Effects of riparian forest harvest on streams: a meta‐analysis

Summary Riparian forest harvesting impacts streams in many ways, from altering temperature regimes, shifting geomorphic structure, increasing sediment fluxes and affecting fish populations. However, we have noted considerable variation in the results between studies that led us to ask whether the effects of forest harvesting on streams were consistent between studies. We used meta‐analysis of 34 replicated studies to address the effects of riparian logging on biological and chemical components of streams in contrast to control sites. We found that the overall effect sizes of several response variables in replicated studies were significantly higher than zero, especially benthic invertebrates, and nitrogen and potassium concentrations. However, there was a very large amount of variation in the effect sizes between studies, and for many measures, the effect sizes from different studies were positive or negative, indicating site‐specific responses. We explored whether stream size, stream gradient and regional potential evapotranspiration could explain some of the effect size variation between studies. Relations with these environmental variables were weak, but suggestive that some of the context‐specific, individual outcomes might be due to underlying environmental differences between sites. Synthesis and applications . Despite relatively low numbers of replicated studies, we found significant overall effects of riparian forest harvesting although the magnitude and direction of responses within individual studies were site specific. This lack of consistency in the direction of effect sizes suggests we need a more context‐dependent approach to the protection of freshwaters from forest management.