Graduation Date

Fall 2021

Document Type

Thesis

Program

Master of Science degree with a major in Natural Resources, option Forestry, Watershed, & Wildland Sciences

Committee Chair Name

Dr. Jeffrey Kane

Committee Chair Affiliation

HSU Faculty or Staff

Second Committee Member Name

Dr. Harold Zald

Second Committee Member Affiliation

Community Member or Outside Professional

Third Committee Member Name

Dr. Rosemary Sherriff

Third Committee Member Affiliation

HSU Faculty or Staff

Fourth Committee Member Name

Eamon Engber

Fourth Committee Member Affiliation

Community Member or Outside Professional

Keywords

Chaparral, Oak mortality, Post-fire mortality, Wildfire, Fuels management

Subject Categories

Natural Resources

Abstract

The issue of wildfires, hazard fuels management, and post-fire tree mortality has become an increasingly common topic in the western United States. This thesis is composed of two studies, with the first study, Chapter 1, examining fuel treatment effectiveness and the second study, Chapter 2, striving to characterize post-fire mortality in oaks.

Prior to wildfire, fuel reduction projects may take place to decrease the likelihood of high severity fire around human infrastructure and communities. Within California’s chaparral ecosystems, common treatment types include hand-thinning, prescribed burning, mechanical mastication, and mechanical mastication followed with prescribed burning. Because chaparral has a longer historical fire return interval and these ecosystems do not need frequent fire disturbance, the efficacy of these treatment types is debated. Our study had the rare opportunity to collect data on fine woody fuel loading, shrub density, and vegetation both immediately before and one year following wildfire in northern California’s Whiskeytown National Recreation Area. Using these comparisons, our goal was to determine the post-fire response of each treatment type and determine an effective fuels treatment in chaparral to mitigate fire behavior, while maintaining ecosystem integrity and supporting native species habitat. The severity of the wildfire was moderate across the study site and did not differ among treatments. Post-fire live shrub density and live shrub height also were not influenced by treatment type, but oak dominated sites had greater live shrub density after wildfire. Fine woody fuel loading levels differed by treatment type, with prescribed burned units having the greatest levels in both chaparral and oak sites. Fine woody fuel consumption was lowest in hand-thinned units. Total plant species richness increased in all treatment types following wildfire, largely driven by an increase in exotic species, as native plant cover decreased and exotic species cover increased across all treatments. This study suggests that areas of chaparral may need to be retreated sooner than this timeframe to reduce fire severity. However, retreating these systems may not be economically feasible and it remains unclear if treatments will meet fuel and fire behavior objectives.

Land managers are concerned about post-fire mortality of trees and rely on statistical models of tree mortality in post-fire decision making. While many studies have evaluated the accuracy of these models in conifers, the performance of these models on hardwood species, specifically oak species, has been understudied. Models, such as FOFEM and FVS-FFE, can help land managers to predict which trees will die following fire and can help in hazard tree removal and post-fire salvage logging operations. These models, however, have been exclusively developed using western United States conifer species, bringing into question the veracity of these models for hardwood species. The purpose of this study was to test current mortality models using observations from wildfire and prescribed burn sites in northern California for two oak species, California black oak (Quercus kelloggii) and canyon live oak (Quercus chrysolepis). Our findings suggest that both modeling approaches performed well, but Random Forest was better at predicting probability of mortality for an imbalanced dataset. When using imbalanced datasets, logistic regression can underpredict mortality, which can have negative repercussions for land managers dealing with recently burned ecosystems containing oaks.

Citation Style

APA

Share

Thesis/Project Location

 
COinS
 
 

To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.