EXTRAPOLATING FROM INDIVIDUAL BEHAVIOR TO POPULATIONS AND COMMUNITIESIN STREAMS

Biological processes, such as predator-prey or competitive interaction s, occur at multiple spatial and temporal scales, but their impacts on the distribution, abundance and fitness of organisms may only be dete ctable at some scales. This is because 1) small scale (local) processe s may be constrained by large-scale (regional) processes, 2) the influ ence of organism movement changes with spatial scale, and 3) multiple small-scale processes may interact and produce variation that obscures large-scale patterns. Most ecologists would like to know the relevanc e of small-scale observations and experiments for large-scale patterns and processes acting over long time periods. In this paper we conside r whether patterns of individual behavior translate in a straightforwa rd way to patterns in population dynamics, community structure or indi vidual fitness at larger spatial and longer temporal scales. We illust rate our discussion with data from communities living in high altitude streams in western Colorado, coastal streams in southern California, and streams on the South Island of New Zealand. We describe patterns o f individual behavior including selective predation, predator-avoidanc e behavior, and competitive interactions. Our examples indicate that l arge scale patterns of distribution and abundance of organisms sometim es deviate from those expected from patterns of individual behavior. W e hypothesize that such discrepancies are often due to multiple confou nding factors, including abiotic disturbances, that have effects on po pulations and communities at larger scales. Similarly, our data illust rate that present behaviors may not have measurable fitness consequenc es over longer time scales. We emphasize that only by studying all lif e history stages of organisms with complex life cycles can we interpre t the long-term fitness consequences of individual behaviors. We concl ude that our ability to extrapolate from individual behaviors to popul ations and communities in streams will improve when we consider the re percussions of individual behaviors and environmental factors over bro ader spatial and temporal scales.