The global phenomenon of climate change has given rise to increasingly frequent and severe extreme weather events, including floods and droughts. The Gulf of Mexico (GOM) region has been notably impacted by such extreme events which not only pose environmental challenges but also bring significant socio-economic repercussions. While Global Climate Models (GCMs) such as Coupled Model Intercomparison Project 6 (CMIP6) offer insights into global climatic trends, their application at local and regional scales is limited due to low spatial resolution. To bridge this gap, this study employs a two-stage downscaling process using Recurrent Neural Network-Long Short-Term Memory (RNN-LSTM) to refine CMIP6 data for localized impact assessment, with a specific emphasis on coastal areas. Data from 1950 to 2100 were downloaded directly via Python and clipped to the region of interest, using custom Python scripts. Historical climate data are available until 2014, with projected data obtained from 11 selected models across four future scenarios. Additionally, we have compiled historical flow data for the region dating back to the 1960s for further water balance analysis. All selected models consistently provide five key climatic parameters: humidity, precipitation, minimum, mean, and maximum temperature. Stationarity and trend analyses are conducted on the climate and flow data to understand their evolving patterns over time. Established indicators such as the Standardized Precipitation Index (SPI) and, Standardized Precipitation-Evapotranspiration Index (SPEI) are employed for a comprehensive analysis of drought and flood risks. The results of this analysis were validated using a physical based water balance method. Preliminary findings suggest that future extremes are predicted to surpass historical events, underlining the need for immediate, localized adaptation and mitigation strategies. This research aims to contribute valuable localized data for water resource planning and management in the Gulf of Mexico region, thereby aiding in the formulation of effective climate adaptation and resilient policies.
