Du et al. (2026) Available-water adjustment improves monthly runoff simulation in cropland-dominated basins: A generalized water–energy complementarity model

Identification

• Journal: Agricultural Water Management
• Year: 2026
• Date: 2026-10-24
• Authors: Shixiong Du, Baowei Yan, Changmei Liang, Yixuan ZOU, Donglin Gu, Dongxu Yang, Jianbo Chang, Xuerui Zhou
• DOI: 10.1016/j.agwat.2026.110888

Research Groups

• School of Civil and Hydraulic Engineering, Huazhong University of Science and Technology, Wuhan, China
• Hubei Key Laboratory of Digital River Basin Science and Technology, Huazhong University of Science and Technology, Wuhan, China
• Hubei Water Saving Research Center, Hubei Water Resources Research Institute, Wuhan, China
• Sichuan Water Development Investigation, Design & Research Co.,Ltd., Chengdu, China

Short Summary

This study enhanced the generalized water–energy complementarity (GWEC) model for monthly runoff simulation by introducing separate adjustments to effective water and energy constraints, demonstrating significantly improved performance and temporal robustness, particularly in cropland-dominated basins.

Objective

• To extend the generalized water–energy complementarity (GWEC) framework to monthly runoff simulation by explicitly accounting for deviations between theoretical and effective water and energy constraints.
• To examine whether the available-water and energy adjustments provide distinct contributions to model performance and exhibit limited practical redundancy.
• To evaluate the robustness and transferability of the proposed framework across contrasting climatic, hydrological, and land-cover conditions.
• To assess its applicability to cropland-dominated basins and its potential value for regional screening and seasonal water-availability assessment.

Study Configuration

• Spatial Scale: 802 basins across the United States (563 Camels-US basins) and Brazil (239 CABra basins).
• Temporal Scale: Monthly runoff simulation using continuous records spanning over 30 years (Camels-US: 1981–2014; CABra: 1981–2010).

Methodology and Data

• Models used: Generalized Water–Energy Complementarity (GWEC) model (proposed), Mezentsev–Choudhury–Yang equation (MCY), Proportionality-Hypothesis Formulation of Wang and Tang (PHWT), and ABCD water-balance model (benchmarks).
• Data sources:
  • Catchment Attributes and Meteorology for Large-sample Studies in the United States (Camels-US) dataset.
  • Catchments Attributes for Brazil (CABra) dataset.
  • Daily meteorological records (precipitation, temperature, solar radiation) aggregated to monthly values.
  • Potential evapotranspiration (Ep) for Camels-US calculated using the Penman–Monteith method with ERA5-Land dataset forcing. Ep for CABra provided directly by the dataset.
  • Runoff data converted to catchment-average water depth in millimeters (mm).

Main Results

• The GWEC model achieved median Nash–Sutcliffe Efficiency (NSE) values of 0.60 (Camels-US) and 0.77 (CABra) during the validation period, outperforming all benchmark models.
• GWEC showed significantly higher basin-wise NSE and Kling-Gupta Efficiency (KGE) and lower absolute volume error (VE) than the MCY, PHWT, and ABCD models in both datasets (p < 0.01).
• In cropland-dominated basins, GWEC achieved a median validation NSE of 0.65, exceeding benchmarks by 0.02–0.11, and exhibited the smallest calibration-to-validation decline among all land-cover groups, indicating enhanced temporal robustness.
• Improvements in GWEC performance were more pronounced in catchments with higher baseflow contributions.
• GWEC achieved the lowest median Structural Risk Minimization (SRM) values (0.52 for Camels-US, 0.35 for CABra), demonstrating a favorable balance between predictive performance and model complexity.
• The separate adjustment parameters for effective water (i) and energy (m) constraints were found to be practically non-equivalent and provided distinct contributions to model performance.

Contributions

• Extended the generalized water–energy complementarity (GWEC) framework to dynamic monthly runoff simulation by explicitly incorporating adjustments for deviations between theoretical and effective water and energy constraints.
• Demonstrated that these separate water and energy adjustments provide distinct and non-redundant improvements in monthly runoff simulation, enhancing model robustness.
• Provided a comprehensive evaluation of the GWEC framework's robustness and transferability across a wide range of climatic, hydrological, and land-cover conditions using large-sample datasets.
• Highlighted the particular applicability and improved temporal robustness of the GWEC model in cropland-dominated basins, offering value for regional screening and seasonal water-availability assessment in agricultural landscapes.

Funding

• National Natural Science Foundation of China (52079054)
• National Natural Science Foundation of Hubei Province (2024AFB646)

Citation

@article{Du2026Availablewater,
  author = {Du, Shixiong and Yan, Baowei and Liang, Changmei and ZOU, Yixuan and Gu, Donglin and Yang, Dongxu and Chang, Jianbo and Zhou, Xuerui},
  title = {Available-water adjustment improves monthly runoff simulation in cropland-dominated basins: A generalized water–energy complementarity model},
  journal = {Agricultural Water Management},
  year = {2026},
  doi = {10.1016/j.agwat.2026.110888},
  url = {https://doi.org/10.1016/j.agwat.2026.110888}
}