Aquatic organisms respond to the physical environmental across a range of spatial scales.  Determining the direct and indirect impacts of environmental factors on aquatic biota poses a significant challenge in understanding habitat quality and in forecasting ecosystem responses to environmental alterations in watersheds.  Whereas avian species are not commonly considered in aquatic studies, they may play a critical role in shedding light on factors that operate at distinct spatial scales.  In particular, the semi-aquatic American dipper (Cinclus mexicanus) may be particularly valuable in elucidating influences of spatially-explicit environmental factors.  In this study, we investigated the influence of large-scale, regional variables (e.g., precipitation, temperature, latitude, etc.); landscape-scale, watershed variables (e.g., land use, land cover, road density, etc.) and local, reach-level variables (e.g., geomorphology, substrate, habitat complexity, riparian buffer, etc.) on measures of American dipper reproductive success, feeding ecology (as measured by stable isotope analysis using δ¹³C and δ¹⁵N), and individual condition.  From 2005 to 2008, we collected these data at 26 American dipper breeding reaches spread across a large swath of Idaho, USA.  Using a combination of principal component analysis and regression modeling, we explored potential environment-American dipper links at each spatial scale.  We then used these results to inform our structural equation modeling (SEM). 

Results/Conclusions

Our results suggest that landscape characteristics cast the widest net, influencing dipper reproductive measures, trophic position, and individual condition.  Regional attributes predominantly influenced dipper condition and weakly predicted δ¹³C values.  Local characteristics appeared to be the least critical, relating only to the number fledged.  Results from structural equations models, which considered the spatial variables in a unified context, showed that whereas dipper productivity is likely influenced by a fairly complex net of scalar factors, energy flows and individual condition may be more directly constrained.  Overall, our results show that relationships between the physical environment and American dippers not only vary by spatial scale, but also by facets of dipper ecology.  Our results also offer evidence that although stream-watershed habitats are spatially-nested, pathways influencing American dippers, and perhaps other aquatic biota, may not always conform to hierarchical constructs.