Summary
The Lake Source Cooling (LSC) concept is designed to provide central campus cooling using the naturally cold waters of Cayuga Lake. Heat removed from campus buildings will be added to Cayuga Lake with the return flow from the heat exchangers. The potential impacts of LSC on the lake's thermal structure, ice cover, and stratification regime have consequently been examined as part of this Draft Environmental Impact Statement (DEIS).
The potential impacts of the LSC project on Cayuga Lake have been examined with respect to two scales: near-field (in the immediate vicinity of the outfall) and lakewide. Near-field impacts are projected using the Cornell Expert Mixing Model (CORMIX), which simulates mixing between the outfall and the ambient lake water. CORMIX projections indicate the boundaries of the plume of warmer (or cooler) water returned through the submerged outfall diffuser. Results of the CORMIX analysis indicate that the LSC return flow will impact only a small region of southern Cayuga Lake. The largest plume (defined as the distance for temperatures to return to within 0.5°C [0.90°F] of background) is projected to occur in the month of August, when the return flow is cooler than natural background conditions in the lake. Even in August, the plume is projected to extend less than 300 meters from the outfall. A very small plume of warmer water is projected near the outfall during the winter.
Lakewide impacts have been projected using the model CE-QUAL-W2, which simulates lake temperatures and thermal stratification. CE-QUAL-W2 was used to examine the lake's heat budget and annual stratification regime with and without the LSC project. Results of a 10-year simulation indicate that the impacts will be negligible. Projected changes in water temperature are small. In southern Cayuga Lake, water temperature changes on the order of 0.08°C (0.14°F) are projected. This difference is well below the natural temporal and spatial variation in temperatures in southern Cayuga Lake. On a monthly average basis, there is no discernable impact of LSC on the lake's thermal structure. No changes will occur in the annual development and breakdown of thermal stratification.
The thermal analyses have led us to conclude that LSC will have no adverse environmental impact on Cayuga Lake's thermal regime. Both near-field and lake-wide impacts will be negligible.
The process of drawing water from the deep area of Cayuga Lake and returning it to shallow, upper waters has the potential to affect the thermal conditions in the lake. The Lake Source Cooling (LSC) investigations have consequently included a focus on thermal characteristics with a comprehensive monitoring and modeling effort. During the winter, the warmed water from the LSC diffuser will have a higher temperature than the surrounding shallow lake waters. During the warmer months, the LSC return flow will be cooler than the surrounding shallow lake waters (Figure 2.3.2-1). In this section of the Draft Environmental Impact Statement (DEIS), we explain the thermal characteristics of Cayuga Lake, and explore how a warmer or cooler LSC return flow might affect the lake's thermal patterns. This issue is a concern because such changes to the lake's ecosystem could potentially alter the habitat of sensitive aquatic species. The following sections examine the thermal impacts of LSC and describe measures that will be taken to mitigate any potential impacts to the lake ecosystem.
2.3.2.1 Existing Conditions.
2.3.2.1.1 Heat Budget.
Annual heat budgets for Cayuga Lake have been estimated by Birge and Juday (1914), Henson, Bradshaw and Chandler (1961), Sunderam et al. (1969), and by J. E. Edinger Associates, Inc. (JEEAI) of Wayne, Pennsylvania as part of the LSC investigation. These annual heat budgets account for the total heat that enters the lake between the times of its lowest and highest heat content. In a typical year, Cayuga Lake stores the lowest amount of heat in March and the greatest amount of heat in August. Table 2.3.2-1 summarizes the heat budgets developed for Cayuga Lake by the investigators. The approach used to develop each budget is different, so direct comparisons between investigations cannot be made. Note the interannual variability in the 10-year projection of JEEAI, which is largely a result of differences in the amount of cloud cover (and thus solar radiation incident on the lake surface) each year. Table 2.3.2-2 presents the heat storage data on a monthly basis over a 10-year period, as modeled by JEEAI. (A complete report of JEEAI's results appears in Appendix C-8.)The budgets demonstrate the periods during which the lake gains and loses heat (Figure 2.3.2-2). In an average year, the lake gains heat from late March through mid-August, and loses heat the rest of the year. The net heat energy enters the lake surface primarily from solar radiation, and is distributed through the water column by lake motions. On a long-term basis, the amount of heat gained by the lake each year is equal to the amount lost, although disequilibria may exist in any given year. During the heating season, the lake gains an average of 2,000 Btu per square foot of surface area each day or 2.3E+07 joules per square meter-day (J/m2-day).
The heat budgets for existing conditions (with the exception of the Birge and Juday calculations) include input from the New York State Electric and Gas Corporation (NYSEG) Milliken Station. Milliken Station is a 387 megawatt (MW) coal-fired power plant located on the east shore of Cayuga Lake, approximately 13 miles north of Ithaca. Cooling water is withdrawn from a depth of 14 meters (m) or 46 feet (ft) using an intake located 183 m (600 ft) off shore. The plant is permitted to reject heat at a rate of 1.3E+09 Btu per hour (Btu/hr); the surface discharge is located in front of the facility. This amount of heat input to the lake reflects a circulation of 169,000 gallons of water per minute, with a 8.3°C (15°F) rise in temperature through the condensers. Milliken contributes approximately 15 Btu of heat per square foot of lake surface area each day (Btu/ft2-day). This amount of heat represents less than 1 percent of the natural heat gain (2,000 Btu/ft2-day) that occurs during the period when Cayuga Lake is warming.
2.3.2.1.2 Stratification and Mixing.
Deep lakes at temperate latitudes, such as Cayuga and the other Finger Lakes, display predictable patterns of water temperature each year. Water temperatures vary with depth in response to seasonal changes in atmospheric temperatures and radiant heating. Moreover, water temperature is a major component of a lake's density stratification (maximum density is at 4°C [39°F]). Understanding the thermal and density structure of Cayuga Lake is central to understanding both the concept and consequences of using the lake as a cooling resource.Considering winter (April in Figure 2.3.2-3) as the beginning of the annual cycle, Cayuga Lake water temperature and density are relatively uniform throughout the water column. Without density stratification, the winds are able to mix the lake waters throughout the water column. As the sun's energy increases in the spring, the lake gains heat and the upper waters begin to warm. Heating at the surface causes the water to expand and the lake becomes stratified, with the less dense (warmer) waters at the surface (May in Figure 2.3.2-3). More work is needed for the winds to overcome the density stratification and mix the warmer water throughout the water column. Depending on meteorological conditions, Cayuga Lake in the springtime alternates between isothermal and weakly stratified conditions.
By June of a typical year, the lake waters exhibit classical thermal stratification. The lake is divided into three layers: a warmer upper layer (the epilimnion), a cold deep layer (the hypolimnion) and a transition layer between the two (the metalimnion, which includes the thermocline). The thermocline is defined as the plane in the metalimnion where the rate of change of temperature with depth is at a maximum. Density stratification is sufficient to impede wind-induced mixing between the hypolimnion and epilimnion. The late July profile in Figure 2.3.2-3 represents classical thermal stratification.
By August, Cayuga Lake ceases to gain heat and the waters begin to cool. The cooling process is reflected by a steady deepening of the epilimnion and a gradual decrease in its temperature. This is due to heat loss at the lake surface. As the epilimnion cools, the metalimnion warms, due to the deeper mixing of warmer waters as density stratification erodes. Maximum depth of the epilimnion occurs in the late fall (refer to the late October temperature profile in Figure 2.3.2-3). As cooling proceeds, the temperature of the epilimnion eventually reaches the temperature of the hypolimnion, and the lake is again isothermal. During most winters, Cayuga Lake remains well mixed and essentially isothermal; stratification is rare and transient.
Because of this temperature cycle, Cayuga Lake is classified as a warm monomictic lake, with one period of thermal stratification (typically, May through December) and one period of lake mixing (January through April). There is some interannual variability in the duration of each stratification period in response to meteorological conditions. Sunderam et al. (1969) reported that thermal stratification averages 200 days each year.
As part of the field investigations in support of LSC, recording thermistors were deployed in Cayuga Lake to measure water temperature in the region of the proposed intake and outfall. These data were supplemented with profiles of temperature obtained during the water quality monitoring program in 1994, 1995, and 1996. Figure 2.3.2-4 plots the lake temperature data at Station S11 (in the region of the proposed intake) and demonstrates the development and breakdown of thermal stratification over the monitoring period.
2.3.2.1.3 Hydrodynamic Motions.
Lake hydrodynamics are essentially a wind-driven system. The long fetch of the lake and its steep-sided valley combine to channel winds down the lake surface. Wind action on the lake's surface causes circulation and mixing of the lake water. This hydrodynamic motion distorts the idealized temperature profiles associated with thermal stratification described above. Three types of hydrodynamic motions are evident in response to the wind-induced turbulence created at the water surface: wind-induced drift current, internal seiche oscillations, and internal waves. Each of these motions (explained below) contributes to water circulation and the distribution of heat and materials.2.3.2.1.3.1 Wind-Induced Drift.
Wind-induced drift current is created by wind blowing over the water surface, moving surface water in the direction of the wind at 2 to 3 percent of the wind speed (Sunderam et al. 1969). A return current flows below the water surface in the direction opposite the wind. During unstratified conditions, the return current may occur anywhere in the water column. During stratified conditions, the return flow is relatively shallow, restricted to the epilimnion and metalimnion. The return flow moves at its highest velocity (approximately half of the velocity of the surface flow) at the depth of the thermocline.2.3.2.1.3.2 Internal Seiche Oscillations.
The drift current moves surface water in the direction of the prevailing wind, which causes a small tilt in the water surface elevation, deepening the epilimnion and depressing the thermocline. In response, the thermocline at the opposite end of the lake tilts upward. This tilting of the water surfaces and layers remains stable as long as the wind maintains its direction and velocity. When the wind stops, the force that maintains the water's tilt is removed, causing the water to rock back and forth in the basin, or oscillate. These oscillations are called internal seiches.The amplitude of the seiche oscillations increases linearly towards the northern and southern ends of the lake. The magnitude of the peak-to-peak variation in water temperature for points at a fixed depth therefore depends on the position of those points (north to south) in the lake. Horizontal velocity of water movement varies with location; it is greatest at the center of the lake and approaches zero at the northern and southern ends (Sunderam et al. 1969).
Data from the continuously recording thermistors, installed to measure temperature with depth as part of the LSC field investigations, demonstrate the significant impacts of internal seiches on the lake's temperature profiles (Figure 2.3.2-4). Note the fluctuations in temperature recorded in the upper waters during the stratified period. Data from the 10 and 20 m depths show a pronounced periodicity in temperature fluctuation. The magnitude of the peak-to-peak variation at Station S11(in the region of the proposed intake) is on the order of 10°C (18°F). Note how the oscillations in temperature are dampened as the lake deepens; only rarely are temperature fluctuations prominent at shallower depths evident in the deepest water. It is also notable that temperature fluctuations at depth become more evident as stratification weakens in the fall.
The periods of seiches are determined solely by the shape of the lake basin; for basins of the same morphology, the seiche periods are directly proportional to the axis of oscillation and inversely proportional to the depth (Hutchinson 1957). During stratified conditions, the theoretical seiche period varies as a function of density differences between the epilimnion and hypolimnion and the thickness of the layers. Several investigators have approached this calculation of Cayuga Lake seiche periodicity after making some simplifying assumptions regarding lake geometry and thermal structure. Treating the lake as a two-layered system, Sunderam et al. (1969) calculated a seiche period on the order of 60 (range 60 to 70) hours, with current reversals every 30 hours. These calculations are generally consistent with those of Henson (1959).
Cayuga Lake has a prominent metalimnion during the stratified period, often without a well-defined thermocline. Therefore, the simplifying assumption that the lake can be modeled as a two-layer system introduces some uncertainty into the theoretical analysis. Moreover, the depth of the epilimnion increases as the summer progresses. These factors will cause the observed period of the seiches to vary from the theoretical. As a further complication, seiche activity at any given time reflects overlapping oscillations in response to past and current wind events.
The LSC thermistor data from the 20 m depth were subjected to a spectral analysis to evaluate the periodicity of seiche activity. A spectral analysis takes a time series data record and examines it over different periods of temporal variation to determine which periods are dominant in the record. The dominant periods have high "wave energy." Data from June to September 1995 were used in a spectral analysis to examine the Cayuga Lake seiche periodicity, since this time reflects a well-established thermocline and relatively uniform depth of the epilimnion. The frequency spectrum for the measured temperatures shows dominant periods extending from 3.5 to 4.3 days (84 to 103 hours). The frequency spectrum is plotted in Figure 2.3.2-5. Peaks around days 6 and 8 are multiples and higher harmonics of the smaller frequencies.
2.3.2.1.3.3 Internal Wave.
The third type of water motion in Cayuga Lake is the progressive internal wave, where all water moves through the same distance, differing only in phase. Sunderam et al. (1969) reported internal waves in Cayuga Lake with a short period (5 minutes) during stratified conditions. Internal waves are caused by bottom irregularities or short-term atmospheric disturbances.2.3.2.1.4 Sources of Natural Variation in Cayuga Lake Water Temperatures.
Significant differences in temperature are measured vertically through the water column during the stratified period as a consequence of surface heat exchange and wind-induced mixing. There are also differences in the horizontal dimension; differences in water temperatures can be measured within individual depth strata of Cayuga Lake. The differences in water temperature in the horizontal dimension are less pronounced than in the vertical.The differences in water temperature within horizontal depth strata are created by two mechanisms: localized inputs of heat (such as tributary inflows, effluent discharges, cooling water discharges, or microclimatic differences) and uneven heat distribution processes (such as wind and internal wave activity). On a large lake such as Cayuga, with tributary inflows concentrated at the southern end and significant seiche activity, horizontal temperature differences can be detected. Simultaneous measurements of surface water temperature at stations located within several hundred meters can vary by more than a degree C (approximately 1.8°F). Temperature measurements at a 1-m depth at Stations P2 (in the region of the proposed LSC outfall) and S11 are plotted in Figure 2.3.2-6. Note that the difference in temperature between these stations is most evident during the fall and spring.
Spatial temperature differences tend to diminish with increasing water depth during the period of thermal stratification. Surface temperatures have the greatest amount of spatial variability, while deeper waters (outside of the influence of seiches) tend to be more homogeneous.
Temporal variation in water temperature in response to meteorological conditions has been described with respect to the annual cycle of thermal stratification. Recall from Figure 2.3.2-4 that the deeper waters (below the influence of seiches) exhibit less variability over the annual cycle. The monthly mean and standard deviation in water temperature measured at discrete depths at Station S11 during 1995 and 1996 are summarized in Table 2.3.2-3. Note the decrease in standard deviation with increasing depth.
2.3.2.2 Impacts of the Proposed Action.
2.3.2.2.1 Lakewide Impacts: Heat Budget.
2.3.2.2.1.1 Model Structure.
The hydrothermal model CE-QUAL-W2 was used to determine the effects of the LSC facility on the seasonal temperature structure of Cayuga Lake. CE-QUAL-W2 is a two-dimensional (longitudinal vertical) simulation model developed for the U.S. Army Corps of Engineers (Buchak and Edinger 1986; Cole and Buchak 1993). The model is considered two-dimensional because temperatures are simulated with depth, and with distance north-to-south in the lake. Projections are laterally averaged in the east-to-west direction.To estimate the lakewide thermal impacts of LSC, the model was first verified by projecting the lake's thermal structure in response to 1995 meteorological and streamflow conditions, then comparing the projections to the temperature conditions measured during the 1995 field program. Once verified, the model was used to project Cayuga Lake's thermal structure with and without the LSC project over a 10-year period (1986 to 1995). This modeling effort was completed by JEEAI. The report of the lakewide thermal analysis using CE-QUAL-W2, along with technical appendices of model assumptions and underlying equations, is provided in Appendix C-8.
Modeling the circulation and temperature fields in Cayuga Lake with CE-QUAL-W2 requires that the lake bathymetry be mapped onto a grid dividing Cayuga Lake into model segments in two dimensions: with depth (vertical) and from south to north (longitudinal). The grid must be designed to represent the significant characteristics of Cayuga Lake: the north-south elongation relative to its width, the rapid increase in depth from the southern end, the extreme depth through the main body of the lake, and the shallow shelf at the northern end. The grid needs sufficient detail in the longitudinal and vertical direction to model the seasonal development and decay of thermal stratification, as well as the characteristic seiche activity. Finally, the grid system must be adequate to represent the area of interest for LSC intake and outfall with sufficient detail.
Figure 2.3.2-7 shows a plan view of Cayuga Lake with the selected CE-QUAL-W2 longitudinal segments superimposed. Note the tighter grid spacing in the southern lake basin, the area of greatest potential LSC impact. The vertical grid segments were designed to provide more detail in the epilimnion relative to the hypolimnion, with 1 m layers through the top 15 m of the lake, and 3-m layers below. The longitudinal-vertical grid that was applied to model Cayuga Lake has a total of 1,103 model cells (Figure 2.3.2-8).
2.3.2.2.1.2 Model Inputs.
Environmental data necessary to run the model CE-QUAL-W2 include tributary inflows and temperatures, meteorological data, and water surface elevation. Heat from Milliken Station was input to the model, as well as the projected flows and heat input from LSC. Meteorological data (short-wave solar radiation, air temperature, dew point temperature, wind speed, wind direction, and atmospheric pressure) were used to determine the surface heat exchange and the wind-driven component of lake circulation, which is an important facet of Cayuga Lake hydrodynamics. Local meteorological data (from Cornell's Game Farm Road facility) were used in the model.2.3.2.2.1.3. Model Verification.
The model was verified using the environmental (meteorological conditions and streamflow) data to simulate 1995 thermal conditions in Cayuga Lake, then comparing the model predictions of water temperature with depth to the measured data at Station S11. Results demonstrate that the model performed well. Compare Figure 2.3.2-9, the model predictions, to Figure 2.3.2-4, the measured results. The ability of the model to reproduce the short-term variability as well as the longer-term development of thermal structure provides strong verification of its adequacy. Table 2.3.2-4 presents the root mean square error between predicted and observed conditions at 10 m depth intervals. These values are well within the expected range for verification of hydrodynamic models. A crucial test of the model's adequacy for use on Cayuga Lake is its success in predicting the lake's characteristic seiche behavior. Figure 2.3.2-10 demonstrates that the spectral frequency of temperature fluctuations predicted by the model match well the spectral frequency of the observed data (provided in Figure 2.3.2-5). Based on these qualitative and quantitative comparisons between predicted and observed conditions, we were confident that the model can be used to project the impacts of LSC on Cayuga Lake.2.3.2.2.1.4 Results of 10-Year Simulation Modeling.
Once the model's capability as a predictive tool was established, it was used to simulate 10 years of Cayuga Lake thermal behavior with and without the LSC project. The purpose of the long-term simulations was to assure that different combinations of meteorological and hydrologic conditions were examined. The 10-year simulation period (1986-1995) also provided an opportunity to evaluate whether there would be any thermal effects of LSC that would carry over from one year to the next.Data from the 10-year simulation were tabulated to enable us to choose critical periods to test thermal impacts with and without LSC. Monthly heat storage data were used as a basis to define the periods that heat from LSC would represent the largest percentage addition to the heat budget. Relative impacts of LSC would be greatest during periods with the lowest natural heat storage. Examination of the record (which was presented in Table 2.3.2-2) reveals that 1992 had the lowest monthly heat budgets over the stratified period and 1989 shows the least amount of heat storage during fall cooling. Therefore, comparisons of Cayuga Lake temperature with and without LSC are presented for spring conditions using the May 1992 results, summer conditions using the August 1992 results, and fall conditions using the October 1989 results.
The CE-QUAL-W2 model simulates water temperatures in each of the 1,103 model cells, thus predicting a temperature profile (surface to bottom) in each of the 40 model segments. We present the model projections of temperatures with and without LSC during the critical months (May 1992, August 1992, and October 1989) for three locations in Cayuga Lake. The three locations correspond to Model Segment 3 (containing the LSC outfall), Model Segment 7 (containing the LSC intake), and Model Segment 12 (uplake, north of the segment containing the LSC intake). These segments were selected to represent the maximum potential impact of LSC on the lake's temperature structure, and to ascertain the spatial extent of thermal impacts.
The time series of differences in temperature at the LSC outfall segment is displayed for May 1992 in Figure 2.3.2-11, for August 1992 in Figure 2.3.2-12, and for October 1989 in Figure 2.3.2-13. These figures show the largest difference in water temperature with and without LSC occurring in May 1992 (spring conditions). The largest temperature differences are approximately 0.3°C (0.5°F).
The time series of differences in temperature at the LSC intake segment are plotted for May 1992 in Figure 2.3.2-14, for August 1992 in Figure 2.3.2-15, and for October 1989 for Figure 2.3.2-16. The time series of differences in temperature for May 1992 varies up to 0.3°C (0.5°F), with the largest variations near the surface. The differences tend to show a periodicity similar to that identified for the internal waves. As depth increases, both the projected temperature differences and their variation tend to decrease. Note that negative differences (cooler water with LSC heat addition) are occasionally projected to occur at a depth of 30 m (98 ft), reflecting the entrainment of cooler hypolimnetic water into the metalimnion. Simulations of Cayuga Lake's thermal characteristics during thermal stratification (August and October projections) demonstrate little impact of the heat added by the LSC return flow. Temporal variation in temperature occurs at periods similar to the seiches, with maximum differences in the range of 0.2 to 0.3°C (0.4 to 0.5°F). Note that temperatures at the 65 m (213 ft) and 70 m (230 ft) depth are projected to be essentially unchanged by implementation of LSC.
The time series of the differences in temperature at the uplake segment are displayed for May 1992 in Figure 2.3.2-17, for August 1992 in Figure 2.3.2-18, and for October 1989 in Figure 2.3.2-19. Note that the magnitude of the largest projected differences has decreased to 0.2°C (0.4°F) in this segment. Similar to the projected changes in the intake segment, the periodicity of the temperature differences is similar to that of the internal waves. Differences and their variation tend to decrease with depth.
From these figures, it is evident that the LSC project will have almost no discernable impact on the lake's thermal structure. Maximum variation is on the order of 0.2 to 0.3°C (0.4 to 0.5°F) in the model segment containing the LSC outfall.
2.3.2.2.2 Lakewide Impacts: Variability in Water Temperature.
The statistics of the temperature differences have been summarized for the 10 years of simulation to indicate variation and range in the differences on a monthly basis. The mean difference at each depth for each month over the 10 years, the standard deviation, the maximum difference found in the 10 years, and the minimum difference have been calculated.The statistics are first presented for the difference in temperatures between the Model Segment 7 (containing the LSC intake) and Model Segment 12 (uplake, north of LSC intake). These statistics (Table 2.3.2-5) are presented to indicate the spatial and temporal variations in temperature that exist due to the natural processes of inflow, outflow, surface heat exchange, and internal waves. The mean of the differences is generally negative, indicating that there is a tendency for the lake to be warmer in the uplake direction. The largest mean differences are in the surface layers in the summer months. The largest standard deviations, 0.65°C (1.2°F) and 1.03°C (1.85°F), occur in June and July between 10 and 30 m (33 and 98 ft) where the thermocline is generally located and the effects of internal waves are largest.
The single maximum value that occurred over the 10 years of simulation and the single minimum value that occurred over the 10 years are presented in Table 2.3.2-5 to indicate the magnitude of potential temperature variation under natural conditions. These single extreme values have a probability of occurrence of about 1 day in 10 years. The maximum differences in temperature between the S11 station and an uplake station occur over the 30 m to 50 m (98 to 164 ft) depths in the winter months of January and February and reach as high as 2.35°C (4.23°F). They are maximum in the 10 m to 20 m depth range during the spring and summer months. The extreme minimum values occur in the 20 m to 30 m range in the spring and summer months.
The statistics of the difference in temperatures with and without the LSC project are shown for a station at the LSC outfall location in Table 2.3.2-6. All of the statistics of the differences with and without the LSC at this location are smaller than shown for the naturally varying conditions in Table 2.3.2-5. The statistics with and without the LSC project are shown for the station at the intake in Table 2.3.2-7 and at the uplake station in Table 2.3.2-8. The summary statistics at these stations are generally smaller than shown for the naturally varying spatial and temporal conditions. Also, comparing the results in Table 2.3.2-8 at the uplake segment with the intake segment shows that the small impact of the LSC on water temperature decreases with distance (north) of the LSC project.
2.3.2.2.3 Lakewide Impacts: Stratification and Mixing.
A final comparison of Cayuga Lake thermal structure with and without LSC is provided by monthly average lake temperature profiles in the region of the proposed intake. In these projections (May in Figure 2.3.2-20, August in Figure 2.3.2-21, and October in Figure 2.3.2-22), the impact of LSC is not detectable. There is no projected impact of LSC on the development, duration, or breakdown of thermal stratification.2.3.2.2.4 Potential Impacts of Global Warming.
The feasibility of LSC relies on the continued performance of Cayuga Lake as a heat sink. The question has been raised whether our conclusions regarding the environmental impacts of LSC would change if global warming were to impact the Ithaca region. As noted in the description of the proposed action (Chapter 1), the mandate to replace CFCs in the central campus cooling system and the decision to reduce the use of fossil fuels reflect concern over global warming. In this section, we provide a qualitative discussion of the potential impacts of global warming on the lake's thermal structure. Scenarios of increased demand for the system in response to warmer air temperatures are not considered, since the thermal impacts of LSC are calculated assuming the system continuously recirculates the maximum design flows. Flow through the heat exchangers is limited by the velocity in the terrestrial chilled water pipelines.There is not yet complete consensus in the scientific community regarding the phenomenon of global warming, although recent atmospheric data strongly support the hypothesis that the earth's climate is warming (see, for example, discussion by Kerr 1996). Inputs of "greenhouse gases" (carbon dioxide, methane, nitrous oxides, and halocarbons) from human activities are projected to increase the global temperature. Current models of global climate predict an annual air temperature increase between 1.5 and 4.5°C (2.7 and 8.1°F) with a doubling of carbon dioxide emissions (Houghton, Jenkins, and Ephraum 1990). Significant uncertainty in these predictions is associated with the unquantified climatic impact of atmospheric aerosols on light scattering, which reduces the amount of solar radiation reaching the earth's surface (Schwartz and Andreae 1996).
Climate warming is likely to impact water resources; to the extent that local precipitation declines and evaporation and evapotranspiration increase, water supplies will diminish. Model predictions indicate that stream flows in central New York could decrease by 30 to 40 percent with a doubling of atmospheric carbon dioxide (Tung and Haith 1995). The monthly distribution of these impacts is not uniform, spring stream flows would decrease the most, in response to a diminished winter snowpack. Late summer and fall stream flows would decline significantly as well, due to the increased evapotranspiration. A decrease in tributary inflows to Cayuga Lake would increase the hydraulic retention time of the system. Impact on the lake's heat budget of diminished tributary inflows is likely to be minor, since surface heat exchange is the major component of the budget.
We conclude that direct effects of global warming on Cayuga Lake's heat budget would be minimal as well. Climate change models predict an increase in cloud cover as carbon dioxide levels rise. Increased cloud cover would effectively shield the lake from increased short-wave solar radiation, which has the largest impact on the heat budget. Direct conduction of heat between the atmosphere and the water is a minor component of surface heat exchange (Edinger, Duttweiler, and Geyer 1968). In general, water temperatures are not very sensitive to air temperatures. An increase in air temperature in the Ithaca area is consequently unlikely to affect the thermal structure of Cayuga Lake. From these qualitative evaluations, we conclude that global warming would not impact the feasibility of LSC.
2.3.2.2.5 Water Temperatures in the Outfall Region.
The Cornell Mixing Zone Expert System (CORMIX) is a software system for the analysis, prediction, and design of aqueous pollutant discharges into diverse water bodies (Jirka, Doneker, and Hinton 1996). The model was developed at Cornell University by Professor Gerhard Jirka and colleagues, and is supported by the United States Environmental Protection Agency (USEPA) for analysis of mixing and dilution of outfalls. The near-field effects of the LSC return flow on the water temperature in the southern lake basin have been projected using version 3.10 of CORMIX2, designed to evaluate the impact of submerged multiport diffusers. In this section, we discuss the thermal impacts on the southern lake basin of LSC return flow under design (permit) conditions. Appendix C-9 provides the detailed CORMIX input and projection files used in the analysis.The mixing model requires that the user specify characteristics of the lake environment in the region of the outfall (water depth, direction and magnitude of ambient currents, temperature, wind speed) and characteristics of the outfall and diffuser (length, location in the lake with respect to shorelines, number, orientation, and size of ports and diffusers). Information describing the nature of the return flow (temperature, density, flow rate, and concentration of any water quality parameters of concern) is also required for the projections.
The model projects the extent of the mixing zone between the return flow and the receiving waters, and provides schematic representations along with estimated dimensions of the idealized "plume" created. In our projections of LSC impacts, the maximum monthly permitted flows were assumed to occur 24 hours each day (Table 2.3.2-9). Under operating conditions, these maximum permitted flows will not occur 24 hours each day.
Projections of temperatures within the mixing zone indicate that LSC will have a minimal impact on the water temperatures in the southern lake basin, even when maximum permitted flows are modeled (Table 2.3.2-10). Note that water temperatures are projected to return to within 2°C of background conditions within a short distance of the outfall diffuser. During most months, the temperatures will return to within 0.5°C (0.9°F) of ambient within several hundred meters of the outfall. The largest plume (a plume of cooler water) is projected to occur in August.
Schematic representations of the region of altered temperatures (the thermal plumes) for the months of April through November and average winter conditions are presented in Figures 2.3.2-23(A-I). These plots indicate whether the plume will rise in the water column (return flow warmer than ambient conditions) or fall (return flow cooler than ambient conditions).
2.3.2.2.6 Sensitivity Analysis.
A sensitivity analysis was performed to assess how changes in the assumptions of ambient conditions might alter the CORMIX projections. The sensitivity analysis was performed by running the CORMIX2 model with different assignments of the input parameters. Overall, the model projections are only slightly sensitive to depth of overlying water at the outfall (which determines initial dilution), wind speed, and ambient velocity in the receiving water (which determine the transport and shape of the plume). No changes in our conclusions regarding the significance of the thermal impacts are necessary based on the results of the sensitivity analyses.Sensitivity analyses of wind speed was also conducted. In the near-field region (close to the outfall), projected thermal plumes were insensitive to assumed wind speed. In the far-field region, however, model projections diverge when high wind versus low wind conditions were assumed. The higher the wind speed, the shorter the distance until complete mixing is achieved. Because the LSC impacts are evident in the near-field, the selection of wind speed in the model projections is not critical. To be conservative, thermal plumes were projected assuming that wind speed is relatively low (2 meters per second [m/sec]).
Sensitivity analyses to ambient current velocity in southern Cayuga Lake were made altering the assumed velocity from a design case 2 centimeters per second (cm/sec) to 0.5 cm/sec and 10 cm/sec. These upper and lower bound numbers were selected based on results of current monitoring of the southern lake basin conducted in the mid-1980s in support of the environmental analysis of the Ithaca Area Wastewater Treatment Plant (Moran 1989). Projected impact of these changes in ambient velocity on the plume size are small (Table 2.3.2-11). Complete output files and plume graphics resulting from the sensitivity analysis are included in Appendix C-9.
2.3.2.2.7 Impact on Ice Cover.
Both CORMIX and CE-QUAL-W2 provide data for predicting the extent to which LSC return flow might alter the amount of winter ice cover in the southern lake basin. The CORMIX projections indicate that the small amount of heat added to the lake during this period of low system demand will rapidly dissipate in the immediate outfall vicinity. No significant impact on ice cover is projected based on this amount of temperature alteration.CE-QUAL-W2 projections compute ice cover at each model segment. Over the 10-year simulation period, the onset and breakup of ice cover (up to two months of cover) and the computed thickness (up to 30 cm, 12 in) of ice appear reasonable as compared to ice diary records maintained over the winters of 1994-1995 and 1995-1996 as part of the LSC field investigations, which are included in Appendix C-10. The model projections with LSC in operation compute a reduction in ice thickness at the outfall of less than 1 cm (maximum daily value over 10 years is less than 3 cm).
2.3.2.3 Mitigating Measures.
Thermal impacts of LSC have been projected using maximum design flows through the system. The system is designed to have variable flow in response to campus demand for cooling, and actual flows will be far less. Even with the highly conservative assumption that flows will be at their maximum permitted values each day, both lakewide and near-field thermal impacts are projected to be minimal.
The submerged outfall diffuser that will be used for the LSC return flow has been designed to minimize the near-field thermal impacts of the LSC system. It does so by locating the outfall under water and maximizing the speed with which the returned lake water is initially diluted into the surrounding waters of the outfall area. The design of the diffuser was interactive with CORMIX2 model projections. Project engineers selected a design that would provide optimal diffusion under the specific conditions present in the LSC system and in Cayuga Lake. No additional mitigating measures are proposed.
2.3.2.4 Unavoidable Impacts.
LSC represents a small incremental increase in the Cayuga Lake heat budget. Compared to the natural input from the sun, and existing industrial input from NYSEG's Milliken Station, LSC contributes substantially less than one tenth of one percent to the annual heat budget (Table 2.3.2-12). Solar energy ranges from 100 to 400 watts per square meter (W/m2) or 2.67 E+08 to 1.07 E+09 J/m2 per day (annual average 250 W/m2). Milliken Station rejects heat to the lake at a rate of 2.42 W/m2 (6.46 E+06 J/m2) per day. LSC input would range from a summer maximum of 0.434 W/m2 (1.16 E+06 J/m2) per day to a winter low of 0.186 W/m2 (4.97 E+05 J/m2) per day (annual average 0.311 W/m2).
The thermal projections indicate that lake water temperatures will be slightly altered in the immediate vicinity of the LSC outfall diffuser. These impacts are limited in size and do not represent a significant change to the near-field thermal regime. Additional discussion of the environmental impacts of the LSC return flow on the southern lake basin is included in Sections 2.3.3 (Phosphorus and Productivity) and 2.3.7 (Cayuga Lake Fish Community).
Lakewide impacts of adding heat from LSC are projected to be almost nondetectable. Subtle changes in water temperature (typically less than 0.08°C [0.14°F]) are projected to occur in the region of the proposed intake. The greatest thermal impact is projected to occur in the shallower waters, and effects would be greatest in the winter. Temperature impacts decrease with distance north in the lake. No impacts on the duration of thermal stratification or the amount of ice cover are projected.
