Year of Award


Document Type


Degree Type

Doctor of Philosophy (PhD)

Degree Name

Forest and Conservation Science

Other Degree Name/Area of Focus

Ecosystem and Conservation Science

Department or School/College

College of Forestry and Conservation

Committee Co-chair

Ronald Wakimoto, Kelsey Jencso, Andrew Wilcox

Commitee Members

Susan Cannon, Carl Seielstad


fire severity, geomorphic threshold, hydrologic connectivity, natural hazards, post-fire erosion, vegetation change


University of Montana


Gully rejuvenation (GR) following wildfire influences landform evolution and generates flooding and debris that alters aquatic habitat and threatens human activities. Fire severity, defined as the degree of vegetation loss by wildfire, is a hypothesized control on this erosion response. I investigated three related aspects of the relationship between fire severity and GR: The capacity of vegetation disturbance to explain the occurrence or non-occurrence of GR; the spatial structure of burn mosaics relative to post-fire erosion; and the relationship between fire severity and threshold conditions required for channel initiation. I surveyed 269 burned catchments and mapped 111 cases of GR across sites in Montana and Idaho. I created the Vegetation Disturbance Index (VDI) derived from LANDSAT images to quantify fire severity and implemented geospatial and statistical analysis to quantify relationships between VDI and post-fire erosion response. Vegetation disturbance strongly explained GR with additional influences from upslope geometry and pre-fire shrub cover. As fire severity increased, the percent of the catchment area covered by continuous patches of high severity burn increased non-linearly. Trends in patch structure defined a threshold of fire severity after which the probability of GR was strongly correlated with the development of large, continuous severely burned patches. Fire severity systematically influenced the relationship between source area and steepness. Threshold conditions for channel initiation, specifically source area steepness and curvature, decreased as vegetation disturbance increased. These results provide inferential evidence that vegetation disturbance exerts first-order controls over post-fire erosion processes. The results of the patch-pattern analysis suggest that progressive loss of vegetation due to wildfire leads to critical thresholds of hydrologic connectivity after which runoff and erosion accelerate. The source area analysis suggests that forces of convergent flow are not fully expressed until a significant proportion of vegetation has been consumed such that flow resistance is minimized. The VDI as a continuous metric of vegetation disturbance may contribute to improved quantitative analysis of landform evolution relative to vegetation disturbance, ecological effects of fire, and ecosystem response to climate change. The assessment methodology outlined herein provides a first step towards a systematic quantification of the potential for GR following wildfire.



© Copyright 2013 Kevin Hyde