Rapid response tools and datasets for post-fire modeling in Boreal and Arctic Environments

Document Type

Conference Paper/Presentation

Publication Date



Preparation is a key component to utilizing Earth Observations and process-based models in order to support post-wildfire mitigation. Post-fire flooding and erosion can pose a serious threat to life, property and municipal water supplies. Increased runoff and sediment delivery due to the loss of surface cover and fire-induced changes in soil properties are of great concern to resource managers. Remediation plans and treatments must be developed and implemented before the first major storms in order to be effective. One of the primary sources of information for making remediation decisions is a soil burn severity map derived from Earth Observation data (typically Landsat) that reflects fire induced changes in vegetation and soil properties. Slope, soils, land cover and climate are also important parameters that need to be considered. Spatially-explicit process-based models can account for these parameters, but they are under-utilized relative to simpler, lumped models because they are both difficult to set up and require spatially-explicit inputs (digital elevation models, soils, and land cover). Our goal is to make process-based models more accessible by preparing spatial inputs before a wild fire, so that datasets can be rapidly combined with soil burn severity maps and formatted for model use. We have built an open source online database (http://geodjango.mtri.org/geowepp /) for the continental United States that allows users to upload soil burn severity maps into the database. The soil burn severity map is then rapidly combined with land cover and soil datasets in order to generate the spatial model inputs needed for hydrological modelling of burn scars. We believe our database could be expanded internationally to support other countries that face post-fire hazards.

This summer we worked with the University of Alberta to model potential erosion from the Fort McMurray fire. We utilized Lidar based DEM, Canadian weather data, Canadian Soil Landscape data, pre and post-fire Landsat imagery, and the Alberta Biodiversity Monitoring Institute land cover map in our modeling. We were able to demonstrate that process based models could be rapidly applied for modeling post-fire effects in Canada. The datasets and modeling developed for the Fort McMurray fire will be refined and utilized under a new NASA SMAP program to help improve Canadian Forest Fire Danger Rating System predictions with SMAP soil moisture data. Data fusion techniques will be used to combine modeled predictions of soil moisture with SMAP observations with the goal of improving the spatial resolution of SMAP.

Publisher's Statement

Presented at the 2017 Remote Sensing Workshop entitled "Opportunities to Apply Remote Sensing in Boreal/Arctic Wildfire Management and Science, held by the Alaska Fire Science Consortium.

Publication Title

Spring 2017 AFSC Remote Sensing Workshop: Opportunities to Apply Remote Sensing in Boreal/Arctic Wildfire Management and Science