Machine Learning-Based Cellular Automata Model and Frame Work for Tsunami Wave Propagation in Indian Coastal Region

Simulating the spread of tsunami waves quantitatively holds significant importance in devising swift risk management strategies and implementing efficient solutions for predicting the earliest arrival time. One effective method for dynamically simulating the spatiotemporal evolution of complex systems is through the use of cellular automata (CA), making it a suitable choice for modeling the intricate process of tsunami wave propagation. Nevertheless, the phenomenon of tsunami wave propagation is intricately linked with numerous interdependent factors, rendering it too intricate for analysis using traditional approaches. In answer to this challenge, we present a brand-new method for simulating the propagation of tsunami waves called LSSVM-CA. In this novel method, we integrate a three-dimensional tsunami wave CA framework with least squares support vector machines (LSSVM). We can now take into account and examine how neighboring waves affect the propagation patterns thanks to this integration. Using a dataset derived from historical data, LSSVM is trained to create sophisticated state transformation rules for tsunami wave propagation. We compared the LSSVM-CA-simulated tsunami wave spread areas with the real observed tsunami wave spread areas in order to evaluate the efficacy of the model. The outcomes unequivocally show how well LSSVM-CA simulates tsunami wave propagation accurately calculates a probability of the waves spreading.