The National Risk Index developed by the Federal Emergency Management Agency provides a relative measurement of community-level natural hazard risk across 50 US states and Washington, DC. The Index leverages authoritative nationwide datasets and multiplies values for exposure, annualized frequency, and historic loss ratio to derive expected annual loss estimates for 18 hazard types and combines this metric with Social Vulnerability and Community Resilience scores to generate Risk Index scores for every Census tract and county. Scores provide a holistic and comparable measure of risk across the US. Risk scores and underlying data are summarized in a custom web application. Geographical and statistical processing techniques were used to reconcile incompatibilities between the spatial and temporal collection of input datasets. The index was developed using a multidisciplinary and collaborative approach and input from subject matter experts across disciplines and target users. The National Risk Index builds upon previous efforts to develop a multi-hazard risk measurement for a large geography by expanding the number of hazard types considered, applying extensive geoprocessing techniques to combine diverse datasets, and combining traditional risk factors with the community risk factors of social vulnerability and community resilience for an enhanced nationwide picture of risk.

Flood risk planning and emergency response at community levels rely on fast access to accurate inundation models that identify geographic areas, assets, and populations that may be flooded. However, limited flood modelling resources are available to support these events and activities. We present a computationally-efficient flood model for facilitating rapid risk analysis across a wide range of scenarios and decision support to operational, crisis action, local flood-fight, and community planning efforts. Our flood depth regression method converts publicly-available river stage heights to flood depths, then downscales the depths from gage locations onto high resolution National Hydrography Dataset flowlines and estimates areas and depths of flooding by subtraction of the National Elevation Dataset from modelled water surface elevations. We demonstrate proof-of-principle analyses for historic 2009 Red River of the North flooding in the United States, achieving comprehensive mainstem flood estimation for the length of the river and depth accuracy of 1.4 ft (0.4 m) compared to gage observations, remote sensing, and higher-resolution hydrologic models. We

No abstract

No abstract