6th UF Water Institute Symposium Abstract

Submitter's Name	Seungwoo Chang
Session Name	Poster Session - Climate Change/Hydrology
Poster Number	18
Author(s)	Seungwoo Chang,  UF Water Institute, University of Florida (Presenting Author)
	Wendy Graham,  UF Water Institute, University of Florida
	Vasubandhu Misra, Department of Earth, Ocean and Atmospheric Science, Florida State University
	Tirusew Asefa, Tampa Bay Water
	Jeffrey Geurink, Tampa Bay Water
	Comparison of dynamic versus statistical downscaling of the CCSM4 General Circulation Model predictions for simulating hydrologic response in the Tampa Bay Region.
	The Coupled Model Intercomparison Project Phase 5 (CMIP5) General Circulation Models (GCMs) have been used to project future climate and assess future climate impacts in many studies. In spite of the fact that bias-corrected, statistically downscaled retrospective CMIP5 GCMs reproduce monthly means and standard deviations of precipitation (P) and reference evapotranspiration (ET0) over the Tampa Bay region perfectly, retrospective streamflow and groundwater predictions using regional hydrologic model (INTB) showed some errors (Chang et al., in progress). Due to important local to regional-scale impacts of Gulf of Mexico and Atlantic Ocean dynamics on Florida’s climate, it is possible that the errors could be removed or reduced by using a coupled ocean-atmosphere Regional Climate Model to dynamically downscale GCMs over Florida rather than statistically downscaling the GCMs directly. This study examined how water supply reliability might be improved and risk might be reduced by using Florida State University’s coupled ocean-atmosphere Regional Climate Model (RSM-ROMS) to downscale CMIP5 CCSM4 predictions and projections for use with Tampa Bay Water’s INTB model. Two dynamically downscaled CCSM4 GCM (coupled and uncoupled ocean-atmosphere RCMs) and four statistically downscaled CCSM4 GCM (LOCA, MACA, BCCA, and BCSA) retrospective and future predictions of P, maximum and minimum temperature (Tmax and Tmin) and ET0 estimated by Hargreaves method were used to drive Tampa Bay Water’s INTB model. Downscaled P, Tmax, Tmin, and ET0, and the resulting INTB predictions of actual ET, streamflows and groundwater levels were compared over the Tampa Bay Region to evaluate hydrologic implications of differences in the downscaling techniques.