Abstract: The wave climate along the Indian coast is predominantly influenced by the Indian Ocean to the south, the Bay of Bengal (BoB) to the east, and the Arabian Sea to the west. With the growing threat of climate change, the impact on wave climate, leading to heightened storm effects and rough sea conditions, is of paramount concern. The study delves into the wind-wave climate under cyclonic storms in the active BoB region, particularly considering three severe cyclonic events: Phailin (2013), Hudhud (2014), and Vardah (2016). To ensure accurate prediction of high-impact waves, the study emphasizes the importance of suitable wind data resolution (Weather Real Money Rummy and Forecasting (WRF) model surface wind data) and a stable wave model grid resolution. Investigating the impact of different spatial resolutions (1/5°, 1/10°, 1/20°, and 1/40°) on wave model (WAM) prediction, the study concludes that resolutions better than 1/10° yield no significant difference in Hs prediction accuracy. Then these cyclones are projected into the future under Representative Concentration Pathway (RCP) scenarios (RCP4.5, RCP6.0, and RCP8.5) for both Near-Future (2035) and Far-Future (2075). Results indicate an increase in intensity for all three cyclones under the Far-Future scenarios of RCP6.0 and RCP8.5, with Hudhud experiencing the greatest intensity of approximately 21%. In addressing the computational challenges associated with physical models, the study incorporates a deep learning model using the Long-Short Term Memory (LSTM) algorithm. Trained with ERA5 reanalysis data and real-time observational data, the deep learning model outperforms traditional physical models, providing an efficient alternative for predicting significant wave height. The overall study emphasizes the need for optimal model resolutions and underscores the potential of data-driven models in advancing our understanding of wave climate under changing climatic conditions.