Share icon
We propose to improve the decision-making activities related to post-fire risk assessment and rehabilitation by developing spatial data and image processing tools needed for modeling the hydrological response of watersheds impacted by wildfires. Post-fire flooding and erosion can pose a serious threat to life, property and municipal water supplies. To respond to this threat the US government forms Interdisciplinary Burned Area Emergency Response (BAER) Teams to access potential erosion and flood risks. BAER teams must quickly determine if expensive remediation treatments are needed and prioritize their spatial application. One of the primary information sources for making decisions is a burn severity map (derived from earth observations) that reflects fire induced changes in vegetative cover and soil properties. Slope, soils, landcover and climate are also important parameters in assessing risk. Process-based hydrological models that could be used to estimate the effects of these other parameters are currently under-utilized because the time and skills required to prepare spatial input data is prohibitive, particularly post-fire. Our proposed solution is to prepare tools and data before a fire occurs and make them readily available to use when needed. The required model inputs (soil, topography, landcover, and climate) for post-fire hydrological models are available, but need additional processing to be utilized. We propose to develop easy-to-use geographic tools for our pilot study area of western Colorado that are needed to modify landcover and soil layers with burn severity. Having the database and processing tools online will make it more feasible for BAER Teams to quickly prepare the needed inputs and take advantage of NASA Earth Science data and spatial process-based models for making critical post-fire management decisions. Additionally we will investigate how other earth observation data could potentially improve post-fire erosion predictions, as well as investigate data visualization tools for viewing model input and output.