Climate change is posing significant challenges to water resources management, particularly with the increasing frequency and severity of drought events. To address these challenges, innovative modelling approaches are necessary to inform effective adaptation and risk management strategies. Hydrological models are critical tools for water resource planning and management, with physical-based fully distributed models being essential for predicting catchment hydrological behaviour. Reliable hydrological data is necessary for the effective use of hydrological models for water resource assessment and management. This study aims to enhance the performance of the fully distributed hydrological model MIKE SHE coupled with MIKE 11 for the transboundary Aa of Weerijs catchment by accurately representing hydrometeorological input and model parameters and using a multi-variate and multi-site approach for model calibration. The existing model was found to require improvements in the spatial representation of precipitation data and saturated zone boundary conditions. The accurate representation of precipitation, both spatially and temporally, is crucial for capturing the hydrological response of the catchment. In the base model rain gauge data for three rain gauge stations namely Ginneken, Zundert, and Leonhout were used considering the Theisen polygons technique. In this study, during pre-processing, temporal lag was identified in the Dutch rain gauge stations (Ginneken and Zundert) due to differences in measurement recording times, which were corrected using time-weighted averages. The corrected data were used in three spatially distributed representations of precipitation: Theisen polygon, Inverse Distance Weighting interpolation, and using radar data for the Dutch part of the catchment. Testing the impact of spatial variability of precipitation on the low-lying meso-scale catchment of Aa of Weerijs showed minimal impact on streamflows, with slight variation in Nash-Sutcliffe efficiency NSE ranging from 0.68% to 4%. ...