Bayesian Approach to Model Silky Shark (Carcharhinus falciformis) Presence Probability From Bycatch in the Eastern Pacific Ocean Tuna Fishery

Abstract

The silky shark (Carcharhinus falciformis) has the highest bycatch in gillnets, longlines, and purse seine globally. Its thermal and depth preferences coincide with the tropical tuna habitat, indicating a high capture vulnerability by these fisheries. The present research seeks to determine the environmental conditions that favor silky shark distribution and identify the areas with the highest probability of occurrence. The data analyzed correspond to silky shark bycatch records in the Eastern Pacific Ocean (EPO) published by the Inter-American Tropical Tuna Commission from 2009 to 2019. The dataset contains information about bycatch (number of organisms), the number of sets made, as well as the fishing indicator (schools not associated [NOA], associated with dolphins [DEL], and with floating objects [OBJ]). The environmental variables sea surface temperature (SST), sea surface height (SSH), mixed layer depth (MLD), chlorophyll-a (Chl-a), and dissolved oxygen (DO) concentrations were considered to perform a Bayesian Additive Regression Trees (BART) to model the entire species range distribution. The best-fitted model included MLD as the most informative variable, followed by Chl-a concentration and SSH. BART model identified the highest presence probabilities in the oceanic area between 15°S and 10°N and −110°W to −150°W, along the North Equatorial Current and around French Polynesia. The BART model results offer significant insights into silky sharks in the EPO, showing that the equatorial and oceanic areas are the most important for its distribution, supporting the findings from previous studies. This information is crucial for developing effective management strategies to reduce silky shark bycatch.