Tu et al. (2026) CO2-driven vegetation responses reduce projected rice irrigation water requirements across Asia

Identification

• Journal: Agricultural Water Management
• Year: 2026
• Date: 2026-10-21
• Authors: Zhuoyi Tu, Changming Li, Ziwei Liu, Xu Lian, Jin Fu, Juntai Han, Hao Liang, Jinghua Xiong, Feng Zhou
• DOI: 10.1016/j.agwat.2026.110879

Research Groups

• Key Laboratory of Soil and Water Processes in Watershed, College of Geography and Remote Sensing, Hohai University, Nanjing, China
• School of Civil Engineering and Transportation, State Key Laboratory of Subtropical Building and Urban Science, South China University of Technology, Guangzhou, China
• CSIRO Agriculture and Food, Black Mountain, ACT, Australia
• Sino-France Institute of Earth Systems Science, Laboratory for Earth Surface Processes, College of Urban and Environmental Sciences, Peking University, Beijing, China
• School of Civil Engineering and Water Resources, Qinghai University, Xining, China
• State Key Laboratory of Hydroscience and Engineering, Department of Hydraulic Engineering, Tsinghua University, Beijing, China

Short Summary

This study quantifies how CO2-driven vegetation responses alter projected rice irrigation water requirements across Asia, revealing that neglecting these dynamic responses in conventional models leads to significantly higher projected irrigation demands under future high-CO2 conditions.

Objective

• To determine how CO2-driven vegetation responses alter projected rice irrigation water requirements across major Asian rice-growing regions.
• To assess how the conventional FAO-56 framework differs from a dynamic-vegetation formulation under rising atmospheric CO2 concentrations.

Study Configuration

• Spatial Scale: Major rice-growing regions in Asia within warm temperate and tropical climate zones, spanning 10.0°S–37.1°N and 60.9°E–141.0°E, covering approximately 11.22 million km².
• Temporal Scale: Historical period: 1982–2020. Future projection: 2020–2100 under Shared Socioeconomic Pathway 5–8.5 (SSP5–8.5).

Methodology and Data

• Models used:
  • Conventional FAO-56 reference-crop framework (IRC)
  • Process-based analytical dynamic-vegetation approach (IVEG), incorporating the Medlyn et al. (2011) optimal stomatal conductance model.
  • FAO water balance approach for irrigation water requirement calculation.
  • 11 Global Climate Models (GCMs) from Coupled Model Intercomparison Project Phase 6 (CMIP6).
• Data sources:
  • ERA5-Land monthly averaged reanalysis dataset (meteorological variables, 0.1° spatial resolution).
  • GIMMS LAI4g product (Leaf Area Index, 1/12° spatial resolution).
  • Gridded rice harvested area dataset (10 km spatial resolution).
  • RiceAtlas (rice phenological information/crop calendar).
  • FLUXNET-CH4 Community Product/V2.0 dataset (observed latent heat flux and meteorological data from seven rice-paddy eddy-covariance sites for validation).
  • CMIP6 SSP5-8.5 simulations (future climate projections).

Main Results

• During 1982–2020, conventional (IRC) and dynamic-vegetation (IVEG) irrigation water requirements showed broadly similar spatiotemporal variability and weak, statistically non-significant trends.
• Under SSP5–8.5 (2020–2100), IRC is projected to increase by 6.7 mm yr⁻¹ per decade, whereas IVEG is projected to decrease by 3.6 mm yr⁻¹ per decade, resulting in an end-of-century gap of approximately 155 mm yr⁻¹.
• Attribution analyses indicate that IVEG additionally captures eco-physiological feedbacks that partially offset the climate-driven increase in irrigation demand projected by IRC.
• Factorial experiments decomposed vegetation effects into a CO2-driven physiological water-saving effect (−2.55 mm yr⁻¹ yr⁻¹) and an LAI-driven structural water-consuming effect (1.28 mm yr⁻¹ yr⁻¹).
• The net vegetation effect leads to reductions in irrigation water requirements, with stronger modeled net reductions associated with grid cells having higher rice harvested-area density.
• The effective crop coefficient (Kc_VEG) derived from the dynamic-vegetation formulation systematically decreases with increasing atmospheric CO2 across all rice growth stages, contrasting with the fixed coefficients in the conventional FAO-56 framework.

Contributions

• Quantifies how CO2-driven vegetation responses fundamentally alter projected rice irrigation water requirements across major Asian rice-growing regions, addressing a critical gap in large-scale irrigation assessments.
• Develops and applies a process-consistent framework to separate and quantify the competing CO2-driven physiological (stomatal) water-saving effect and the LAI-driven structural (canopy) water-consuming effect.
• Highlights that neglecting CO2-driven vegetation responses is a significant source of structural uncertainty in future rice irrigation-demand projections, potentially leading to overestimations under high-CO2 conditions.
• Demonstrates that the effective crop coefficient for rice is not constant but systematically decreases with rising atmospheric CO2, providing a model-based reinterpretation within the FAO-56 framework.
• Identifies that the modeled net water-saving effect is more pronounced in grid cells with higher rice harvested-area density, indicating regions where vegetation representation is particularly crucial for accurate projections.

Funding

• National Natural Science Foundation of China (Grant No. 42501032, 42361144876)
• Fundamental Research Funds for the Central Universities (Grant No. B250201124)
• Open Research Fund Program of the State Key Laboratory of Hydroscience and Engineering (Grant No. sklhse-KF-2025-A-01)

Citation

@article{Tu2026CO2driven,
  author = {Tu, Zhuoyi and Li, Changming and Liu, Ziwei and Lian, Xu and Fu, Jin and Han, Juntai and Liang, Hao and Xiong, Jinghua and Zhou, Feng},
  title = {CO2-driven vegetation responses reduce projected rice irrigation water requirements across Asia},
  journal = {Agricultural Water Management},
  year = {2026},
  doi = {10.1016/j.agwat.2026.110879},
  url = {https://doi.org/10.1016/j.agwat.2026.110879}
}