2020 ESA Annual Meeting (August 3 - 6)

COS 224 Abstract - Burn severity, repeat fires, and forest management interact to influence forest structure in northeastern Washington, USA

Van Kane1, C. Cansler2, Jonathan T. Kane1, Bryce Bartl-Geller1, Nicholas A. Povak3, James A. Lutz4, Derek Churchill5, Paul F. Hessburg3 and Andrew Larson6, (1)School of Environmental and Forest Sciences, University of Washington, Seattle, WA, (2)Rocky Mountain Research Station, USDA Forest Service, Missoula, MT, (3)USDA-FS, Pacific Northwest Research Station, Wenatchee, WA, (4)Department of Wildland Resources, and the Ecology Center, Utah State University, Logan, UT, (5)Washington Department of Natural Resources, Olympia, WA, (6)College of Forestry & Conservation, University of Montana, Missoula, MT
Background/Question/Methods

Wildfire is the most important disturbance agent affecting forest landscape structure across the inland western United States. In recent decades, rates of annual area burned have outpaced restorative treatments across the region, and there is a need for more reliable models of fire severity, fire effects, and post-fire structural development to support strategically focusing forest management efforts. We examine drivers of burn severity and the effects of pre-fire and post-fire management across hundreds of wildfires that burned in eastern Washington state, USA, in the last three decades. We address the following research questions using 30-m remotely-sensed burn severity data and other geospatial predictors: (1) how do biophysical setting, antecedent climate, and fire weather influence burn severity in initial fires and reburns? (2) what types of pre-fire silviculture treatments effectively reduce burn severity, and are those treatments effective under different burning conditions? We address an additional research question using LiDAR-derived measurement of forest structure: (3) what types of post-fire management leads to forest recovery and resilience a decade or more after fire? We use random forest modeling explore relationships between predictors of burn severity. We test for statistical differences in distributions to compare post-fire management actions.

Results/Conclusions

Question 1: Best predictors of whether a location burned at high severity or became a fire refugia (unburned and low-severity fire) differed. The former was best predicted by annual precipitation, annual temperature, AET and deficit, and fire resistance score, (abundance of species’ fire-adaptive traits) and the latter by fire resistance score, and fire-weather. Previous fire severity and time since fire were important predictors in reburns. Question 2: The effects of treatments varied by fire weather. Under moderate fire weather, pre-fire hazard fuel and prescribed fire treatments had slightly lower severity than untreated areas, silvicultural harvest, and tree plantings. Under severe fire weather, prescribed fire resulted in lower severity than untreated areas, silvicultural harvest, and tree plantings. Question 3: In dry mixed-conifer forests, post-fire structure in untreated areas had lower canopy cover, particularly in the 2-8 meter height stratum, but similar overstory tree structure for higher strata. In areas that burned with low severity, and then were treated, treatments had lower overstory tree height, lower vertical complexity, and lower canopy cover than untreated areas, but higher density of individual trees. These results help quantify the interactions of factors affecting burn severity and resulting forest structure and can help guide managers.