelengths that penetrate deepest into Cayuga Lake. This finding is consistent with measured spectral sensitivity of M. relicta collected from the Baltic Sea (Lindstrom and Nilsson 1988).
2.3.4.3.3.1.3 Depth Distribution as a Function of Light.
The wavelength-specific light extinction coefficients and the results of the spectral sensitivity analysis enables researchers to estimate the depth to which the critical wavelengths of light penetrate into Cayuga Lake. This calculated light level may provide a stronger correlation to the depth at which M. relicta are located during daylight conditions. The June 20, 1996 data set has been examined in this manner. The upper limit of the M. relicta band was detected at a water depth of 60 m (197 ft). Based on the wavelength-specific light extinction coefficient, the intensity of the 500-520 nm wavelength of light at 60 m was 2.72 * 10-4 lux. This finding is consistent with the literature values reported for the light intensity to which mysids are sensitive (Rudstam and Hansson 1990).In summary, the LSC field program determined that M. relicta are not consistently associated with the lake sediments during the day in the region of the proposed intake; at times they are present in the water column. Daily variations in irradiance, and perhaps seasonal illumination patterns, dictate the depth to which M. relicta descend. During summer conditions, M. relicta were observed off the bottom at water depths of 70 m to 90 m (230 to 295 ft) (Figure 2.3.4-14). During the spring survey, M. relicta were observed suspended off the bottom at water depths between 50 and 65 m (164 and 213 ft) (Figure 2.3.4-15). Since the LSC intake would be located at a water depth of approximately 76 m (250 ft) and oriented to draw water at approximately 2.8 m (9.2 ft) above the lake bottom, M. relicta would be potentially vulnerable to entrainment throughout the year. These depths are based on the mean summer lake level of 382.4 ft above mean sea level (AMSL), based on the United States Geological Survey (USGS).
2.3.4.3.3.2 Diurnal Migration.
In order to characterize M. relicta's daily pattern of vertical migration and its relationship to LSC, we conducted diurnal sampling events in the region of the proposed intake. High frequency (420 kHz) hydroacoustic signals were used to follow the upward migration of the animals from the hypolimnion or benthos to shallower water at dusk, and their downward migration at dawn. Both ascent and descent were observed during different seasons and lunar stages. An example of output of the hydroacoustic display (an echogram) depicting the downward migration of M. relicta is included as Figure 2.3.4-16.Data from the August 7,1995 event were used to estimate the speed of migration through the water column. During this survey, M. relicta were observed beginning the vertical migration from a 72 m depth at 8:30 p.m. The top of the layer stabilized at 25 m at 9:45 p.m. The downward migration started from 20 m (65.6 ft) of water at 5:00 a.m. the following morning and ended at 5:50 a.m., when no M. relicta were discernable in the water column. Swimming speeds of 0.010 and 0.017 m/sec (36 and 61 m/hr) were calculated for the upward and downward migration during this event. Swimming speeds between 30 and 40 m/hr were calculated for M. relicta in Lake Michigan (Moen and Langeland 1989). The calculated swimming speeds in Cayuga Lake are consistent with Moen and Langeland's calculations for upward migration (actively swimming), but downward migration (sinking and swimming) occurred at a faster rate in the LSC study.
We have based our estimates of the volitional swimming speed of Mysis relicta on the speed of upward migration. This estimate is used in our assessment of the induced flow field that the animals would be able to actively avoid, and provides the necessary linkage to estimate the illumination of light required to repel M. relicta from the LSC intake. To avoid entrainment, the illumination must be sufficient to create an exclusion zone around the intake to a distance beyond which the induced velocity falls below the volitional swimming speed of M. relicta.
2.3.4.3.4 Response to Artificial Light.
2.3.4.3.4.1 Response to Artificial Light of Different Intensity.
The LSC team investigated the effectiveness of light of different intensity in repelling M. relicta in a series of experiments conducted during the 1995 field season. The experiments were conducted at night at a depth of 25 m (82 ft) where M. relicta tend to form well-defined layers in the water column. Data were gathered for light bulbs of three intensities: 3.36, 7.54, and 26.88 watts.Details of the experimental procedures are included in Appendix C-2. The effectiveness of artificial light as a mitigating measure was confirmed during this series of experiments. The layer of M. relicta was depressed in the zone of influence of the light. The response to light and the effect of the light source on the depth of the layer are displayed in Figure 2.3.4-17. Increasing wattage excluded the M. relicta to a greater distance from the light source. The radius of the exclusion zone was measured vertically from the hydroacoustic display.
The 3.36-watt light bulb produced a 20-m (65.6-ft) radius exclusion zone, the 7.54-watt bulb produced a 25-m radius exclusion zone, and the 26.88-watt light bulb created a 30-m (98-ft) exclusion zone. The field team observed that the 26.88-watt light source appeared to attract larger hydroacoustic targets (presumably fish) to the test apparatus.
2.3.4.3.4.2 Response to Artificial Light in Benthic Region.
Another field experiment was designed to test the response of M. relicta to artificial light at a depth comparable to that proposed for the LSC intake. A 7.97-watt bulb was placed on an apparatus at a depth of 65 m (213 ft). Diurnal migration of M. relicta in the vicinity of the light was observed using hydroacoustics. Two complete diurnal migration cycles (one with the apparatus in place, light off, and the second with the apparatus in place, light on) were observed. The field team was able to document the efficacy of the light in repelling M. relicta.Details of the experimental procedures are included in Appendix C-2. During the experiment, M. relicta were observed stopping their dawn descent through the water column at a depth of 20 m over the light source, moving laterally to a distance corresponding to the zone of influence, and proceeding downward towards the sediments once outside of the zone. Observations of the evening migration documented that the animals continued to avoid the lighted area as they migrated towards the surface. Once the M. relicta reached a water depth of 40 to 45 m (131 to 148 ft) (20 to 25 m above the light source) they began to form their characteristic layer. From these data, we conclude that the radius of the exclusion zone was approximately 20 to 25 m (66 to 82 ft). The exclusion zone is sketched in Figure 2.3.4-18.
2.3.4.3.4.3 Long-Term Response to Artificial Light in Benthic Region.
A final field experiment was designed to assess whether M. relicta would acclimate to an artificial light source after prolonged exposure. In 1996, a long-term benthic light experiment was designed to test the behavior of M. relicta when exposed to an artificial light source over time. A light source was installed in the region of the proposed LSC intake and operated from late May to November 1996.Four diurnal surveys were conducted over this period to assess the response of M. relicta to the presence of the light. The LSC team monitored the dusk migration of M. relicta through the water column toward the lake surface. An exclusion zone was consistently observed during each of the four monitoring events. The animals formed a layer 25 m (82 ft) above the light source. The horizontal distance (25 m) was determined from a boat speed of 4.2 knots and the number of pings per second as measured from the hydroacoustics, converted to horizontal distance. The area of influence was determined to be spherical. The exclusion zone created by the long-term benthic light source is displayed in Figure 2.3.4-19.
During one sampling event, the LSC team obtained a net tow to confirm the hydroacoustic results within the zone of influence of light during the upward migration. The net tow captured 6 animals. A control net tow taken from the same depth just outside the zone of influence as depicted on the hydroacoustic display captured 120 animals.
After one month, the light source became fouled, as diatoms began to grow on the surface of the waterproof housing. Since there is no other light available for photosynthesis at this depth in Cayuga Lake, we concluded that the microclimate formed by the light created favorable conditions for algal growth. It will therefore be necessary to incorporate a design feature that enables the lights to be cleaned.
2.3.4.4 Impacts of the Proposed Action.
We anticipate no impact on the population of M. relicta during the LSC construction phase. The littoral lake sediments to be disturbed do not represent habitat for the hypolimnetic crustacean. Deep water construction activities will not affect the organism.
During the LSC operation phase, the potential impact on M. relicta is entrainment. The LSC field investigations have demonstrated that M. relicta is present at the depth and location of the proposed intake under some conditions of light, season, and water temperature. Hydroacoustic and net surveys have been utilized to define the animal's distribution over a range of conditions of thermal stratification and ambient light. While the distribution of M. relicta throughout Cayuga Lake is patchy, the lakewide survey results have led us to conclude that the hypolimnetic crustacean is not proportionately more concentrated in the southern lake basin. Daytime surveys have demonstrated that the animals can be present in the water column as well as associated with the benthos at the depth of the proposed LSC intake (76 m [250 ft]). From these results, we have concluded that the M. relicta population is potentially vulnerable to entrainment by the LSC intake. We have assumed that any M. relicta drawn into the system would be killed by passage through the pumps and heat exchangers.
The LSC system will be designed and permitted to circulate varying amounts of water over the annual cycle, based on campus demand for cooling. The approximate volume of the lake and of the hypolimnion (during stratification) that would be circulated through the system are summarized in Table 2.3.4-6. Values in this table reflect permit conditions, assuming that permitted flows are circulated 24 hours a day, each day of the month. Actual operating conditions will always circulate less water each month. Even with the assumption of maximum permitted flows, the cumulative percent of the lake's hypolimnetic volume that would be circulated through LSC is very low, approximately one-half of 1 percent annually.
Assuming that entrainment of M. relicta would be proportional to the circulated water volume, and that M. relicta are distributed uniformly throughout the hypolimnion, we would conclude that, without mitigating measures, LSC entrainment of M. relicta would be likely to kill less than one percent of the lakewide population annually.
However, the assumption that M. relicta are distributed uniformly throughout the hypolimnion is not supported by the results of the field investigation. The animals are restricted by their response to light to defined regions of the water column during the day, and migrate through a significant vertical range at dusk and dawn. Frequently, the animals are distributed as well-defined bands in the water column. The proposed LSC intake is located at a depth and location where M. relicta spend a significant portion of the day. The proportional impact on the population could therefore be greater than calculated from percentage of hypolimnetic volume, since the intake would selectively draw from a depth representing their preferred habitat.
The size and strength of the hydrodynamic flow field induced by the LSC intake has been estimated as part of this environmental investigation. Analytical calculations based on stratified flow theory were used. The analytical calculations demonstrate that the thickness of the withdrawal layer in the hypolimnion is very narrow (on the order of a few meters) during stratified conditions, and that the induced velocity decreases rapidly with distance from the intake. Within 10 m of the intake, the induced currents are on the order of 1 to 2 centimeters per second (cm/sec) and cannot be distinguished from the background currents in the hypolimnion.
Because of the inherent difficulty in quantifying the potential entrainment and associated biological impact, the LSC team has decided to incorporate artificial light as a means to repel M. relicta from the region of the LSC intake. The light intensity will be sufficient to repel M. relicta from a region corresponding to the induced flow field around the LSC intake. The potential for entrainment is therefore greatly reduced.
2.3.4.5 Mitigating Measures.
The strong and consistent response of M. relicta to artificial light indicates that the use of a light near the intake would be an effective means of reducing the potential for entrainment. The 1995 and 1996 field data demonstrated that the hypolimnetic crustacean could be repelled to a distance beyond the calculated hydrodynamic flow field induced by the LSC intake.
LSC system design will therefore include continuous lighting of the intake. A light source will be suspended above the intake structure. The light source will be reliable, long-lived, repairable, and cleanable, and will provide the level of illumination necessary to keep M. relicta from entering the LSC intake's hydrodynamic field of influence. The light source will be redundant.
Test light bulbs have been shown to produce the required effect. The light was designed by Nova District Energy, Inc. (formerly Nova-Industra, Inc.) of Los Angeles, CA, and is graphically represented in Figure 2.3.4-20. Four incandescent light sources contained in waterproof housings will be attached to a retrievable spool. Power willbe supplied to the light source via an underwater cable. The lights will be illuminated alternately (two lights on at any one time) to reduce the surface warming and algal growth on the waterproof housings.
Maintenance of the light mitigation system requires the periodic removal of the light source, replacement of the bulbs, and cleaning of the waterproof housings. Field data collected in 1996 suggest that cleaning should occur approximately once every four months. The maintenance schedule will be adjusted based on operating experience. A SCUBA diver will retrieve the light source by attaching a lifting cable to the guideline located at a depth of 18 m (60 ft). After cleaning and bulb replacement, the diver will relocate the assembly on the guide tube and lower it into position. Maintenance will be performed during the ice-free season (April, July, and November).
The LSC system will also be designed to enable biomonitoring of the numbers of M. relicta drawn into the system. Based on conceptual design, we anticipate being able to divert a portion of the intake flow stream through a removable strainer apparatus that would capture organisms of this size class (3-20 mm). Biomonitoring will be used to assess the efficacy of the light source as a mitigation method.
2.3.4.6 Unavoidable Impacts.
We anticipate that LSC might entrain a small number of M. relicta, even with the light mitigation system in place. This residual entrainment could be associated with rare occurrences of high velocity internal currents in the region of the LSC intake that would drive the organisms towards the intake, or temporary failure of the light mitigation system. This small amount of residual entrainment would have no significant impact on the population of M. relicta.
As discussed further in the section on the Cayuga Lake fish community (2.3.7), a light may attract some fish to the region of the LSC intake. During the stratified period, few fish are present at the depth of the proposed intake. However, a hydroacoustic survey conducted during the unstratified period of spring 1996 detected some small fish at the intake depth. Based on the target strength of the hydroacoustic signals, the fish were the size of young-of-the-year or yearling alewife and/or rainbow smelt. These fish may be present at the depth of the proposed intake during the unstratified period. High frequency sound is proposed as a mitigating measure to repel alewife from the region of the intake. It has been reported in the literature that rainbow smelt would avoid light at the level proposed for the LSC intake (Appenzeller and Leggett 1995). Biomonitoring of fish entrainment will be conducted.
