RECOVER: An Automated, Cloud-Based Decision Support System for Post-Fire Rehabilitation Planning

RECOVER is a site-specific decision support system that automatically brings together in a single analysis environment the information necessary for post-fire rehabilitation decision-making. After a major wildfire, law requires that the federal land management agencies certify a comprehensive plan for public safety, burned area stabilization, resource protection, and site recovery. These burned area emergency response (BAER) plans are a crucial part of our national response to wildfire disasters and depend heavily on data acquired from a variety of sources. Final plans are due within 21 days of control of a major wildfire and become the guiding document for managing the activities and budgets for all subsequent remediation efforts. There are few instances in the federal government where plans of such wide-ranging scope and importance are assembled on such short notice and translated into action more quickly. RECOVER has been designed in close collaboration with our agency partners and directly addresses their high-priority decision-making requirements. In response to a fire detection event, RECOVER uses the rapid resource allocation capabilities of cloud computing to automatically collect Earth observational data, derived decision products, and historic biophysical data so that when the fire is contained, BAER teams will have a complete and ready-to-use RECOVER dataset and GIS analysis environment customized for the target wildfire. Initial studies suggest that RECOVER can transform this information-intensive process by reducing from days to a matter of minutes the time required to assemble and deliver crucial wildfire-related data.

Using fuzzy C-means and local autocorrelation to cluster satellite-inferred burn severity classes

Burn severity classifications derived from multitemporal Landsat Thematic Mapper images and the Normalised Burn Ratio (NBR) are commonly used to assess the post-fire ecological effects of wildfires. Ongoing efforts to retrospectively map historical burn severity require defensible, objective methods of classifying continuous differenced Normalised Burn Ratio (dNBR) data where field data are often unavailable. For three fires, we compare three methods of classifying pre- and post-fire Landsat data: (1) dNBR classification using Composite Burn Index (CBI) field data to assign severity classes; (2) fuzzy C-means classification of a dNBR image; (3) local Getis–Ord statistic (Gi*) output applied to a dNBR image, classified using fuzzy C-means clustering. We then use a Kappa statistic to evaluate the agreement of severity classes assigned to a pixel with its corresponding CBI plot. For two of the three fires, the C-means clustering of the dNBR and the Gi* output performed as well or better than dNBR images classified using CBI data, with strong agreement for moderate- and high-severity classes. These results suggest that clustering of dNBR data may be a suitable approach for classifying burn severity data without field data. This method may also be useful as a tool for rapid post-fire assessments (e.g. Burned Area Emergency Response and Burned Area Reflectance Classification maps), where images must often be classified quickly and subjectively. Further analysis using additional field data and across different vegetation types will be necessary to better understand the importance of localised spatial variability in classifying burn severity data or other remote sensing change-detection analyses.