Spatio-temporal pattern in a city during a disaster recovery process: Based on aggregated positioning data of mobile phones

When a city is hit by a huge disaster, activities and functions of the city necessary for people are lost for several and damaged. Total loss of the disaster is not only determined by thedirect damage but also by recovering speed. The aim of this analysis is to extract spatio-temporal patterns in a city during a disaster recovery process based on aggregated positioning data of mobile phones. In this analysis, we classified activities and functions decreased through spatial patterns ofpopulation distribution during a disaster recovery process based on a factor analysis, and confirmeddifference in recovering speed. Moreover, we suggested that each spatialpattern is related to spatial pattern of activitiesin normal time.