COMING SOON! PQDT Open is getting a new home!

ProQuest Open Access Dissertations & Theses will remain freely available as part of a new and enhanced search experience at www.proquest.com.

Questions? Please refer to this FAQ.

Dissertation/Thesis Abstract

Understanding Determinants of Rapid Wildfire Spread Across California’s Diverse Ecosystems
by Scaduto, Erica, M.A., University of California, Davis, 2020, 74; 28157554
Abstract (Summary)

Wildland fires at varying intensities and frequencies are a critical ecological process across ecosystems within the western United States. However, in the recent decade, with 2020 being the largest wildfire season on record in California’s modern history, there is an urgency to understand high-intensity fire behavior dynamics. Variability in fire behavior is heavily influenced by dynamic and often complex interactions between meteorological and biophysical components. Humans also affect wildfire characteristics directly via fire suppression and indirectly via fuel management and human settlement patterns. In response to this, we propose a novel method to derive satellite-based daily fire perimeters to be used in investigating rapid fire spread across large wildfires, allowing for increased temporal and spatial fidelity than in previous methods. Chapter 1 will elaborate on the geospatial interpolation techniques in mapping continuous daily fire spread and area burned derived from MODIS and VIIRS active fire products. For the first portion of the study, we found that the estimated fire progression areas generated by the natural neighbor method with the combined MODIS and VIIRS active fire input layers performed the best, with R2 of 0.7 ± 0.31 and RMSE of 1.25 ± 1.21 (103 acres) at a daily time scale; the accuracy was higher when assessed at a two day rolling window, e.g., R2 of 0.83 ± 0.20 and RMSE of 0.74 ± 0.94 (103 acres). Furthermore, Chapter 2 consists of a machine learning approach to investigate what factors caused the rapid spread of large wildfires in the recent decade, across six distinct ecoregion's in California. The gradient boosting machine learning model explained 87% of variability in area burned and 66% of variability in the magnitude of spread at the daily time scale. Across ecoregion's, the Klamath Mountains (KM) performed the best overall with R2 of 0.87 and RMSE of 0.26, followed by Southern Coastal Sage (SCS), and Sierra Nevada (SN). Model diagnosis also showed distinct drivers for fire spread varying across regions statewide. The results from this study will ultimately seek to provide insights on the efficacy of fuel management on reducing the rate of fire progression across ecologically diverse regions and help communities and managers to better anticipate and mitigate future risk of fast moving wildfires in the coming decades.

Indexing (document details)
Advisor: Jin, Yufang
Commitee: Hijmans, Robert, Ustin, Susan
School: University of California, Davis
Department: Geography
School Location: United States -- California
Source: MAI 82/8(E), Masters Abstracts International
Source Type: DISSERTATION
Subjects: Remote sensing, Environmental management, Physical geography
Keywords: Active fire detection, Fire behavior, Satellite-based fire detection, Wildland fires, Rapid wildfire spread, California
Publication Number: 28157554
ISBN: 9798582526582
Copyright © 2021 ProQuest LLC. All rights reserved. Terms and Conditions Privacy Policy Cookie Policy
ProQuest