Vulnerability of eastern US tree species to climate change

Abstract

Climate change is expected to alter the distribution of tree species because of critical environmental tolerances related to growth, mortality, reproduction, disturbances, and biotic interactions. How this is realized in 21st century remains uncertain, in large part due to limitations on plant migration and the impacts of landscape fragmentation. Understanding these changes is of particular concern for forest management, which requires information at an appropriately fine spatial resolution. Here we provide a framework and application for tree species vulnerability to climate change in the eastern United States that accounts for influential drivers of future distributions. We used species distribution models to project changes in habitat suitability at 800 m for 40 tree species that vary in physiology, range, and environmental niche. We then developed layers of adaptive capacity based on migration potential, forest fragmentation, and propagule pressure. These were combined into metrics of vulnerability, including an overall index and spatially explicit categories designed to inform management. Despite overall favorable changes in suitability, the majority of species and the landscape were considered vulnerable to climate change. Vulnerability was significantly exacerbated by projections of pests and pathogens for some species. Northern and high-elevation species tended to be the most vulnerable. There were, however, some notable areas of particular resilience, including most of West Virginia. Our approach combines some of the most important considerations for species vulnerability in a straightforward framework, and can be used as a tool for managers to prioritize species, areas, and actions.

Projected Future Distribution of Tsuga canadensis across Alternative Climate Scenarios in Maine, U.S

“…Such changes are already being reported and modeling results predict a general trend of habitat suitability expansion northward for tree species [11,13,28]. Overall, our findings are consistent with this trend, suggesting that Tsuga canadensis will find more suitable habitat across Maine, which will generally support relatively low tree density.…”

“…Climate variables (i.e., temperature, precipitation) were consistently among the most important predictors for Tsuga canadensis (R 2 = 0.47) and these were the only variables changed in the future scenarios, while the other predictor variables remained constant. This approach was used to limit range shifts and ensure realistic predictions [11].…”

“…To identify the most suitable corridors, we only included pixels with an estimated Tsuga canadensis relative basal area of ≥16%. The model using RCP 8.5 for the year 2050 was used in the corridor assessment since this year provides management relevance and RCP 8.5 more closely matches current emission trends [11]. The conserved lands data layer reflects Maine land in federal, state, municipal and non-profit ownership with easements, obtained from Maine GIS data catalog [26].…”

“…Suitable habitat and tree species distributions have recently been produced for the U.S. at varying resolutions [11][12][13], often at landscape scale, for example, Iverson's [13] tree atlas has a resolution of 20 km. Forest management often takes place at the regional scale (i.e., 30 m) and forest managers, although they have been increasingly able to incorporate climate change considerations into their decision making, require practical ways in which to implement modeling results [14].…”

Classifying forest inventory data into species-based forest community types at broad extents: exploring tradeoffs among supervised and unsupervised approaches

“…We overlaid FIA plots on these spatial data layers to extract the values for plot locations. We focused on variables that have been shown to be important in other recent studies of tree distributions, vulnerability, and climate change (e.g., Iverson et al 2008;Rogers et al 2017). Soil variables were taken from the 10-m resolution Gridded Soil Survey Geographic Database (gSSURGO; Soil Survey Staff 2017a), where available.…”