Future Projections of Sea Surface Temperature (SST) in the MDR and Wider Caribbean Region: Utilising CMIP6 GCM Ensembles Climatology

This study evaluates the performance of CMIP6 global climate models (GCMs) in simulating and projecting sea surface temperature (SST) variations across the Caribbean and the tropical Main Developing Region (MDR) that are climatically sensitive and of significant ecological and socio-economic importance. Particular focus is given to the Late Rainfall Season (LRS; August to November), which is critical for regional rainfall and hurricane activity. Using a suite of 14 CMIP6 models to create a multi-model ensemble mean, we assess SST projections under two Shared Socioeconomic Pathways (SSP2-4.5 and SSP5-8.5). Model validation was conducted against the NOAA Optimum Interpolation SST (OISST) dataset using the historical period 1982–2011. Model performance for the LRS and annual was robust, with Pearson correlation coefficients (R) of 0.84, respectively, indicating that the models capture over 71% of the observed SST variability. A low root-mean-square error (RMSE) of 0.17 further confirms their fidelity in reproducing historical SSTs. Projected SST trends indicate pronounced warming in the Caribbean Sea and Gulf of Mexico, with mid-century (2036–2065) increases during the LRS and more dramatic warming by late-century (2071–2100), especially under SSP5-8.5. A spatial gradient of warming is evident, with stronger warming in the western Atlantic and a relatively cooler trend in the eastern basin. These findings underscore the strong performance of CMIP6 models in representing regional SST dynamics and highlight significant projected warming that may alter rainfall regimes, hurricane characteristics and marine ecosystems. By combining model evaluation with future projections, this study provides a reliable foundation for understanding SST-driven climate impacts and supports informed adaptation planning in the Caribbean and broader tropical Atlantic. Grant No. NA22SEC4810016 Grant No. NA22OAR4310545