Improving Soil Moisture and Evapotranspiration Simulations in an Intensively Irrigated Arid Region Using Noah-MP

Intensively managed irrigation districts in arid regions pose major challenges for land surface and hydrological modeling due to strong anthropogenic disturbances and highly nonlinear soil–vegetation–atmosphere interactions. The Hetao Irrigation District (HID), one of the largest irrigated areas in northern China, exemplifies such complexities, where conventional land surface models often struggle to realistically represent soil moisture （SM） dynamics and evapotranspiration (ET) processes. In this study, we improve the performance of the Noah Land Surface Model with Multi-Parameterization options (Noah-MP) by integrating global sensitivity analysis and parameter optimization. The model was driven by long-term meteorological forcing, and dominant parameters related to soil hydraulic properties and vegetation phenology were identified as key controls on simulated soil moisture and ET. These parameters were subsequently optimized using the Shuffled Complex Evolution (SCE-UA) algorithm, jointly constrained by in-situ observations and remotely sensed SM and ET products. The calibrated model shows a consistent improvement in reproducing observed soil moisture dynamics and better captures the seasonal variability of ET associated with irrigation practices. In particular, the optimized parameter set enhances the representation of irrigation-induced soil wetting and crop growth cycles, leading to more realistic land–atmosphere exchange processes. This study highlights the importance of multi-source observational constraints and parameter sensitivity-informed calibration for land surface modeling in human-dominated environments. The proposed framework provides a transferable approach for improving hydrological simulations in heavily managed arid irrigation districts.