Search

The physiological basis for well-known correlations between summer air temperature indices and year-class strength in northern smallmouth bass (Micropterus dolomieui) populations was examined. Field and laboratory studies demonstrated the existence of two critical stages in early life when smallmouth bass are particularly vulnerable to features characteristic of many natural water temperature regimes. The first stage extends from fertilization until the young leave the nest; high mortality results from exposure to extreme temperatures. The second stage extends over the first winter, when the young subsist on accumulated energy reserves. Because the ratio of energy stored to basal metabolic rate increases with size, large fish can withstand winter starvation better than small fish. The results from these and other studies were incorporated into a deterministic model of the relations between temperature and first-year survival of small-mouth bass. Analysis of water temperature time series data from many locations led to the development and parameterization of a stochastic model capable of simulating variations in water temperature characteristic of the littoral zones of typical North American lakes. The stochastic physical and deterministic biological models were used together to assess the effects on first-year survival of changes in climate and of realistic changes in the magnitude and frequency of short-term temperature fluctuations. The model successfully predicted the observed northern limit of the species' range. It also generated approximate environmental criteria for judging when year-to-year variation in survival, over one or both of the critical life history stages identified, is likely to reach a level sufficiently great to determine the pattern of year-to-year variation in recruitment to the adult stock. The combined model was also used to assess the effects of thermal loading from a nuclear power plant on a particular population.

We report on a 20-year integrated field, laboratory, and simulation study of the impact of thermal effluent from a nuclear power plant on the smallmouth bass Micropterus dolomieui population in Baie du Doré, Lake Huron. A quantitative model was constructed to forecast the range of likely effects of the plant on the population. The model was based on the findings of 50 years of basic research on smallmouth bass ecology in Ontario and on 15 years of environmental and biological data collected at Baie du Doré before the power plant became operational. The model embodies detailed descriptions of both the natural variation in water temperature at Baie du Doré, and the mechanisms whereby variations in temperature generate variations in cohort strength. Five years of monitoring data, collected after the plant became operational, were used to evaluate the forecasts of the model. Observed effects on timing of spawning and individual growth were generally within the range defined by the forecasts. The thermal effluent also altered the distribution of adult fish and this, in turn, led to an increase in the effectiveness of the angling fishery. Data from the postoperational period were used to refine the assessment model. The revised model was then used to evaluate the long-term effects on the population of the changes observed to date. We describe how the availability of extensive site-specific and species-specific data helped us to deal with technical problems that have plagued other impact assessment projects. We also discuss the contributions of basic research and quantitative modelling to the effectiveness of the study.

As part of a study of the life cycle of the smallmouth bass Micropterus dolomieui, young of the year were sampled in the Baie du Doré, Lake Huron, during six summers to determine growth after dispersal from nests. Length-frequency distributions were plotted approximately weekly each of the summers. The changes through time of the mean size of fish in each cohort were described by a mathematical model that assumed a size-dependent growth rate and no size-dependent mortality for the fish. The constraint on the model to fit the empirical mean fish sizes through the growing season automatically produced time-varying standard deviations of the size distributions that compared closely to the empirical standard deviations for 5 out of the 6 years. The poor fit to the standard deviations of the sixth year was explicable in terms of unusual data on one sampling date. These results suggest that size dependence of the growth rate was an important mechanism in the spread in the length-frequency distribution and that size-dependent mortality played no significant role in determining the changes in this distribution through the growing season. Because winter mortality of first-year smallmouth bass in northern waters is important in their population dynamics and is size-dependent, the model provides a link in the understanding of smallmouth bass populations and their management.