Large Sample Basin Experiments for Hydrological Model Parameterization: Results of the Model Parameter Experiment–MOPEX. IAHS Publ. 307, 2006, 188–195.

 

Simulating runoff from MOPEX experimental river basins using the SWAP land surface model and different parameter estimation techniques

 

YEUGENIY M. GUSEV & OLGA N. NASONOVA

 

Institute of Water Problems, Russian Academy of Sciences, Gubkina St. 3, 119991 Moscow, Russia

gusev@aqua.laser.ru

 

Abstract Different techniques for the a priori estimation and calibration of parameters in the SWAP land surface model for the 12 MOPEX experimental river basins were investigated. Two approaches for a priori parameter estima­tion were applied: (1) derivation of parameters from spatial distribution of different soil and vegetation classes within a basin; (2) application of a global parameter data set provided within the framework of the Second Global Wetness Project (GSWP-2). Two parameter calibration techniques were also used: (1) manual calibration of hydraulic conductivity at saturation that is the most crucial for runoff generation; (2) automatic optimization of several soil parameters. The results of river flow simulations with all the derived sets of parameters were compared with observations to reveal optimal values of soil parameters. Since the optimal values differed from a priori estimations using the USDA soil classification, an attempt was made to connect the optimal parameter values with the USDA soil texture classes.

 

Key words  a priori parameter estimation; land surface modeling; MOPEX; parameter calibration; river runoff