Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Raising the Level of a Subarctic Lake A common form of environmental intervention is damming a river to produce an artificial lake or raising the level of an existing lake to create a larger one. The environmental problems arising from these manipulations are more complex than those resulting from pollution of lakes, because the manipulations establish new flow patterns, inundate existing shore- lines, initiate erosional processes that create new shorelines, and change the morphologic characteristics of the lakes. For these reasons, prediction of the consequences of altering the dimensions of lakes is more difficult and more uncertain than prediction of the responses of lakes to pollutants and their elimination. Southern Indian Lake, in northern Manitoba, was raised 3 m in 1976, and rivers were diverted so that the flow of water through the lake was reduced by about 75%. The case study reveals that, despite careful plan- ning of the project, many of the predictions were seriously wrong. The limnologists who carried out the preproject studies later analyzed the disparities between their predictions and the outcomes. As a result, future studies for similar projects will be based on better models. 317
Case Study JOHN T. LEHMAN, Division of Biological Sciences, University of Michigan, Ann Arbor Southern Indian Lake (57° N. 99° W), in northern Manitoba, is a riverine basin along the Churchill River that drains northward to Hudson Bay. In 1976, the natural lake outlet was dammed and the lake level raised 3 m, so that the Churchill could be diverted southward across a drainage divide. Of the natural river flow of almost 1,000 m3/second, 75% was diverted into the Nelson River and no longer flows through the main lake. The diversion scheme permitted the combined flow of drainage from over 1.4 million square kilometers of the northern Great Plains to flow through a single series of dams and hydroelectric generating stations. The project had been planned in principle since the 1950s, and the lake had been the object of careful impact assessments. Extensive and detailed knowledge of the later ecological events has come through the efforts of scientists at the Freshwater Institute in Winnipeg. Institute scientists were charged in 1976 with a long-term study to evaluate the predictive capabilities of impact assessments and to generate new capabilities with increased quan- titative precision. The preimpoundment assessment work was done by professional lim- nologists who used existing modeling techniques and a literature-based paradigm regarding reservoir dynamics. The project was treated as a large- scale experiment to test the hypotheses of the impact assessments and to modify the paradigm. Many predictions were realized, and others differed from expectation only in quantitative detail. Most biological responses above the primary trophic level, however, were predicted either incorrectly or not at all. Assessments based on analogy and scaling were often mis- leading. The most striking unpredicted change involved the commercial whitefish fishery and the local economy that depended on it. Southern Indian Lake supported the largest commercial fishery in north- ern Manitoba. About 85% of the catch was Coregonus clupeaformis, the lake whitefish. During the 3 decades before impoundment, the annual catch averaged 334 metric tons, almost exclusively high-quality fish of "export" grade (Bodaly et al., 1984b). The dendritic basin of Southern Indian Lake consists of several subbasins that are separated by islands and channel constrictions. The high quality of the fishery was maintained by selective fishing in the most profitable basins. Elsewhere, stocks were dominated by darker fish with high incidences of muscle cysts of the parasitic cestode Triaenophorus crassus. 318
RAISING THE LEVEL OF A SUBARCTIC LAKE 319 Southern Indian Lake is in a region underlain by Precambrian shield bedrock. The overlying sedimentary deposits are glaciolacustrine or gla- ciofluvial, tracing either to the glacial Lake Agassiz or to outwash during the retreat of the last continental glaciation. Before the 3-m impoundment, 76% of the shoreline was exposed bedrock. Mean annual temperature in the region is-5°C, so permafrost is widespread in the glacial deposits. Vegetation is characteristic of boreal forest or taiga (Newbury et al., 19841. On the basis of federal and provincial feasibility studies, Manitoba Hydro in the 1960s proposed raising Southern Indian Lake by 10 m. This high-level impoundment was selected to optimize the generation of elec- tricity, without regard for environmental issues or a thorough study of the cost-effectiveness of hydrologic storage in the lake. In response to public concern, the utility contracted with Underwood-McLellan and Associates Ltd., a consulting engineering firm, to produce a predevelopment impact assessment of several diversion scheme options. Published in 1970, that study caused Manitoba Hydro to opt instead for a less costly low-level (3-m) impoundment, on the grounds that it would have proportionally smaller effects on watershed and shoreline. A federal-provincial study board was later commissioned to investigate the effects of the new plan. The Lake Winnipeg, Churchill and Nelson Rivers Study Board undertook its work concurrently with construction activities and issued its final report in April 1975, shortly before the diversion. The two studies differed considerably in their objectives and data bases. The first involved no substantial field work and was aimed at determining costs and benefits of different diversion schemes. The second was initiated after the configu- ration and construction schedule of the project were fixed; scientists were charged with providing reliable baseline data, predicting future conditions, and then measuring them. The federal-provincial study was comprehensive. It included studies of hydrology, soils, fisheries, limnology, wildlife, geology, recreation, for- estry, archaeology, navigation, and socioeconomic factors. University scientists, private consultants, and government laboratories were involved in different aspects of the investigations. The fisheries-limnology com- ponent was summarized by Hecky and Ayles (1974), who described the existing conditions of the basin in concise, quantitative terms and then presented a series of predictions. The limnologists recognized that the diversion scheme would alter flow patterns in the lake and that some regions would be flushed more quickly than others. In their opinion, the altered flow would have a greater impact on water chemistry than would inundation itself, because Churchill River water greatly influenced the main basins of the lake under natural con- ditions. If normal flowthrough were reduced by 80%, which was the
320 SELECTED CASE STUDIES proposed long-term average, local drainage could dominate conditions in a large area of the lake. Projections were based in part on conceptual and numerical models that related nutrient loading and water exchange with algal abundance and productivity. These were new, sophisticated versions of the approach that Edmondson had used to evaluate nutrient loading in Lake Washington (see Chapter 20 and, e.g., Dillon and Rigler, 1974, and Vollenweider, 1975~. The models had sound empirical support and widespread accep- tance as a means to predict algal abundance and trophic state. Predictions were for steady-state conditions that would follow transient inundation effects. Reduced inflow of Churchill River water would lead to diminished turbidity and nutrients from fluvial sediment loading in the long run, but water residence time would increase. Overall annual algal production was expected to decline by 33-50% in the main basins. The diversion was expected to have a positive effect in one basin along the diversion route, which previously had received little Churchill water directly. Inundation was expected to affect water quality, but not as drastically in the long term as diversion. The conventional reservoir paradigm pre- dicted that nutrient release from soils and vegetation would increase post- impoundment production at all levels in the food chain. The overburden of frozen glaciolacustrine clays around Southern Indian Lake, however, had great potential for erosion. Increased turbidity would negate benefits of increases in nutrients, according to surveys of light penetration and primary productivity. The inundation was thus expected to have no net favorable effects and to be detrimental in littoral regions. The limnologists expected higher trophic levels generally to track changes in the algae, although not necessarily immediately or in strict proportion. Declines in planktonic and zoobenthic production in main basins were expected to diminish fish production gradually. Suitable spawning sites for whitefish in the inundated areas were of concern, as was the chance that diversion could disorient migrating fish. Hecky and Ayles (1974) foresaw difficulties for the commercial fishery that would cause it to decline more than overall biological production would decline. The tra- ditional fishery was concentrated in the largest basin, which was certain to experience the greatest deprivation from diversion. As production po- tential fell, fishermen would see their yields per unit of fishing effort decline. No evidence of overfishing yet existed, but many marginal op- erators would probably leave the fishery if faced with a need to increase their effort substantially. The scientists expected an overall decrease of 13% in commercial yield. Lakewide average biological production was expected to decline by 10%. The declines were forecast on the basis of expected redistributions of nutrient income from the Churchill River. They
RAISING THE LEVEL OF A SUBARCTIC LAKE 321 applied to the long-term, steady-state outcome of the manipulation. Greater uncertainty was attached to short-term responses, and statements on that matter were largely qualitative. Early in the study, the federal-provincial study board had identified the need for long-term ecological monitoring of the impact areas. Hecky and his colleagues at the Freshwater Institute initiated their own case study of the Southern Indian Lake reservoir in 1974 with the preimpoundment work. The operational regimen has been roughly that considered by the impact study, so predictions can legitimately be compared with results. The lake provides a good opportunity to evaluate scientific judgments retrospectively. The technical results of the decade-long effort were pub- lished in 17 papers in Volume 41 of the Canadian Journal of Fisheries and Aquatic Sciences (1984~. Quantitative predictions had been generally lacking in the impact as- sessments of 1970 and 1975. The 10% reduction in biological production offered by Hecky and Ayles was cited repeatedly in the 1975 summary report as one hard prediction (Lake Winnipeg, Churchill and Nelson Rivers Study Board, 19751. The lack of postimpoundment case studies in northern Canada and the absence of verified models for important processes pre- cluded other precise estimates. The limnologists recommended research into process models as one way to improve predictions. They had been forced to rely, as had those who conducted the 1970 study, on results from reservoirs in temperate and tropical regions. Excellent studies had been done in Siberia, but those reservoirs were deep, steep-sided im- poundments, and their physical characteristics differed from those of Southern Indian Lake. The Siberian reservoirs, for instance, are deep enough to develop thermal stratification with reduced oxygen in the bottom water during the summer, but Hecky and Ayles expected no such thermal strat- ification or deoxygenation in Southern Indian Lake. The potential for extensive shoreline erosion caused their predictions to differ from con- ventional wisdom regarding productivity and fish yield in new reservoirs. In retrospect, the predevelopment predictions were extremely good with regard to nutrients and algal production. No thermal stratification devel- oped, and shoreline erosion was extensive (Hecky, 1984; Hecky and McCullough, 1984; Newbury and McCullough, 1984. The increased nu- trients from erosion and decay of vegetation increased phosphorus con- centrations as expected, but turbidity increased also. Physiological studies revealed that phosphorus deficiency among the algae had been replaced by light limitation in turbid regions of the lake, and no significant increase in productivity occurred (Hecky and Guildford, 1984; Planas and Hecky, 1984~. Regions with high transparency showed increased productivity. Few changes were predicted incorrectly. The notable exception was in
322 SELECTED CASE STUDIES fish recruitment. Both impact assessments had predicted that northern pike (Esox lucius) would benefit from the increase in habitat associated with inundation and that the sport fishery would improve. Some short-term spawning problems were anticipated for walleye (Stizostedion vitreum vitreum) and the important whitefish, but the fish were expected to exploit new spawning grounds and to recover. Esox in fact experienced no increase in survival or growth, and the commercial whitefish fishery collapsed (Bodaly and Lesack, 1984; Bodaly et al., 1984b). Loss of the viable commercial fishery in Southern Indian Lake was a blow to the local economy, and it had been unexpected. The reasons for the change are complex, but they were anticipated in part by Hecky and Ayles (1974~. The catch per unit effort indeed declined on the traditional fishing grounds, to about half the preflooding value. Total catches were maintained by major increases in total fishing effort. The fishermen also began to exploit regions that had previously been avoided because the fish were of lower marketability, being darker and having higher rates of cestode infestation. The lower-quality fish made up 12-72% of the summer catch in the years after impoundment, whereas they were insignificant components of the original fishery. In 1982, the lake was changed from "export" to "continental" classification, and the commercial value of the catch plummeted. Total catch eventually fell to one-third of its preim- poundment size as many operators abandoned the fishery (Bodaly et al., 1984b; Wagner, 19841. The catch declines apparently stemmed from migrations of fish away from traditional fishing grounds. Genetic markers measured among whitefish stocks before and after impoundment indicate that stocks became redistributed when normal flow patterns were altered. There is an indi- cation of net emigration of whitefish from Southern Indian Lake. Com- pensation payments to commercial fishermen by Manitoba Hydro subsidized their efforts from 1977 to 1982. Without the payments, production costs would have exceeded fishery revenues. In 1982, Manitoba Hydro provided a one-time cash settlement of Can$2.5 million for all future losses (Bodaly et al., 1984b). The fishery problems were not confined to collapse of the commercial enterprise. Changes had been triggered by physical manipulations that proved more far-reaching and persistent than anyone expected. The shore- line of Southern Indian Lake had been bedrock-controlled; 76G%o of the shore was exposed granite or gneiss. After flooding, only 14% of the new shore consisted of bedrock. Erosion had been expected by everyone, but not of the scale and duration produced when lakewater melted the perma- frost. The shoreline retreated at up to 12 m/year. Lakewater melted and undercut the backshore zone, and that resulted in massive faulting of the
RAISING THE LEVEL OF A SUBARCTIC LAKE 323 overhanging shoreline. Newbury and McCullough (1984) predicted that it would take at least 35 years for the shoreline to be eroded to preflooding conditions and that high rates of sediment input from inundation would persist for decades. The continuing flooding of terrestrial areas posed a new threat to fish- eries, this time presenting a health hazard for domestic consumers. The flooded soils released mercury, which became concentrated particularly in the piscivorous northern pike and walleye (Bodaly et al., 1984a). Increased mercury concentrations are a common consequence of reservoir creation, but the phenomenon has been recognized only within the last decade (Abernathy and Cumbie, 1 977; Cox et al., 1979; Kent and Johnson, 1979; Potter et al., 1975~. Soils need not be rich in mercury, and indeed the Southern Indian Lake source materials have a low or average mercury content (Bodaly et al., 1984a). Bacterial methylation of inorganic mercury salts mobilizes the element, and it then associates with particles. Muscle mercury concentrations in the walleye and northern pike now exceed the Canadian marketing standards (0.5 ppm) and will probably remain in- creased for years. A few other unexpected changes occurred. Mean lake temperatures decreased, owing to increased mean depth, river diversion, greater surface reflectance, and backscattering of incident solar radiation from the turbid water (Hecky, 19841. The zooplankton community changed, too (Patalas and Salki, 19841. The large-bodied crustacean Mysis relicta, formerly rare, became common. Large-bodied calanoid copepods like Limnoca- lanus macrurus increased. Cladocerans and small copepods decreased. Patalas and Salki ascribed the changes primarily to lower water temper- atures and increased water depth, which enlarged the habitat of generally deep-dwelling species. They regarded an overall decrease in the abundance and biomass of crustacean zooplankton to be a consequence of the di- version-related temperature decrease. Lower temperatures slow growth and development rates of both eggs and juveniles. The patterns suggest, moreover, that predation pressure from planktivorous fish decreased and that predatory invertebrates became an important force in the case of the plankton. The decline in predation by fish could reflect a decrease in stocks or unsuccessful foraging due to turbidity. The changes in the zooplankton community in Southern Indian Lake parallel some of the changes that occurred in Lake Tahoe when Mysis was introduced (Goldman et al., 1979; Richards et al., 19751. It seems reasonable to ascribe the changes in zooplankton, including Limnocalanus and Mysis, to declines in fish stocks and foraging success. The findings are united with recent events in Lake Washington by the major theme in zooplankton community ecology of the last 2 decades, which emerged in
324 SELECTED CASE STUDIES the seminal works of Hrbacek (Hrbacek, 1962; Hrbacek et al., 1961) and Brooks and Dodson (19651: that the introduction of planktivorous fish causes wholesale alterations of species composition and size distribution in zooplankton communities. That zooplankton communities are shaped by the size-selective predatory behavior of fish is well established (Gal- braith, 1967; Hall et al., 1976; Lynch, 1979; O'Brien, 1979; Wells, 1970; Zaret, 19801. The proximate mechanism is prey conspicuousness (Zaret, 1972; Zaret and Kerfoot, 1975; Zaret and Suffern, 19761. Much of the recent excitement in zooplankton community ecology has come from in- vestigations of community structure in the absence of fish. Brooks and Dodson had used allometric estimates of metabolism and feeding to support their "size-efficiency hypothesis"-i.e., large animals simply outcom- pete small ones through economies of scale. Others contested that claim and argued with convincing empirical support that predation by inverte- brates could eliminate small species (Dodson, 1974; Kerfoot, 1977; Zaret, 19751. Unlike vertebrates, these predators rely on mechanoreception to detect their prey and manipulate the prey to consume it. This type of predation by invertebrates, mostly insect larvae and predatory crustacean zooplankton, has been linked to much morphological variation including cyclomorphosis-within the taxa that they prey on (Halbach, 1971; Jacobs, 1965; Zaret, 1972~. Large invertebrate predators can thrive in the absence of fish, feeding on small-bodied animals and the juvenile stages of larger taxa. Large- bodied zooplankton have an advantage, particularly if they can produce large eggs and boost their offspring past the vulnerable juvenile period. When planktivorous fish abound, the invertebrate predators are held in check; so, too, are large herbivores. The small-bodied animals are released from predation. They might come to dominate the assemblage, even though the fish might eventually be forced to turn to them as prey (Brooks, 19681. This explanation probably figures in the events in Southern Indian Lake, ~ltho.lah P~t~ln~ ~nr1 Salki (1984) regarded the direct physiological effects ~ ~-~ ~ _ O ~ ~ ~ , ~ of temperature to be dominant. Hecky and colleagues undertook in 1984 their own retrospective analysis of environmental impact prediction and assessment based on their expe- rience with Southern Indian Lake (Hecky et al., 19841. Scientists who had formulated the original predictions and then used the experiment to advance understanding provided their scientific insights into the limitations and capabilities of the prediction process. They conceded, for instance, that the 1970 "office" study was nearly as effective in forecasting impacts as was the 1975 study after a year of field work. Improved precision in the 1975 study resulted in part from having a better definition of the project. Hecky and colleagues concluded that both assessments were of
RAISING THE LEVEL OF A SUBARCTIC LAKE 325 only marginal overall utility, because significant impacts had not been predicted and because even the correct predictions had often been quali- tative. Qualitative statements could not enter the quantitative cost-benefit analyses that influence major resource development. The field study had proved invaluable, nonetheless, primarily as the means to document the baseline conditions in the lake. Without such data, any followup study might have been meaningless, and compensation pay- ments might have been more difficult to arrange. The field study, in essence r~rovided insurance against unpredicted adverse environmental -7 rim impacts (Hecky et al., 1984~. Three-meter impoundment was selected because of the high cost of accessory "saddle" dams that are needed to contain high-level impound- ment and because the 1970 study had suggested that the effects of the originally planned 10-m impoundment might be too severe. The authors of the 1970 study had assumed that effects were proportional to the mag- nitude of manipulation. A 3-m change was assumed to have smaller effect than a 10-m change. In fact, impoundment levels greater than 3 m would have produced effects only slightly worse than actually occurred, because severe shoreline erosion began as soon as the natural range of water levels was exceeded and the impounded water entered the previously frozen backshore zone. The Freshwater Institute scientists bemoaned the lack of suitable analog studies and the limitations of current reservoir paradigms. Conventional wisdom and expectations were based on experience with deep riverine basins with limited wind fetches. Indigenous biological processes had been emphasized nearly to the exclusion of physical forces. The most likely analog candidates either lacked permafrost features or had little fine- grained erodable material on the shores. In the absence of examples or experience, it was difficult for the scientists to judge the importance of these novel characteristics. However, they developed empirical relations between energy and erosion and can now use shoreline maps and mete- orological records to predict lake volumes after erosion (Newbury and McCullough, 19841. Southern Indian Lake has changed the understanding of reservoir dynamics. Responses at higher trophic levels showed the greatest failures of pre- diction. Hecky et al. (1984) rejected the notion that the responses are intrinsically unpredictable. They argued instead that possible responses, unless parts of existing paradigms, tend to be overlooked or considered only rarely. Hecky et al. seemed to blame themselves for not anticipating the problems with fishery quality and with mercury contamination, in- asmuch as elements of the logical puzzle were already in their grasp. They suggested that they were on surest ground when predicting energy flow,
326 SELECTED CASE STUDIES biomass, and general trophic relations, but that species-level predictions are elusive. Their most severe criticism was that they learned that current impact assessment was incomplete and unacceptable. They could also have mentioned that their studies opened the door to new understanding of reservoir processes and functions. REFERENCES Abernathy, A. R., and P. M. Cumbie. 1977. Mercury accumulation by largemouth bass (Micropterus salmoides) in recently impounded reservoirs. Bull. Environ. Contam. Tox- icol. 17:595-602. Bodaly, R. A., and L. F. W. Lesack. 1984. Response of a boreal northern pike (Esox lucius) population to lake impoundment: Wupaw Bay, Southern Indian Lake, Manitoba. Can. J. Fish. Aquat. Sci. 41: 706-714. Bodaly, R. A., R. E. Hecky, and R. J. P. Fudge. 1984a. Increases in fish mercury levels in lakes flooded by the Churchill River diversion, northern Manitoba. Can. J. Fish. Aquat. Sci. 41 :682-691. Bodaly, R. A., T. W. D. Johnson, R. J. P. Fudge, and J. W. Clayton. 1984b. Collapse of the lake whitefish (Coregonus clupeaformis) fishery in Southern Indian Lake, Man- itoba, following lake impoundment and river diversion. Can. J. Fish. Aquat. Sci. 41:692- 700. Brooks, J. L. 1968. The effects of prey size selection by lake planktivores. Syst. Zool. 17:272-291. Brooks, J. L., and S. I. Dodson. 1965. Predation, body size, and the composition of the plankton. Science 150:28-35. Cox, J. A., J. Carnahan, J. DiNunzio, J. McCoy, and J. Meister. 1979. Source of mercury in fish in new impoundments. Bull. Environ. Contam. Toxicol. 23:779-783. Dillon, P. J., and F. H. Rigler. 1974. A test of a simple nutrient budget model predicting the phosphorus concentration in lakewater. J. Fish. Res. Bd. Can. 31:1771-1778. Dodson, S. I. 1974. Adaptive change in plankton morphology in response to size-selective predation: A new hypothesis of cyclomorphosis. Limnol. Oceanogr. 19:721-729. Fudge, R. J. P., and R. A. Bodaly. 1984. Postimpoundment winter sedimentation and survival of lake whitefish (Coregonus clupeaformis) eggs in Southern Indian Lake, Manitoba. Can. J. Fish. Aquat. Sci. 41:701-705. Galbraith. M. G.* Jr. 1967. Size-selective predation on Daphnia by rainbow trout and yellow perch. Trans. Am. Fish. Soc. 96:1-10. Goldman, C. R., M. D. Morgan, S. T. Threlkeld, and N. Angell. 1979. A population dynamics analysis of the cladoceran disappearance from Lake Tahoe, California-Nevada. Limnol. Oceanogr. 24:289-297. Halbach, U. 1971. Zum Adaptivwert der zyklomorphen Dornenbildung von Brachionus calyciflorus Pallas (Rotatoria). Oecologia 6:267-288. Hall, D. J., C. W. Burns, and P. H. Crowley. 1976. The size-efficiency hypothesis and the size structure of zooplankton communities. Annul Rev. Ecol. Syst. 7:177-208. Hecky, R. E. 1984. Thermal and optical characteristics of Southern Indian Lake before, during, and after impoundment and Churchill River diversion. Can. J. Fish. Aquat. Sci. 41 :579-590.
RAISING TI1E LEVEL OF A SUBARCTIC LAKE 327 Hecky, R. E., and H. A. Ayles. 1974. Summary of fisheries-limnology investigations on Southern Indian Lake. Lake Winnipeg, Churchill and Nelson Rivers Study Board, 1971- 1975. Technical Report Appendix 5, Volume 1A. Environment Canada Fisheries Service, Winnipeg, Man. Hecky, R. E., and S. J. Guildford. 1984. Primary productivity of Southern Indian Lake before, during, and after impoundment and Churchill River Diversion. Can. J. Fish. Aquat. Sci. 41:591-604. Hecky, R. E., and G. K. McCullough. 1984. Effect of impoundment and diversion on the sediment budget and nearshore sedimentation of Southern Indian Lake. Can. J. Fish. Aquat. Sci. 41:567-578. Hecky, R. E., R. W. Newbury, R. A. Bodaly, K. Patalas, and D. M. Rosenberg. 1984. Environmental impact prediction and assessment: The Southern Indian Lake experience. Can. J. Fish. Aquat. Sci. 41:720-732. Hrbacek, J. 1962. Species composition and the amount of the zooplankton in relation to the fish stock. Rozpr. Cesk. Akad. Ved Rada Mat. Prir. Ved. 10:1-116. Hrbacek, J., M. Dvorakova, M. Korinek, and L. Prochazkova. 1961. Demonstration of the effect of fish stock on the species composition of zooplankton and the intensity of metabolism of the whole plankton association. Verh. Int. Verein. Limnol. 14:192-195. Jacobs, J. 1965. Significance of morphology and physiology of Daphnia for its survival in predator-prey experiments. Naturwissenschaften 52:141. Kent, J. C., and D. W. Johnson. 1979. Mercury, arsenic and cadmium in fish, water, and sediment of American Falls Reservoir, Idaho, 1974. Pestic. Monit. J. 13:35-40. Kerfoot, W. C. 1977. Implications of copepod predation. Limnol. Oceanogr. 22:316-325. Lake Winnipeg, Churchill and Nelson Rivers Study Board. 1975. Summary Report: Canada- Manitoba Lake Winnipeg, Churchill and Nelson Rivers Study. Allied Printing, Winnipeg, Man. Lynch, M. 1979. Predation, competition, and zooplankton community structure: An ex- perimental study. Limnol. Oceanogr. 24:253-272. Newbury, R. W., and G. K. McCullough. 1984. Shoreline erosion and restabilization in the Southern Indian Lake reservoir. Can. J. Fish. Aquat. Sci. 41:558-566. Newbury, R. W., G. K. McCullough, and R. E. Hecky. 1984. The Southern Indian Lake impoundment and Churchill River diversion. Can. J. Fish. Aquat. Sci. 41:548-557. O'Brien, W. J. 1979. The predator-prey interaction of planktivorous fish and zooplankton. Am. Sci. 67:572-581. Patalas, K., and A. Salki. 1984. Effects of impoundment and diversion on the crustacean plankton of Southern Indian Lake. Can. J. Fish. Aquat. Sci. 41:613-637. Planas, D., and R. E. Hecky. 1984. Comparison of phosphorus turnover times in northern Manitoba reservoirs with lakes of the Experimental Lakes Area. Can. J. Fish. Aquat. Sci. 41:605-612. Potter, L., D. Kidd, and D. Standiford. 1975. Mercury levels in Lake Powell: Bioampli- fication of mercury in man-made desert reservoir. Environ. Sci. Tech. 9:41-46. Richards, R. C., C. R. Goldman, T. C. Frantz, and R. Wickwire. 1975. Where have all the Daphnia gone? The decline of a major cladoceran in Lake Tahoe, California-Nevada. Verh. Int. Verein. Limnol. 19:835-842. Vollenweider, R. A. 1975. Input-output models with special reference to the phosphorus loading concept in limnology. Schweiz. Z. Hydrol. 37:53-84. Wagner, M. W. 1984. Postimpoundment change in financial performance of the Southern Indian Lake commercial fishery. Can. J. Fish. Aquat. Sci. 41:715-719. Wells, L. 1970. Effects of alewife predation on zooplankton populations in Lake Michigan. Limnol. Oceanogr. 14:556-565.
328 SELECTED CASE STUDIES Zaret, T. M. 1972. Predators, invisible prey, and the nature of polymorphism in the Cladocera (Class Crustacea). Limnol. Oceanogr. 17:171-184. Zaret, T. M. 1975. Strategies for existence of zooplankton prey in homogeneous env~ron- ments. Verb. Int. Verein. Limnol. 19: 1484-1489. Zaret, T. M. 1980. Predation and freshwater communities. Yale University Press, New Haven, Conn. Zaret, T. M., and W. C. Kerfoot. 1975. Fish predation on Bosmina longirostris: Body- size selection versus visibility selection. Ecology 56:232-237. Zaret, T. M., and J. S. Suffern. 1976. Vertical migration in zooplanlcton as a predator avoidance mechanism. Limnol. Oceanogr. 21:804-813. Committee Comment L.ike the Lake Washington work (Chapter 20) Q^~3th^~ I; ~ ~1 ~ ~_11_ J ~ , the investigation of -~11~111 ~llu~al~ ~c `;~11~(1 on scientists to use nearly all the sources of ecological knowledge identified in this case study. Freshwater Institute scientists were explicit at the outset about using impoundment and diver sion as an experiment in reservoir processes. They coordinated and focused scientific and technical talent on a scale rarely achieved in such ventures. Productivity of fish and plankton communities was a dominant theme of the biological investigations. The observational program was designed to study spatial differences in the dendritic basin and to identify regions that were influenced, either positively or negatively, by the physical manip ulation. The specific incorporation of genetic markers into the study of whitefish populations (Bodaly et al., 1984), for instance, made it possible to document the redistribution and migration of stocks. Productivity and biomass were the measures of choice for all trophic levels, because they reduce biological entities and rates to a common currency. For a study that encompasses an entire ecosystem, such a com- mon means of presentation of data is necessary. Nonetheless, events in Southern Indian Lake are traceable to the life histories and physiological and behavioral characteristics of individual interacting populations. In- creased turbidity and lowered lake temperatures triggered biological re- sponses in ways not explainable strictly by thermodynamics or mass flux alone. Temperature affects the rates of metabolic and developmental pro- cesses, and inorganic particles in suspension attenuate light that would otherwise be available for photosynthesis. Turbidity, moreover, obscures the vision of predators that hunt by sight and reduces the risk to their preferred prey. The experiment in Southern Indian Lake apparently intro- duced manipulations at both the base and the apex of the food web. Primary producers were affected by the altered light and by changes in nutrient
RAISING THE LEVEL OF A SUBARCTIC LAKE 329 input. Whereas selection might have operated for species adept at nutrient acquisition before impoundment, the postimpoundment community clearly was governed by light limitation (Hecky and Guildford, 19844. One would expect changes in species composition, and the phytoplankton community showed such changes (H. Kling, personal communication, cited by Patalas and Salki, 19841. The increase in large-bodied crustaceans suggests that predatory pres- sure on the species had relaxed. Deeper, more turbid water provided a refuge from the fish. The predatory behavior of these large zooplankters, in turn, could have contributed to the decline in cladocerans and small copepods reported by Patalas and Salki (19841. Effects of the manipulation thus became focused on the small herbivores, beset on the one side by abundant carnivorous zooplankton and on the other by reduced abundance of phytoplankton owing to the new light regime. The cladocerans declined by 75% and small cyclopoid copepods by 50% (Patalas and Salki, 19841. In contrast with the alterations of the plankton communities, little or no change occurred among the profundal macrobenthos (Wiens and Ro- senberg, 1984~. The benthic organisms seemed to be influenced principally by the input of organic material from the Churchill River or from shoreline erosion and thus to be divorced from biological interactions in the turbid waters. The scientific papers include no discussion of meroplanktonic taxa like Chaoborus, which can enter the plankton each night and are known to be influenced by planktivorous fish (e.g., Northcote et al., 19781. If fish predation were substantially relaxed, these species would likely be affected. The professional publications produced as a result of the Southern Indian Lake study are valuable contributions to current knowledge about reservoir processes. The environmental manipulation was an experiment that had never been tried previously. Most of the results could be guessed only qualitatively, if at all. From a strictly scientific point of view, the project was exploited very profitably, in that new principles were uncovered and present paradigms have been enriched. But the knowledge was gained at a cost: the native peoples of the Southern Indian Lake region who suffered the loss of livelihood and threats to the quality of their food did not share in the scientific adventure. Hecky and colleagues seemed to regret most their inability to forecast the fisheries problems the eventual compen- sation program was somewhat arbitrary and inequitable. Had they pre- dicted the decline in fishery quality and the hazards associated with mercury, the forecast need not have halted the experiment, but might have led to a well-planned agreement for adequate and just compensation. Such a plan could very easily have entered the cost-benefit analysis of the project.
330 SELECTED CASE STUDIES References Bodaly, R.A., T. W. D. Johnson, R. J. P. Fudge, and J. W. Clayton. 1984. Collapse of the lake whitefish (Coregonus clupeaformis) fishery in Southern Indian Lake, Manitoba, following lake impoundment and river diversion. Can. J. Fish. Aquat. Sci. 41:692-700. Hecky, R. E., and S. J. Guildford. 1984. Primary productivity of Southern Indian Lake before, during, and after impoundment and Churchill River Diversion. Can. J. Fish. Aquat. Sci. 41:591-604. Northcote, T. G., C. J. Walters, and J. M. B. Hume. 1978. Initial impacts of experimental fish introductions on the macrozooplankton of small oligotrophic lakes. Proc. Int. Assoc. Theor. Appl. Limnol. 20:2003-2012. Patalas, K., and A. Salki. 1984. Effects of impoundment and diversion on the crustacean plankton of Southern Indian Lake. Can. J. Fish. Aquat. Sci. 41:613-637. Wiens, A. P., and D. M. Rosenberg. 1984. Effect of impoundment and river diversion on profundal macrobenthos of Southern Indian Lake, Manitoba. Can. J. Fish. Aquat. Sci. 41 :638-648.