Influência das variáveis ambientais na redistribuição dos juvenis de Atum de barbatana amarela (Thunnus albacares) no nordeste do Oceano Índico

Abstract

The main objective of this study was to investigate the effects of environmental variables on the redistribution of juvenile tuna (Thunnus albacares) in the western Indian Ocean. For this study, purse seine catch data from the YFT were obtained from the database of the Spanish Oceanographic Institute (IEO) for the period from January 2005 to December 2013. Data on environmental variables for the 2005-2013 study period were acquired from the MyOcean-Copernicus EU consortium website (marine.copernicus.eu) in netCDF format and were extracted using specific codes from the netCDF4, tidyverse, and lubridate packages, as well as other basic functionalities of the R program, version 4.3.1. The generalized additive GAM model was used through the regression method, which consists of adjusting all previously available variables and eliminating one variable at a time according to the significance level (p-value <0.05). Subsequently, the k-fold cross-validation method was used to validate the selected GAM in the study area. The climatic variables included in the model accounted for approximately 33.2% of the variation in yellowfin tuna (Thunnus albacares) catches in the Northeastern Indian Ocean. The model evaluation, using k-fold cross-validation, indicated satisfactory performance, with a root mean square error (RMSE) metric of approximately 0.6, a Pearson correlation (rho) of approximately 0.47, and a similarity index (D) of 0.83. According to the data provided, tuna presence was closely related to seasonal conditions, with a predominance in summer, especially in the northernmost latitudes (7ºN to 12ºN). This translates into relatively higher catch rates during this season, while in central and southern latitudes (from -13°S to 2°N), catches were significantly lower, especially during the transition and winter periods. Furthermore, the analysis of environmental variables and catch rates for the studied species demonstrated that primary productivity and seasonality were the most prominent variables, that is, those that most influenced the redistribution of these juvenile tuna. This study may contribute to improving management and conservation measures for this species, as our model predicts the locations of greatest occurrence of the species in correlation with oceanographic variables.