ABSTRACTThis paper presents an application of an automated calibration technique for the United States Geological Survey Method of Characteristics (USGS‐MOC) code. The model was applied to a shallow sand aquifer where contamination due to leakage from a solvent recycling plant was detected. Using available water‐table observations, transmissivity parameters for the USGS‐MOC model were calibrated by an automated parameter identification (PI) technique. Dispersivity was determined from trial‐and‐error simulations. For comparison purposes, the transmissivity parameters were also independently calibrated by trial‐and‐error simulations. The study results show that, although the PI‐calibrated model can produce water‐table contours that are in good agreement with observations and the contaminant plumes produced by the USGS‐MOC using parameters determined from the PI technique and the trial‐and‐error approach are not vastly different, the parameters determined by this technique are not considered to be physically plausible in all cases. The best strategy in model calibration would be to use bot
展开▼
