- In this study, we used a combination of conventional statistical analyses and mechanistic bioenergetics models to understand the potential effects of environmental variability on growth of stream-living fishes in four headwater streams subject to upstream forest harvest. We focused on two common fish consumers in headwater streams in the Pacific Northwest, USA: trout (Oncorhynchus clarkii clarkii and O. mykiss) and sculpin (Cottus spp.). Statistical analyses focused on patterns of summer growth in relation to upstream forest harvest, and the effects of annually variable stream temperatures, stream discharges, and biomasses of conspecifics. Statistical and bioenergetics models were based on 2,524 individual measurements of growth for trout and 208 measurements for sculpin observed over 8 summers (July-Sept) in four headwater streams in the northwestern Oregon Coast Range. Statistical analyses of patterns of growth indicated no effect of forest harvest on growth of either species. However, increased temperature was associated with increased growth rates, after adjusting for stream discharge and biomass of conspecifics. To further elaborate processes driving growth of fish in the streams we studied, bioenergetics simulations were applied to estimate relative consumption (observed consumption / maximum consumption) based on energy inputs (prey energy densities), temperature and size-dependent physiological costs, and measured growth. Simulations yielded relative consumption values that ranged from 0 to 0.19 (mean = 0.076) for trout and from 0.016 to 0.63 (mean 0.33) for sculpin. Although sculpin exhibited higher levels of relative consumption than trout, 66% of individuals lost weight. In contrast, most trout gained weight (53%), even with lower relative consumption levels. These contrasts reflected higher modelled physiological maintenance costs for sculpin relative to trout. We applied the bioenergetics model to evaluate warming and cooling of stream temperatures within and outside of the range of observed variation and estimated that trout would continue to have positive summer growth, if relative consumption remained unchanged, over a large range of temperatures. Sculpin, on the other hand, were expected to have increased growth rates if mean water temperature increased by up to 1°C, but would be sensitive to further temperature perturbations (positive or negative), unless consumption levels increased. Additionally, sculpin were the dominant fish in terms of biomass and abundance in three of the four stream sites. Collectively, these findings point to a suite of key processes driving growth of fish in headwater streams of the Pacific Northwest Coast, variability in sensitivity of species to environmental conditions, and the importance of sculpin: a group of typically overlooked species that appear to be more abundant and sensitive to environmental conditions than more commonly-studied trout.