The research, led by Ruige Dong and Yuan Zhang of Central South University of Forestry and Technology together with colleagues, tackles a question that has long frustrated ecologists and climate scientists. Aerosols—tiny solid or liquid particles suspended in the atmosphere, ranging from a thousandth of a micron to 100 microns across—do more than pollute the air. They scatter and absorb solar radiation, altering the radiation balance at the surface, shifting temperature and precipitation patterns, and changing the very light environment in which plants photosynthesize. Whether these effects help or hurt vegetation has remained stubbornly uncertain, because aerosols act through competing pathways that can push productivity in opposite directions at the same time.

The core of the ambiguity lies in the physics of scattered light. When aerosol concentrations rise, they intercept direct beam radiation, reducing the total solar energy reaching the ground and potentially starving plants of the photons they need. Yet the same particles convert a portion of that direct radiation into diffuse radiation, which arrives from all directions of the sky rather than in a single beam. Diffuse light penetrates deeper into plant canopies, illuminating shaded lower leaves that direct beam would never reach, and thereby raising the overall light-use efficiency of the canopy. Earlier work, including a landmark 2009 analysis published in Nature, suggested that increases in diffuse radiation can enhance the global land carbon sink. But the boost from diffuse light does not necessarily compensate for the loss of direct radiation, and at high concentrations aerosols can also deposit on leaf surfaces, clog stomata, and induce physiological stress. The net effect, in other words, depends on dose—and on whether gains in good years can offset losses in bad ones.

To disentangle this balance, the team turned to a statistical tool called the asymmetry index, or AI, borrowed from earlier ecological studies of how ecosystems respond to precipitation variability. The index works by comparing the magnitude of productivity gains during favorable years against the magnitude of losses during unfavorable years, standardized against the long-term mean. When the index is positive, the gains exceed the losses—the system has been effectively compensated. When it is negative, the damage done in bad years outweighs the recovery in good ones. Applied pixel by pixel across China, the index allowed the researchers to ask not just whether vegetation productivity correlates with aerosols, but whether the relationship is symmetric: whether a hazy year costs the ecosystem exactly what a clear year gives back.

The data underpinning the analysis were formidable. The team used the LGHAP dataset, the first long-term, gap-free, 1-kilometer-resolution AOD product for China, which fuses six satellite aerosol retrievals, ground-based measurements, and reanalysis simulations using random-forest bias correction and a tensor-reconstruction technique. Validation against AERONET ground observations yielded a correlation coefficient of 0.91. On the vegetation side, the researchers tracked four complementary indicators: aboveground carbon storage derived from field-calibrated biomass maps, leaf area index from the MODIS MOD15A2H product, net primary productivity from MOD17A3HGF, and gross primary productivity upscaled from the China FLUX eddy-covariance network using random forest regression. All datasets were harmonized to a common 1-kilometer grid spanning 2002 to 2020, and a five-year sliding window—chosen to smooth interannual climate noise while matching the roughly five-year horizon of China’s major air-pollution policies—was used to track how the asymmetry evolved over time.

The headline finding is that China’s vegetation productivity has, on the whole, been living on aerosol-related compensation that is now running out. During the study period, all four productivity indicators rose steadily: aboveground carbon storage climbed from 1.835 to 2.196 kilograms of carbon per square meter, leaf area index and gross primary productivity each increased by roughly 0.15 to 0.16 kilograms of carbon per square meter, and net primary productivity edged up from 0.412 to 0.453. Meanwhile, national mean AOD fluctuated between 0.293 and 0.397, rising until 2011 and then declining steadily after a turning point in 2012, coinciding with the Twelfth Five-Year Plan’s push for clean energy and the 2016 revision of the Air Pollution Prevention and Control Law. Regression analysis showed significant negative correlations between AOD and all four productivity parameters, with coefficients ranging from −0.49 for gross primary productivity to −0.61 for aboveground carbon storage, and partial correlation analysis controlling for temperature and precipitation confirmed that negative relationships dominated across 78 to 90 percent of significantly correlated areas.

The asymmetry analysis revealed where the compensation was happening and where it was failing. Nationally, positive asymmetric responses covered 64.67 percent of the area for leaf area index, 58.87 percent for net primary productivity, and 51.20 percent for gross primary productivity, meaning that in most of China, the productivity gains of clear years outweighed the losses of hazy ones. But aboveground carbon storage told a different story, with 43.77 percent of the country showing negative asymmetry—losses uncompensated by subsequent gains. Regionally, northeastern, eastern, and central China stood out as zones of persistent negative response, where high-AOD years inflicted damage that low-AOD years could not repair. The sliding-window analysis added a temporal dimension that may prove the study’s most consequential result: the asymmetry index for all four parameters peaked during the 2008–2013 window, when compensatory capacity was strongest, and has declined steadily since. In other words, the aerosol-related subsidy to vegetation productivity is diminishing, and the balance is tipping from positive toward negative asymmetry as China’s air continues to clear.

The response also varied sharply by ecosystem type and environment. Grasslands proved the most sensitive to aerosol fluctuations, showing the largest positive pulses and negative declines in productivity, a pattern the authors attribute to the shallow root systems, rapid growth cycles, low light saturation points, and structurally simple canopies of herbaceous vegetation. Croplands and shrublands responded strongly as well, while forests were remarkably muted—their complex, multi-layered canopies, enormous biomass stocks, and long life cycles buffer them against short-term swings in the light environment. Environmental gradients modulated the pattern further. In regions receiving less than 900 millimeters of annual precipitation, productivity losses from hazy years were generally compensated; beyond that threshold, and especially above 2,100 millimeters, compensation weakened or vanished, likely because excess water drives anaerobic respiration in plant roots while increased cloud cover further suppresses sunlight. Temperature told a similar story: below 15 degrees Celsius mean annual temperature, vegetation showed positive asymmetry, but above that threshold compensation faded, and in the warmest zones the index for net primary productivity and gross primary productivity turned negative, plausibly because elevated temperatures accelerate respiration and transpiration, eroding the photosynthetic gains. Altitude worked in the opposite direction, with compensation strengthening at higher elevations, peaking above 4,500 meters for carbon storage.

The implications reach well beyond Chinese borders. As countries worldwide pursue air-quality improvement and carbon neutrality, the study suggests that cleaner skies will progressively remove a diffuse-radiation bonus that has quietly supported terrestrial carbon uptake, and that the trajectory of vegetation productivity will increasingly be governed by temperature, precipitation, and drought rather than by aerosol dynamics. The authors note that frequent extreme droughts in recent years and phenotypic adjustments of plants to historically low-light conditions—such as increased specific leaf area and lowered light saturation points—may partly explain the weakening compensation. Interaction analyses confirmed that precipitation, temperature, elevation, vegetation type, and policy period all significantly modified the statistical relationship between AOD and gross primary productivity. The researchers acknowledge limitations inherent in remote-sensing-based statistical association, including scale-matching errors among resampled datasets, the difficulty of separating aerosol effects from rising carbon dioxide and nitrogen deposition, and spatial autocorrelation among neighboring pixels. Future work, they argue, should combine flux-tower observations, process-based ecosystem models, and radiation measurements to disentangle the direct and diffuse pathways quantitatively. For now, the study offers a sobering counterpoint to the celebration of cleaner air: the atmosphere’s murk was never purely a burden, and China’s ecosystems are already adjusting to a world with less of it.

Subject of Research: Asymmetric responses of vegetation productivity to atmospheric aerosol optical depth changes across China from 2002 to 2020

Article Title: Diminishing aerosol-related compensation of vegetation productivity across China: asymmetric responses of AGC, LAI, NPP, and GPP from 2002 to 2020

Article References: Dong, R., Duan, X., Yin, C., Zhang, M., Jin, W., & Zhang, Y. (2026). Diminishing aerosol-related compensation of vegetation productivity across China: asymmetric responses of AGC, LAI, NPP, and GPP from 2002 to 2020. Frontiers in Remote Sensing, 7, Article 1941508. https://doi.org/10.3389/frsen.2026.1941508

Image Credits: AI Generated

DOI: 10.3389/frsen.2026.1941508

Keywords: aerosol optical depth, vegetation productivity, asymmetry index, China, gross primary productivity, diffuse radiation, air pollution control, remote sensing, carbon storage, grasslands, forests, climate change