Research
I'm working on probabilistic reconstructions of past temperature and precipitation, conditioned on observed tree-ring width data. My goal is to include information on the nonlinear, possibly nonstationary processes by which these climate controls leave their imprint on the natural proxy archive. My approach is to embed a process-based model of tree-ring growth inside a Bayesian hierarchical model (BHM) of monthly temperature and precipitation. The BHM framework combines uncertainty from both data and model sources naturally and rigorously.This work is being carried out in the UofA Laboratory of Tree Ring Research. My advisor is Mike Evans (now at U. Maryland Dept. of Geology).
Publications
Tolwinski-Ward, S.E., M.P. Tingley, M.N. Evans, A probabilistic technique for reconstructing local temperature and soil moisture using a nonlinear, biologically-motivated model of tree-ring width variability (in prep.)Tolwinski-Ward, S.E., M.N. Evans, M.K. Hughes, K.J. Anchukaitis, An efficient forward model of the climate controls on interannual variation in tree-ring width, Climate Dynamics, doi:10.1007/s00382-010-0945-5, (2010).
[Preprint] [Online First version] [Erratum] [Model code, documentation, and test data] (Code last updated Nov. 2011)
Selected Presentations
Tolwinski-Ward, S.E., M.N. Evans, M.K. Hughes, K.J. Anchukaitis, Toward Process-Based Modeling of Tree-Ring Width for Bayesian Climate Reconstructions, contributed talk at the 11th International Meetings of Statistical Climatology, Edinburgh, Scotland, July 12th, 2010.Tolwinski-Ward, S.E., M.N. Evans, M.K.Hughes, K.J. Anchukaitis, An Efficient Forward Model of Tree-Ring Width Foramtion for Bayesian Climate Reconstructions, poster presentation at the Statistical and Mathematical Sciences Institute Climate Change Workshop, Research Triangle Park, NC, February 17-19, 2010, and at the Institute for Pure and Applied Mathematics Workshop on Data Hierarchies for Climate Modeling, Los Angeles, CA, May 24-28, 2010.