Time-Varying Epipelagic Community Seascapes: Assessing and Predicting Species Composition in the Northeastern Pacific Ocean

The vast spatial extent of the ocean presents a major challenge for monitoring changes in marine biodiversity and connecting those changes to management practices. Remote-sensing offers promise for overcoming this problem in a cost-effective, tractable way, but requires interdisciplinary expertise to identify robust approaches. In this study, we use generalized additive mixed models to evaluate the relationship between an epipelagic fish community in the Northeastern Pacific Ocean and oceanographic predictor variables, quantified in situ as well as via remote-sensing. We demonstrate the utility of using MODIS Rrs555 fields at monthly and interannual timescales to better understand how freshwater input into the Northern California Current region affects higher trophic level biology. These relationships also allow us to identify a gradient in community composition characteristic of warmer, offshore areas and cooler, nearshore areas over the period 2003–2012, and predict community characteristics outside of sampled species data from 2013 to 2015. These spatial maps therefore represent a new, temporally and spatially explicit index of community differences, potentially useful for filling gaps in regional ecosystem status reports and is germane to the broader ecosystem-based fisheries management context.