Gillnets and trammel nets are widely used in artisanal fisheries in developing countries due to their relatively high catch efficiency, low cost, and ability to target commercially valuable species. However, gillnets are among the most selective fishing gears in terms of both species and size retention. The objective of the current study was to determine which selectivity curve was most representative of the gear from which the data were collected, and to analyze the effect of different selectivity models on the estimated mortality. From March, 2015 to March, 2016, 58 landings surveys, directed towards artisanal gillnet fishermen, were collected in the Bay of Tela, Honduras. In each survey length and weight data, gear characteristics, fishing effort, and fishing location data were collected. Because of the importance of Lane Snapper (Lutjanus synagris) for the artisanal fisheries in Honduras, this study focused on the data collected for it, providing insights into the management of this important fishery resource. Data from 1261 Lane Snapper were collected, and subsequent selectivity, age at length, and catch curve analyses were performed. Three selectivity models (Normal, Lognormal, and Skewed Normal) were investigated, as well as two mortality estimation thresholds.
Results/Conclusions
Based on the AIC, skewed normal was the best model for the selectivity correction. Adjusting for selectivity had a large effect on mortality estimation for all scenarios that were tested; for the data adjusted by the skewed selectivity curve the mortality estimations (Z values) ranged from 0.77 to 1.93, with unadjusted Z values ranging from 1.6 to 2.07. The two different thresholds that were tested (Threshold A and Threshold B) in each scenario for mortality estimation had only a minor effect on the resultant mortality estimate. Combining the data from both mesh sizes (mesh size 2” and mesh size 3”) versus analyses from the separate meshes had an effect on the mortality estimation, with results from using only data from the 3” mesh size tending to overestimate mortality compared to the results for 2” mesh size only, or combined data. Overall, adjusting for selectivity has a clear effect on mortality estimates, which in turn has an effect on our understanding of the ecology of a targeted species. The ecological implication of ignoring gear selectivity when estimating mortality is that the mortality may be vastly overestimated, risking overexploitation of the species of interest.