Do surviving trees foretell forest growth declines under warm drought climates?

Future warm droughts are anticipated to exacerbate forest decline either directly by affecting plant hydraulics and carbon balance, or indirectly by increasing the prevalence of disease and insect attacks. The roles these stressors play on forest health are often studied singly, but knowing how they affect tree mortality in combination is critical. They generally are useful for making inferences because proximal causes of tree death are unknown. In addition, heterogeneity in abiotic controls over tree mortality is poorly characterized. One way to connect the stressors to potential forest decline is to study stands of trees that have survived landscape scale mortality events. The hypothesis is that trees that grow roots to where they can access reliable groundwater water experience modulated responses to atmospheric drought, making them less susceptible to the distal causes of mortality. This idea was tested in landscapes exposed to drought, heat, insects, and their combinations, using the coupled ecophysiology-hydrology model TREES to quantify the modulating role of groundwater water on carbon budgets and hydraulic status. TREES has a transient hydraulic system coupled to carbon allocation, which allows it to account for the carbon costs of growing and maintaining water absorbing fine root growth. To cover a breadth of stressors we synthesized results using multiple seasonally or periodically dry landscapes spanning riparian woodlands to subalpine forests, and natural experiments as well as experimentally manipulated systems. The carbon costs of maintaining fine roots near reliable water sources were clear, as there was high mortality risk associated with slow root grow growth rates after the onset of stress. Access to a reliable source of water, if only for a small fraction of total plant water use, significantly explained tree survival regardless of applied stressor(s). But warming alone increased carbon costs suggesting a possible future tilt towards unsustainable forest cover, even where there are reliable groundwater sources. Future research should combine controlled experiments and multiple stressors with well-tuned simulation modeling to elucidate mechanisms of forest health decline.