The vast majority of theses in this collection are open access and freely available. There are a small number of theses that have access restricted to the WWU campus. For off-campus access to a thesis labeled "Campus Only Access," please log in here with your WWU universal ID, or talk to your librarian about requesting the restricted thesis through interlibrary loan.

Date Permissions Signed

7-29-2016

Date of Award

Summer 2016

Document Type

Masters Thesis

Degree Name

Master of Science (MS)

Department

Environmental Studies

First Advisor

Medler, Michael J.

Second Advisor

Wallin, David O.

Third Advisor

Smith, Douglas F.

Abstract

One of the continuing challenges in wildland fire management is maintaining accurate vegetation and fuel data of an adequate resolution on an ever-changing landscape. The USGS’s LANDFIRE program produces national, mid-level resolution datasets of fuel, vegetation, and fire regime data useful in the modeling of wildland fire behavior. One of the most effective and least expensive ways for maintaining the accuracy of these layers is to incorporate area updates by detecting landscape changes. While many algorithms exist for detecting change and disturbances, these algorithms are often tuned for a particular landscape and require very precise training data or rely heavily on scene statistics. This research looks at a method for detecting wildland fire across a broad array of landscapes using a collection of computer-generated rules built from hundreds of thousands of points of training data. Verification of the results were assessed by visual comparison to a time series of high spatial resolution imagery through Google Earth and cross-referenced to fires from various historical databases. A majority of the fires detected in this assessment were in either a conifer or grassland landscape. The methods outlined in this thesis performed best in those two landscapes, detecting 73% to 78% of conifer fires correctly and 79% to 83% of grassland fires correctly.

Type

Text

Publisher

Western Washington University

OCLC Number

956465363

Digital Format

application/pdf

Genre/Form

Academic theses

Language

English

Language Code

eng

Rights

Copying of this thesis in whole or in part is allowable only for scholarly purposes. It is understood, however, that any copying or publication of this thesis for commercial purposes, or for financial gain, shall not be allowed without the author's written permission.

Share

COinS