Application of Different Weighting Schemes and Stochastic Simulations to Parameterization Processes Considering Observation Error: Implications for Climate Change Impact Analysis of Integrated Watershed Models

We investigated the potential impact of observation error on calibration performance an integrated watershed model. A three-dimensional model was constructed using HydroGeoSphere and applied to Sabgyo in South Korea assess groundwater–surface water interaction process. During calibration, three different weighting schemes that consider variances were parameter estimation tool (PEST). The compared with results from stochastic models, which errors surface discharges considered a random variable. Based calibrated model, interactions between groundwater predicted under climate change scenarios (RCP). Comparisons models showed observation-error-based contributed improvement parameterization. Analysis exchange flux highlighted significance delaying hydrological response systems. Predictions based RCP suggested increasing role dynamics. concluded comparison for determination covariance could contribute improved characterization processes reduce uncertainty arising errors.