Supporting information for "Can upwelling regions be potential thermal refugia for marine fishes during climate warming? "

Supporting information of the manuscript "Can upwelling regions be potential thermal refugia for marine fishes during climate warming? "
These are the files created for the project titled " Can upwelling regions be potential thermal refugia for marine fishes during climate warming" The "Average_richness" file shows the total pixel richness, which is the sum of the different binary maps developed. The "Average Suitability" folder contains the average value of all species suitability maps. The "Difference_suitability" and "Difference_richness" files show the subtraction results between suitability and richness, respectively, for the present and different RCP scenarios for the periods 2040-2050 and 2090-2100. The "Exploratory_figures" folder displays the maximum, mean, and minimum temperatures at which the species are reported. The Jaccard turnover results are stored in the "Jaccard" folder. The filenames indicate the period (Present, 2040-2050 (2050), and 2090-2100 (2100)), the RCP scenarios (6.0 (6) and 8.5 (8)), and the filtered used (0,50,100 and 500 for 0 km filter, 50 km filter, 100 km filter, and 500 km filter respectevely). The two folders named "Species_figure" and "Species_raster" contain separate results for each species. The "Species_figure" folder displays the results graphically, while the "Species_raster" folder has files in raster format (.tif). Both folders have subfolders indicating which database was used to create the maps, such as 0, 50 and 100 km.
The folder named "Calibration_results" contains the results of the model calibration achieved in this study. Within this folder, the subfolder "train" displays the data used for every species in each of the ten runs to create the minimum volume ellipses. The subfolder "raw" displays all the results from every possible combination of models. Lastly, the subfolder "weight" displays the average results of the chosen model. Finally, Pearson correlationship of databases with more than 30 occurrences are shown in the folder “Correlationship”. To replicate our results, please access the "Code_replication" folder, which includes the required scripts and databases. Please, open first the “rProj” file in rstudio and follow the numbered scripts. Note that the unnumbered scripts are functions to run the models.
Finally, we stored the occurrences downloaded from the Ocean Biodiversity Information System (OBIS) and Vertnet, along with their references, using the 'robis' and 'vertnet' packages, respectively (“Occurrences” folder). For the data obtained from the Global Biodiversity Information Facility (GBIF), we stored it as data keys and the number of records to create a derived dataset, which can be accessed and cited following the next link: https://doi.org/10.15468/dd.8wacqq.