Rising Aridity Could Expand Salty Soils Across Global Drylands by 2100

Some of the world’s most productive farmland may be quietly turning toxic from within. A new study published in Nature Communications projects that as the planet’s dry regions grow hotter and drier, the area of salt-affected soils across global drylands could expand by roughly 8 percent by the end of the century under a high-emissions scenario, threatening crops that feed hundreds of millions of people. The research, led by Dietian Yu and Zhou Shi of Zhejiang University together with an international team spanning institutions from Tsinghua University to the Food and Agriculture Organization of the United Nations, offers one of the most comprehensive forward-looking maps yet of where and how soil salinization is likely to intensify as aridity increases.

Soil salinization is one of those slow-motion environmental crises that rarely makes headlines but steadily erodes the foundations of agriculture. When salts accumulate in the root zone, plants struggle to take up water, nutrients become imbalanced, and yields fall. In severe cases, land is abandoned altogether. The problem is particularly acute in drylands, the arid and semi-arid regions that cover roughly 40 percent of Earth’s land surface and are home to a large share of the world’s population. In these water-limited environments, evaporation routinely exceeds rainfall, drawing dissolved salts upward through the soil profile and leaving them behind at or near the surface.

What has been missing, the authors argue, is a clear picture of how climate change itself will reshape the global geography of soil salinity. Previous assessments have tended to snapshot the problem at a single point in time, without accounting for the dynamic interplay between shifting climate zones, changing land use, and human activity. The new study set out to close that gap by asking a deceptively simple question: as drylands expand and grow more arid through the twenty-first century, where will salt follow?

To answer it, the team first assembled a global soil salinity dataset compiled from observations over the past two decades, a relatively comprehensive baseline by the standards of this field. Building reliable global salinity data is notoriously difficult. Soil salinity varies enormously over short distances, responds to seasonal water tables and irrigation practices, and is measured inconsistently across countries. By pulling together two decades of observations, the researchers created a training foundation broad enough to support global-scale modeling.

Onto that foundation they built climate-driven machine learning models that predict soil salinity from multiple covariates. Crucially, the models incorporated not just climatic variables but also a dynamic representation of dryland extent, which itself shifts as aridity classifications change under warming, and indicators of human influence such as land management pressures. That combination matters because salinization is never purely a climate story. Irrigation, deforestation, and agricultural intensification all accelerate salt accumulation, and any credible projection has to let those human fingerprints interact with the physical drivers rather than treating them as static background conditions.

Running the models forward under SSP5-8.5, the high-emissions scenario that represents a future with limited climate mitigation, the researchers found an 8 percent expansion of saline soils across global drylands by 2100 relative to the 2004 to 2023 reference period. That figure may sound modest, but in the context of drylands, where saline soils already blanket vast areas, it represents a substantial addition of newly degraded land. The expansion is driven by increasing aridity: as precipitation patterns shift and evaporative demand climbs, the climatic conditions that concentrate salts in surface soils spread into regions that have so far escaped the worst of the problem.

The study’s risk maps identify the places where this intensification will bite hardest. Southwestern United States, Mexico, southern Europe, central South America, and eastern and southern Africa emerge as the highest-risk hotspots. Several of these regions sit at the intersection of multiple stressors: they are projected to become markedly drier, they support significant agricultural production, and in many cases they already face water scarcity and land degradation pressures. Southern Europe’s irrigated basins, the croplands of Mexico and the American Southwest, and the rain-fed farming systems of eastern and southern Africa all fall within the projected danger zones, which means the findings carry direct implications for regional food policy and land management planning.

Perhaps the most consequential findings concern croplands. The researchers show that dryland croplands face serious salinization in the coming decades, with irrigated systems especially vulnerable. That vulnerability is in some ways a cruel irony: irrigation is what makes agriculture possible in many arid regions, but irrigation water carries dissolved salts, and without adequate drainage and leaching, every application leaves a residue behind. As aridity increases, evaporative concentration of those salts accelerates, tightening the squeeze on farmers who depend on the very water deliveries that are poisoning their soil.

To translate soil chemistry into human terms, the team estimated potential production losses for four staple crops in newly salinized irrigated areas. By 2100, they project that more than 4.9 percent of maize production, 17.2 percent of dryland rice production, 3.3 percent of sorghum production, and 21.2 percent of wheat production could be lost in these affected zones. Wheat, the world’s most widely grown cereal and a dietary cornerstone across the Middle East, North Africa, and Central Asia, shows the steepest projected losses, a result that should ring alarm bells for food-importing nations in precisely the regions flagged as hotspots. Rice losses approaching a fifth of production in newly salinized dryland areas are similarly striking, given the crop’s importance to food security across Asia and Africa.

The authors frame their work as a tool for long-term adaptation rather than a doomsday forecast. Because the models identify where salinization pressure will build decades in advance, the results can inform the siting of drainage infrastructure, the selection of salt-tolerant crop varieties, the design of irrigation schemes, and the prioritization of land for conservation or restoration. The study was supported by funding from China’s Ministry of Education, the National Natural Science Foundation of China, and the Xinjiang Key Laboratory of Soil and Plant Ecological Processes, and it appears in open-access form, making the projections available to the planners and policymakers who need them most.

There are, of course, uncertainties inherent in any century-scale projection. Machine learning models are only as good as their training data, and soil salinity observations remain sparse in some of the world’s most remote drylands. The choice of SSP5-8.5, a high-emissions pathway, means the projections represent something closer to an upper bound than a central estimate; a world that decarbonizes more aggressively would likely see less aridity-driven salinization, though the direction of the effect would remain. The authors also note that their published version is subject to final editorial updates. Still, the core message stands on solid physical ground: salt moves where water moves, and as climate change redraws the global water balance, the salty edges of the world’s drylands are set to creep inward, toward some of the most important farmland on Earth. The question for the coming decades is not whether dryland salinization will intensify, but whether governments and farmers will act on the warning while there is still time to adapt.

Subject of Research: Projected expansion of soil salinization in global drylands under increasing aridity and its impacts on crop production

Article Title: Projected intensification of soil salinization in global drylands under future aridity increase

Article References: Yu, D., Wang, N., Hassani, A., Perri, S., Peng, Y., Chen, S., Peng, J., Shangguan, Y., Zhang, C., Dong, J., Zhang, G., & Shi, Z. (2026). Projected intensification of soil salinization in global drylands under future aridity increase. Nature Communications. https://doi.org/10.1038/s41467-026-78622-6

Image Credits: AI Generated

DOI: 10.1038/s41467-026-78622-6

Keywords: soil salinization, drylands, aridity, climate change, machine learning, crop yields, irrigation, food security, SSP5-8.5, wheat, Nature Communications, land degradation