HR: 1340h
AN: H43A-0486 [Abstracts]
TI: Application of a multi-objective autocalibration method in watershed-scale hydrological and water
quality model
AU: * Debele, B
EM: bd58@cornell.edu
AF: University of Arkansas, Department of Agricultural and Biological Engineering, University of Arkansas,
203 Eng. Hall, Fayetteville, AR 72701
United States
AU: Chaubey, I
EM: chaubey@uark.edu
AF: University of Arkansas, Department of Agricultural and Biological Engineering, University of Arkansas,
203 Eng. Hall, Fayetteville, AR 72701
United States
AB:
Hydrological and water quality parameters vary numerously at a watershed scale. Many more parameters come into play when the
physics of water and solute movements are examined at a larger scale, such as watershed, compared to what would be the case
under laboratory scale studies of similar processes. Moreover, larger study areas and more complex hydrological and water
quality models entail investigation of more likely number of model parameters, which renders manual calibration infeasible.
Thus, one should resort to a more robust yet meaningful aoutocalibration method for model parameters' optimization.
We implemented a fairly new approach of multi-objective autocalibration that incorporates the Shuffled Complex Evolution
algorithm. We used a two-stage multi-objective autocalibration scheme to enable robust parameter optimization. We first ran
the sensitivity analysis using the One-factor-At-a-Time (OAT) approach to select important parameters for optimization. We
then implemented parameters' optimization on selected parameters. The optimization uses a global optimization criterion,
whereby several output variables are taken into account simultaneously.
We applied our autocalibration method with the SWAT (version 2003) model at two watersheds in the Sate of Arkansas, US. The
SWAT model is a well-documented semi-distributed and physically based hydrological and water quality model. We calibrated the
SWAT model for flow, sediment and nutrients, and the results will be reported.
DE: 1846 Model calibration (3333)
DE: 3333 Model calibration (1846)
SC: Hydrology [H]
MN: Fall Meeting 2005