Abstract. While free-tropospheric ozone (O3) over western North America (WNA) has increased since the mid-1990s, research has primarily focused on the mean. We investigate the lower tail (O3 < 33rd percentile) to characterize the evolving remote background state. Because these air masses are minimally affected by episodic extremes, they offer a clearer window into long-term shifts in background O3, transport, and photochemistry. Using FLEXPART-ERA5 source–receptor relationships (SRRs) from 1994 to 2021, we analyze the transport history of air masses reaching WNA (25–55° N, 130–90° W). Despite no robust SRR trends within the lower and mid-troposphere (0-8 km), changing emission patterns suggest an intensifying remote influence. Specifically, WNA's surface NOX emissions have decreased while lower-tail O3 continues to rise, aligning with increasing surface emissions from Southeast Asia and intensified shipping. In contrast, UTLS (8–13 km) SRRs show a clear increase, indicating growing influence from high-altitude sources, including enhanced transport from Southeast Asia and the tropical Pacific, and rising global aircraft emissions. GMI chemical simulations corroborate these findings, revealing that net O3 production over Southeast Asia increased by 157 % in the lower troposphere and 7 % in the free troposphere between 2007 and 2019. The rise in WNA's low O3 percentiles is driven by the combined influence of intensified transport from Southeast Asia and the tropical Pacific, along with increasing global aircraft and shipping emissions. Ultimately, these trends reflect both the rapid growth of Southeast Asia emissions and shifting trans-Pacific transport.

The operation of geostationary (GEO) instruments such as the Tropospheric Emissions: Monitoring of Pollution (TEMPO) provides unprecedented hourly nitrogen dioxide (NO2) observations compared to the once-daily data from a low-Earth orbit (LEO) platform like the TROPOspheric Monitoring Instrument (TROPOMI). This study investigates the performance and challenges of using TEMPO versus TROPOMI measurements to constrain anthropogenic nitrogen oxides (NOx) emissions. The accuracy of TEMPO and TROPOMI NO2 tropospheric columns are assessed using Pandora observations, finding a low bias of 9%-12.3% in TEMPO, and TROPOMI data during August 2023, while TEMPO midday and late afternoon observations are less of low bias. Top-down NOx emissions derived by midday TEMPO and TROPOMI data are generally consistent over urban areas, being 5%-20% lower than bottom-up emissions provided by the 2021 GReenhouse gas And Air Pollutants Emissions System (GRA2PES), and align with 2023 GRA2PES emissions, demonstrating the reliability of using satellite data for timely updates of bottom-up inventories. However, assimilating additional morning/late afternoon TEMPO data leads to the poorest top-down NOx emissions, likely resulting from larger negative measurement biases. NOx emission inversions effectively mitigate NOx overprediction, though the top-down NOx emissions might be over-corrected in urban cores. NOx emissions optimization also improves ozone forecasts by reducing the model's positive biases, especially when assimilating midday TEMPO data. Our study suggests that TEMPO midday observations provide better constraints on the magnitude and spatiotemporal variation of anthropogenic NOx emissions than TROPOMI, while morning TEMPO v3 data should be used cautiously due to potential negative impact on NOx emissions inversion.

Wildfires are the largest terrestrial source of atmospheric ammonia (NH3), yet their impacts on NH3 concentrations and ammonium (NH4 +) deposition remain poorly quantified. In this study, we evaluate the effects of the record-breaking 2023 Canadian wildfire season on NH3 concentration and NH4 + deposition across the Upper Midwest. This study integrates satellite observations, ground-based data, and in situ aircraft measurements. In May-June 2023, NH3 concentrations increased at 83% of ground sites, and NH4 + deposition flux rose at 100% of ground sites in the Upper Midwest. Satellite data showed significantly higher column-averaged NH3 in 47% of grid cells in the Upper Midwest. On 1 August, a smoke plume over the Midwest corresponded with an AEROMMA flight observing enhanced NH3, NH4 +, carbon monoxide, and acetonitrile. These findings highlight the substantial impact of wildfire smoke on NH3 and NH4 + at regional scales, with implications for nitrogen cycling, air quality, and atmospheric modeling.

Long-term atmospheric ozone observations in Western North America (WNA) provide essential data for assessing tropospheric ozone trends. Backward atmospheric simulations based on these observations establish the source–receptor relationships (SRRs) to improve our understanding of the factors driving ozone trends across different regions, time periods, and atmospheric layers. In this study, we integrated 28 years of ozone observations (1994–2021) from ozonesondes, lidar, commercial aircraft, and aircraft campaigns across WNA, spanning the upper atmospheric boundary layer, free troposphere, and upper troposphere (i.e., 900 to 300 hPa). We integrated the multiplatform datasets using a data fusion framework to generate 553 608 gridded ozone receptors. For each receptor, we use the FLEXible PARTicle (FLEXPART) dispersion model, driven by ERA5 reanalysis data, to produce the SRRs calculations, providing global simulations at high temporal (hourly) and spatial (1°×1°) resolution from the surface up to 20 km a.g.l. (above ground level). This SRR database retains detailed information for each receptor, including the gridded ozone value product, which enables user to illustrate and identify source contributions to various subsets of ozone observations in the troposphere above WNA over nearly 3 decades at different vertical layers and temporal scales, such as diurnal, daily, seasonal, intra-annual, and decadal. More generally, the calculated SRRs are applicable to any study looking to evaluate origins of airmasses reaching WNA. As such, this database can support source contribution analyses for other atmospheric components observed over WNA, if other co-located observations have been made at the spatial and temporal scales defined for some or all of the gridded ozone receptors used here. The entire dataset is publicly available at https://doi.org/10.5067/ASDC/WNA-BackTraj (Cui et al., 2025).

Quantifying long-term free-tropospheric ozone trends is essential for understanding the impact of human activities and climate change on atmospheric chemistry. However, this is complicated by two key challenges: the differences among existing satellite-derived tropospheric ozone products, which are not yet fully understood or reconciled, and the limited temporal and spatial coverage of ground-based reference measurements. Here, we explore if a more consistent understanding of the geographical distribution of tropospheric ozone column (TrOC) trends can be obtained by focusing on regional trends from ground-based measurements. Regions were determined with a correlation analysis between modeled TrOCs at the site locations. For those regions, TrOC trends were estimated with quantile regression for the Trajectory-mapped Ozonesonde dataset for the Stratosphere and Troposphere (TOST) and with a linear mixed-effects modeling (LMM) approach to calculate synthesized trends from homogenized HEGIFTOM (Harmonization and Evaluation of Ground-based Instruments for Free-Tropospheric Ozone Measurements) individual site trends. For different periods (1990-2021/22, 1995-2021/22, 2000-2021/22), both approaches give increasing (partial) tropospheric ozone column amounts over almost all Asian regions (median confidence) and negative trends over Arctic regions (very high confidence). Trends over Europe and North America are mostly weakly positive (LMM) or negative (TOST). For both approaches, the 2000-2021/22 trends decreased in magnitude compared to 1995-2021/22 for most regions; and for all time periods and regions, the pre-COVID trends are larger than the post-COVID trends. Our results enable the validation of global satellite TrOC trends and assessment of the performance of atmospheric chemistry models to represent the distribution and variation of TrOC.

The North American component of the Geo-Ring for Air Quality, the Tropospheric Emissions: Monitoring of Pollution (TEMPO) instrument, began collecting measurements on August 2, 2023. Multiple airborne field-intensives were conducted over the US during this TEMPO first-light period, including the NOAA Atmospheric Emissions and Reactions Observed from Megacities to Marine Areas (AEROMMA) and Coastal Urban Plume Dynamics Study (CUPiDS) campaigns, coordinated with the NASA Synergistic TEMPO Air Quality Sciences (STAQS) campaign. The North American cities targeted included New York City, Los Angeles, Chicago, and Toronto. Here, we present an overview of the AEROMMA / CUPiDS collected summer 2023 datasets, relevant for calibration and validation activities of TEMPO, including for ozone (O3), nitrogen dioxide (NO2), formaldehyde (CH2O), sulfur dioxide (SO2), aerosol optical depth (AOD), and aerosol layer height (ALH). Ground-based lidar and airborne in-situ vertical profiling by the NASA DC-8 and NOAA Twin Otter aircraft are available for evaluating TEMPO Level 2 O3 profile products (tropospheric and 0-2 km column retrievals). Airborne measurements of NO2 (photolytic conversion of NO2 into NO followed by laser-induced fluorescence, cavity enhanced spectroscopy, and multi-axis differential optical absorption spectroscopy (MAX-DOAS)) are available for evaluating TEMPO Level 2 NO2 vertical column density products. Airborne measurements of formaldehyde and glyoxal (in-situ and MAX-DOAS remote sensing) can be evaluated similarly as other volatile organic compounds (VOCs). Lastly, a wide array of aircraft-based in-situ measurements of composition, size distribution, optical properties can be utilized to derive aerosol optical depth (AOD) and aerosol extinction profiles for evaluating TEMPO AOD and ALH products, along with TROPOMI and stereoscopic aerosol layer height products from GOES-16/18. Preliminary evaluation of TEMPO NO2 will be presented as an initial calibration / validation test case, employing best practices to facilitate direct comparisons between airborne data with TEMPO Level 2 observations.

Ozone changes in the upper troposphere–lower stratosphere (UTLS) resulting from dynamical and chemical processes strongly affect the atmosphere's radiative forcing. This study analyzed intra- and inter-hemispheric ozone differences in the UTLS within the 45–60° latitude band, distinguishing between years disrupted by sudden stratospheric warming (SSW) events from 2002 to 2022. We utilized measurements from IAGOS commercial aircraft (45–60° N), long-term ozonesonde records (45–60° S) and unique in situ measurements from the high altitude and long range (HALO) research aircraft deployed over the southernmost region of South America (45–60° S) in September–November 2019 as a part of the Southern Hemisphere Transport, Dynamics, and Chemistry (SouthTRAC) research campaign. The mission period enabled us to examine the impact of two Southern Hemisphere (SH) SSW events that developed during 2002 and 2019. To enhance the spatial coverage, we incorporated Copernicus Atmosphere Monitoring Service reanalysis data. Stratospheric air origin was assigned using relative humidity (<20 %) and carbon monoxide (<50 nmol mol−1) thresholds. Our results show that air masses of stratospheric origin had higher ozone abundances in the Northern Hemisphere (NH) UTLS than in the SH (between 300–200 hPa and 45–60° latitude): in high ozone depletion years in the stratospheric vortex, the SH ozone median (184 nmol mol−1) was only 54 % of that in the NH (341 nmol mol−1), while in low depletion years, SH ozone median (214 nmol mol−1) reached 57 % of the NH values (371 nmol mol−1). Notably, the SSW events (2002 and 2019) increased SH UTLS ozone by 24 % (43 nmol mol−1) compared to high depletion years, while in the NH, the increase was 9 % (31 nmol mol−1).

Machine learning (ML) is transforming atmospheric chemistry, offering powerful tools to address challenges in tropospheric ozone research, a critical area for climate resilience and public health. As in adjacent fields, ML approaches complement existing research by learning patterns from ever-increasing volumes of atmospheric and environmental data relevant to ozone. We highlight the rapid progress made in the field since Phase 1 of the Tropospheric Ozone Assessment Report (TOAR), focussing particularly on the most active areas of research, namely short-term ozone forecasting, emulation of atmospheric chemistry and the use of remote sensing for ozone estimation. This review provides a comprehensive synthesis of recent advancements, highlights critical challenges, and proposes actionable pathways to develop ML in ozone research. Further advances hinge on addressing domain-specific issues such as the dependence of ozone concentrations on several poorly observed precursor species, as well as making progress on generic ML challenges such as the definition of suitable benchmarks and developing robust, explainable models. Reaping the full potential of ML for ozone research and operational applications will require close collaborations across atmospheric chemistry, ML and computational science and vigilant pursuit of the rapid developments in adjacent fields.

We present a comprehensive regional analysis of trends and variability in daily maximum 8-hour average ozone across the contiguous United States over 1990-2023. At the first stage, we evaluate the trends based on various seasonal percentiles at all available monitoring sites. Secondly, the overall regional trends (Western and Eastern USA) are derived at various seasonal percentiles. Results show that consistent and strong negative trends can be found in the eastern USA at the 95th and 50th percentiles in spring, summer and fall since the early 2000s, while winter trends are increasing. The similar seasonal trends are found in the Western USA, but with weaker magnitudes of trends. Throughout the analysis implications of the correlations between heatwave frequency/intensity and ozone variability are discussed.

From May 20 through July 25, 2023, large boreal forest fire plumes were transported across the central and eastern United States of America. On many days the U.S. Environmental Protection Agency (EPA) air quality monitoring network detected co-located ozone and PM2.5 anomalies, which indicate a contribution from the smoke to surface ozone production. We apply a general additive mixed model (GAMM) to interpolate the observed ozone and PM2.5 observations onto daily maps of the continental United States. Comparison to a recent baseline period with below-average wildfire activity (2014, 2016, 2019) allows for the identification of smoke events that contributed to enhanced surface ozone levels.

This paper outlines a comprehensive trend assessment of surface ozone observations across the conterminous USA over 1990–2023. A change point detection algorithm is applied to evaluate seasonal trends at various percentiles. We found that highly consistent and robust negative trends in extreme values have occurred in spring, summer, and fall since the 2000s across the eastern USA. A less strong but similar picture is found in the western USA, while increasing winter trends are commonly observed in the Southwestern and Midwestern regions of the country. The impact of a potential climate penalty that might offset some of the improvement in the ozone extremes is also investigated based on various heat wave metrics. By comparing ozone threshold exceedances, we found that the exceedance probabilities during heat waves are higher than those under normal conditions; moreover, the differences have decreased over time, as the effectiveness of emission controls has led to a great reduction in ozone extremes under both heat wave and normal conditions. When the increasing heat wave trends are accounted for, we find evidence that decreases in exceedances during heat waves have likely halted at 20 %–40 % of sites, depending on heat wave definitions. By identifying monitoring sites with (1) reliably decreasing ozone exceedances and (2) reliably increasing co-occurrences of ozone exceedances and heat wave events, we can show that several sites in California have been impacted by the ozone climate penalty (1995–2022). These findings are limited by the availability of long-term continuous ozone records, which are sparsely distributed across the USA and typically less than 30 years in length.

The MIXv2 Asian emission inventory is developed under the framework of the Model Inter-Comparison Study for Asia (MICS-Asia) Phase IV and produced from a mosaic of up-to-date regional emission inventories. We estimated the emissions for anthropogenic and biomass burning sources covering 23 countries and regions in East, Southeast and South Asia and aggregated emissions to a uniform spatial and temporal resolution for seven sectors: power, industry, residential, transportation, agriculture, open biomass burning and shipping. Compared to MIXv1, we extended the dataset to 2010–2017, included emissions of open biomass burning and shipping, and provided model-ready emissions of SAPRC99, SAPRC07, and CB05. A series of unit-based point source information was incorporated covering power plants in China and India. A consistent speciation framework for non-methane volatile organic compounds (NMVOCs) was applied to develop emissions by three chemical mechanisms. The total Asian emissions for anthropogenic/open biomass sectors in 2017 are estimated as follows: 41.6/1.1 Tg NOx, 33.2/0.1 Tg SO2, 258.2/20.6 Tg CO, 61.8/8.2 Tg NMVOC, 28.3/0.3 Tg NH3, 24.0/2.6 Tg PM10, 16.7/2.0 Tg PM2.5, 2.7/0.1 Tg BC (black carbon), 5.3/0.9 Tg OC (organic carbon), and 18.0/0.4 Pg CO2. The contributions of India and Southeast Asia were emerging in Asia during 2010–2017, especially for SO2, NH3 and particulate matter. Gridded emissions at a spatial resolution of 0.1° with monthly variations are now publicly available. This updated long-term emission mosaic inventory is ready to facilitate air quality and climate model simulations, as well as policymaking and associated analyses.

High-quality long-term observational records are essential to ensure appropriate and reliable trend detection of tropospheric ozone. However, the necessity of maintaining high sampling frequency, in addition to continuity, is often under-appreciated. A common assumption is that, so long as long-term records (e.g., a span of a few decades) are available, (1) the estimated trends are accurate and precise, and (2) the impact of small-scale variability (e.g., weather) can be eliminated. In this study, we show that the undercoverage bias (e.g., a type of sampling error resulting from statistical inference based on sparse or insufficient samples, such as once-per-week sampling frequency) can persistently reduce the trend accuracy of free tropospheric ozone, even if multi-decadal time series are considered. We use over 40 years of nighttime ozone observations measured at Mauna Loa, Hawaii (representative of the lower free troposphere), to make this demonstration and quantify the bias in monthly means and trends under different sampling strategies. We also show that short-term meteorological variability remains a cause of an inflated long-term trend uncertainty. To improve the trend precision and accuracy due to sampling bias, two remedies are proposed: (1) a data variability attribution of colocated meteorological influence can efficiently reduce estimation uncertainty and moderately reduce the impact of sparse sampling, and (2) an adaptive sampling strategy based on anomaly detection enables us to greatly reduce the sampling bias and produce more accurate trends using fewer samples compared to an intense regular sampling strategy.

During summer 2023 Canada experienced its most intense wildfire season on record. Smoke plumes from these fires advected across the United States (U.S.) Upper Midwest, producing regional scale surface enhancements of PM2.5 and ozone, as recorded by the U.S. surface monitoring network. These events are notable because they occurred early in the fire season (May 15-June 30), and they produced the highest regional-scale surface ozone levels ever recorded across the northern tier of the U.S. during early (May-June) or late (July-August) summer. Specifically, the Upper Midwest 50th ozone percentile was greater than in any other year since 1995, when the ozone monitoring network had sufficient coverage to assess regional-scale ozone levels; the 90th percentile was the highest since 2002. Satellite and aircraft measurements demonstrate the availability of ozone precursors and ozone production within the smoke plumes.

Tropical tropospheric ozone (TTO) is important for the global radiation budget because the longwave radiative effect of tropospheric ozone is higher in the tropics than midlatitudes. In recent decades the TTO burden has increased, partly due to the ongoing shift of ozone precursor emissions from midlatitude regions toward the Equator. In this study, we assess the distribution and trends of TTO using ozone profiles measured by high-quality in situ instruments from the IAGOS (In-Service Aircraft for a Global Observing System) commercial aircraft, the SHADOZ (Southern Hemisphere ADditional OZonesondes) network, and the ATom (Atmospheric Tomographic Mission) aircraft campaign, as well as six satellite records reporting tropical tropospheric column ozone (TTCO): TROPOspheric Monitoring Instrument (TROPOMI), Ozone Monitoring Instrument (OMI), OMI/Microwave Limb Sounder (MLS), Ozone Mapping Profiler Suite (OMPS)/Modern-Era Retrospective analysis for Research and Applications version 2 (MERRA-2), Cross-track Infrared Sounder (CrIS), and Infrared Atmospheric Sounding Interferometer (IASI)/Global Ozone Monitoring Experiment 2 (GOME2). With greater availability of ozone profiles across the tropics we can now demonstrate that tropical India is among the most polluted regions (e.g., western Africa, tropical South Atlantic, Southeast Asia, Malaysia and Indonesia), with present-day 95th percentile ozone values reaching 80 nmol mol−1 in the lower free troposphere, comparable to midlatitude regions such as northeastern China and Korea. In situ observations show that TTO increased between 1994 and 2019, with the largest mid- and upper-tropospheric increases above India, Southeast Asia, and Malaysia and Indonesia (from 3.4 ± 0.8 to 6.8 ± 1.8 nmol mol−1 decade−1), reaching 11 ± 2.4 and 8 ± 0.8 nmol mol−1 decade−1 close to the surface (India and Malaysia–Indonesia, respectively). The longest continuous satellite records only span 2004–2019 but also show increasing ozone across the tropics when their full sampling is considered, with maximum trends over Southeast Asia of 2.31 ± 1.34 nmol mol−1 decade−1 (OMI) and 1.69 ± 0.89 nmol mol−1 decade−1 (OMI/MLS). In general, the sparsely sampled aircraft and ozonesonde records do not detect the 2004–2019 ozone increase, which could be due to the genuine trends on this timescale being masked by the additional uncertainty resulting from sparse sampling. The fact that the sign of the trends detected with satellite records changes above three IAGOS regions, when their sampling frequency is limited to that of the in situ observations, demonstrates the limitations of sparse in situ sampling strategies. This study exposes the need to maintain and develop high-frequency continuous observations (in situ and remote sensing) above the tropical Pacific Ocean, the Indian Ocean, western Africa, and South Asia in order to estimate accurate and precise ozone trends for these regions. In contrast, Southeast Asia and Malaysia–Indonesia are regions with such strong increases in ozone that the current in situ sampling frequency is adequate to detect the trends on a relatively short 15-year timescale.

&lt;p&gt;The first phase of the Tropospheric Ozone Assessment Report (TOAR-I), an activity of the International Global Atmospheric Chemistry Project (IGAC), provided the first comprehensive view of surface ozone&amp;#8217;s global distribution and trends, based on all available surface ozone observations. &amp;#160;TOAR-I focused on a present-day period of 2010-2014, and calculated trends for a range of periods, but primarily focused on the most recent years of 2000-2014, plus long-term trends from the 1970s/1980s through 2014. &amp;#160;Subsequent studies of ozone trends using data after the TOAR-I cut-off of 2014, have shown a wide range of trends, both positive and negative, at monitoring sites around the world. &amp;#160;To keep up with the rapid changes of ozone at urban, rural and remote locations this study provides current world-wide ozone trends using observations through 2021, archived in the newly updated TOAR-II Database of Surface Observations. &amp;#160;Focus is placed on two ozone metrics relevant to human health impacts: &amp;#160;1) the annual peak of the 6-month running mean of maximum daily 8-hour average ozone; this metric is used by Global Burden of Disease (GBD) to estimate mortality due to long-term ozone exposure; 2) the number of days per year that exceed 70 ppbv, based on the maximum daily 8-hour average ozone value; this value corresponds to the primary U.S. National Ambient Air Quality Standards for ozone and is relevant to short-term ozone exposure. &amp;#160;Global maps will indicate the regions of the world where the potential for ozone impacts on human health are greatest (and least), and will show regions where ozone air quality is either improving or degrading. &amp;#160;Despite our effort to use all available surface ozone observations, large data gaps exist across many regions of the world, especially in developing nations, and GBD maps generated by data fusion will be used to identify, and to estimate ozone levels in the data-poor regions.&amp;#160;&lt;/p&gt;

This study has produced an improved percentile and seasonal (median) trend estimate of free tropospheric ozone above western North America (WNA), through a data fusion of ozonesonde, lidar, commercial aircraft, and field campaign measurements. Our method combines heterogeneous data sets according to the consensus data characteristics and inherent uncertainty in order to produce our best fused product. In response to different data collection environments (in situ or ground-based), we investigate the ozone variability based on a wide range of percentiles, which is preferable for trend detection due to tropospheric ozone's high degree of heteroscedasticity (i.e., inconsistent trends and variability between different ozone percentiles). We then compare the ozone trends and variability above the California sub-domain to the full WNA region for better understanding of the correlations between different regional scales. In California, the 1995-2021 percentile (from the 5th to 95th) and seasonal trends are clearly positive in terms of high signal-to-noise ratios. The magnitude of the trends is generally weaker over WNA compared to California, but reliable positive trends can still be found between the 10th and 70th percentiles, as well as winter and summer, whereas autumn shows a negative trend over the same period. In addition, dozens of rural surface sites across the region are selected to represent the boundary layer variability. In contrast to increasing free tropospheric ozone, we find overall strong negative surface trends since 1995, with the greatest divergence found in summer. Throughout the analysis implications of the COVID-19 economic downturn on ozone variability are discussed. Plain Language Summary Free tropospheric ozone above western North America has increased since the mid-1990s. Despite an observed drop of ozone in 2020 due to the COVID-19 economic downturn, this observation-based study shows the overall free tropospheric ozone trends have not been offset and continued to increase over 1995-2021, mainly driven by strong positive trends in winter and summer. In combination with the strong negative trends observed at rural surface sites over the same period, this study adds to the growing body of evidence that surface trends are frequently disconnected from the general increases observed in the free troposphere.