Characterizing Boreal Forest Fire Disturbance Boundaries Through Space And Time In Ontario

Abstract

The Ontario boreal forest contains vast natural resources but is increasingly threatened by wildland fires, which are becoming more frequent and affecting larger areas due to climate change. In response, this thesis compares wildland fire boundaries derived from vegetation index slopes with those provided by BorealDB a newly developed database that compiles consistent disturbance maps from 1972 to the present. BorealDB includes various attribute combinations and an ensemble confidence measure that shows how often different data sources agree. By examining which attribute combinations produce fire boundaries that most closely match remote sensing data, this research offers practical guidance for BorealDB users on selecting the most reliable disturbance points for their analyses.