HR: 1340h
AN: H43A-0485    [Abstracts]
TI: Toward Improved Calibration of a Semi-arid Distributed Flash-Flood Model: A Hierarchical Sensitivity Scheme for Model Evaluation
AU: * Yatheendradas, S
EM: soni@hwr.arizona.edu
AF: SAHRA NSF-STC & Department of Hydrology and Water Resources, University of Arizona, Tucson, AZ 85721 United States
AU: Wagener, T
EM: thorsten@engr.psu.edu
AF: Department of Civil and Environmental Engineering, Pennsylvania State University, Pennslyvania, PA 16802 United States
AU: Gupta, H
EM: hoshin.gupta@hwr.arizona.edu
AF: SAHRA NSF-STC & Department of Hydrology and Water Resources, University of Arizona, Tucson, AZ 85721 United States
AU: Unkrich, C
EM: cunkrich@tucson.ars.ag.gov
AF: USDA-ARS-SWRC, 2000 E. Allen Rd., Tucson, AZ 85719 United States
AU: Schaffner, M
EM: Mike.Schaffner@noaa.gov
AF: National Weather Service, Tucson Weather Forecast Office, 520 North Park Ave., Suite 304, Tucson, AZ 85719 United States
AU: Goodrich, D
EM: dgoodrich@tucson.ars.ag.gov
AF: USDA-ARS-SWRC, 2000 E. Allen Rd., Tucson, AZ 85719 United States
AU: Goodrich, D
EM: dgoodrich@tucson.ars.ag.gov
AF: ARS Water Conservation Lab, 4331 E. Broadway Road, Phoenix, AZ 85040 United States
AB: Arid and semi-arid regions currently span approximately one-third of the earth's surface, with possibly greater coverage in the future, for example due to current global warming. Many of these regions are particularly affected by flash floods caused by summertime convective storm systems, often resulting in significant risk to life and property. The short spatial and temporal extent of these events makes predicting the subsequent floods extremely difficult. To improve our predictive capability, an established event-based semi-arid rainfall-runoff model KINEROS2 is modified to ultimately allow for the continuous simulation of the basin response driven by high-resolution precipitation measurements in an uncertainty framework. The model contains process descriptions required to represent semi-arid and arid regions, including a dynamic infiltration algorithm and the ability to account for channel transmission losses. The complex process description and the spatially distributed system representation require a large number of parameters to be estimated. Reliable calibration is essential for effective operational forecasting and parameter regionalization studies, but is typically hampered in semi-arid regions due to uncertain or lacking input and output data. Strategies to reduce the calibration burden in a sensible way need to be found. Starting with the simple paradigm of reducing the model parameter dimensionality due to spatial distribution using multipliers, a hierarchical global sensitivity analysis scheme is implemented. This scheme merges variance-based and regional sensitivity analysis techniques, incorporating parameter, input, initial state and objective function interactions. This talk will discuss the application of this technique to a semi-arid basin in the southwestern USA.
DE: 1804 Catchment
DE: 1816 Estimation and forecasting
DE: 1821 Floods
DE: 1846 Model calibration (3333)
SC: Hydrology [H]
MN: Fall Meeting 2005