Context
The Washington, D.C. Integrated Flood Model is a multi-year effort to model flood risks from riverine, interior, and coastal sources under current and future climate conditions. The project required digital workflows that could transform model outputs and building inventory data into usable datasets for analysis, planning, and risk communication.
Challenge
Flood model outputs are technically rich but not always directly usable for downstream spatial analysis. Mesh-based model results, parcel assessment records, and building inventory data must be cleaned, validated, transformed, and aligned before they can support flood exposure and loss analysis.
My Role
I led AI and digital application support for geospatial data processing tasks. My work focused on converting flood mesh outputs into raster flood surfaces and supporting Hazus occupancy classification from parcel assessment information.
Approach
I developed workflows to validate and repair flood mesh triangle geometry, process elevation and depth attributes, and convert model outputs into raster flood surfaces. I also supported classification logic that linked D.C. property use codes to Hazus occupancy descriptions so that building exposure could be analyzed using consistent risk categories.
Output
The project produced processed flood raster surfaces and building classification inputs suitable for flood risk analysis. These outputs connected model results, parcel data, and Hazus-compatible building information into a more usable analytical structure.
Impact
The work strengthened the bridge between flood modeling and applied risk assessment. It helped prepare technical model outputs for use in planning, exposure analysis, and future resilience decision-making.