Quantifying, Comparing Human Mobility Perturbation during Hurricane Sandy, Typhoon Wipha, Typhoon Haiyan

Abstract

Climate change has intensified tropical cyclones, resulting in several recent catastrophic hurricanes and typhoons. Such disasters impose threats on populous coastal urban areas, and therefore, understanding and predicting human movements plays a critical role in disaster evacuation, response and relief. Despite its critical roles, limited research has focused on tropical cyclones and their influence on human mobility. Here, we studied how severe tropical storms could influence human mobility patterns in coastal urban populations using individuals’ movement data collected from Twitter. We selected three significant tropical storms, including Hurricane Sandy, Typhoon Wipha, and Typhoon Haiyan. We analyzed the human movement data before, during, and after each event, comparing the perturbed movement data to movement data from steady states. We also used different statistical analysis approaches to quantify the strength and duration of human mobility perturbation. The results suggest that tropical cyclones can significantly perturb human movements by changing travel frequencies and displacement probability distributions; however, the nature-derived Lévy Walk model still predominantly governs human mobility. Also, human mobility exhibits a surprisingly mild and brief perturbation for Hurricane Sandy and Typhoon Wipha, while the duration of disturbance was much longer for Typhoon Haiyan. Our finding that the Lévy Walk model can still predict human mobility suggests that bio-inspired examinations of human mobility patterns may uncover solutions to improve disaster evacuation, response and relief plans.