Machine learning projections of Iran’s water scarcity response to climate-land use synergies
Year: 2025
Type: article
Abstract: Study region Iran faces severe water stress with over 88 million people and 14.2 million hectares of cultivated land. Agriculture consumes over 87 % of national water, with total withdrawals exceeding 80 % of renewable resources. From 2003–2019, total water storage declined by over 200 BCM, exceeding annual national consumption. Study focus This study analyzed the impacts of climate change and cultivated area on Iran’s total water storage. Country-wide actual evapotranspiration (ET a ) in agricultural and non-agricultural sectors was simulated using machine learning and statistical analysis. Changes in total water storage were then projected via the water balance equation using bootstrap sampling, under combined climate change and cultivated area management scenarios. New hydrological insights Projections indicate that by 2100, climate change will increase nationwide ET a by up to 7.7 % and agricultural ET a by 23.8 %. Even 10 % reduction in cultivated area cannot offset the impacts of changing climate. Under high emissions scenarios, nationwide ET a will increase by 37.1 BCM (21 %) and agricultural ET a by 17.5 BCM (39 %) compared to the 2009–2014 baseline. Given current consumption trends and high withdrawal ratios (>80 %), aquifer deficits are projected to increase by 300 BCM (42 %) to 578 BCM (174 %) over the next two decades, depending on climate scenarios. These findings warn that failure to manage demand and improve irrigation efficiency in agriculture will seriously threaten the country’s water and food security.
Authors: Nooshdokht Bayat-Afshary, Mohammad Danesh-Yazdi, Fereshteh Shakeri
Institution: Sharif University of Technology
Language: English
Subfield: Ocean Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: gold
Estimated APC paid: $2,600
Funder: Iran National Science Foundation
Award: G7512308297
OpenAlex ID: W4412626153