Impacts of stratosphere-to-troposphere transport on tropospheric ozone in southeastern China: insights from ozonesonde observations

Tropospheric ozone pollution poses a major environmental challenge in China. As its primary natural source, Stratosphere-to-Troposphere Transport (STT) has been recognized as a significant contributor to tropospheric ozone in western, northeastern, and eastern China. However, the extent of STT’s influence on southeastern China has been less studied due to data limitations. Using a recently available one-year dataset of ozonesonde observations from a regional background station, we find that STT contributes significantly to tropospheric and surface ozone elevation in southeastern China. Our results show that STT plays a more substantial role in shaping tropospheric ozone during spring than previously believed, accounting for over 30% of ozone concentrations above 4 km. Without the stratospheric contribution, the spring seasonal peak almost disappears. STT can also significantly influence ozone concentrations at the surface. For example, a distinct ozone profile was observed on 4 May 2022, with a notable increase in tropospheric ozone. This tropospheric ozone increase was caused by a STT event triggered by a robust horizontal trough and subsequent southward movement of subtropical jets in the upper troposphere. According to a stratospheric tracer derived from an atmospheric chemistry model, this STT event contributed to 25%–30% of the surface ozone increase. Overall, this study highlights the important role of STT in driving tropospheric ozone variations, even in regions with comparatively lower ozone levels in southeastern China.


Introduction
Tropospheric ozone serves as a significant atmospheric pollutant, functioning as a potent oxidant that poses adverse effects on human health and vegetation growth when present in elevated concentrations (Van Dingenen et al. 2009, Rider andCarlsten 2019).Despite the implementation of the clean air plan in 2013 and its subsequent second phase initiated in 2018, which specifically targeted ozone pollution, tropospheric ozone levels in China continue to increase substantially (Gaudel et al 2018, Li et al 2020).This currently makes it one of the most serious environmental challenges in the country.Tropospheric ozone primarily results from photochemical reactions involving nitrogen oxides (NO x ) and volatile organic compounds under sunlight exposure (Wang et al 2017).Additionally, ozone can be transported from the lower stratosphere to the troposphere, a process termed as stratosphere-to-troposphere transport (STT).STT accounts for approximately 20% to 30% of mid-latitude tropospheric ozone in the Northern Hemisphere (Lelieveld and Dentener 2000), potentially introducing ozone-rich stratospheric air directly into the near-surface layer and leading to ozone pollution episodes.
The impacts of STT on tropospheric ozone in China have been less studied compared to other regions such as North America and Europe (Elbern et al 1997, Sprenger and Wernli 2003, Stohl et al 2003, Lin et al 2012, Wang et al 2020b).Recently, the importance of STT to tropospheric ozone variations has been increasingly recognized, leading to a growing number of studies on this topic.The significant contribution of STT events to tropospheric ozone has been reported in western, northeastern, and eastern China, associated with processes such as tropopause folding, cut-off lows, Rossby wave breaking, and typhoons (Ding and Wang 2006, Li and Bian 2015a, Jing and Banerjee 2018, Li et al 2015b, 2018, Wang et al 2020a, Zhang et al 2021, Meng et al 2022, Wang et al 2023, Chen et al 2023a).However, due to data limitations, especially regarding vertical measurements of ozone profiles, the studies mentioned above mostly focus on mid-to-high latitudes (30˚further North).The importance of STT to tropospheric ozone at the lower latitudes has only been studied in Hong Kong andTaiwan (Ou-Yang et al 2022, Zhao et al 2021).To what extent does STT contribute to tropospheric ozone variations at lower latitudes in the mainland of China, e.g.southeastern China, is an interesting question and awaits further studies.
From November 2021 to November 2022, the China Meteorological Administration conducted ozonesonde observations at a regional background station in southeastern China.In this area, ozone levels at the surface and throughout the troposphere exhibit a distinct seasonal cycle, with two peaks occurring in spring and autumn (Chen et al 2023b).Although the ozone levels in this area are lower than in highly polluted regions in China, episodes of high ozone concentrations exceeding national standards still occur frequently, particularly in spring and autumn, which have not been fully explained (Ge et al 2022).This study uses valuable data to explore the potential impacts of STT on ground-level ozone variations at lower latitudes.Especially, we will focus on spring, when the annual ozone peak occurs and STT events are frequent.Combined with reanalysis data and model simulations, we will give the first assessment of the specific contribution of STT to tropospheric ozone variations in southeastern China, and intend to offer a novel understanding into the seasonal variations of ozone in this area.

Ozonesonde data
Shaowu observation station (27.32 • N, 117.49• E, altitude: 218.9 m, as marked by the red star in figure S1) is one of China's national sounding stations, located in Nanping City, Fujian Province, in southeastern China.It serves as a regional background station with low air pollution levels.Conducting ozonesonde observations here is of significant scientific importance for elucidating the vertical distribution of ozone in the southeastern coastal areas of China.The ozone-sounding system, incorporating the CYT-1 ozonesonde and the HYDF-MCRS1 satellite navigation sounding receiver, among other components, demonstrates a reliable capability in measuring ozone profiles (Zheng et al 2022, Zhu et al 2022).The CYT-1 ozonesonde, developed by the Institute of Atmospheric Physics, Chinese Academy of Sciences, has a measurement accuracy of ±10% for tropospheric ozone and ±15% for stratospheric ozone below 10 hPa.This system can measure ozone partial pressure (units: mPa) and obtain meteorological data such as air pressure (units: hPa) and relative humidity (RH, units: %).Ozonesonde observations are conducted every Wednesday at 14:00 (Local Time), with potential postponements or cancellations due to rain or other severe weather.From November 2021 to November 2022, a total of 51 ozone profiles were obtained.In this study, we converted ozone partial pressure into the more commonly used ozone mixing ratio (OMR, units: ppbv) using the following formula: OMR = 10 000 × P O3 P , where the P represents air pressure and P O3 the ozone partial pressure.

Reanalysis data
The  (Emmons et al 2012).In the stratosphere, its concentration mirrors ozone levels, while in the troposphere, O 3 S undergoes only chemical loss and does not undergo chemical production.Therefore, O 3 S can represent the contribution of stratospheric ozone transport to tropospheric ozone at various altitudes in the troposphere.

Impacts of STT on tropospheric ozone
Figure 1(a) shows the profile-to-profile tropospheric ozone evolution of the one-year measurements.As the Shaowu station is situated in a regional background area, ozone concentrations near the surface are relatively low and increase with altitude.This differs from polluted areas, e.g. in Xiamen and Nanjing, where ground-level ozone concentrations exceed those in the free troposphere in August and September (Chen et al 2023b).This study presents, for the first time, the one-year evolution of ozone throughout the entire troposphere at a regional background station, which may provide further insights into explaining ozone variations at different altitudes in the troposphere.Consistent with previous studies, ozone concentrations peak in spring and summer in the troposphere, particularly in the middle-toupper troposphere (Chen et al 2023b).This trend is further elucidated in the monthly mean ozone evolution (figure S2(a)).
The CAM-Chem model effectively simulates the profile-to-profile tropospheric ozone evolution (figure 1(b)), exhibiting a high level of consistency with observations.In monthly averages, the model also demonstrates a strong agreement in seasonal variation with the observations (figure S2(b)).This provides us with a high level of confidence in utilizing the model to interpret the observed ozone variations.Since observations are only available for one year, the climatological mean  seasonal cycle based on model simulations is also depicted in figure S3(a), confirming the tropospheric ozone peaks in spring and summer.
Figure 1(c) gives the profile-to-profile evolution of the stratospheric ozone tracer (O 3 S).It is clear that the stratospheric ozone contribution (indicated by O 3 S) peaks in spring, which is more evident in monthly mean data (figure S2(c) for 2022 and figure S3(b) for climatology) and consistent with previous studies (Chen et al 2023b).Note that the stratospheric contribution in the CAM-Chem model (figure S3(b)) is notably higher than that in the GEOS-Chem model estimated by the tagging method (Chen et al 2023b).Previous studies have shown that GEOS-Chem simulations of STT are weaker compared to actual observations (Hu et al 2017).Given the very consistent ozone evolution between the CAM-Chem model and observations, we have confidence in the reliability of the values presented in our study.While subtracting the O 3 S values (indicating the stratospheric contribution) from the model simulated ozone, ozone concentrations in spring are strongly reduced and the seasonal peak in spring disappears (figure 1(d)).This can also be seen in the climatological seasonal cycle from the model (figure S3(d)).The relative contribution of O 3 S to ozone is up to 50% in spring at 4-6 km and even larger at upper levels for the year 2022 (figure S2(d)).In the climatological mean, the ratio of O 3 S/O 3 is also high in spring, exceeding 30% above 4 km (figure S3(c)).These findings indicate that STT plays a crucial role in determining ozone values in spring through the middle-to-upper troposphere.In contrast, the high values of tropospheric ozone in late summer and early autumn may primarily result from active photochemical reactions and enhanced regional ozone transport (Chen et al 2023b).

Impacts of STT on ozone at the surface
The above analysis highlights the important role of STT in tropospheric ozone variation.However, it remains unclear the mechanisms underlying STT events, and the extent to which STT contributes to surface ozone.To address this question, we selected and analyzed a case with evident STT using ozonesonde data, ERA5 reanalysis, the CAM-Chem model, as well as site-based ozone measurements.Among the 51 ozone sounding profiles, the data from 4 May 2022, stands out as particularly noteworthy, drawing our attention with notably elevated tropospheric ozone levels across almost all altitudes (figure S4(a)).In fact, the total tropospheric ozone observed on 4th May ranked first among the entire datasets.Compared to the average ozone sounding profiles in April, May, and June, ozone levels on 4th May are significantly higher than the average at almost all altitudes except 5-8 km (figure 2).This consistent pattern is also observed in the ERA5 data, confirming that it is not an artifact or deviation linked to this particular measurement (Figure S4(b)).The model also simulates the profile relatively well, with consistent high ozone values at ∼4 km.At the same time, the relative humidity at 9-10 km and ∼4 km is significantly lower than the average.Such anomalously high ozone and low water vapor at 9-10 km and ∼4 km suggests a possible occurrence of STT on 4th May 2022.
To confirm the occurrence of STT on 4th May 2022, the PV, which is a conserved quantity in the absence of friction and heat sources and widely used as a tracer to distinguish stratospheric and tropospheric airs, is analyzed.Figure 3 depicts the evolution of PV at 200 hPa from May 1st to 6th.Initially, high PV values were predominantly observed in the northeastern region of China and its eastern oceanic areas, extending towards the Mongolian Plateau.Following that, the high-PV belt gradually shifted eastward and southward, approaching the region of interest examined in this study.Between May 3rd and 5th, PV values over this region surpassed 2 PVU.This confirms the occurrence of STT over China from May 1st to 6th, which influenced the targeted region around May 4th.
We then analyze the meteorological conditions for this STT event around May 4th, 2022.Figure 4 presents the 500 hPa geopotential height and 200 hPa wind field, including the jet stream, from May 1st to May 6th, 2022.As shown in the Figure, a powerful low-pressure system is observed over northeastern China and its eastern regions on May 1st.Following this system, there is an extensive north-east to southwest oriented upper-level trough.The subsequent development and eastward movement of the lowpressure system and trough led to their bifurcation into two separate troughs.The northern trough advances more rapidly, gradually establishing a configuration characterized by a north ridge and a south trough.On May 3rd and 4th, the southeastern part of China was positioned in front of the southern branch of the trough.Subsequently, all systems continued their eastward movement and dissipated.At the same time, an upper-level jet stream accompanies the bottom and front of the trough, shifting eastward in tandem with the development and progression of the trough.Combining figures 3 and 4, the position of the high-PV belt, situated north of the jet stream axis, correlates closely with the location of the upper-level trough, undergoing evolution and shifts concurrent with their development.The STT event is therefore caused by a potent horizontal trough and upper-level jet stream.
To further investigate the vertical transportation of ozone-rich air from the lower stratosphere to the troposphere during this STT event, we present latitude-altitude profiles of ozone anomaly, PV, and RH obtained from ERA5 data at Shaowu observation station for the period of May 1st to May 6th (figures 5(a)-(c), respectively).On May 1st, substantial positive ozone anomalies were observed mainly above 300 hPa and north of 35 • N, with a well-defined southward-extending branch.As time progressed, the region of positive ozone anomalies and associated branches shifted southward.By May 4th, significant positive ozone anomalies were evident at 100-200 hPa and 500-700 hPa over Shaowu station, consistent with the ozonesonde profiles depicted in figure 2. In figure 5(b), the 2 PVU line descended southward and downward from regions of high altitude and high latitude, reaching below 200 hPa at Shaowu's latitude on May 4th, indicating tropopause folding and an STT event.The patterns of high PV and positive ozone anomalies are in concordance, supporting the link between tropospheric ozone enhancement and STT events.STT events are characterized not only by an increase in ozone and PV but also by a reduction in RH.According to earlier studies, the RH threshold for detecting STT events ranges from 30% to 50% (Itahashi et al 2020, Ou-Yang et al 2022).As depicted in figure 5(c), above 200 hPa, RH consistently remains below 50% between 20 • N and 40 • N.This indicates that, with tropopause folding and decreased altitude, dry stratospheric air descends to lower levels.Simultaneously, a noticeable intrusion of dry air between 500-700 hPa extends continuously southward, signifying the oblique downward flux of stratospheric air.
The simulated O 3 S tracer corroborates the transport of ozone-rich air from the stratosphere to the troposphere.The O 3 S/O 3 ratio serves as an indicator of stratospheric influence in the troposphere.prominent increase and peaks in ozone observed above 15 km, as measured by the ozonesonde.Similarly, a descending high O 3 S/O 3 ratio plume, analogous to ozone anomaly and PV, progressively moves southward with the southward movement of the trough and jet stream.On May 4th, the O 3 S/O 3 ratio above 500-700 hPa exceeds 55%, confirming that the peaks around 4 km in the ozone sounding profiles result from the downward transport of stratospheric ozone.Notably, the high O 3 S/O 3 ratio belt began extending downward to the boundary layer and even the surface starting from May 1st.This implies that the STT event may have impacted surface ozone.
To further explore the impact of STT events on surface ozone in southeastern China, we present surface ozone and CO time series for the city of Nanping, where the ozonesonde launches occur, as well as for two larger cities in southeastern China, Fuzhou and Xiamen (figure 6).Ozone levels in all three cities exhibited an increase from the 3rd to the 6th, reaching or approaching 80 ppbv during peak ozone days, indicating a pollution event in an otherwise clean southeastern region of China.Elevated ozone levels were observed in Nanping during the nights of May 1st, 2nd, 4th, and 5th, with concurrent lower CO concentrations on May 1st and 2nd.The increased nighttime ozone levels and reduced CO concentration suggest that, beyond the impact of photochemical processes, stratospheric transport plays a role in influencing ground-level ozone during May 1st to May 5th.This pattern is also depicted in figure 5.Both Fuzhou and Xiamen also experienced elevated nighttime ozone from May 1st to 3rd.However, the highest nighttime ozone occurred on the 4th, accompanied by a noticeable decrease in CO concentration.Temporal disparities are linked to the westward location of Fuzhou and Xiamen relative to Nanping, resulting in a lag in the impact of stratospheric transport on ground-level ozone.The CAM-Chem simulations partially captured the evolution of surface ozone and elevated nighttime ozone in the three cities.Simulated O 3 S values from May 1st to May 5th showed higher levels, particularly on May 3rd and May 4th, consistent with earlier findings.According to CAM-Chem, STT events contributed approximately 15-20 ppbv, accounting for 25%-30% of surface ozone.This significant contribution underscores the necessity for heightened attention to the potential impact of STT events on air quality.
Note that the case of STT 's impact on surface ozone mentioned above is not the only one.Using one standard deviation above the mean as the threshold, we identified cases exhibiting significantly elevated tropospheric ozone levels.To ensure a robust selection process and prevent including cases with limited data points meeting the threshold, we specified that ozone concentrations exceeding one standard deviation of the mean must persist continuously for at least 1 km in altitude range.The final selection comprised 14 cases, as depicted in Figure S4a.We further checked that, in 9 out of 14 selected cases, the observed ozone values are beyond one standard deviation of the multiyear mean (figure S5), indicating a close relationship between the anomalously high ozone levels in the free troposphere and the high-ozone episodes at the surface.Furthermore, among these 9 cases, 6 exhibited relatively high model-simulated O 3 S values, particularly concentrated in the spring season.This signifies the significant contribution of STT to surface ozone elevations.

Conclusion
In this study, we present, for the first time, the annual evolution of tropospheric ozone at a regional background station in Shaowu, southeastern China.It indicates peak ozone concentrations during spring and summer, particularly in the mid-to-upper troposphere.The CAM-Chem model effectively simulates tropospheric ozone evolution, showing consistent seasonal variations with observations.Additionally, this study highlights the significant contribution of STT to tropospheric ozone in spring, with contributions reaching 30% in climatology and even 50% in 2022 above 4 km.Subtracting this contribution leads to a considerable reduction in springtime ozone concentrations, eliminating the seasonal peak.These findings emphasize the more crucial role of STT in shaping tropospheric ozone levels during spring than previously believed.
We also examined an anomalous ozone profile observed on 4 May 2022.The distinctive feature lies in the notable enhancement of tropospheric ozone, particularly with a significant increase in concentrations above 15 km and below 5 km.Analysis of ERA5 reanalysis data and the CAM-Chem model reveals that the rise in tropospheric ozone is primarily attributed to a STT event induced by a deep horizontal trough and the subtropical jet.Ozone-rich stratospheric air crosses the tropopause, resulting in distinctive transport to both the upper and lower troposphere.This STT event has also led to heightened surface ozone levels in the southeastern region of China, including elevated values during both nighttime and daytime.Model evaluations suggest that it contributed roughly to a 15-20 ppbv (25%-30%) increase in surface ozone.Such influences of STT on surface ozone can be seen frequently during the one-year measurements.Among the 14 selected cases with anomalously high ozone plumes in the troposphere, observed surface ozone was present in 9 cases, while 6 cases exhibited relatively high simulated O 3 S notably concentrated in the spring season.This indicates that STT events have a significant impact on surface ozone in southeastern China.
This study employed ozonesondes to directly measure authentic atmospheric ozone data in the southeastern region of China, investigating STT events and providing an additional perspective for STT research in China.This underscores the importance of vertical sounding observations in atmospheric composition studies, suggesting that China should consider incorporating them into operational observations.Meanwhile, the contribution of STT events to the increase in surface ozone is notably more pronounced in the relatively less polluted southeastern region compared to heavily polluted areas.This highlights the potential role of STT events as a key factor contributing to ozone pollution in southeastern China.Consequently, ozone forecasting and analysis of ozone pollution causes in this region should consider the influence of STT events.

Figure 1 .
Figure 1.Profile-to-profile evolution of tropospheric ozone (a) observed in 51 ozonesonde measurements and (b) simulated by the CAM-Chem model.The evolution of (c) the stratospheric ozone tracer (O3S) and (d) the difference between O3 and O3S.
Following previous studies, 2 PVU is commonly considered indicative of the dynamical tropopause (Wang et al 2020a).Rapid and irreversible deformation of PV often indicates the breaking of Rossby waves and the folding of the tropopause, associated with STT events (Wernli and Sprenger 2007, Kumar et al 2020, He et al 2024).

Figure 2 .
Figure 2. Profiles of ozone mixing ratio (red line) and relative humidity (blue line) measured by ozonesonde.The solid lines correspond to data collected on May 4th, while the dashed lines indicate the average values for the months of April, May, and June.
Figure Potential vorticity (shaded) from ERA5 data at 200 hPa for 1 May 2022, to 6 May 2022.The white lines denote the contour of the 2 PVU.The red star marks the launch location of the ozonesonde.

Figure 4 .
Figure 4. Geopotential height (contour lines, units: gpm) at 500 hPa, horizontal winds (vectors), and wind speeds exceeding 30 m s −1 (shaded) at 200 hPa.The depiction is based on ERA5 data collected between 1 May 2022, and 6 May 2022.The red star denotes the launch location of the ozonesonde.

Figure 5 .
Figure 5. Latitude-altitude cross-section of daily mean ozone mixing ratio anomaly (a), potential vorticity (b), relative humidity (c) using ERA5 data, and the O3S/O3 ratio (d) simulated by CAM-Chem model.The depicted profiles are specific to the longitude of the ozonesonde launch and cover the period from 1 May to 6 May 2022.The orange dashed line indicates the latitude corresponding to the ozonesonde launch site.The solid black lines in (b) represent the contour of the 2 PVU.

Figure 6 .
Figure 6.Time series of surface O3 (red lines) and CO (blue lines) concentrations from MEE (hourly), along with O3 (orange dots) and O3S (green dots) simulated by CAM-Chem model (every six hours) for 1 May 2022 to 8 May 2022 over Nanping (a), Fuzhou (b) and Xiamen (c).The O3 levels at 6:00 (Local Time) daily, corresponding to the nadir of the diurnal cycle, are indicated by red dotted lines.The left Y-axis represents O3 or O3S concentration, while the right Y-axis represents CO concentration.
2013: influence of convective transport and stratospheric intrusions Atmos.Chem.Phys.18 17979-94 Li K, Jacob D J, Shen L, Lu X, De Smedt I and Liao H 2020 Increases in surface ozone pollution in China from 2013 to 2019: anthropogenic and meteorological influences Atmos.Chem.Phys.20 11423-33 Lin M et al 2012 Springtime high surface ozone events over the western United States: quantifying the role of stratospheric intrusions J. Geophys.Res.Atmos.and Carlsten C 2019 Air pollution and DNA methylation: effects of exposure in humans Clin.Epigenetics 11 1-15 Sprenger M and Wernli H 2003 A northern hemispheric climatology of cross-tropopause exchange for the ERA15 time period (1979-1993) J. Geophys.Res.Atmos.108 8521 Stohl A et al 2003 Stratosphere-troposphere exchange: a review, and what we have learned from STACCATO J. Geophys.Res.Atmos.108 8516 Van Dingenen R, Dentener F J, Raes F, Krol M C, Emberson L and Cofala J 2009 The global impact of ozone on agricultural crop yields under current and future air quality legislation Atmos.Environ.43 604-18 Wang H, Wang W, Huang X and Ding A 2020a Impacts of stratosphere-to-troposphere-transport on summertime surface ozone over eastern China Sci.Bull.65 276-9 Wang H, Wang W, Shangguan M, Wang T, Hong J, Zhao S and Zhu J 2023 The stratosphere-to-troposphere transport related to rossby wave breaking and its impact on summertime ground-level ozone in Eastern China Remote Sens. 15 2647 Wang T, Xue L, Brimblecombe P, Lam Y F, Li L and Zhang L 2017 Ozone pollution in China: a review of concentrations, meteorological influences, chemical precursors, and effects Sci.Total Environ.575 1582-96 Wang X, Wu Y, Randel W and Tilmes S 2020b Stratospheric contribution to the summertime high surface ozone events over the western united states Environ.Res.Lett.15 1040a6 Wernli H and Sprenger M 2007 Identification and ERA-15 climatology of potential vorticity streamers and cutoffs near the extratropical tropopause J. Atmos.Sci.64 1569-86 Zhang J, Li D, Bian J, Xuan Y, Chen H, Bai Z, Wan X, Zheng X, Xia X and D L 2021 Long-term ozone variability in the vertical structure and integrated column over the North China Plain: results based on ozonesonde and Dobson measurements during 2001-2019 Environ.Res.Lett.16 074053 Zhao K et al 2021 Impact of stratospheric intrusions on ozone enhancement in the lower troposphere and implication to air quality in Hong Kong and other South China regions J. Geophys.Res.Atmos.126 e2020JD033955 Zheng Y, Deng H, You H, Qiu Y, Zhu T, Cheng X and Wang H 2022 A study of the vertical distribution and sub-peaks of ozone below 12 km over wuyishan region based on ozone sounding in winter Atmosphere 13 979 Zhu T, Deng H, Huang J, Zheng Y, Li Z, Zhao R and Wang H 2022 Analysis of ozone vertical profiles over wuyishan region during spring 2022 and their correlations with meteorological factors Atmosphere 13 1505