2017 ESA Annual Meeting (August 6 -- 11)

COS 4-10 - Learning from the recent pPast: Modelling present suitability of Cross Timbers

Monday, August 7, 2017: 4:40 PM
B118-119, Oregon Convention Center
William Hammond1, Henry D. Adams1, Monica Papes2, Stephen W. Hallgren3 and Alissa J. Freeman1, (1)Plant Biology, Ecology, and Evolution, Oklahoma State University, Stillwater, OK, (2)Department of Ecology and Evolutionary Biology, University of Tennessee, Knoxville, TN, (3)Natural Resource Ecology and Management, Oklahoma State University, Stillwater, OK
Background/Question/Methods

Anthropogenic climate change forecasts increasing temperatures, drought intensity, and drought duration in the Cross Timbers, an ecoregion covering some 7 million hectares in the south central US. Drought and heat associated with increased tree mortality of forested ecosystems is well documented globally. For the Cross Timbers there is evidence from the 1950s and 2000s of elevated background mortality of codominant Q. marilandica and Q. stellata, with greater mortality in Q. marilandica often observed. The Cross Timbers, which occupy a westernmost position on the dry edge of the US temperate deciduous hardwood forest, represent a study system for early detection of drought impacts from climate change in this ecosystem. Identifying areas of reduced environmental suitability will better inform field study of stress responses for these keystone species of the Cross Timbers. Utilizing occurrence data from a mid-1950s survey, we developed an ecological niche model for these co-dominant Cross Timbers oak species. We used bioclimatic, soil, and topographic variables to train a MaxEnt species distribution model and project species ranges to the present-day. Projecting present suitability from past occurrence and environmental data allows for rigorous model validation with observational studies, field measurements of tree stress, and remote sensing information.

Results/Conclusions

We report that suitability for codominant oaks in this ecosystem is on the decline. Overall, 25% of sites from the 1950s are no longer predicted suitable in the present day. This model serves to estimate areas that are most sensitive to tree mortality events of codominant oaks, and field validation is underway to assess recent mortality and tree stress at projected Cross Timbers sites that vary in suitability to test model predictions. Preliminary analysis has shown increasing change in sapling density, tree density, and basal area at sites predicted most unsuitable by our model.