Modelling the spatio-temporal bycatch dynamics in an estuarine small-scale shrimp trawl fishery

The Patos Lagoon estuary in Southern Brazil is an important nursery ground for many commercially important fish and invertebrate species, with shrimp fisheries sustaining most of the surrounding small-scale fishermen. Although they are illegal, trawl gears are widely used by small-scale fishermen in order to catch shrimp. Shrimp trawl fisheries in tropical waters have high bycatch rates and cause a variety of problems, spanning from conflicts with other fisheries to changes in ecosystem trophic dynamics. A better understanding of the factors driving the bycatch process is therefore needed to establish reliable management measures. Here we used hierarchical Bayesian spatio-temporal models (HBSMs) to describe the bycatch of a small-scale shrimp trawl fishery in the Patos Lagoon estuary relative to a set of environmental and spatio-temporal predictors. Two different models were used to estimate bycatch trends: an effort-based method (BPUE) and a ratio-based method (BR, bycatch rate). A third model was used to evaluate the dynamics of shrimp catch per unit of effort (CPUE) as this measure is equally important to describe the underlying bycatch process. Overall, the bycatch consisted of 61 species of which 45 have market value. The mean bycatch rate was 86.2% for the entire period, which represents 6 kg of bycatch for every 1 kg of shrimp caught. Although the contribution of environmental variables differed among the models, the yearly spatial effect was identified as a common that explained bycatch dynamics. Moreover, the predictive maps revealed consistent hot and cold-spots from year-to-year, which varied in intensity for BR, BPUE and CPUE, and, therefore, represent a valuable benchmark for stakeholders and managers.