The research team, led by Anees Akhtar and Chunsheng Fang of Jilin University together with colleagues, set out to fill a stubborn gap in air quality science. Most of what is known about China’s ozone and PM2.5 problems comes from national assessments, regional campaigns, or studies of megacities such as Beijing and Shanghai. Medium-sized industrial cities in Northeast China, despite hosting dense clusters of chemical and energy infrastructure, have remained poorly resolved in the scientific record. Jilin City offered an ideal natural experiment: the COVID-19 pandemic suppressed emissions in 2020, creating a low-emission baseline, while 2023 represented the post-pandemic recovery, when industrial activity and traffic had largely returned.

To compare the two years, the researchers combined surface observations from monitoring stations with high-resolution atmospheric modeling. They ran the Weather Research and Forecasting model coupled with the Community Multiscale Air Quality modeling system, known as WRF-CMAQ, at a fine 3-kilometer grid spacing, allowing pollution patterns to be resolved at the level of individual districts. The team then applied correlation analysis, multiple linear regression, and a technique called meteorological normalization, which statistically removes the influence of weather so that underlying changes in emissions can be glimpsed. Their focus was the May-to-July window, the warm season when photochemical ozone production peaks in this part of Northeast China.

The model evaluation itself carried an important caution. Simulations performed acceptably for meteorological variables and for PM2.5, but systematically underestimated ozone concentrations. That bias means the absolute magnitude of simulated ozone should be treated with uncertainty, even though the model’s spatial and temporal patterns remain informative. It is a reminder that photochemical models, which must track hundreds of reactions between nitrogen oxides, volatile organic compounds, and sunlight, still struggle to reproduce ozone faithfully in complex terrain, where valley winds, boundary layer dynamics, and local emission plumes interact in ways that coarse grids and simplified chemistry can miss.

The observational record told a story of divergence. Ozone concentrations rose in May and June of 2023 compared with 2020, then dipped slightly in July, with persistent hotspots concentrated in the district of Panshi. PM2.5, by contrast, increased in May, changed little in June, and fell markedly in July, while remaining relatively elevated in the districts of Huadian, Chuanying, and Jiaohe. District-level anomaly and decomposition analyses revealed a uniform increase in May ozone across the city, heterogeneous responses in June, and clearer pollutant-specific patterns separating urban cores from periurban areas. In other words, the two pollutants did not simply rise and fall together; they traced distinct spatial and seasonal fingerprints.

The most striking technical result came from the regression analysis. Meteorological predictors, dominated by air temperature, explained roughly 35 percent of the variability in observed ozone, a substantial fraction that underscores how sensitive photochemistry is to heat and sunlight. For PM2.5, however, the same suite of meteorological variables accounted for less than 1 percent of the variability. That near-total failure of weather to explain particulate levels implies that PM2.5 in Jilin City is governed by other forces, most plausibly emissions, secondary aerosol chemistry, and transport processes not captured by the simple predictor set. The authors are careful to note that driver interpretation for PM2.5 should therefore remain qualitative rather than quantitative.

Meteorological normalization sharpened the picture further. After statistically controlling for weather differences between the two years, the analysis suggested that the May ozone enhancement was largely meteorologically associated, meaning warmer and more photochemically favorable conditions could account for much of the rise. The June ozone enhancement, however, persisted as an unexplained residual, hinting at changes in emissions or chemistry between the pandemic baseline and the recovery period. Together with the weak meteorological control on PM2.5, this points to a genuine decoupling of the two pollutants: ozone in Jilin City dances to the rhythm of temperature and sunlight, while fine particles follow a different, less weather-dependent beat.

The authors are admirably explicit about the limits of their design. Because meteorology and emissions changed simultaneously between the 2020 and 2023 simulation years, the observed signals reflect statistical associations rather than controlled causal attribution. The COVID-19 period was not a clean experiment; weather varied, emission reductions were uneven across sectors, and the recovery in 2023 restored activity at different rates in different industries. What the study can legitimately claim is a pattern that is statistically consistent with decoupled mechanisms, not a proof that specific emission changes caused specific pollution responses. That honesty is rare and valuable in a field where pandemic-era comparisons are often overinterpreted.

The broader implications reach into policy. China’s clean air campaigns have achieved dramatic reductions in PM2.5 over the past decade, yet ozone has proven far more stubborn, and in many regions has worsened as nitrogen oxide reductions altered the photochemical balance. A well-known two-pollutant strategy published in Nature Geoscience argued that controlling ozone and particulates requires coordinated cuts in both nitrogen oxides and volatile organic compounds, tuned to local chemical regimes. The Jilin City findings add a Northeast China, medium-sized-city perspective to that debate, showing that in such settings ozone’s meteorological dominance and PM2.5’s apparent emission sensitivity demand different diagnostic tools and possibly different control levers.

There is also a health dimension that gives the work urgency. Both ozone and PM2.5 are linked to respiratory and cardiovascular harm, and their combined toxicity, including potential synergistic mechanisms, is an active research frontier. Co-occurring extremes of the two pollutants have been documented with increasing frequency in the western United States and southern China, and the Jilin study’s demonstration of decoupled behavior suggests that compound pollution episodes there may arise through distinct pathways rather than a single shared driver. For a city ringed by petrochemical facilities and cradled by terrain that can trap stagnant air, understanding which pollutant responds to which lever is not an academic exercise. It is the difference between effective intervention and wasted effort. As climate change continues to raise summer temperatures across Northeast Asia, the temperature-driven component of ozone risk is likely to grow, making studies like this one, grounded in careful local observation and honest uncertainty, essential templates for the many industrial cities that science has yet to resolve.

Subject of Research: Spatiotemporal variability and meteorological sensitivity of ozone and PM2.5 pollution in Jilin City, China

Article Title: Spatiotemporal characteristics of ozone and PM2.5 pollution in Jilin City: insights from observations in 2020 and 2023

Article References: Akhtar, A., Fang, C., Wang, J., Abbasi, A., & Basharat, U. (2026). Spatiotemporal characteristics of ozone and PM2.5 pollution in Jilin City: insights from observations in 2020 and 2023. Environmental Monitoring and Assessment, 198(11), Article 1138. https://doi.org/10.1007/s10661-026-15929-3

Image Credits: AI Generated

DOI: 10.1007/s10661-026-15929-3

Keywords: ozone, PM2.5, air quality, WRF-CMAQ, meteorological normalization, Jilin City, Northeast China, COVID-19 emissions, temperature sensitivity, multiple linear regression, petrochemical emissions, spatiotemporal analysis