A synthesis of variability in nearshore Alaskan marine populations.

One of the primary goals of any monitoring program is to detect anthropogenic changes; however natural variability can be so high that it prevents detection of human-induced effects. This project synthesized existing data to identify patterns of temporal variation within nearshore marine habitats in the Gulf of Alaska and was motivated by the need for monitoring in this environment vulnerable to oil spills, such as the Exxon Valdez. I collected 786 time series that were greater than 2 years in length from unimpacted (control) sites for 226 species. Temporal variability (CV) varied widely from 1% to 447%, averaged 89%, and was not significantly different among marine populations of birds, mammals, fish, algae and invertebrates. Temporal variability of invertebrates and algae were not significantly different when abundance was measured as biomass, percent cover or density. Both invertebrates and algae showed higher variability in low intertidal habitat compared to high intertidal habitat. For invertebrates, populations on bedrock substrates had greater temporal variability than those on cobble or soft sediment substrates, while there was not a significant difference for algae on either cobble or bedrock substrates. Many of these results are surprising, given differences in movement patterns, survey methods, life histories, and so forth. These results suggest that no single group, measure or habitat will be better or worse for monitoring and detecting change. Understanding variability is a difficult task, but until we tackle it, we will likely not understand or have predictive capabilities in ecological populations and communities.