Karinja Thejaswi

Tropical storm surge poses significant risks to coastal areas, necessitating precise prediction for effective emergency preparedness and mitigation. Recent advances in numerical models such as SLOSH, ADCIRC, and FVCOM have revolutionized storm surge forecasting by accurately simulating complex hydrodynamic processes, bolstered by ADCIRC's use of high-resolution grids and parallel computing for enhanced predictive capabilities crucial in emergency management. Hybrid models that integrate numerical simulations with statistical and machine learning techniques further refine forecasts, utilizing real-time observational data to correct biases and improve initial conditions dynamically. Artificial Intelligence (AI) and machine learning, particularly neural networks like CNNs and RNNs, play pivotal roles in improving prediction accuracy by analyzing spatial and temporal data from diverse sources, thus facilitating real-time data assimilation critical for operational forecasting systems. Future advancements are expected to deepen AI integration, refine data assimilation, and enhance global accessibility to real-time storm surge data, promising increased resilience and improved disaster response strategies worldwide.