Floods affect more people than any other natural hazard. To expand urban development, preserve property values, and avoid paying for flood insurance, real estate developers in the US request to remove properties from the Federal Emergency Management Agency (FEMA) regulatory floodplain, often through flood protection infrastructures such as levees and drainage. However, such infrastructures frequently divert floodwaters to others downstream, an effect that has not been well studied. Underserved communities are the most exposed but may not have the tools to challenge their underrepresentation in flood maps or zoning changes. This Faculty Early Career Development Program (CAREER) award supports research which aims to advance foundational knowledge by generating observed flood data to measure if and where flood protection infrastructures change flooding patterns as well as who is affected. The project builds capacity for using satellite-based maps to counter unjust actions from flood maps and advocate for more equitable allocation of mitigation funding. This project examines past policies that have contributed to increased flood exposure of underserved communities and sheds light on their long-standing injustice. As floods impact more people and properties in a changing climate, public policies must evolve to address this growing risk and prioritize protection for the most vulnerable. Current regulatory floodplain maps underrepresent flood risk in part due to reductions though the Letters of Map Revision (LOMR) policy. To evaluate the accuracy of redrawn maps, this project documents where floodplains have been reduced because of LOMR policies and builds a spatial database of flood events from multiple satellite sources between 2001 and 2023. The work will address the question of the extent to which flood protection infrastructures are responsible for amplifying inequality in flood exposure by combining census data with satellite observations and statistical models to assess if the burden of flood risk has shifted among different demographics and communities. The project trains students and staff from environmental justice organizations to co-produce satellite-based AI tools for flood mapping. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.