Droughty times in mesic places: factors associated with forest mortality vary by scale in a temperate subalpine region

Abstract

AbstractUnderstanding how drivers of ecological disturbance operate across scales is important in an era of increasing disturbance activity. Severe and extensive Dendroctonus bark beetle outbreaks across western North America have left in their wake dominance by shade‐tolerant and commonly late‐seral trees such as subalpine fir (Abies lasiocarpa), which can foster resilience of forest cover. However, subalpine fir decline (SFD) is a poorly understood phenomenon that has killed trees across millions of hectares in western North America with unknown consequences for future forest resilience. How different factors (e.g., climate, topography, host‐tree characteristics, and abundance) govern SFD presence and severity across spatial scales from individual trees to a subcontinental scale has not been explored in a single framework. Here, we combine broad‐scale geospatial data on SFD occurrence, stand‐scale field data on SFD severity, and fine‐scale individual tree data on mortality to test the relative importance of factors related to SFD across spatial scales spanning >10 orders of magnitude (<1 m to >10 M hectares). At the broadest scale (subcontinental, ~25 M ha), annual areal extent of SFD over time increased sharply with antecedent drought. At regional‐ (~6 M ha) and stand‐ (0.1–0.25 ha) scales, the occurrence and severity of SFD were spatially associated with more mesic topographic positions and greater host‐tree abundance. Finally, at the individual tree‐ and tree‐neighborhood‐ (<1 m) scale, the probability of mortality increased for larger trees and trees closer to dead conspecific neighbors. The positive temporal association of SFD with drought at broad scales versus the positive spatial association of SFD with mesic sites at fine scales suggests strong importance of local biotic processes in mediating drought‐driven forest decline and highlights the need for understanding multi‐scale drivers of ecological disturbance.