Climate Risk Modelling of Balsam Woolly Adelgid Damage Severity in Subalpine Fir Stands of Western North America

The balsamwoolly adelgid ( Adelges piceae(Ratzeburg) ( Homoptera: Adelgidae)) (BWA) is a nonnative, invasive insect that threatens Abies species throughout North America. It is well established in the Pacific Northwest, but continues to move eastward through Idaho and into Montana and potentially threatens subalpine fir to the south in the central and southern Rocky Mountains. We developed a climatic riskmodel and map that predicts BWA impacts to subalpine fir using a two-step process. Using 30-year monthly climate normals from sites with quantitatively derived BWA damage severity index values, we built a regression model that significantly explained insect damage. The sites were grouped into two distinct damage categories (high damage and mortality versus little or no mortality and low damage) and the model estimates for each group were used to designate distinct value ranges for four climatic riskcategories: minimal, low, moderate, and high. We then calculated model estimates for each cell of a 4-kilometer resolution climate raster and mapped the risk categories over the entire range of subalpine fir in the western United States. The spatial variation of risk classes indicates a gradient of climatic susceptibility generally decreasing from the Olympic Peninsula in Washington and the Cascade Range in Oregon and Washington moving eastward, with the exception of some high risk areas in northern Idaho and western Montana. There is also a pattern of decreasing climatic susceptibility from north to south in the Rocky Mountains. Our study provides an initial step for modeling the relationship between climate and BWA damage severity across the range of subalpine fir. We showed that September minimum temperature and a metric calculated as the maximum May temperature divided by total May precipitation were the best climatic predictors of BWA severity. Although winter cold temperatures and summer heat have been shown to influence BWA impacts in other locations, these variables were not as predictive as spring and fall conditions in the Pacific Northwest.