No abstract

No abstract

Understanding the spatial structure of individual variation is critical to assessing the mechanisms and scale of species‐level ecological and evolutionary responses. But the continuity and complexity of streams can make it difficult to discern spatial structure, and data on individual variation spanning spatial scales are rare. Our goal was to objectively resolve the scale of ecological and evolutionary processes affecting the salamander Gyrinophilus porphyriticus by testing for spatial structure in individual variation across a headwater stream network. We used 3 years of spatially explicit capture–mark–recapture data from four headwater streams in the Hubbard Brook Experimental Forest, New Hampshire, USA, to test for individual variation in survival at three scales: (a) among c. 2‐km2 watersheds, (b) in 500‐m downstream and upstream reaches within watersheds and (c) in riffle and pool habitats along reaches. We also tested for corresponding spatial variation in individual body condition and morphology to gain insight on the causes and implications of variation in survival. Survival of G. porphyriticus larvae was constant among watersheds, but differed by reach and habitat. Larval survival was higher in upstream reaches than downstream reaches, matching the distribution of predatory brook trout (Salvelinus fontinalis), which were restricted to downstream reaches. Larval survival was also higher in pools than riffles. Survival of G. porphyriticus adults was constant at all scales. Larval body condition was higher in downstream reaches than upstream reaches. Both larvae and adults differed in morphology at the habitat scale: individuals in pools had longer limbs than individuals in riffles. Negative covariation in larval survival and body condition at the reach scale suggests that there is a trade‐off between fitness and proximate performance along streams. The surprising differences between habitats in survival and morphology show the potential for fine‐scale ecology–evolution interactions in streams, potentially driven by differences in flow and gradient conditions in riffles and pools. Our results suggest that the larval stage is key to understanding individual, population and community‐level processes affecting G. porphyriticus. More broadly, this research provides a novel empirical link between traditional studies of fine‐scale ecological complexity in streams and recent studies showing consistent spatial dynamics at the network scale. By documenting differences in survival, body condition and morphology at reach and habitat scales, we hope this work shows the feasibility and value of spatially explicit approaches to species‐level ecological and evolutionary questions in streams.

No abstract