Search

With the ongoing climate change affecting the ocean, there is a need to understand and predict the future distributions of marine species in order to assess the sustainability of marine ecosystem. In this study, remotely-sensed satellite oceanographic data together with Indian mackerel (Rastrelliger kanagurta) fishery dataset were used to predict potential fishing grounds in the Exclusive Economic Zone (EEZ) of Peninsular Malaysia using boosted regression trees (BRT) model. The model was developed using three years (2008–2010) fish catch data and environmental variables of chlorophyll-a (chl-a), sea surface temperature (SST) and sea surface height (SSH). Result indicated that potential fishing grounds were closely associated with SSH, followed by SST and chl-a. The performance of the BRT model indicated acceptable fishing grounds prediction accuracy (AUC value of 0.749). Seasonal variability in fishing grounds was related to favorable environmental conditions of SSH (1.1–1.3 m), SST (29–32 °C) and chl-a (0.3–0.6 mg/m3). The projection of increases in SST due to climate change according to IPCC-AR5-RCPs was observed to influence the spatial and temporal distributions R. kanagurta. Increased temperature at 1.80 °C resulted in high potential catch areas for R. kanagurta in the EEZ. Meanwhile, elevated temperature at 2.60 °C and 3.30 °C showed decreased in potential catch areas for R. kanagurta in the EEZ. Most of the future fishing grounds area were projected to decline, and it was observed to shift outside the EEZ off South China Sea. Hence, by understanding these relationships, this analysis identifies where strategies can be adapted to face the ecological impacts under changing environmental conditions.