AI Driven Analysis and Prediction of Mobility Pattern Changes in Response to Atmospheric River Events

Atmospheric rivers are increasingly recognized as major drivers of extreme weather, particularly along the U.S. West Coast, where they frequently lead to severe flooding. These events cause significant disruptions in mobility patterns, presenting substantial challenges for emergency response and public safety. This study employs advanced artificial intelligence (AI) architectures to analyze and predict changes in mobility patterns, using the 2024 atmospheric river event in Southern California as a case study. Our focus is on understanding the implications for community resilience and disaster preparedness.

We utilize a comprehensive mobility dataset, comprising information from mobile devices and vehicle navigation histories, aggregated into 100-meter resolution geographic tiles to ensure individual anonymity. Spanning from 2021 to 2024, this dataset offers critical insights into mobility behavior before, during, and after atmospheric river events. By applying various AI architectures, we aim to capture temporal dependencies and spatial patterns, building a robust AI framework that enhance es our understanding of how people respond to atmospheric river events.

The study begins with an exploratory data analysis (EDA) to identify key trends, anomalies, and spatio-temporal patterns within the mobility data. Subsequently, we develop and validate our AI framework using mobility data recorded during the 2024 atmospheric river event in Southern California, focusing on the framework's ability to predict mobility pattern changes induced by such events.

Our findings are expected to provide valuable insights into how different communities, characterized by varying levels of vulnerability and resources, respond to atmospheric river-induced flooding. These insights will inform the optimization of emergency management strategies, particularly in evacuation planning and resource allocation. Additionally, the study addresses the ethical considerations inherent to applying AI to this context, ensuring that the models are developed and deployed in a manner that promotes fairness and equity.

By leveraging AI to enhance the understanding and prediction of mobility pattern changes, this research paves the way for more resilient and informed decision-making processes in response to the increasingly frequent and severe atmospheric river events.