Exploring the Relationship Between Pre-fire Hazard and Post-fire Fractional Vegetation Cover – A Preliminary Comparison for the Beachie Creek Fire, OR

Keywords:  Fire, GIS, Remote Sensing
Abstract Type: Poster Abstract

Abstract

During Labor Day 2020, an extreme wind event caused a series of wildfires, known collectively as the 2020 Labor Day Fires, to rapidly spread in Oregon. These fires impacted hundreds of thousands of hectares within and adjacent to the Cascade Mountains, destroying many towns and claiming several lives. As several years have now passed, there is growing interest in the emerging post-fire landscape. This study compares pre-fire hazard levels to post-fire vegetation growth for one of these large wildfires, the Beachie Creek Fire. The objective of this study is to apply geospatial technologies to map pre-fire hazards and post-fire fractional vegetation cover (FVC) for areas affected by the Beachie Creek Fire. To accomplish this objective, a combination of various GIS datasets were used to estimate pre-fire hazard and Landsat 8 OLI imagery was obtained to calculate various spectral vegetation indices (SVI). Additionally, field data related to FVC was gathered via hemispherical photography in the summer of 2023. We found that the Normalized Difference Vegetation Index (NDVI) SVI best explained the variance in FVC, with an R2 of 0.77. The regression equation was used in a GIS to convert the NDVI dataset into the estimated FVC. This was then compared to the pre-fire hazard levels via spatial statistics to explore the relationship between these two datasets. While more robust statistical analysis is still required, the preliminary results suggest that the level of fire hazard is moderately predictive of the 3-year post-fire fractional vegetation cover for the Beachie Creek fire.