Modeling the ongoing impact of Hurricane Maria using human mobility data

Hurricane Maria has recently caused extensive damages in the Caribbean including Puerto Rico’s Arecibo Observatory. While we have now gained an understanding of the physical damages, the impact on human behavior has not been described quantitatively. For example, even though evacuation orders were issued, the extent and destinations of population migration are still unclear. With the aim to support relief efforts and help direct the delivery of aid, we use near-real time aggregate mobility data from Google location services to quantify the impact of the disaster on people in Puerto Rico.We show that movement out of Puerto Rico increased by 20% the day before the landfall of Hurricane Maria on September 20, 2017. Immediately after the landfall, travel in and out of Puerto Rico virtually stopped, and only slow recovery has been recorded until October 11 (last day of analysis) with more people leaving …