 The study uses Landsat Archive and other data to estimate burn probability in forested ecosystems of Saskatchewan, Canada by analyzing the interactions between contributing factors such as forest structure, disturbance history, and land cover. The results show that forests composed of coniferous species with shorter trees, low canopy height variability, open stand structure, and low timber volumes have higher burn probabilities. A 2015 burn probability map indicates that forests that burned in 2015 had a median predicted burn, probability of 81%, while the median burn probability for unburned forest area was 19%. The study demonstrates the potential to generate detailed and spatially explicit burn probability maps from time series remote, sensing data to inform wildland fire risk modeling, management, and mitigation. This article was authored by Chen Xiong, Michael A. Wooder, Nicholas C. Coops, and others.