Patterns and Predictors of Tree Mortality Amidst Exceptional Drought

Abstract:

Incredibly small changes in vital rates have tremendous impacts on forest ecosystems. Even a short-lived shift in the pace or scale of mortality can substantially alter forest structure, habitat availability, and a range of ecological functions, making mortality trends a strong indicator of the status and condition of terrestrial ecosystems. Today, we continue to witness extensive mortality events in forests worldwide, often driven by novel climatic shifts. However, in many forest ecosystems, annual and periodic mortality estimates for both typical and anomalous conditions are unavailable or unspecific. Our work explores mortality patterns in detail, using tree, forest, and landscape features to 1) understand major influences on individual and population-scale tree death and 2) explore the transformative impacts of exceptional drought. Analysis focuses on California forests from 2001-2015, spanning a major statewide drought from 2012-2015 and the more moderate and locally varied droughts preceding it. Estimates are based on a comprehensive ground-based inventory by the U.S. Forest Service Forest Inventory and Analysis program.

In initial analysis, we found that tree and stand-specific factors were more predictive of survival outcomes than anomalous climate conditions. For example, middle diameter classes, higher live crown ratios, less bole damage, and higher stand basal area were associated with lower odds of mortality statewide (predictions in attached figure). Mortality rates increased markedly during the study period across key species, forested regions, and diameter classes. Extending this work, we ask: does exceptional drought change the influence of tree and stand characteristics on survival? Is there an extremity at which drought and its agents, such as bark beetles, might render tree and forest features relatively unimportant? We also explore whether the key drivers of tree death are specific to certain regions or populations of trees, and contrast individual vs. population scale patterns. Mortality outcomes are a fundamental measure of how forests are being rapidly reshaped, and understanding these patterns is an essential tool for accurately modeling vegetation dynamics and predicting future climate-driven mortality events.