Urban Flood Analytics Using Community Reporting

Floods are becoming increasingly prevalent causing injuries and death, damage, and adverse socioeconomic impacts. Increasing access to mobile communication is resulting in increased information provided by the public, particularly in cities, on variety of concerns. We carried out a study on over 5 million flood related calls in New York City. A novel approach was used to create a spatial density map of reports across New York City’s five boroughs. Using the top 25 most frequent complaint types ranging from noise to rats, the overall propensity of calling was estimated and mapped. This map was then used to normalize the flood-related complaints.  The temporal-spatial analysis was highly correlated with monthly rainfall intensities.  The methodology highlights an approach by which useful information may be obtained from open data.