Global meta-analysis and machine learning uncover region-specific salinity thresholds and risks for sustainable wheat production under saline irrigation
- 1Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences
- 2University of Chinese Academy of Sciences
Abstract
Saline irrigation represents a pivotal alternative strategy for enhancing crop production under freshwater scarcity. However, previous assessments have mainly quantified average yield responses, while how interacting environmental and management conditions shape region-specific irrigation-water salinity thresholds and associated spatial risks remain elusive across global major breadbaskets. This study combined a meta-analysis of 551 field observations with machine learning (ML) to quantify global wheat yield responses to saline irrigation, identify optimal region-specific thresholds, and evaluate spatial risks in four major wheat-producing areas: the North China Plain (NCP), Indo-Gangetic Plain (IGP), Middle East (ME), and Central California Valley (CCV). Meta-analysis revealed that saline irrigation led to significant average yield reductions ranging from 12.4% in the NCP to 31.4% in the CCV. ML further showed that irrigation water salinity, growing-season temperature, and irrigation amount emerged as the dominant drivers, collectively explaining more than 54% of yield variation. An optimum irrigation amount of 500 mm was optimal for maximizing yield under saline conditions. Region-specific best practices included border irrigation at flowering (FL) and grain-filling stages for the NCP, drip irrigation at these stages for the IGP, and drip irrigation at the FL stage for the ME and CCV. Spatial risk assessment revealed a mismatch between salinity tolerance thresholds and actual exposure. Under a conservative 5% yield-loss scenario, the IGP exhibited the highest model-derived salinity threshold of 4.14 dS m ⁻¹ but the largest risk area proportion due to elevated background salinity, whereas the ME faced the lowest risk, with a risk area of 7.7%, indicating high potential for saline water utilization. These findings highlight the limitations of universal water quality guidelines and demonstrate the necessity of localized management. By establishing a data-driven framework, this study provides actionable insights to safeguard global wheat supply, build climate resilience, and ensure agricultural sustainability in salt-affected environments.
Related
- Sustainable energy for agricultural irrigation in Ethiopia: techno-economic and life cycle analysis of solar versus diesel pumps
- Depth-dependent soil water-salt relationships in the Hetao Irrigation District, arid Northwest China: Implications for sustainable management
- Every drop counts: optimizing water use efficiency for sustainable tomato production
- IrriTwin: Integrating reinforcement learning with soil water and crop models toward digital-twin-oriented irrigation scheduling
- A Spatially Explicit Societal Life Cycle Cost Framework for Sustainable Biographite Production in the United States.
BibTeX
@article{fang2026global,
title = {{Global meta-analysis and machine learning uncover region-specific salinity thresholds and risks for sustainable wheat production under saline irrigation}},
author = {Yimin Fang and Guoyong Leng and Linfei Yu and Jiali Qiu and Xiaoyong Liao},
journal = {Environmental Research Letters},
year = {2026},
month = oct,
doi = {10.1088/1748-9326/aeaf6c},
url = {https://iopscience.iop.org/article/10.1088/1748-9326/aeaf6c},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Global meta-analysis and machine learning uncover region-specific salinity thresholds and risks for sustainable wheat production under saline irrigation
AU - Yimin Fang
AU - Guoyong Leng
AU - Linfei Yu
AU - Jiali Qiu
AU - Xiaoyong Liao
PY - 2026
DA - 2026/10/12
JO - Environmental Research Letters
DO - 10.1088/1748-9326/aeaf6c
UR - https://iopscience.iop.org/article/10.1088/1748-9326/aeaf6c
AB - Saline irrigation represents a pivotal alternative strategy for enhancing crop production under freshwater scarcity. However, previous assessments have mainly quantified average yield responses, while how interacting environmental and management conditions shape region-specific irrigation-water salinity thresholds and associated spatial risks remain elusive across global major breadbaskets. This study combined a meta-analysis of 551 field observations with machine learning (ML) to quantify global wheat yield responses to saline irrigation, identify optimal region-specific thresholds, and evaluate spatial risks in four major wheat-producing areas: the North China Plain (NCP), Indo-Gangetic Plain (IGP), Middle East (ME), and Central California Valley (CCV). Meta-analysis revealed that saline irrigation led to significant average yield reductions ranging from 12.4% in the NCP to 31.4% in the CCV. ML further showed that irrigation water salinity, growing-season temperature, and irrigation amount emerged as the dominant drivers, collectively explaining more than 54% of yield variation. An optimum irrigation amount of 500 mm was optimal for maximizing yield under saline conditions. Region-specific best practices included border irrigation at flowering (FL) and grain-filling stages for the NCP, drip irrigation at these stages for the IGP, and drip irrigation at the FL stage for the ME and CCV. Spatial risk assessment revealed a mismatch between salinity tolerance thresholds and actual exposure. Under a conservative 5% yield-loss scenario, the IGP exhibited the highest model-derived salinity threshold of 4.14 dS m ⁻¹ but the largest risk area proportion due to elevated background salinity, whereas the ME faced the lowest risk, with a risk area of 7.7%, indicating high potential for saline water utilization. These findings highlight the limitations of universal water quality guidelines and demonstrate the necessity of localized management. By establishing a data-driven framework, this study provides actionable insights to safeguard global wheat supply, build climate resilience, and ensure agricultural sustainability in salt-affected environments.
SN - 1748-9326
ER -