Abstract
Climate change, as one of the most pressing challenges of the current century, has had profound implications for marine and coastal ecosystems, particularly coastal arid lands. Increases in sea surface temperature (SST), the occurrence of extreme climate events, and intensified thermal stresses are among the consequences of this global phenomenon. In this context, the application of downscaling numerical models such as the Weather Research and Forecasting (WRF) model represents a novel approach for modeling SST in such sensitive regions. The Persian Gulf, as one of the world's most important coastal arid ecosystems characterized by high annual evaporation, elevated salinity, shallow depth, and strong seasonal temperature fluctuations exhibits particular ecological vulnerability to climate change. The primary objective of this study is to identify the most optimal configuration of physical parameterization schemes within the WRF model for simulating SST in the Persian Gulf. SST data were obtained from the Iran Meteorological Organization. The WRF model was executed using eight different configurations, and the outputs were validated using RMSE, MAE, and MAPE statistical indices. The results indicated that Configuration No. 5 comprising the WSM 5-class microphysics, RRTM longwave radiation, Dudhia shortwave radiation, Revised MM5 Monin-Obukhov surface layer, Unified Noah land surface model, YSU planetary boundary layer, and Kain-Fritsch cumulus parameterization exhibited the lowest error rates and the highest correlation coefficient (0.985), and was thus identified as the optimal combination. RMSE values for spring, summer, autumn, and winter were calculated as 0.036°C, 0.012°C, 0.028°C, and 0.011°C, respectively. Based on the findings, WRF model outputs can serve as a suitable, high-precision alternative to Bushehr meteorological buoy data for climate change monitoring studies, fisheries management, coral reef protection against bleaching induced by global warming, a
