Abstract
Visionmaker NYC is a web-based, user-friendly environmental modeling system for New York City (NYC) created by the Wildlife Conservation Society (WCS). We empirically evaluate the floodwater component of the hydrological model contained within Visionmaker against observed flooding using reports from the NYC 311 help phone line and historical precipitation data. Of the 26 flood events identified from 2015, we tested 10 in Visionmaker and found that in no cases did Visonmaker NYC predict flooding correctly. The simplified “bucket model” for precipitation and drainage used by Visionmaker is the likely cause of this disparity and the hydrological model should be adjusted to include time dependence. We also critically evaluate the storm models used by Visionmaker and find significant disparities with other climatic data that should be addressed.
