Precision Tsunami Prognostication: A Machine Learning Expedition for Predictive Accuracy

In coastal regions, the ever-present threat of tsunamis necessitates rigorous disaster preparedness strategies. This study represents a groundbreaking advancement in tsunami forecasting accuracy, employing various machine learning models to predict tsunami characteristics and arrival times. Comprehensive data spanning historical earthquake, tsunami, and landslide records from 1800 to 2023 were meticulously collected from reputable sources. Machine learning models, including Random Forest (RF), Support Vector Machine (SVM), Linear Regression, and neural networks, were applied to predict tsunami wave speed and travel time. Notably, the Enhanced Random Forest model demonstrated an exceptional predictive accuracy, achieving a Mean Squared Error (MSE) of 84.40 and an R-squared value of 0.9989, which significantly outperforms the other models evaluated in this study.This research contributes to achieving Sustainable Development Goal 11 (Sustainable Cities and Communities) and SDG 13 (Climate Action) by integrating climate change considerations into tsunami prediction and preparedness. The outcomes underscore the robustness of the models in accurately forecasting tsunami characteristics and arrival times, surpassing previous benchmarks. By harnessing the power of these advanced predictive models, we have achieved unparalleled accuracy, setting a new standard in tsunami forecasting. This advancement enhances our understanding of tsunami dynamics and empowers disaster management authorities and coastal communities with actionable insights for targeted preparedness and mitigation strategies. Through global collaboration and the application of cutting-edge predictive models, this research aims to safeguard vulnerable coastal regions, offering indispensable contributions to disaster risk reduction and coastal resilience as a process innovation.