Using Remote Sensing and Climate Data to Map the Extent and Severity of Balsam Woolly Adelgid Infestation in Northern Utah, USA

Abstract

Balsam woolly adelgid (Hemiptera: Adelges picea Ratzeburg; BWA) is a nonnative, invasive insect that has infested fir trees in the US for over a century, yet robust methods for mapping BWA have remained elusive. We compare three approaches to mapping BWA in the subalpine fir forests of northern Utah, the forefront of BWA spread in the western US: (1) using moderate-resolution, multispectral satellite imagery; (2) using terrain and climate data; and (3) using a combination of imagery, terrain, and climate data. While the spectral data successfully detected forest degradation, they failed to distinguish between causal agents of change (R2mean = 0.482; RMSEmean = 0.112). Terrain and climate data identified landscape conditions that promote BWA infestation but lacked the ability to characterize local-scale tree damage (R2mean = 0.746; RMSEmean = 0.078). By combining spectral, terrain, and climate data, we present a repeatable approach for accurately mapping infestation severity that captures both regional abiotic drivers and the local damage signals of BWA (R2mean = 0.836; RMSEmean = 0.065). Highly infested areas featured increased visible and shortwave infrared reflectance over time in the spectral data. The terrain bore little influence on severity, but climate variables indicated that warmer areas are more prone to severe infestation. This research study presents an analytical framework upon which future BWA monitoring efforts can be built.