Reveal Global Seasonality and Climate Controls of Air Pollution

A research team led by Professor Faming Wang at the South China Botanical Garden, Chinese Academy of Sciences, has characterized the seasonal patterns, spatial regimes and recent directional trends of global atmospheric pollution by integrating satellite observations with meteorological reanalysis.

The study, entitled “Characterizing global air pollution seasonality and trends through integrated satellite and meteorological analytics,” was published in Environmental Pollution.

Air pollution remains a major challenge to public health, ecosystem integrity and sustainable development. Nitrogen dioxide (NO₂), ozone (O₃), carbon monoxide (CO), sulfur dioxide (SO₂) and formaldehyde (HCHO) arise from diverse sources, and their formation, transport and removal are jointly regulated by anthropogenic emissions and meteorological conditions.

Previous assessments have often focused on individual pollutants or restricted regions, limiting a unified understanding of global multi-pollutant seasonality, spatial coherence and climate–pollution interactions.

To address this gap, the researchers integrated 72 months of Sentinel-5P TROPOMI observations and ERA5-Land meteorological data covering 2019–2024. Five major pollutants were assessed together with temperature, precipitation, relative humidity and wind speed.

One-way analysis of variance, the Seasonal Mann–Kendall test, Sen’s slope estimation, pixel-wise correlation analysis and K-means clustering were combined to identify seasonal differences, emerging trends, coherent pollution regimes and pollutant–meteorology relationships within a single analytical framework.

Figure. Conceptual framework linking environmental drivers, atmospheric processes, analytical methods, health and ecological impacts, and policy implications.（Image by WANG Faming）

Distinct seasonal rhythms shape global atmospheric pollution

All five pollutants exhibited pronounced seasonal variability. Wintertime NO₂ increased by as much as 40% across South and East Asia, reflecting the combined effects of greater energy demand, shallow atmospheric boundary layers and weaker pollutant dispersion.

By contrast, summertime O₃ increased by up to 35% across the mid-latitudes, demonstrating the importance of high temperature and strong solar radiation in accelerating photochemical ozone production.

Biomass-burning regions in equatorial Africa, the Amazon Basin and other parts of South America experienced CO peaks exceeding background levels by more than two-fold. HCHO hotspots closely followed tropical regions dominated by biogenic emissions and biomass burning.

High NO₂ and SO₂ levels were mainly concentrated in urban, industrial and energy-producing regions of East Asia, South Asia, Europe and North America, reflecting their close association with fossil-fuel combustion, transportation, power generation and industrial activity.

Climate warming may amplify photochemical pollution

Seasonal Mann–Kendall analyses identified significant emerging increases in global NO₂, HCHO, O₃ and temperature during 2019–2024, with HCHO showing the strongest upward signal. CO and SO₂ displayed regionally mixed responses and no statistically uniform global trend.

Temperature was positively correlated with O₃ and NO₂, with global correlation coefficients of 0.52 and 0.68, respectively. These relationships suggest that warming can enhance precursor reactivity, photochemical production and boundary-layer effects, thereby increasing ozone and related pollution risks in susceptible regions.

Precipitation, relative humidity and wind generally suppressed the accumulation of primary pollutants. Rainfall promotes wet scavenging, while stronger winds enhance atmospheric ventilation, dispersion and dilution.

SO₂ was strongly negatively related to temperature and precipitation, illustrating how removal processes, atmospheric oxidation and changes in atmospheric residence time regulate its spatial distribution.

The findings therefore demonstrate that pollution risks are shaped not only by emission intensity but also by climate and weather conditions. Continued warming may intensify photochemical pollution, while changes in rainfall, humidity and wind can alter the transport, accumulation and removal of primary pollutants.

Pandemic disruption and emission controls produced detectable regional signals

The analysis captured the short-term effects of COVID-19 restrictions in 2020. During the main lockdown months, NO₂ column densities decreased by approximately 20–30% across large parts of Western Europe, 15–25% over eastern North America and 15–35% across parts of South and East Asia.

Local reductions exceeded 40% in several major urban and industrial areas. Subsequent increases in NO₂ across many regions reflected the rapid atmospheric response to the recovery of transportation, industrial production and broader socioeconomic activity.

These observations demonstrate that satellite monitoring can identify both continuing structural changes in emissions and short-lived atmospheric disruptions associated with major socioeconomic events.

At the same time, statistically significant declines in NO₂ and SO₂ were detected across Europe and parts of East Asia, consistent with the effects of emission-control policies targeting power generation, industrial facilities and transportation.

These patterns demonstrate that targeted regulation can generate measurable air-quality benefits. However, differences in energy structure, development stage, emission sources and climate mean that no single mitigation strategy will be appropriate for all regions. Air-quality management must therefore be both region-specific and seasonally adaptive.

A transferable framework for climate-aware air-quality management

By combining multi-pollutant satellite observations, meteorological reanalysis and statistical clustering, the study provides a reproducible and transferable framework for diagnosing global atmospheric pollution.

The approach can support seasonal forecasting, early-warning assessments, pollution-hotspot identification and targeted emission mitigation, particularly in regions where ground-based monitoring networks remain sparse.

The authors emphasized that the six-year record is best interpreted as evidence of recent directional change rather than a fully established long-term climatological trend. In addition, satellite column retrievals cannot be directly equated with near-surface human exposure.

Future research should integrate ground observations, regional field campaigns, emission inventories and atmospheric chemical transport models, extend the observational record, and develop satellite–ground–model data-fusion and nonlinear machine-learning approaches to reduce uncertainty.

Dr. Muhammad Naveed, a postdoctoral researcher at the South China Botanical Garden, is the first author. Professor WANG Faming is the corresponding author and contributed to the conceptualization, supervision, investigation, validation, review and editing, and funding acquisition.The study was conducted with collaborators from Central South University, Government College University Faisalabad in Pakistan, Mary Immaculate College at the University of Limerick in Ireland, and Guangdong Eco-engineering Polytechnic.

The research was supported by the Alliance of National and International Science Organizations for the Belt and Road Regions, the National Key Research and Development Program of China, the National Natural Science Foundation of China, the Guangdong Basic and Applied Basic Research Foundation, the Key-Area Research and Development Program of Guangdong Province, the Southern Marine Science and Engineering Guangdong Laboratory (Zhuhai), the CAS Youth Innovation Promotion Association, and the Guangdong Provincial Key Laboratory of Applied Botany.Article link:DOI: 10.1016/j.envpol.2026.127728