Aerosols and Greenhouse Gases Drive Opposing Rainfall Shifts in the Yellow River Basin

For a river that sustains some 100 million people across northern China, the Yellow River’s fate hangs on a deceptively simple question: why does it rain where it rains, and why does that change? A new study published in Climate Dynamics by Mengyao Cai, Linjing Qiu, and colleagues at Xi’an Jiaotong University, together with collaborators at the China University of Geosciences and the Chinese Academy of Sciences, dissects that question with unusual precision. Rather than treating rainfall as a single undifferentiated quantity, the team separated it into two physically distinct flavors: convective precipitation, produced by vigorous vertical updrafts that build towering storm clouds, and non-convective precipitation, the gentler, widespread rain and snow that forms when air is lifted gradually along large-scale weather systems. Each type, the researchers show, responds to human and natural influences through entirely different atmospheric machinery.

The study region could hardly be more consequential. The Yellow River Basin is a water-limited region where precipitation is the master regulator of the hydrological cycle, feeding agriculture, industry, and urban water supplies across a vast swath of semi-arid China. Convective and non-convective precipitation exert distinct influences on regional hydrological processes: convective rain tends to arrive in intense bursts that can trigger floods and landslides, while non-convective rain soaks more gently into soils and reservoirs. Understanding which type is waxing or waning, and why, is therefore not an academic exercise but a matter of water security for one of the world’s most densely populated river systems.

To untangle the mechanisms, the researchers combined two powerful lines of evidence. The first was ERA5, the European Centre for Medium-Range Weather Forecasts’ state-of-the-art reanalysis, which blends decades of observations into a physically consistent record of the atmosphere. The second came from CMIP6, the latest generation of global climate models used in international assessments, which allowed the team to run so-called single-forcing experiments. In these experiments, models are driven by one influence at a time: anthropogenic aerosols alone, greenhouse gases alone, or natural variability alone. By comparing the responses, the scientists could attribute changes in total, convective, and non-convective precipitation over the basin to specific culprits, a technique at the heart of modern climate attribution science.

The headline finding is sobering. Between 1960 and 2014, precipitation decreased over most parts of the Yellow River Basin. When the team decomposed that decline using the single-forcing experiments, anthropogenic aerosol pollution emerged as the dominant driver. Aerosol forcing reduced total precipitation by 0.92 millimeters per year, of which 0.51 millimeters per year came from suppressed convective rain and 0.39 millimeters per year from suppressed non-convective rain. In other words, the tiny particles emitted by industry, power plants, and vehicles did not merely trim rainfall at the margins; they throttled both major modes of precipitation generation across the basin, with the larger share of the damage falling on the convective component.

The mechanism behind the aerosol effect is a story about large-scale circulation, not just cloud microphysics. The analysis revealed that aerosol forcing weakened two pillars of East Asian climate: the East Asian Subtropical Westerly Jet, the high-altitude river of air that steers weather systems across the region, and the East Asian Summer Monsoon, the seasonal circulation that pumps moisture from the oceans onto the Asian continent. With the jet and monsoon enfeebled, less moisture was transported into the basin, and vertical ascent, the upward motion of air that is the essential engine of all precipitation, was suppressed. Deprived of both fuel and lift, convective storms and stratiform rain systems alike retreated, compounding the basin’s water stress during the decades of rapid industrialization.

Greenhouse gases tell the opposite story, and a more nuanced one. The single-forcing experiments showed that greenhouse gas forcing increased total precipitation by 0.56 millimeters per year, but almost all of that gain, 0.55 millimeters per year, came from a pronounced rise in convective precipitation. Non-convective precipitation, by contrast, slightly decreased, by 0.06 millimeters per year. This asymmetry reflects two competing pathways described by the study’s thermodynamic and dynamic analysis. Warming allows the atmosphere to hold more water vapor, a thermodynamic moistening that boosts the moisture available to convective storms. At the same time, greenhouse gas forcing increased moisture input into the basin and strengthened vertical advection, the upward transport of humid air, further energizing convection.

Yet the same greenhouse-gas pathway that feeds convective storms undermines non-convective rain. The analysis found that under greenhouse gas forcing, horizontal advection weakened, and it is precisely this horizontal transport of moisture within large-scale weather systems that sustains non-convective precipitation. The result is a basin whose rainfall becomes increasingly dominated by intense convective events even as the steadier, soaking rains that historically replenished soils and reservoirs slowly fade. For water managers, that shift matters enormously: a millimeter of torrential convective rain running off baked ground is not hydrologically equivalent to a millimeter of gentle stratiform rain percolating into an aquifer. The study’s decomposition suggests that the character of Yellow River rainfall is changing, not merely its total amount.

Natural forcing, encompassing influences such as solar variability and volcanic eruptions, induced relatively weak changes in the basin’s precipitation by comparison. That finding carries a pointed implication: the observed drying of the Yellow River Basin over the second half of the twentieth century was not primarily an act of nature but a consequence of human emissions, with aerosols doing most of the damage and greenhouse gases partially, but unevenly, offsetting it. As air quality regulations reduce aerosol emissions over China, the masking effect of particulate pollution may diminish, potentially unmasking the greenhouse-gas-driven intensification of convective rainfall, with attendant risks of more frequent flash flooding in a basin already grappling with drought.

The study’s methodological rigor deserves attention in its own right. Trend detection relied on well-established statistical tools, including the Mann-Kendall nonparametric test and Sen’s slope estimator, which are robust to outliers and non-normal data distributions common in hydroclimatic records. Model performance was evaluated using Taylor diagram diagnostics, and the team drew on a rich body of prior work on monsoon indices, westerly jet dynamics, and precipitation attribution across Asia. By anchoring the analysis in both reanalysis data and coordinated model experiments, the researchers built a chain of evidence that connects observed rainfall trends to specific, physically explicable pathways, rather than relying on correlations alone.

What emerges is a template for understanding regional hydroclimate change far beyond China’s borders. Precipitation is not a monolith; it is the sum of distinct physical processes, each with its own sensitivity to warming, pollution, and natural variability. The Yellow River Basin study demonstrates that the same external forcing can strengthen one precipitation type while weakening another, and that the sign of a regional rainfall trend can hinge on the balance between aerosol cooling and greenhouse warming. For the hundreds of millions who depend on the Yellow River, the message is clear: managing the water future of this basin will require watching not just how much it rains, but how it rains, and recognizing that the atmosphere’s response to human activity is written in two very different scripts, one written by rising thermometers and one by drifting plumes of particulate haze.

Subject of Research: Attribution of convective and non-convective precipitation changes in the Yellow River Basin to aerosol, greenhouse gas, and natural forcings

Article Title: Thermodynamic and dynamic pathways of convective and non-convective precipitation under multiple forcings in the Yellow River Basin, China

Article References: Cai, M., Qiu, L., Wu, Y., Ma, Y., Chang, Y., & Zhao, D. (2026). Thermodynamic and dynamic pathways of convective and non-convective precipitation under multiple forcings in the Yellow River Basin, China. Climate Dynamics, 64(11), Article 464. https://doi.org/10.1007/s00382-026-08429-0

Image Credits: AI Generated

DOI: 10.1007/s00382-026-08429-0

Keywords: Yellow River Basin, convective precipitation, non-convective precipitation, aerosol forcing, greenhouse gases, East Asian Summer Monsoon, subtropical westerly jet, CMIP6, ERA5 reanalysis, climate attribution, hydroclimate change, moisture transport