Few generalizable patterns of tree-level mortality during extreme drought and concurrent bark beetle outbreaks

摘要

Tree mortality associated with drought and concurrent bark beetle outbreaks is expected to increase with further climate change. When these two types of disturbance occur in concert it complicates our ability to accurately predict future forest mortality. The recent extreme California USA drought and bark beetle outbreaks resulted in extensive tree mortality and provides a unique opportunity to examine questions of why some trees die while others survive these co-occurring disturbances. We use plot-level data combined with a three-proxy tree-level approach using radial growth, carbon isotopes, and resin duct metrics to evaluate 1) whether variability in stand structure, tree growth or size, carbon isotope discrimination, or defenses precede mortality, 2) how relationships between these proxies differ for surviving and now-dead trees, and 3) whether generalizable risk factors for tree mortality exist across pinyon pine (Pinus monophylla), ponderosa pine (P. ponderosa), white fir (Abies concolor), and incense cedar (Calocedrus decurrens) affected by the combination of drought and beetle outbreaks. We find that risk factors associated with mortality differ between species, and that few generalizable patterns exist when bark beetle outbreaks occur in concert with a particularly long, hot drought. We see evidence that both long-term differences in physiology and shorter-term beetle-related selection and variability in defenses influence mortality susceptibility for ponderosa pine, whereas beetle dynamics may play a more prominent role in mortality patterns for white fir and pinyon pine. In contrast, incense cedar mortality appears to be attributable to long-term effects of growth suppression. Risk factors that predispose some trees to drought and beetle-related mortality likely reflect species-specific strategies for dealing with these particular disturbance types. The combined influence of beetles and drought necessitates the consideration of multiple, species-specific risk factors to more accurately model forest mortality in the face of similar extreme events more likely under future climates.