HR: 1340h
AN: A53B-0890    [Abstracts]
TI: The Use of Bayesian Statistical Modeling to Aid in the Validation of Regional Climate Model Output
AU: * Snyder, M A
EM: msnyder@es.ucsc.edu
AF: Dept. of Earth Sciences, University of California, Santa Cruz, 1156 High St., Santa Cruz, CA 95064
AU: Sanso, B
EM: bruno@ams.ucsc.edu
AF: Dept. of Applied Mathematics and Statistics, University of California, California, Santa Cruz, 1156 High St., Santa Cruz, CA 95064
AU: Sloan, L C
EM: lcsloan@es.ucsc.edu
AF: Dept. of Earth Sciences, University of California, Santa Cruz, 1156 High St., Santa Cruz, CA 95064
AB: The application of regional climate modeling to assessments of future climate has grown dramatically in recent years. The need for high spatial resolution estimates of future climate has driven the use of these models. As a consequence, the need for validation, an assessment of how well the regional climate model represents a known climate, has also grown. Previous validation efforts have used gridded climate datasets such as those produced by the Climate Research Unit (CRU) of the University of East Anglia for comparison to regional model output. The disadvantages of this method are that the gridded climate datasets often exist at lower spatial resolution than the climate model output. In addition, some of these datasets only sample short time periods (i.e. 30 years or less) or don't include years of interest (i.e. 1990-2000). We have created a method of validation for regional climate model output using a Bayesian statistical model derived from observational data. We took temperature data from 115 observational stations in California, all with records of approximately 50 years in length, and created a statistical model based on this data. The statistical model takes into account the elevation of the station, distance from coastline, and the NOAA climate region in which the station resides. Initial results indicate that the statistical model provides reliable estimates of the mean monthly temperature at any given station. This statistical model is then used to estimate average temperatures corresponding to each of the climate model grid cells. These estimates are compared to the output of the regional climate model to assess how well the model matches the observed climate.
DE: 9350 North America
DE: 3337 Numerical modeling and data assimilation
DE: 3309 Climatology (1620)
DE: 1600 GLOBAL CHANGE (New category)
SC: Atmospheric Sciences [A]
MN: 2004 AGU Fall Meeting