Year of Award

2025

Document Type

Thesis

Degree Type

Master of Science (MS)

Degree Name

Forestry

Department or School/College

W.A. Franke College of Forestry & Conservation

Committee Chair

Dr. Alina Cansler

Commitee Members

Dr. Zachary H. Hoylman, Dr. Andrew J. Larson, Dr. Saron M. Hood

Keywords

Fuel, treatment, effectiveness, Northern, Rocky, Mountains

Subject Categories

Forest Management | Other Forestry and Forest Sciences

Abstract

Fire severity in mesic mixed-conifer forests is shaped by complex interactions among climate, weather, fuels, and topography, yet treatment effectiveness in these systems remains poorly quantified across broad landscapes. We evaluated drivers of burn severity and fuel treatment effectiveness across four Northern Rocky Mountain ecoregions (Canadian Rockies, Idaho Batholith, Middle Rockies, and Northern Rockies) using an empirical, landscape-scale modeling framework. We compiled burn severity, treatment history, and environmental predictors for 700 wildfires (2001–2023), and developed Random Forest models using both gridded weather (gridMET) and higher-resolution TopoFire predictors. Treatment effectiveness was evaluated using matched-control comparisons and a counterfactual approach that estimated expected severity under a no-treatment scenario. Across ecoregions, fire severity was most strongly regulated by climate and short-term weather conditions, particularly predictors related to fuel moisture (e.g., dead fuel moisture, soil moisture, and climate water deficit), while static topographic and fuel metrics played secondary, context-dependent roles. Elevation was the only topographic predictor consistently ranked among the strongest variables, likely reflecting covariation with bioclimatic gradients rather than independent controls on fire behavior. Incorporating higher-resolution TopoFire weather data modestly improved representation of within-fire variability but did not substantially improve model performance or transferability to independent fires, underscoring challenges in generalizing severity models across heterogeneous landscapes. Treatment effectiveness varied by treatment type, age, ecoregion, and evaluation method. Combined treatments that reduced both canopy and surface fuels—particularly removal + surface reduction and removal + prescribed fire—showed the most consistent reductions in severity. In several categories, older treatments occasionally reduced severity more than recent treatments, suggesting that treatment longevity in mesic forests may be mediated by fuelbed decomposition and moisture dynamics. Across the study extent, prior wildfire and prescribed fire comprised the largest treatment footprints within subsequent wildfire perimeters, indicating that wildfire itself is a dominant driver of landscape fuel modification. These findings highlight the context dependence of treatment outcomes in mesic forests and support evaluation frameworks that integrate counterfactual modeling and broader management objectives beyond severity reduction.

Available for download on Friday, January 15, 2027

Share

COinS
 

© Copyright 2025 Mikaela J. Balkind