<p>Agricultural ecosystems play a crucial role in sustaining food production, regulating carbon cycling, and supporting sustainable water resource use. However, under global warming and increasingly frequent extreme weather events, the stability of cropland water–carbon processes faces substantial challenges. To better understand these processes, this study evaluates the performance of the STEMMUS-SCOPE model using in-situ observations from a winter wheat cropland in the Guanzhong Plain. Model parameters were calibrated using observed meteorological variables and water–carbon fluxes. The results show that the model accurately simulates net radiation (Rn), latent heat flux (LE), gross primary productivity (GPP), and net ecosystem exchange (NEE) with high reliability (R² &gt; 0.75), whereas the simulation of sensible heat flux (H) exhibits noticeable deviations, primarily attributable to uncertainties in the estimation of aerodynamic and thermal resistances. The model also reliably captures the temporal dynamics of soil moisture and temperature across multiple depths, with RMSE values of 0.01–0.04 m³ m^− 3 for soil moisture and 0.76–3.85℃ for soil temperature, indicating strong capability in representing soil hydrothermal processes. To address missing observational data, simulated GPP and LE were used to fill data gaps and construct complete water–carbon flux time series. Based on the reconstructed dataset, seasonal evapotranspiration, carbon uptake, and water–carbon coupling characteristics were quantified. The estimated water use efficiency (WUE) reached up to 3.15&#xa0;g C kg^− 1 H₂O, providing meaningful insights for optimizing water resource management and improving crop productivity. Overall, STEMMUS-SCOPE demonstrates strong potential for simulating ecological–hydrological processes in winter wheat croplands, though future work should extend validation to more agricultural ecosystems, additional sites, and longer time series to further enhance model robustness and regional applicability.</p>