Event-scale impact and recovery of forest cover following wildfire in the Northern Rocky Mountains

Anthropogenic climate change is expected to lead to forest conversion to grass and shrubland due to more extreme fire behavior and hotter and drier post-fire conditions. However, field surveys of wilderness areas in the Northern Rocky Mountains of the United States indicate robust conifer regeneration on burned sites, indicating this region might be more resistant to conversion than dryer southerly forests. This study utilizes a machine learning approach to monitor canopy cover on burned sites in two large wilderness areas from 1985 to 2021 and quantify fire impact and recovery. We used a GBM model to integrate LiDAR and Landsat observations of the region to predict canopy cover. A time series was then calculated for 352 distinct sites. Fire impact was highly correlated with severity and was consistent with the mixed severity fire regime characteristic of the region. Of the burned area studied, 85% showed evidence of recovery. Sites that are failing to recover are occurring more recently than their recovering counterparts, with 60% of non-recovering sites burning for the first time after 2003. Extended secondary mortality was common even on recovering sites, so more monitoring time might be required to determine which sites are truly not going to recover. While median recovery time has remained stable through the study period, in recent years there have been more outlier fires with long recovery times. Of the 18 sites with projected recovery times greater than 150 years, 14 occurred since 2002. While fires that are failing to recover or recovering slowly make up proportionally small portions of the landscape, they are of particular management interest because they may be harbingers of future forest conversion.