Hidden Acid in City Air Emerges as Key Driver of Ozone and Smog Trends

In the crowded chemistry of urban air, one molecule has long been treated as a supporting actor. Nitrous acid, or HONO, is fleeting, reactive, and notoriously difficult to measure, so most air-quality models have relegated it to a minor role. A new study published in Nature Communications argues that this understated compound deserves top billing. By combining two decades of field measurements with a deep neural network trained on campaigns spanning the globe, researchers led by Xuan Li, Can Ye, Keding Lu and Yuanhang Zhang of Peking University have reconstructed how surface HONO concentrations evolved in major cities between 2005 and 2021, and they show that these invisible shifts have been quietly steering the trajectories of both ozone pollution and fine particulate matter.

The importance of HONO lies in what it does at sunrise. When sunlight strikes the molecule, it breaks apart to release hydroxyl radicals, the highly reactive fragments known collectively as HOx that initiate nearly all daytime atmospheric oxidation. Hydroxyl radicals attack volatile organic compounds and nitrogen oxides, setting off the chain reactions that manufacture ground-level ozone. They also oxidize sulfur dioxide and nitrogen dioxide into the acids and nitrates that condense onto particles, swelling the mass of secondary PM2.5, the fine aerosol most strongly linked to cardiovascular and respiratory disease. In other words, HONO is not merely another pollutant; it is the ignition switch for the photochemical engine that produces the smog cities breathe.

Despite this pivotal chemistry, HONO has been a blind spot in long-term air-quality assessment. Ground networks routinely monitor ozone, nitrogen dioxide and particulate matter, but HONO is rarely measured continuously, and its sources remain contested. It forms directly in combustion exhaust, heterogeneously on wet surfaces from nitrogen dioxide, through photolysis of nitric acid deposited on surfaces, and via reactions of NO2 with water vapor enhanced by sunlight. Because each pathway responds differently to emission controls, the net HONO burden of a city can move in directions that conventional pollution metrics do not predict. The new work tackles this problem by training a neural network on the scattered but rich record of worldwide field campaigns, allowing the team to reconstruct consistent, city-scale HONO histories where direct measurements are absent.

The reconstructed trends are striking for how sharply they diverge. Beijing, the poster child of aggressive clean-air policy, saw its surface HONO fall by 47.4 percent between 2013 and 2021, coinciding with sweeping controls on coal combustion, vehicle emissions and industrial output. Delhi moved in the opposite direction, with HONO rising 10.0 percent from 2005 to 2018 as its vehicle fleet and biomass burning expanded faster than regulation could contain. London, New York and Los Angeles, which had already cleaned much of their air in earlier decades, recorded slight decreases over the study period. These contrasting trajectories gave the researchers a natural experiment: cities whose HONO burdens changed at very different rates and for very different reasons.

What the team found when they traced the consequences is the study’s central revelation. In both Beijing and Delhi, changes in HONO emerged as a major driver of variations in ozone and particulate nitrate, outweighing the contributions of conventional nitrogen oxide and volatile organic compound abatement. In the other megacities examined, HONO’s influence was comparable in magnitude to that of the traditional precursors. This reorders the hierarchy of urban pollution chemistry. For two decades, ozone and secondary aerosol trends have been interpreted almost entirely through the lens of NOx and VOC emissions, with famously counterintuitive results: cutting NOx sometimes raises ozone because of the nonlinear chemistry of radical cycling. HONO, by controlling the radical pool itself, sits upstream of those nonlinearities and can therefore dominate the outcome.

The mechanistic logic is worth unpacking. Hydroxyl radical production in the morning boundary layer depends critically on HONO photolysis, because HONO accumulates overnight on surfaces and in shallow air layers and then releases radicals precisely when the atmosphere begins to mix. A city whose HONO source strength falls, as Beijing’s did, starts each day with a smaller radical reservoir, slowing ozone formation and reducing the oxidation of nitrogen dioxide to nitric acid, which ultimately becomes particulate nitrate in PM2.5. A city whose HONO rises, like Delhi, experiences the reverse: amplified radical production that accelerates both photochemical ozone production and secondary aerosol formation, even if primary emissions of the precursors themselves are not rising in lockstep. The radical budget, in effect, transmits and amplifies the signal of surface chemistry changes into the pollutants that health authorities measure.

To convert this understanding into policy guidance, the researchers built a fused data-driven framework that links HONO formation regimes to the most effective control strategies. The analysis reveals that the optimal lever depends strongly on season. In summer, when photochemistry is intense and radical cycling is rapid, prioritizing NOx reductions is the most effective way to curb HONO formation and, with it, ozone production. In winter, the picture changes: low temperatures, weak sunlight and stable boundary layers favor heterogeneous HONO production on particle and ground surfaces, so the framework points instead toward co-reduction of NOx, PM2.5 and direct combustion emissions together. Suppressing the particulate surfaces and the fresh combustion exhaust that feed winter HONO production, the study suggests, is the faster route to joint ozone and PM2.5 mitigation during the cold months when haze episodes are most severe.

The seasonal split carries a practical warning for cities designing control programs. A strategy tuned for one regime can underperform or even backfire in another. Summer-focused NOx cuts that look successful in July may leave winter HONO largely untouched, because the heterogeneous pathways that dominate cold-season production respond more to particle loading and direct combustion sources than to ambient NO2 alone. Conversely, combustion controls aimed at winter haze will also trim the direct HONO emissions that accompany incomplete burning, delivering benefits that compound across seasons. The regime-dependent framework formalizes this intuition into a quantitative tool, allowing regulators to ask, for a given city and season, which combination of precursor reductions yields the largest joint decrease in ozone and secondary aerosol.

The Beijing and Delhi case studies illustrate both ends of the spectrum of what is achievable. Beijing’s 47.4 percent HONO decline over eight years demonstrates that determined, multi-sector emission control can rapidly shrink the radical engine of photochemical pollution, and that the resulting ozone and nitrate benefits are real and measurable. Delhi’s 10.0 percent rise over thirteen years shows the cost of allowing combustion growth to outpace controls: the city’s HONO-driven radical production has been pushing ozone and secondary particles upward even as other factors fluctuate. For the mature cities of Europe and North America, where HONO drifted only slightly downward, the study implies that remaining HONO sources, particularly those tied to surface chemistry and residual combustion, may now be among the more stubborn obstacles to further ozone improvement.

For a field that has spent decades building policy around NOx and VOC control, the message is a call to widen the lens. Long-term HONO evolution, the authors conclude, modulates ozone and secondary aerosol variations across global megacities, and any credible pathway to simultaneous ozone and PM2.5 reduction must account for it. The deep learning reconstruction offers a template: rather than waiting for sparse, campaign-based measurements to accumulate, models can now be trained on the existing global record to fill the gaps and track this critical species where monitoring networks do not reach. As cities from Beijing to Delhi to Los Angeles confront the stubborn tail of air pollution, the molecule that ignites each day’s photochemistry may prove to be the lever that finally brings both smogs, the gaseous and the particulate, under control at once.

Subject of Research: Long-term urban nitrous acid evolution and its control of ozone and secondary aerosol formation in global megacities

Article Title: Long-term nitrous acid evolution modulates ozone and secondary aerosol variations in global megacities

Article References: Li, X., Ye, C., Dong, H., Xu, X., Wang, H., Chen, X., Tan, Z., Ma, X., Chen, S., Lu, K., & Zhang, Y. (2026). Long-term nitrous acid evolution modulates ozone and secondary aerosol variations in global megacities. Nature Communications. https://doi.org/10.1038/s41467-026-78349-4

Image Credits: AI Generated

DOI: 10.1038/s41467-026-78349-4

Keywords: nitrous acid, HONO, ozone, PM2.5, secondary aerosol, hydroxyl radicals, air quality, megacities, Beijing, Delhi, NOx control, atmospheric chemistry