Speaker: Beth Tellman, Assistant Professor, Nelson Institute for Environmental Studies, UW–Madison
Floods affect more people than any other hazard, and building in the floodplain plays a larger role in increasing exposure than a changing climate. Improvements in the resolution of satellite data and machine learning provide new opportunities to monitor floods locally and globally. In this talk, Tellman will show how the Social Pixel Lab leverages AI to make flood maps from multiple satellite sensors.
Mapping flood exposure globally reveals that settlement growth is higher in places that have experienced floods than elsewhere. But what and who drives urban expansion in risky places — and what actors are ultimately responsible? Current regulatory floodplain maps in the US from FEMA underrepresent flood risk, which can be due to failure to produce or update maps, not incorporating different flood causes (e.g. pluvial flooding), or by a policy called Letters of Map Change (LOMC).
Tellman will show initial results of where we see LOMCs across the US and Wisconsin and how satellite data can be leveraged for adaptation,
This seminar can also be viewed via our live stream.
Hosted by the Climate, People and the Environment Program (CPEP).