Accessible Navigation. Go to: Navigation Main Content Footer
Recent News

Shawn successfully defended his Ph.D.!

Solomon has been awarded NSF funding to study glacial refugia

Currently accepting applications for an MS position

Dobrowski (2011) and Crimmins et al. (2011) were selected for Faculty of 1000 Biology and Crimmins et al. (2011) was featured in Nature Climate Change.

Our paper on plant species' elevation shifts was published in the January 21 issue of Science. News coverage is listed here.

Spatial Patterns in Niche Dynamics and SDM Uncertainty

Much of our understanding about the response of forest species distributions to climate change comes from species distribution models (SDMs). SDMs are empirical models that relate field observations of species to environmental predictors based on statistically or theoretically derived response functions. SDM climate change projections are largely untested and rely on a number of critical assumptions. Existing approaches to the validation of SDM projections partition contemporary datasets into calibration and validation sets. The working assumption is that a model with a strong goodness-of-fit will accurately project distributions under novel climates. An alternative approach to this form of validation is to assess the predictive performance of SDMs directly by fitting models retrospectively with historic data to produce projections of the present day. We are working with historic climate conditions, historic vegetation distribution data, and measured climate change, to predict current species distributions. Our overall objectives are to put real numbers on estimates of SDM climate-change projection accuracies, determine if projection accuracies vary by species and are related to species autecological traits, and assess the role of land-use type, intensity, and disturbance on projection accuracies.

In addition to developing our work, we organized a special session at the 95th Ecological Society of America (ESA) annual meeting to highlight the use of historical data in projecting climate change impacts. The special session, Predicting the present from the past: using historic datasets to understand and predict climate change impacts, brought together researchers from around the world to discuss relevant research. Building upon this session, we are now organizing a special issue of Global Ecology & Biogeography related to this topic.

Publications and Presentations

Swanson, A., and S. Dobrowski (2011) Spatial predictive process models give improved forecasts of vegetation response to climate change. US-IALE 26th Annual Landscape Ecology Symposium, Portland, Oregon.

Dobrowski, S.Z., J.H. Thorne, J.A. Greenberg, H.D. Safford, A.R. Mynsberge, S.M. Crimmins, and A.K. Swanson (2011) Modeling plant distributions over 75 years of measured climate change in California, USA: Relating temporal transferability to species traits. Ecological Monographs 81(2): 241-257.

Swanson, A., S.Z. Dobrowski, A.R. Mynsberge (2010) Spatial predictive process models yield improved forecasts of vegetation response to climate change. 2010 Fall Meeting, AGU, San Francisco, California.

Dobrowski, S.Z., J.H. Thorne, J.A. Greenberg, H.D. Safford, A.R. Mynsberge, S.M. Crimmins, A.K. Swanson (2010) Predicting the present from the past: modeling plant species distributions over 75 years of measured climate change in California, USA. The 95th ESA Annual Meeting, Pittsburgh, PA.