Modelling and mapping regional‐scale patterns of fishing impact and fish stocks to support coral‐reef management in Micronesia

Abstract Aim Use a fishery‐independent metric to model and map regional‐scale fishing impact, and demonstrate how this metric assists with modelling current and potential fish biomass to support coral‐reef management. We also examine the relative importance of anthropogenic and natural factors on fishes at biogeographical scales. Location Reefs of five jurisdictions in Micronesia. Methods A subset of 1,127 fish surveys (470 surveys) was used to calculate site‐specific mean parrotfish lengths (a proxy for cumulative fishing impact), which were modelled against 20 biophysical and anthropogenic variables. The resulting model was extrapolated to each 1 ha reef cell in the region to generate a fishing impact map. The remaining data (657 surveys) were then used to model fish biomass using 15 response variables, including fishing impact. This model was used to map estimated current regional fish standing stocks and, by setting fishing impact to 0, potential standing stocks. Results Human population pressure and distance to port were key anthropogenic variables predicting fishing impact. Total fish biomass was negatively correlated with fishing, but the influence of natural gradients of primary productivity, sea surface temperature, habitat quality and larval supply was regionally more important. Main conclusions Mean parrotfish length appears to be a useful fishery‐independent metric for modelling Pacific fishing impact, but considering environmental covariates is critical. Explicitly modelling fishing impact has multiple benefits, including generation of the first large‐scale map of tropical fishing impacts that can inform conservation planning. Using fishing impact data to map current and potential fish stocks provides further benefits, including highlighting the relative importance of fishing on fish biomass and identifying key biophysical variables that cause maximum potential biomass to vary significantly across the region. Regional‐scale maps of fishing, fish standing stocks and the potential benefits of protection are likely to lead to improved conservation outcomes during reserve network planning.