Chevuru et al. (2025) Relevance of feedbacks between water availability and crop systems using a coupled hydrological–crop growth model

Identification

• Journal: Hydrology and earth system sciences
• Year: 2025
• Date: 2025-09-09
• Authors: Sneha Chevuru, Rens van Beek, Michelle T. H. van Vliet, Jerom Aerts, Marc F. P. Bierkens
• DOI: 10.5194/hess-29-4219-2025

Research Groups

• Department of Physical Geography, Utrecht University, Utrecht, the Netherlands
• Water Resources Section, Faculty of Civil Engineering and Geosciences, Delft University of Technology, Delft, the Netherlands
• Department of Hydraulic Engineering, Faculty of Civil Engineering and Geosciences, Delft University of Technology, Delft, the Netherlands
• Unit Subsurface & Groundwater Systems, Deltares, Utrecht, the Netherlands

Short Summary

This study quantifies the fine-grained spatiotemporal feedback between crop systems and hydrology using a coupled hydrological-crop growth model. It finds that two-way coupling, incorporating dynamic feedback of crop phenology, significantly improves performance for rainfed crops compared to one-way coupling, highlighting its necessity for capturing interannual climate variability impacts on food production.

Objective

• To quantify the feedback between crop systems and hydrology on a fine-grained spatiotemporal level.
• To test the hypothesis that coupling a crop growth model with a hydrological model will improve both crop yield and hydrological predictions by incorporating dynamic feedbacks between water availability and crop processes.
• Specifically, to test if a more realistic representation of soil moisture dynamics and water availability leads to better estimates of water stress and yield, and if integrating crop growth information into hydrological models enhances the accuracy of predictions regarding irrigation needs and water resource allocation.

Study Configuration

• Spatial Scale: Contiguous United States (CONUS) region at 5 arcminute (approximately 10 kilometer) resolution.
• Temporal Scale: Daily time step, covering the period from 1979 to 2019.

Methodology and Data

• Models used:
  • PCR-GLOBWB 2 (hydrological and water resources model)
  • WOFOST (crop growth model)
  • Coupling implemented using the Basic Model Interface (BMI) framework.
• Data sources:
  • W5E5 forcing data (Lange et al., 2021) for meteorological variables (temperature, radiation, precipitation, vapor pressure, wind speed, humidity).
  • MIRCA2000 (Portmann et al., 2010) for global monthly irrigated and rainfed crop area datasets and growing seasons.
  • WOFOST crop parameter dataset (WOFOST Crop Parameters, 2024) for crop parameters, atmospheric CO2 concentrations, and fertilizer application.
  • FAO soil map (FAO, 2007) for harmonized soil parameters.
  • United States Department of Agriculture (USDA) National Agricultural Statistics Service (USDA, 2024) for reported crop yields (for validation).
  • USGS Water Use Database (USGS, 2023) for reported annual irrigation withdrawals (for validation).

Main Results

• The coupled framework effectively replicates observed irrigated and rainfed crop yields and irrigation water use when evaluated against reported data.
• Two-way coupling, which includes dynamic feedback mechanisms, significantly outperforms one-way coupling for rainfed crops. This improvement is attributed to the feedback of WOFOST crop phenology (e.g., leaf area index, rooting depth, evapotranspiration) to the crop parameters in the hydrological model.
• For irrigated crops, one-way and two-way coupled model runs yield nearly identical results to stand-alone WOFOST simulations, indicating that irrigation generally meets crop water demands.
• One-way coupling generally overestimates rainfed yields compared to stand-alone and two-way simulations, particularly for maize and wheat, due to the prescribed phenology in the hydrological model not dynamically responding to crop development.
• Rainfed crops exhibit larger inter-annual variability (higher coefficient of variation) in the mid-western and western CONUS, with two-way coupling showing a larger coefficient of variation than one-way, underscoring the importance of dynamic crop phenology for accurate soil moisture simulation.
• Simulated irrigation water withdrawals are correct in order of magnitude but generally underestimated compared to corrected USGS reported data.

Contributions

• Developed a novel two-way coupled hydrological–crop growth model framework by integrating PCR-GLOBWB 2 and WOFOST at a fine spatial (5 arcminute) and temporal (daily) resolution for the CONUS region.
• Quantified the intricate two-way interactions and feedback mechanisms between crop growth and hydrological systems, addressing a previous knowledge gap in understanding these dynamics.
• Demonstrated the critical importance of two-way coupling for accurately simulating rainfed crop yields and their interannual variability, especially under drier conditions, by allowing crop phenology to dynamically influence hydrological processes.
• Enhanced the understanding of hydrological and crop growth interactions and their implications for agricultural productivity and water resource management at a continental scale.
• Provided a framework capable of evaluating large-scale water use management strategies and simulating impacts of informed decision-making under changing hydroclimatic conditions.

Funding

• European Union Horizon program GoNexus project (grant agreement number 101003722).
• Netherlands Scientific Organisation (NWO) VIDI grant (VI.Vidi.193.019).
• European Research Council (ERC) under the European Union’s Horizon Europe Research and Innovation program (grant agreement 101039426 B-WEX).

Citation

@article{Chevuru2025Relevance,
  author = {Chevuru, Sneha and Beek, Rens van and Vliet, Michelle T. H. van and Aerts, Jerom and Bierkens, Marc F. P.},
  title = {Relevance of feedbacks between water availability and crop systems using a coupled hydrological–crop growth model},
  journal = {Hydrology and earth system sciences},
  year = {2025},
  doi = {10.5194/hess-29-4219-2025},
  url = {https://doi.org/10.5194/hess-29-4219-2025}
}