HR: 17:30h
AN: H54B-07    [Abstracts]
TI: Automatic Calibration of a Semi-Distributed Hydrologic Model Using Particle Swarm Optimization
AU: * Bekele, E G
EM: elias@siu.edu
AF: Southern Illinois University at Carbondale, Department of Civil and Environmental Engineering MC 6603, Carbondale, IL 62901 United States
AU: Nicklow, J W
EM: nicklow@engr.siu.edu
AF: Southern Illinois University at Carbondale, Department of Civil and Environmental Engineering MC 6603, Carbondale, IL 62901 United States
AB: Hydrologic simulation models need to be calibrated and validated before using them for operational predictions. Spatially-distributed hydrologic models generally have a large number of parameters to capture the various physical characteristics of a hydrologic system. Manual calibration of such models is a very tedious and daunting task, and its success depends on the subjective assessment of a particular modeler, which includes knowledge of the basic approaches and interactions in the model. In order to alleviate these shortcomings, an automatic calibration model, which employs an evolutionary optimization technique known as Particle Swarm Optimizer (PSO) for parameter estimation, is developed. PSO is a heuristic search algorithm that is inspired by social behavior of bird flocking or fish schooling. The newly-developed calibration model is integrated to the U.S. Department of Agriculture's Soil and Water Assessment Tool (SWAT). SWAT is a physically-based, semi-distributed hydrologic model that was developed to predict the long term impacts of land management practices on water, sediment and agricultural chemical yields in large complex watersheds with varying soils, land use, and management conditions. SWAT was calibrated for streamflow and sediment concentration. The calibration process involves parameter specification, whereby sensitive model parameters are identified, and parameter estimation. In order to reduce the number of parameters to be calibrated, parameterization was performed. The methodology is applied to a demonstration watershed known as Big Creek, which is located in southern Illinois. Application results show the effectiveness of the approach and model predictions are significantly improved.
DE: 1805 Computational hydrology
SC: Hydrology [H]
MN: Fall Meeting 2005