The research, conducted by Sachin Budakoti of Lund University’s Department of Physical Geography and Ecosystem Science, tackled a long-standing question in atmospheric chemistry: is ozone formation in a given region limited by nitrogen oxides (NOx) or by volatile organic compounds (VOCs)? The answer matters enormously for policy, because cutting the wrong pollutant can actually make ozone pollution worse. To resolve this, the study used a diagnostic known as the formaldehyde-to-nitrogen dioxide ratio, or FNR, which compares the abundance of formaldehyde (HCHO), a by-product of VOC oxidation, with nitrogen dioxide (NO2), a key component of NOx. This ratio acts as a chemical fingerprint of the ozone production regime.

The underlying chemistry is deceptively simple yet notoriously non-linear. Ozone is not emitted directly; it forms when sunlight drives reactions between NOx and VOCs. In a VOC-limited regime, ozone production is saturated with NOx, so the limiting ingredient is the supply of VOC-derived radicals. In a NOx-limited regime, the opposite holds: VOCs are plentiful and NOx availability controls ozone formation. Between these extremes lies a transitional regime where both matter. The FNR captures this balance because it reflects the relative abundance of VOC-derived radicals versus NOx. Ratios below 1 indicate VOC-limited conditions, ratios above 2 indicate NOx-limited conditions, and values between 1 and 2 signal a transitional regime.

Budakoti drew on the Copernicus Atmosphere Monitoring Service (CAMS) global reanalysis, which combines satellite observations with chemical transport modeling to produce continuous, spatially consistent fields of ozone, HCHO, and NO2 at a resolution of 0.75 by 0.75 degrees. To validate the dataset, the CAMS ozone values were compared against AIRS satellite observations over Sweden, yielding a correlation coefficient of 0.85, a root mean square error of 3.45 ppbv, and an index of agreement of 0.66, indicating reasonable agreement. Meteorological variables, including surface temperature, relative humidity, and wind speed, came from the ERA5 reanalysis produced by the European Centre for Medium-Range Weather Forecasts at finer 0.25-degree resolution.

Long-term trends were assessed with the Mann-Kendall test, a non-parametric statistical method applied independently at each grid cell across Sweden using monthly mean concentrations. The results revealed a striking spatial pattern. Northern Sweden showed a statistically significant ozone increase greater than 0.3 ppbv per year at the 95 percent confidence level, while trends in NO2, which decreased, and HCHO, which increased, over the same region fell short of statistical significance. Seasonally, ozone rose during spring and summer months, driven by enhanced photochemical activity under longer daylight hours and warmer temperatures, as well as an earlier onset of the ozone season linked to changes in snow cover and vegetation phenology. During autumn and winter, ozone declined as reduced solar radiation, weaker photochemical production, and more frequent atmospheric inversions limited formation and trapped pollutants near the surface.

Correlation analysis added further depth to the picture. Ozone displayed a strong and statistically significant negative association with NO2 across Sweden, with correlation coefficients below -0.4, reflecting the well-known titration effect in which nitric oxide destroys ozone by converting it to nitrogen dioxide. In contrast, ozone showed a moderate positive correlation with HCHO, between 0.35 and 0.5, and a strong positive correlation with the FNR ratio, exceeding 0.6. Together, these relationships indicate that higher formaldehyde and lower nitrogen dioxide levels combine to produce elevated ozone across the country, a pattern shaped by both anthropogenic emissions and the biogenic VOC flux from Sweden’s extensive forests.

The meteorological analysis underscored the pivotal role of temperature. Surface ozone exhibited a strong, statistically significant positive association with temperature across central, northern, and southern Sweden, while associations with relative humidity were weaker and not statistically significant, and wind speed showed no consistent relationship. High temperatures and strong sunlight accelerate photochemical reaction rates and boost biogenic VOC emissions, creating a feedback that climate warming is likely to amplify. Both temperature and wind speed themselves showed statistically significant increasing trends over the study period, a signal consistent with the pronounced warming Sweden has experienced in recent decades in line with broader Arctic climate trends.

The core finding of the sensitivity analysis is unambiguous: FNR values below 1 dominate across the urbanized southern and central parts of Sweden, including cities such as Stockholm and Gothenburg, indicating that ozone production there is VOC-limited, meaning the system is saturated with NOx from traffic, shipping, and industry. Only the rural areas of the northern Norland region, where FNR values fall between 1 and 2, occupy a transitional regime shifting from VOC-limited toward NOx-limited conditions. An ozone isopleth analysis reinforced this conclusion, showing that the highest ozone concentrations occurred under conditions of rising HCHO and falling NO2, the hallmark of VOC-limited chemistry. In such an environment, increasing VOCs supplies more peroxy radicals that convert nitric oxide to nitrogen dioxide and sustain ozone formation, while decreasing NO2 simultaneously weakens the titration effect that would otherwise destroy ozone, creating an optimal chemical environment for ozone accumulation.

These findings carry substantial implications for air quality management. In VOC-limited urban environments, reducing NOx emissions alone could initially increase ozone concentrations by removing the titration effect, a counterintuitive outcome that has complicated mitigation efforts in many cities worldwide. The study therefore argues that VOC-focused strategies are essential for reducing ground-level ozone across most of Sweden’s urban areas. The results also align with earlier Scandinavian and European assessments, which have reported that ozone levels in European cities are rising even as peak values decline, and that ozone in Sweden is shaped substantially by long-range transport from continental Europe rather than by local emissions alone. Sweden, despite not being a major emitter of ozone precursors, is increasingly affected by imported pollution and by climate-driven increases in biogenic VOC emissions that can shift ozone formation regimes even where anthropogenic NOx is low.

The study is the first of its kind to apply FNR-based sensitivity diagnostics to the Swedish region over a two-decade span, and it supports the growing evidence that region-specific thresholds are more appropriate than universal cutoffs when diagnosing ozone chemistry. Because the analysis used surface-level concentrations rather than satellite column measurements, the threshold magnitudes differ from those derived from space-based studies, since HCHO has a substantial free-tropospheric contribution while NO2 is strongly surface-confined. The author cautions that reanalysis data smooth out local emission gradients, particularly in urban areas, and that HCHO fields at high latitudes carry known uncertainties, so the results should be read as indicating broad spatial regimes rather than precise local chemical controls. Future work integrating in situ observations and higher-resolution chemical transport modeling will help refine the diagnosis, but the message is already clear: as Sweden warms and its forests exhale more reactive carbon, the fight against ground-level ozone will increasingly be a fight against VOCs.

Subject of Research: Long-term surface ozone trends and NOx-VOC sensitivity regimes in Sweden using the HCHO/NO2 ratio and CAMS reanalysis data

Article Title: Identification of surface ozone sensitivity for NO₂ and secondary HCHO in Sweden

Article References: Identification of surface ozone sensitivity for NO₂ and secondary HCHO in Sweden. (n.d.). https://doi.org/10.1186/s44329-026-00047-9

Image Credits: AI Generated

DOI: 10.1186/s44329-026-00047-9

Keywords: surface ozone, Sweden, CAMS reanalysis, HCHO/NO2 ratio, VOC-limited regime, NOx-limited regime, Mann-Kendall test, biogenic emissions, air quality, ozone pollution, climate warming, ERA5 reanalysis