Quantifying habitat associations in marine fisheries: a generalization of the Kolmogorov-Smirnov statistic using commercial logbook records linked to archived environmental data

摘要

Understanding species-habitat associations is critical for designing marine reserves, defining essential fish habitat, and predicting the impacts of climate change on fisheries. For many species, however, there is a paucity of fisheries-independent data that simultaneously track abundance and environmental variables, as is the case for widow rockfish (Sebastes entomelas), a commercially important fishery off the west coast of the United States. In this paper, I generalize a previous approach to identifying habitat associations so that fisheries-dependent data can be used. In analyzing Oregon commercial logbook records and archived environmental data from the National Oceanographic Data Center, I found three environmental variables (bottom depth, vertical depth of fish in the water column, and temperature) to be statistically adequate. Using a generalized Kolmogorov-Smirnov test statistic, I compared an empirically derived cumulative distribution function (CDF) of the habitat sampled to a CDF weighted by widow rockfish catch. Results suggest that the significant habitat association for widow rockfish includes bottom depths between 136 and 298 in, vertical depths between 101 and 197 in, and temperatures between 7.1 and 8.1degreesC. This novel use of commercial logbook data, which links disparate data sources and explicitly accounts for unequal spatial sampling, is a methodological advance that also provides initial insights into widow rockfish habitat preferences.