Skip to main content
Boise State University
Sign Up

Dissertation Defense: Kendra Fallon

Wednesday, November 12, 2025 3:00pm MST

1435 W University Dr, Boise, ID 83706

View map

Title: LEVERAGING DISPARATE WILDLAND FIRE DATASETS TO ADDRESS LONGSTANDING FIRE MANAGEMENT QUESTIONS

Program: Geosciences PhD

Committee Chair: Jen Pierce

Committee Co-Chair: Moji Sadegh

Committee: Jen Pierce, Moji Sadegh, Jim McNamara, Megan Cattau

Abstract: While wildfire season is perceived as occurring in the warm months, and expanding, wildfire management is a year-long process. Juxtaposed to wildfire seasons are the reactive seasons of mitigating for post-fire effects, namely first-order effects of potential hydrological impacts in burn footprints, and the proactive implementation of fuels treatments to modify fire behavior to lessen the reactive workload. In this dissertation, I focus on answering lingering questions that are critical to the management of the shoulder seasons of wildfire season. To assist in resource allocation and identification of post-hydrological risks, I present a model identifying the major drivers of soil burn severity, which cannot be dependably parameterized for modeling efforts using remotely sensed vegetation severity. This model is couched within a larger distillation of applied fire management concepts as they pertain to hydrologic modeling to assist land managers, and serves as a cautionary demonstration for academic application. This dissertation also presents a novel classification schema to track the compounded contributions of multiple fuels treatments on effective fire behavior modification across a large geography, all National Forest System lands in California. The retention of past treatment combinations afforded the granularity of fire-specific studies while leveraging a large amount of monitored interactions, allowing for a substantial interaction sample size to analyze for drivers of effective fire behavior modification within treated areas. This methodology was expanded upon with statistical modeling and machine learning to evaluate the strength and ranking of top-down (i.e., weather and climate) variables and bottom-up (e.g., topography, vegetation, treatment) in fire behavior modification. Given the projected increase in wildland fire acres and severity predicted across a majority of the Western United States, it is imperative that land managers have tools to move from a reactive to a proactive management stance. This pivot will not happen fast; however, understanding how to effectively increase and maintain treated acres will afford great dividends over time in reducing undesirable wildfire effects.