| Summary: | One of the most remarkable effects of climate change in marine realm is the shift in the distribution of species. On the other hand, marine ecosystems are also being modified by commercial fisheries (e.g., decline of species with low reproductive rates, alterations in population size structure, habitat loss and decreased habitat quality). Fisheries combined with climate change may cause the redistribution of marine species commercially exploited by fisheries and these changes can cause considerable socio-economic impacts. Therefore, it is crucial to anticipate these changes aiming a sustainable management of marine living resources and food security. The European hake (Merluccius merluccius) is a demersal species while the horse mackerel (Trachurus trachurus) is a pelagic-neritic species, and both are among the most important species commercially exploited in the Northeast Atlantic. In this work, I estimated the habitat suitability along the Iberian Atlantic waters for the European hake and horse mackerel using an ensemble ecological niche model (ENM) approach. Three machine-learning algorithms (Generalized Boosting Model, Random Forest, Maximum Entropy) were selected to estimate the habitat suitability and related uncertainty maps. The predictions of habitat suitability were made for present ocean conditions (2002-2014) and then projected to future scenarios of climate change (RCP 2.6 and 8.5) for the years of 2050 and 2100. The models were fitted using both oceanographic variables (temperature, salinity, and current velocity) and physiographic variables (bathymetry, bathymetric slope, and sea bottom curvature). The variables that most contribute to the model’s predictions were bathymetry, bottom temperature, and salinity for the European hake and bathymetry and salinity for the horse mackerel. Both the ensemble and single models showed high predictive performance, with the ensemble model estimates consistently performing better than the single models. According to the model for the European hake, the most ...
|
|---|