Fish activity costs are often estimated by transforming their swimming speed in energy expenditures with respirometry models developed while forcing fish to swim against a flow of constant velocity. Forced swimming models obtained using a procedure that minimizes flow heterogeneity may not represent the costs of swimming in rivers characterized by turbulence and by a wide range of instantaneous flow velocities. We assessed the swimming cost of juvenile Atlantic salmon (Salmo salar) in turbulent flows using two means (18 and 23 cm·s–1) and two standard deviations of flow velocity (5 and 8 cm·s–1). Twenty respirometry experiments were conducted at 15 °C with fish averaging 10 g. Our results confirmed that swimming costs are affected by the level of turbulence. For a given mean flow velocity, swimming costs increased 1.3- to 1.6-fold as turbulence increased. Forced swimming models under estimated actual swimming costs in turbulent flow by 1.9- to 4.2-fold. Spontaneous swimming models overestimated the real cost of swimming in turbulent flow by 2.8- to 6.6-fold. Our analyses suggest that models in which both the mean and the standard deviation of flow velocity are explicitly represented are needed to adequately estimate the costs of swimming against turbulent flows.

We used the Kitchell et al. (J. Fish. Res. Board Can. 34: 1922–1935) bioenergetics model and field derived estimates of growth and consumption rates to estimate the quantity of energy allocated to activity by 28 combinations of yellow perch (Perca flavescens) age class and population. Activity costs among populations ranged from 0 to 40% of the perch bioenergetics budget. We further evaluated the influence of activity rates on the food consumption estimates predicted by the Kitchell et al. model and the model proposed by Kerr (Can. J. Fish. Aquat. Sci. 39: 371–379). As suggested by Kerr, activity costs increased as food consumption increased. However, we found no significant relationship between predicted and observed food consumption estimates for either model. The magnitude of, and the among-population variance in, the quantity of energy allocated to activity is consistent with our hypothesis that this component of the bioenergetics budget of fishes has the potential to contribute meaningfully to the explanation of inter-population differences in perch growth and, by extension, to the variance in growth of other actively foraging fish species.

Papers and panel discussions given during a 1992 symposium on bioenergetics models are summarized. Bioenergetics models have been applied to a variety of research and management questions relating to fish stocks, populations, food webs, and ecosystems. Applications include estimates of the intensity and dynamics of predator–prey interactions, nutrient cycling within aquatic food webs of varying trophic structure, and food requirements of single animals, whole populations, and communities of fishes. As tools in food web and ecosystem applications, bioenergetics models have been used to compare forage consumption by salmonid predators across the Laurentian Great Lakes for single populations and whole communities, and to estimate the growth potential of pelagic predators in Chesapeake Bay and Lake Ontario. Some critics say that bioenergetics models lack sufficient detail to produce reliable results in such field applications, whereas others say that the models are too complex to be useful tools for fishery managers. Nevertheless, bioenergetics models have achieved notable predictive successes. Improved estimates are needed for model parameters such as metabolic costs of activity, and more complete studies are needed of the bioenergetics of larval and juvenile fishes. Future research on bioenergetics should include laboratory and field measurements of key model parameters such as weight‐dependent maximum consumption, respiration and activity, and thermal habitats actually occupied by fish. Future applications of bioenergetics models to fish populations also depend on accurate estimates of population sizes and survival rates.

We evaluated the ability of numerical habitat models (NHM) to predict the distribution of juveniles of Atlantic salmon (Salmo salar) in a river. NHMs comprise a hydrodynamic model (to predict water depth and current speed for any given flow) and a biological model (to predict habitat quality for fish using water depth, current speed, and substrate composition). We implemented NHMs with a biological model based on (i) preference curves defined by the ratio of the use to the availability of physical conditions and (ii) a multivariate logistic regression that distinguished between the physical conditions used and avoided by fish. Preference curves provided a habitat suitability index (HSI) ranging from 0 to 1, and the logistic regression produced a habitat probabilistic index (HPI) representing the probability of observing a parr under given physical conditions. Pearson's correlation coefficients between HSI and local densities of parr ranged from 0.39 to 0.63 depending on flow. Corresponding values for HPI ranged from 0.81 to 0.98. We concluded that HPI may be a more powerful biological model than HSI for predicting local variations in fish density, forecasting fish distribution patterns, and performing summer habitat modelling for Atlantic salmon juveniles.

We compared estimates of daily ration developed using the theoretically rigorous and logistically demanding Elliott and Persson model and the more easily applied Eggers model which is infrequently used because of its assumptions about rigid fish feeding periodicity. Comparisons were based on ten 24-h samplings of six different yellow perch (Perca flavescens) populations. Daily ration estimates from the two models did not differ significantly. This consistency occurred in spite of the fact that in some cases the observed feeding periodicity violated the assumptions of the Eggers model. A simulation model demonstrated that 95% confidence intervals were smallest for the Eggers estimates and that the Eggers model was more robust than the Elliott and Persson model to changes in both sampling frequency and number offish sacrificed at each sampling event. The latter proved particularly sensitive to changes in sampling frequency. We concluded that the two models provide estimates of daily ration comparable in magnitude and accuracy and consequently that the restriction of the Eggers model to fish with rigid feeding periodicity is not justified. Furthermore, the Eggers model, because of its robustness, reduces the sampling requirements to determine daily ration, and hence, permits its estimation on a more frequent basis.

We tested the hypothesis that the energetics of swimming in a flume accurately represent the costs of various spontaneous movements using empirical relationships between fish swimming costs, weight, and speed for three swimming patterns: (1) ‘forced swimming’ corresponded to movements adopted by fish forced to swim against a unidirectional current of constant velocity; (2) ‘directed swimming’ was defined as quasi‐rectilinear movements executed at relatively constant speeds in a stationary body of water and (3) ‘routine swimming’ was characterized by marked changes in swimming direction and speed. Weight and speed explained between 76% (routine swimming) and 80% (forced swimming) of net swimming cost variability. Net costs associated with different swimming patterns were compared using ratios of model predictions (swimming cost ratio; SCR) for various weight and speed combinations. Routine swimming was the most expensive swimming pattern (SCR for routine and forced swimming =6.4 to 14.0) followed by directed (SCR for directed and forced swimming =0.9 to 2.8), and forced swimming. The magnitude of the difference between the net costs of forced and spontaneous swimming increases with movement complexity and decreases as fish weight increases.

We tested the hypothesis that activity rates can represent a large and variable component of a fishˈs energy budget. We executed five experiments between July 15 and August 2, 1991, to estimate growth, consumption, and activity rates of five young‐of‐the‐year brook trout Salvelinus fontinalis kept in field enclosures. Mean fish wet weight decreased from 3.47 to 2.89 g during our study. This represented a total loss of 3,544 cal (all fish combined). Consumption rates averaged 0.18 g dry·100 g wet–1·d–1. The total quantity of food consumed by the fish inside the enclosure ranged from 112.9 to 188.8 cal·d–1. Median swimming speed ranged from 3.9 to 7.9 cm·s–1. Energetic costs associated with spontaneous swimming (all fish combined) ranged from 94.0 to 498.1 cal·d–1. Spontaneous activity metabolic rate (standard metabolism + net activity rate) represented 1.6–3.8 times the costs associated with standard metabolism. Our study supported the hypothesis that activity rates can be a large and variable component of a fishˈs energy budget.

Quantifying the role of spatial patterns is an important goal in ecology to further understand patterns of community composition. We quantified the relative role of environmental conditions and regional spatial patterns that could be produced by environmental filtering and dispersal limitation on fish community composition for thousands of lakes. A database was assembled on fish community composition, lake morphology, water quality, climatic conditions, and hydrological connectivity for 9885 lakes in Ontario, Canada. We utilized a variation partitioning approach in conjunction with Moran's Eigenvector Maps (MEM) and Asymmetric Eigenvector Maps (AEM) to model spatial patterns that could be produced by human‐mediated and natural modes of dispersal. Across 9885 lakes and 100 fish species, environmental factors and spatial structure explained approximately 19% of the variation in fish community composition. Examining the proportional role of spatial structure and environmental conditions revealed that as much as 90% of the explained variation in native species assemblage composition is governed by environmental conditions. Conversely on average, 67% of the explained variation in non‐native assemblage composition can be related to human‐mediated dispersal. This study highlights the importance of including spatial structure and environmental conditions when explaining patterns of community composition to better discriminate between the ecological processes that underlie biogeographical patterns of communities composed of native and non‐native fish species.