A Stratospheric Intrusion-Influenced Ozone Pollution Episode Associated with an Intense Horizontal-Trough Event

Ozone pollution is currently a serious issue in China. As an important source of tropospheric ozone, the stratospheric ozone has received less concern. This study uses a combination of ground-based ozone measurements, the latest ERA5 reanalysis data as well as chemistry-climate model and Lagrangian Particle Dispersion Modeling (LPDM) simulations to investigate the potential impacts of stratospheric intrusion (SI) on surface ozone pollution episodes in eastern China. Station-based observations indicate that severe ozone pollution occurred from 27 April to 28 April 2018 in eastern China, with maximal values over 140 ppbv. ERA5 meteorological and ozone data suggest that a strong horizontal-trough exists at the same time, which leads to an evident SI event and brings ozone-rich air from the stratosphere to the troposphere. Using a stratospheric ozone tracer defined by NCAR’s Community Atmosphere Model with Chemistry (CAM-Chem), we conclude that this SI event contributed about 15 ppbv (15%) to the surface ozone pollution episode during 27–28 April in eastern China. The potential impacts of SI events on surface ozone variations should be therefore considered in ozone forecast and control.


Introduction
Tropospheric ozone is an important greenhouse gas contributing to global warming [1,2] and also a main pollutant harmful to human health and crop productivity [3,4]. A significant decrease of ozone concentrations has been observed in the eastern United States and parts of Europe due to precursor emission controls [2,5,6]. At the same time, although the Chinese government has strictly controlled industrial emissions since 2011, ozone concentrations in eastern China have increased over the past years [7,8]. Ozone pollution episodes have been reported frequently in eastern China, and ozone pollution is now a serious issue in China [6,7,9]. Surface ozone is related to complex photochemical reactions involving anthropogenic and biogenic emitted volatile organic compounds (VOCs) and nitrogen oxides (NOx) [10][11][12]. While the anthropogenic emitted NOx decreased significantly due to government control and the VOCs emissions changed little [7,8,12], the exact reason for elevated ozone and frequently occurring ozone pollution episodes in eastern China are still not clear.
Besides chemical production, tropospheric ozone can also be transported from the stratosphere through the process of stratospheric intrusion (SI) [13][14][15][16][17]. As another important source of tropospheric ozone, SI contributes 20%-30% to the tropospheric ozone budget in the mid-latitudes of the Northern Hemisphere [14,18,19]. In cases of intense SIs, the stratospheric ozone-rich air may be transported rapidly from the lower stratosphere to the lower troposphere and near the surface [20][21][22][23]. Such deep SI events may cause a steep ozone increase near the surface and lead to ground-level ozone pollution [24][25][26][27][28]. Whether such deep SI events contribute to the recently reported ozone pollution episodes in eastern China is an important question waiting to be answered. Quantifying the relative contribution of SI events to the elevated ozone is important for surface ozone forecasting and control in eastern China.
In the extratropics, SIs always form near the vicinity of extratropical cyclones or baroclinic eddies, through processes like tropopause folding, wave-breaking or cut-off lows [13,29,30]. SIs are often associated with westerly jet stream or frontal activities [17,31,32]. Most of these SI processes occur in areas with high latitudes or high altitudes [33][34][35]. In East Asia, studies of STE mainly focus on the Qinghai-Tibet Plateau and northeast China [15,[36][37][38][39]. The potential effects of SI events on surface ozone in eastern China, however, has been less investigated. A recent study found that SI events occur frequently during the summer and contribute about 10 ppbv to surface ozone in eastern China [40]. This poses a question about the potential impacts of SI events on surface ozone variations in eastern China, whereas human activities are particularly active and heavy air pollution happens frequently [6,9,41].
This study investigates a severe ozone pollution episode that occurred during 28-29 April 2018 using the station-based surface ozone measurements as well as the latest ERA5 reanalysis ozone data. In particular, the potential contribution of SIs to surface ozone in the spring season (MAM ) is estimated with the aid of a stratospheric ozone tracer of the CAM-Chem model. The transport mechanism is also analyzed using the ERA5 meteorological reanalysis data.

Observational Data Sets
Hourly surface ozone and CO concentrations for the year 2018 were obtained from the public website of the China Ministry of Ecology and Environment: beijingair.sinaapp.com/. The network has had 1500 monitoring stations since 2017 including about 330 cities.

ERA5 Reanalysis
ERA5 (the fifth generation of ECMWF atmospheric reanalysis) [42] ozone and meteorological data are used in this study for analyzing the ozone and meteorological conditions during the SI events. ERA5 was produced using 4D-Var data assimilation in CY41R2 of ECMWF's Integrated Forecast System (IFS), with 137 hybrid sigma/pressure (model) levels from the surface to 0.01 hPa. The horizontal resolution of ERA5 for the high resolution realisation (HRES) is about 31 km, 0.28125 degrees. More details of the ERA5 data can be found at its website https://confluence. ecmwf.int/display/CKB/ERA5%3A+data+documentation. Within the framework of the Copernicus Atmosphere Monitoring Service (CAMS) atmospheric composition forecast, IFS was found capable of predicting ozone increases in the troposphere during deep SI events [17]. ERA5 assimilated many ozone observations (satellite and in-situ) including MLS (Microwave Limb Sounder), OMI (Ozone Monitoring Instrument) ozone data, which is of great importance to the determination of STE (Stratosphere Troposphere Exchange). ERA5 ozone shows good agreement with in-situ and satellite measurements in the upper troposphere and lower stratosphere [40]. However, its accuracy in the lower troposphere is relatively poor [40]. That is because the chemical scheme of ozone in ERA5 is completely dependent on the parameterization of ozone source/sink (https: //www.ecmwf.int/en/elibrary/16648-part-iv-physical-processes). In addition, due to the lack of in-situ and satellite observations, the tropospheric ozone in ERA5 is not well constrained.

CAM-Chem Simulations
The Community Atmosphere Model with Chemistry (CAM-Chem) simulation used in this study is performed by NCAR , which is publicly available at https://www.acom.ucar.edu/cam-chem/cam-chem.shtml. It is driven by specified dynamics, with meteorological fields from MERRA2 reanalysis. It is run at 0.9 • × 1.25 • horizontal resolution with 56 vertical levels. The chemistry mechanism used is the MOZART-T1. In particular, the O 3 S variable, a stratospheric ozone tracer relative to the tropopause, is defined by the CAM-Chem model to estimate the stratospheric contribution to tropospheric ozone. O 3 S is set to the ozone mixing ratio in the stratosphere and is destroyed below the tropopause at the same rate as ozone [43]. More details of the simulation are described on the website https://wiki.ucar.edu/display/camchem/CESM2.1%3ACAM-chem+as+Boundary+Conditions.

LPDM Simulations
LPDM (Lagrangian Particle Dispersion Modeling) is conducted using the HYSPLIT (Hybrid Single-Particle Lagrangian Integrated Trajectory) model [9,44]. In the model, the residence time and position of particles released at a receptor can be calculated according to the meteorological field. In this study, the model was run 24-h backwards with 3000 particles released at an altitude of 100 m above Hangzhou at 16:00 (local time) on 27 April 2018. The residence time of particles was used to identify the "footprint" retroplume. More details of the LPDM can be found in Ding et al. [45].

Meteorological Conditions
The synoptic conditions over east Asia in the upper atmosphere (at 300 hPa) are shown in    Figure 4, two troughs can be seen at 500 hPa on 25 April 2018 with an earlier strong one over the region between northeastern China and Japan and a later weak one over Mongolia and north China. The later one moved faster and was enhanced during its moving. From 26 April to 27 April 2018, the two troughs merged together and formed a very deep trough. On 27 April 2018, the area of interest was right behind and influenced by the deep trough, which brought air from higher latitudes.  Figure 6 shows the spacial distribution of potential vorticity (PV) at 300 hPa. In the absence of friction and heat sources, PV is conserved along trajectories, which allows it to be used as a tracer in upper-level dynamics. In particular, PV is broadly used to separate the stratospheric and tropospheric air masses due to the significant difference of the static stability between them. Following previous studies [40], we used 2 PVU to mark the dynamical tropopause. On 25 April 2018, high PV values can be seen over Mongolia and north China as well as the region between northeastern China and Japan, associated with the two upper-level troughs as seen in Figure 3. Along with the horizontal-trough development, a broad quasi-horizontal band of high PV values can be seen from the west to the east of China on 26 and 27 April 2018. It is noteworthy that there are two belts in this high PV band, with one between 30 • and 40 • N and the other between 20 • and 30 • N. The later one crossed the area of interest in this study. This suggests a potential impact of stratospheric ozone intrusion on the tropospheric ozone over the area of interest.

Stratospheric Ozone Intrusion to the Troposphere
To illustrate the downward transport, the ERA5 ozone data are used for analysing the ozone distribution in the upper troposphere and lower stratosphere (UTLS) region. Figure 7 shows the horizontal distribution of ozone at 300 hPa from 25 April 2018 to 28 April 2018. The spatial pattern of high ozone values is very similar to the PV pattern shown in Figure 5. This confirms the transport of ozone-rich air from the stratosphere to the troposphere associated with the upper-level trough development (Figure 3)   The transport of ozone-rich air from the stratosphere to the troposphere is also confirmed by the O 3 S tracer. Figure 9 shows the longitude-height cross-section of the O 3 S/O 3 ratio from the CAM-Chem simulation. Consistent with ERA5 ozone shown in Figure 8,

SI Impacts on Surface Ozone
So far, we find that the severe ozone pollution episode from 27 to 28 April 2018 might be related to an SI event. From the discussion above, the SI event is evident and the transport of ozone-rich air from the lower stratosphere to the troposphere is clear. The stratospheric ozone-rich air reached the middle troposphere below 400 hPa. However, whether such enhanced ozone concentrations in the troposphere can be mixed with the planetary boundary layer (PBL) and influence the surface is still not clear.
To illustrate the vertical mixing between the free troposphere and the PBL, the divergence of water vapour as well as the vertical velocity are shown in Figure 10. Positive values of water vapour divergence dominated from 26 to 27 April 2018. In particular, a strong divergence centre from 600 hPa down to the surface can be seen on 26 and 27 April 2018, which indicates the sink of the dry air. This suggests that the ozone-rich and dry air may move further down from the middle troposphere to the lower troposphere. On 28 April 2018, weak convergence occurred in the lower troposphere, which indicates a possible vertical mixing between the PBL and the upper levels.
Consistent with the water vapour divergence, downward motion of air masses is evident from 27 to 28 April 2018 as shown in the vertical distribution of vertical velocity. In particular, positive values of vertical velocity (indicating sink of air) can be seen from 400 hPa to the lower troposphere from 27 April to the morning of 28 April 2018. This strengthens the case that the ozone-rich and dry air is transported from the upper troposphere down to the PBL. On the afternoon of 28 April 2018, upward motion can be seen indicating unstable conditions and subsequent vertical mixing between the PBL and upper levels.
The vertical mixing between the free troposphere and the PBL can also be confirmed by the 'footprint' of air mass as shown by LPDM simulations. A 24-h backward integration of LPDM indicates that particles released at 16:00 (local time) in Hangzhou mainly originated from northern China (Figure 11a). Vertically, seen from the northwest-southeast transect of the retroplume (Figure 11b), the particles originated from up to 5 km (close to 500 hPa). Especially, the retroplume distributes vertically from the surface to about 700 hPa, which indicates strong vertical-mixing between the PBL and the free troposphere. The stratospheric ozone-rich air was therefore transported from the free troposphere to the surface.  To quantify the potential contribution of SI events to surface ozone, Figure 12 Figure 13. The CAM-Chem model simulates the ozone pollution episode relatively well, although the absolute value is lower than observations. According to the CAM-Chem simulation, the stratospheric ozone intrusion contributes about 15 ppbv (15%) to the ozone pollution episode from 27 to 28 April 2018. Such a significant contribution cannot be neglected and should be considered in surface ozone forecasting and developing emission control strategies.

Conclusions and Discussion
Ozone pollution has been reported frequently and has become a serious issue in China [6,9,11]. Currently, most of the research has focused on anthropogenic and biogenic emitted ozone precursors from the surface [7,8,12], while the stratospheric ozone intrusion, another important source of tropospheric ozone, has been less studied. A recent study poses the question of to what extent do the SI events influence the surface ozone at relatively lower latitudes, that is, in eastern China [40]. This study is a follow-up work of a previous study [40], and investigates the potential contribution of SI events to surface ozone pollution episodes in eastern China during spring using ground-based measurements, the latest ERA5 reanalysis data as well as chemistry-climate model simulations.
Results indicate that a strong SI event occurred, associated with a horizontal-trough that brings ozone-rich air from the lower stratosphere to the surface and contributes to severe ozone pollution, during the period from 27 April to 28 April 2018 in eastern China. According to a CAM-Chem simulation, SI contributed about 15 ppbv (15%) to this surface ozone pollution episode in eastern China. Although such a contribution may not fully explain the mechanism of this severe ozone pollution episode, this study highlights the importance of SI events to the surface ozone variations. The potential impacts of SI events on tropospheric and surface ozone should be considered in ozone variation attribution and surface ozone forecasting.