The time- and distance-decay effects of hurricane relevancy on social media: an empirical study of three hurricanes in the United States

Abstract

ABSTRACT Hurricane activity has been increasing in frequency and severity in recent years. This has serious implications for coastal and nearby communities who, when recovering from hurricanes, seek outside assistance from relevant government, non-governmental agencies, and nearby communities. The ever-increasing popularity of social media offers a new medium through which such social relevancy can be derived to inform targeted assistance-seeking efforts. This study utilizes Twitter to develop an understanding of disaster relevancy across space and time to establish a clearer context for impacted communities as to when and where assistance may be derived. Tweets were collected for three hurricanes within the contiguous United States (Hurricane Harvey in 2017, Florence in 2018 and Laura in 2020) and examined over a 12-week period following hurricane landfalls. The relationships between tweets and time and between tweets and distance were examined through correlation analysis. Results show statistically significant time- and distance-decay effects of hurricane relevancy on social media, though the time-decay effect was stronger. Most tweets occurred during the first week following hurricane landfall within the states wherein the hurricanes made landfall as well as around large cities. These findings could inform aid-seeking efforts in the event of hurricanes and other disasters.