Investigation of the behavior of the atmospheric dynamics during occurrences of the ozone hole's secondary effect in southern Brazil

The Antarctic ozone hole (AOH) directly influences the Antarctic region, where its levels can reach values below 220 DU. The temporary depletion of ozone in Antarctica generally occurs between the beginning and middle of August, during the austral spring, and extends to November, when a temporary reduction in ozone content is observed in a large region over the Antarctic continent. However, masses of ozone-depleted air can break away from the ozone hole and reach mid-latitude regions in a phenomenon known as the secondary effect of the Antarctic ozone hole. The objective of this paper is to show how atmospheric dynamics behave during the occurrence of this type of event, especially in mid-latitude regions, such as southern Brazil, over a 12year observation period. For the analysis and identification of the events of influence of the AOH on the southern region of Brazil, data from the total ozone column were used from ground-based and satellite experiments, the Brewer Spectrophotometer (MkIII no. 167), and the Ozone Monitoring Instrument (OMI) on the Aura satellite. For the analysis of the stratospheric and tropospheric fields, the ECMWF reanalysis products were used. Thus, 37 events of influence of the AOH that reached the southern region of Brazil were identified for the study period (2006–2017), where the events showed that in approximately 70 % of the cases they occurred after the passage of frontal systems and/or atmospheric blocks over southern Brazil. In addition, the statistical analysis showed a strong influence of the jet stream on mid-latitude regions during the events. Among the 37 identified events, 92 % occurred in the presence of the subtropical and/or polar jet stream over the region of study, possibly explaining the exchange of air masses of ozone deficient in the upper troposphere–lower stratosphere (UT–LS) region.


Introduction
Discovered in 1840 by Christian F. Schonbein, ozone is the most important constituent of stratospheric gas traces which, together with Water Vapor (H 2 O) and Carbon Dioxide (CO 2 ), is responsible for the Earth's energy balance (SEINFELD most important component in the stratosphere from the point of view of the skin protection against the harmful UVB solar radiation (even considering that a small portion this spectrum can pass through the O 3 layer and hit the ground surface). Most of its atmospheric content (about 90%) is concentrated in the stratosphere between 15 and 35 km altitude (LONDON et al., 1985) in the region known as "ozone layer".
The concentration of ozone in a particular region of the Earth is mainly determined by the meridional transport of this 40 element in the stratosphere (GETTELMAN et al., 2011). The explanation for the higher concentration of ozone found in polar rather than equatorial regions (where there is greater production) is precisely a special type of poleward transport known as the Brewer-Dobson circulation, in which air masses are transported quasi-horizontally from the stratospheric tropical reservoir to polar regions (BREWER, 1949;DOBSON, 1968;BENCHERIF et al., 2007 andBENCHERIF et al., 2011). The poleward transport of stratospheric ozone is one of the essential factors for the concentration of this atmospheric 45 constituent in a certain region of the planet (PLOEGER et al., 2012), being much studied from the use of Potential Vorticity, which correlates with the transport of chemical constituents traces such as Ozone (O 3 ), Nitrous Oxide (N 2 O) and Water Vapor (H 2 O) on isentropic surfaces in the lower stratosphere. The Potential Vorticity (PV) acts as a dynamical tracer for large-scale air mass transport, behaving as a material surface where the potential temperature is conserved (HOSKINS et al., 1985). In the lower reaches of the stratosphere the lifetime of O 3 molecules is longer and therefore they can be used as a 50 tracer in the study of air mass flow in the Stratosphere-Troposphere Exchange events (BUKIN et al., 2011).
The first studies discussing the ozone concentration on Polar Regions showed that during the spring of the Southern Hemisphere there was a massive reduction of the O 3 content, being known as Antarctica Ozone Hole (AOH) (CHUBACHI et al., 1984;FARMAN et al., 1985 andSOLOMON et al., 1999).The ozone hole area is defined when there is a region with values below 220 DU, less than two thirds of the historical level (HOFMANN et al., 1997). Nevertheless, temporary 55 destruction directly influences ozone content around the Polar Regions due to the crossing of the polar vortex boundary over these regions, causing drastic reductions in the ozone content and increase of the levels of surface ultraviolet radiation (CASICCIA et al., 2008). Studies by Guarnieri et al. (2004) have shown that reductions of up to 1% in total ozone content in southern Brazil cause an average 1.2% increase in surface ultraviolet radiation. In addition, increased ozone-related ultraviolet radiation may also affect aquatic and terrestrial systems, helping to explain the decline in amphibian species 60 associated with genetic malformations caused by increased radiation levels received (SCHUCH et al., 2015). However, their effects can affect regions of mid-and low-latitudes, causing temporary decreases in the total columns of ozone.
Poor ozone air masses are released from the interior of the Antarctic polar vortex, the edge of the Ozone Hole, being carried by the polar filaments on these regions (MARCHAND et al., 2005), in a phenomenon called "Secondary effect of the Antarctic Ozone Hole" causing a temporary fall in ozone content, first observed by Kirchhoff and collaborators (1996) over 65 the South of Brazil. Peres et al., (2014) and Peres et al., (2016) showed the effects of this secondary event on middle latitude regions such as the southern region of Brazil, where ozone content falls over the region from August to November. Recently, Bittencourt et al. (2018) reported on the second most intense event ever recorded in the southern region of Brazil. According to the latest World Meteorological Organization (WMO) reports (2014 and 2018) there is a growth trend between the 80's and 90's, stabilizing at high rates since the years 2000, despite indications of declining trends in the Antarctic ozone in recent 70 years (SOLOMON et al., 2016).
Unlike other regions of Brazil, the weather conditions in southern Brazil are strongly influenced by transient meteorological systems (REBOITA et al., 2010). Examples of such systems are cold and hot fronts, which carry strong west winds at high tropospheric levels called jet streams. Moreover, the Upper Troposphere -Lower Stratosphere (UT-LS) region in southern Brazil seems to be the home place of many dynamical processes such as stratosphere-troposphere exchanges and isentropic 75 transport between the tropical stratosphere reservoir, polar vortex and mid-latitude. Indeed, understanding the patterns of the UT-LS is important in understanding transport and exchange processes, and the links with tropospheric meteorology (OHRING et al., 2010).
Because of this, the main objective of this paper is to show how atmospheric dynamics behave during the occurrence of events of influence of the Antarctic ozone hole over mid-latitude regions such as southern Brazil, during a 12-year 80 observation period. In addition, recent works (Peres et al., 2014;Bittencourt et al., 2018) show that there is a secondary influence of the ozone hole with these regions where the tropospheric dynamics behavior can influence most cases in the occurrence of these events.

Region of Study and Instruments 85
The region of study was the central portion of Rio Grande do Sul, comprising the city of Santa Maria -RS (29.72°S; 53.72°W). In this work two instruments were used for the analysis of the total ozone content over the southern region of Brazil for a period of 12 years of data (2006)(2007)(2008)(2009)(2010)(2011)(2012)(2013)(2014)(2015)(2016)(2017). The ground based instrument denominated Brewer Spectrophotometer, # 167), now on referred only as Brewer (BREWER MODELS), operated in the 90 municipality of São Martinho da Serra -RS (about 30 km from the city of Santa Maria -RS), and data from OMI instrument were used for the days when there were no Brewer measurements, completing the database for the region of study.  (SCI TEC, 1988).

100
Brewer is a fully automated instrument designed for terrestrial measurements of spectral irradiations in the UV-B range of the solar spectrum at five wavelengths, 306.3; 310.1; 313.5; 316.8 and 320.1 nm (KERR et al., 2010), with an approximated resolution of 0.5 nm, allowing to obtain the Total Column of Ozone (O 3 ) and Sulfur Dioxide (SO 2 ) (FIOLETOV et al., 2005). The MKIV instrument also measures intensity of radiation in the visible part of the spectrum (430-450nm) and uses differential absorption in this region to infer Total Column Nitrogen Dioxide (BREWER MANUAL). This instrument also 105 permits to obtain the optical thickness of atmospheric aerosols and the vertical O 3 profile by the Umkehr technique (KERR, 2002; and BREWER MANUAL). Peres et al. (2019) analyzed a long-term (35-year) Total Ozone Column (TOC) series for southern Brazil using Brewer data and satellites, where they showed good agreement between ground-based and satellite instruments. Vanicek (2003) presented 110 a history of calibrations of the Dobson and Brewer instruments in Prague, Czech Republic, where the correct function of the Brewer spectrophotometer was shown and, consequently, its accurate observation process, which depends only on the precise adjustments of the optics as components brewer's electronics and mechanics. In South America, the accuracy and quality of TOC data is ensured by cross calibration using the Brewer # 017 calibrator, allowing a deviation of only about 0.58% from daily averages (FIOLETOV et al., 2005). measures the TOC besides other atmospheric parameters related to ozone chemistry and climate (e.g., NO 2 , SO 2 ). OMI also 120 can distinguish distinct types of aerosols (such as smoke, dust, and sulfates) and measures the atmospheric pressure and cloud coverage. The Earth's atmosphere is observed in 740 spectral bands of wavelength along the satellite path, with a band large enough to provide global coverage in 14 orbits (1day). The 13 x 24 km spatial resolution can be expanded to 13 x 12 km to detect and track sources of pollution on an urban scale, and observes the atmosphere in two ultraviolet bands, named UV-1 (270 to 314 nm) and UV-2 (306 to 380 nm), with a spectral resolution of 0.45 and 1 nm, respectively (LEVELT et al., 125 2006). In terms of the TOC (in Dobson Unit, DU), the OMI presents an absolute accuracy (the root sum of square of all (1) errors) of 3% and a relative accuracy of 1%, respectively for the full vertical column and for the 13 x 24 km in horizontal resolution (LEVELT et al., 2006).
We also used reanalysis data available at the Era-Interim/ECMWF, and described by Dee et al. (2011), where meteorological 130 fields were prepared for the analysis of the stratospheric and tropospheric dynamics. Due to radiosonde limitations on the region of study, the spatial resolution used was 2.5º x 2.5º latitude/longitude, responding well to the objectives of this work, where a higher resolution is not necessary for further details. For the stratospheric dynamics analysis, data of potential vorticity and ozone mixing ratio were used at the potential temperature levels of 265 K and 850 K. For the analysis of tropospheric dynamics it was used winds data (components u, v and w), geopotential height and layer temperature available 135 from 1000hPa to 1 hPa, at the pressure level, in addition to pressure data at mean sea level were used. With these data, potential vorticity fields were made for the potential temperature levels of 600 K and 700 K. For tropospheric fields, sea level pressure and layer thickness between 1000 and 500 hPa, horizontal layer cut showing the jet at 250 hPa and Omega at 500 hPa, and a vertical cut of the layer between 1000 and 50 hPa of potential temperature and wind (m/s) for the longitude of 54ºW. 140 The HYSPLIT/NOAA model was used to help identifying the events of influence of the Antarctic Ozone Hole over the region of study (ROLPH et al., 2017). The Lagrangian HYSPLIT model is a complete system for calculating simple trajectories of air parcels as well as complex transport simulations, chemical transformation and deposition (HYSPLIT, 2017). The model assumes that a particle follows the wind flow passively its trajectory is the integration of the particle 145 position vector in space and time. In this study, which aims to show the air mass behavior for four days before the events, the HYSPLIT model, an isentropic vertical velocity model, was used to assist in the observation of possible event days over the region, available in HYSPLIT (2017).

Identification of AOH Influence Events 150
The identification of the events of influence of the AOH is done first by the analysis of the average daily data of the TOC through instruments on board of satellites and on the ground. In this work, 12 years of satellite data were analyzed with the aim of identifying days where the mean daily value of TOC is less than the climatological average for the month of analysis, i.e., the climatological average minus 1.5 of its standard deviation value (μ -1,5σ), where, μis a climatological average for 155 the month of interest, σ is the standard deviation, and the value of -1.5 is the criterion chosen from the normal frequency distribution tests (WILKS et al., 2006). This criterion was also used by Peres (2016), where it was observed that the variations around the mean value can represent well the influences of the ozone content in the region of investigation.
After identifying the possible days of influence of the AOH over the region of study, the analysis of isentropic surfaces was 160 conducted, where Absolute Potential Vorticity (APV) fields were made. For the analysis of the stratospheric dynamics, we used reanalysis data available on the Era-Interim/ECMWF platform and the APV fields were analyzed on isentropic surfaces for the potential temperature levels of 600 K and 700 K. Potential Vorticity (PV) was used in previous studies that correlated PV with chemical constituents such as ozone, water vapor and nitrous oxide on isentropic surfaces (adiabatic surfaces where the potential temperature remains constant) in the lower troposphere (SCHOEBERL et al., 1989). 165 In these case studies, the PV acts as a dynamic large-scale air mass tracer and can be used as a horizontal coordinate (HOSKINS et al., 1985). In this way, this type of analysis aims to verify the origin of the air masses in which an APV increasing is observed when the air mass originates from highest latitudes (e.g., Antarctica). For this analysis, it is needed to consider the previous days of the event or equatorial origin when a decrease of Absolute Potential Vorticity occurs 170 (SEMANE et al., 2006). Bittencourt et al., (2018) and Bresciani et al., (2018) showed the analysis of an extreme event of influence of the AOH on regions of middle latitudes through the analysis of the stratospheric dynamics with the fields of PV.
In the analysis of potential vorticity fields, the air mass trajectory was observed. When there is an increase in the APV it can be stated that the mass of air had polar origin, and other wise (decrease in the APV) the air mass has equatorial origin. As 175 described above, the APV acts as a dynamic marker for large-scale air masses and, thus, observations are made to identify the Secondary Effect of AOH, where reductions in O 3 content are observed from intense to moderate (BITTENCOURT et al., 2018 andPERES et al., 2016).

Tropospheric Analysis 180
After identifying all the events for the study period (2006)(2007)(2008)(2009)(2010)(2011)(2012)(2013)(2014)(2015)(2016)(2017), meteorological fields were prepared for the analysis of tropospheric dynamics, which aims to show how the troposphere behaved before, during and after the occurrence of every event of Secondary Effect of the AOH identified in southern Brazil. Peres et al., (2014) showed a case study of 2012, presenting two events of influence of the AOH on the southern region of Brazil, in which the synoptic analysis was done for 185 the region on the day of the event. The results showed that one of the events occurred just after the passage of a frontal stationary system, where then the arrival of a high pressure system helped to stabilize the region and in the advance of the air masses poor in O 3 , configuring the occurrence of the AOH influence event.
The meteorological data for the construction of the pressure fields at Sea level and the layer thickness between 1000 hPa and 190 500 hPa were obtained by the ECMWF, and the purpose here is to check which synoptic systems were predominating during the events. A presence of the subtropical jet is intended to be displayed in the field of horizontal winds at 250 hPa and Omega at 500 hPa. In addition, ascending and descending surface movements were identified. Another field analyzed was the vertical cut of the atmosphere at different levels of potential temperature (in Kelvin) and wind components (in m/s) for the longitude of 54ºW. In this case, the jet stream was present at higher levels of the troposphere which may aid in air 195 exchanges from the stratosphere to the troposphere (SANTOS, 2016).

Statistical Analyzes
The average daily data of the ECMWF for the horizontal wind components (zonal-u, and meridional -v) and also the vertical 200 movement velocity (Omega -w) were used for this analysis. In addition, temperature and geopotential height or the available pressure levels between 1000 and 1 hPa, and the component of PV and O 3 for the level of 700 K potential temperature were used. These data were correct read and organized into matrices with the daily averaged values for each of these variables (temperature,u,v and w,geopotential,PV and O 3 ) in a grid of 2.5° latitude by 2.5° longitude at the levels previously used.

205
As the set of data used in the preparation of the stratospheric and tropospheric analysis fields was only for the months of interest in this analysis, from August to November in 12 years, the reduction of this dataset allowed the separation of the days of interest (days of occurrence of events of influence of the AOH on the southern region of Brazil), and the subsequent calculation of the monthly averages.

210
For stratospheric analysis, mean fields of all identified events were made for a period of 3 days before and after each event.
For the anomaly analysis of the potential vorticity fields, the following expression was used:

Results and Discussion
In this work, the daily average data of the total ozone column were analyzed from the two instruments described above

230
After the identification of the mean daily data of the TOC and analysis of the monthly climatology of the data observed at SSO, the first step for the identification of AOH side effect events over the study region is the analysis of the climatological average for the reference months and the occurrence of AOH during the extended austral spring period (August to daily value of the total ozone column is less than the average climatological value (for the respective month) minus 1.5 of its 235 standard deviation (μ-1,5σ). Table 1 shows the TOC monthly climatological values with monthly standard-deviation, together with the lower TOC limit, for the extended spring season.
After the confinement of the drop limits presented in Table 1, we analyzed 90 days where the TOC value of the respective day was lower than this limit minus 1,5σ. From these days, using a methodology described above, a total of 37 events were 240 identified as important events that reached the southern region of Brazil from August to November in the years from 2006 to 2017. As expected, based on previous works and climatology, the identified events occurred mostly during October, and it is in good agreement with the results found by Peres (2016). To exemplify the analysis developed in this paper, we present in the next section a case study that took place on September 18, 2017, as this is the most recent event identified throughout the observed period, showing a side effect event of AOH in the region of South Brazil. Other interesting, and even more 245 prominent, events have already been reported by Bresciani et al., (2018) and Bittencourt et al., (2018).

Case Study: Event Observed on September 18, 2017
The event that occurred on 18 Sept. 2017 presented a TOC value, measured by the Brewer Spectrophotometer, of 271.5 DU, 250 representing a decrease of approximately 8.5% in comparison with the climatological average for the month of September, as reported in Table 1. The observed decrease in TOC could be attributed to isentropic transport in the stratosphere. Figure 1 show the PV fields obtained from ECMWF data at 600K and 700K isentropic levels in the stratosphere. One can see from Figure 1 that Chile, Argentina, Uruguay, South of Brazil and Paraguay are under the influence of the passage of 255 stratospheric air masses characterized by APV values greater than 100. We obtained almost the same PV pattern at the 600K isentropic level. As explained above, PV is a conservative dynamical parameter and indicates the transport of air masses which takes place on isentropic surfaces (HOSKINS et al., 1985). Therefore, PV distributions could be used to determine the origin of air masses. Since PV values are positive in the north hemisphere and negative in the south hemisphere, for convenience, we refer here after to the APV, which is positive regardless of the latitude. From Figure 1 we can observe that 260 PV values are higher than 100 and can be associated to air masses of polar origin, what suggests that the observed decrease in the total ozone column at SSO, in South of Brazil, is a result of the transport of air masses with low ozone concentration from high southern latitudes. In order to corroborate this hypothesis, the Lagrangian HYSPLIT model was initialized on September 18, 2018 at SSO location and run for back-trajectory retrievals in the lower stratosphere (see Figure 2.a). All the stratospheric back-trajectories show that air masses observed over SSO in the South of Brazil travelled northward and 265 eastward over the polar region. This confirms the polar origin of the observed air masses. Moreover, Figure 2.b illustrates the global distribution of TOC recorded by OMI experiment on September 18, 2018. It shows that transport of polar air is characterized by reduction in TOC distribution extending from the polar region up to mid-latitude region. This well illustrates the side effect of AOH, resulting in a decrease in stratospheric ozone concentrations during the analyzed event.

270
After the identification of the secondary effect of the AOH on the southern region of Brazil on September 18, 2018, the tropospheric dynamics analysis, to observe how the troposphere was behaving during the occurrence of this event. Figure 3 presents the atmospheric fields used in this work for the study of tropospheric dynamics. This type of analysis was used by Bittencourt et al., 2018, where the study was done only for an extreme event of influence of AOH. In the days leading up to the confirmation of the side effect event, the region remained unstable from September 11, 2018 until one day before the 275 event, which can be explained by the isentropic tapering corresponding to a more compressed layer thickness, besides the presence of an intense temperature gradient. For the day of the event September 18, 2018, was observed in the region the formation of a system of high pressure, which moved to the ocean in the following days. Under these circumstances, on the surface, we have a post-frontal high-pressure system near the region of interest, which may have helped to carry this O 3 air mass to reach mid-latitude regions such as central southern Brazil. 280 The horizontal wind and temperature fields ( atmosphere showed that the presence of the polar jet dominates the region until the day of the event confirmation. For this reason, the arrival of O 3 poor air masses in the region may be associated with the performance of a frontal system that passed through the region days before the event was confirmed, as well as the presence of the jet at higher atmospheric levels, helping in the exchanges air from the stratosphere to the troposphere contributing to a temporary reduction in O 3 content on September 18, 2018. The vertical section of the atmosphere between 1000 and 10 hPa of potential temperature and wind at 290 54º longitude, Figures 3.e and 3.f shows the presence of polar jet current at higher levels of the atmosphere, as well as the isentropic funnel near longitude. from 30ºS to September 17, 2018, indicating a frontal ramp, which helps in the air exchange from the highest to the lowest levels on the day of the event. Figure 4 shows the mean field of the 37 AOH influence events identified in this work, where potential vorticity fields were used for the 700 K isentropic level, for three days before and up to three days after the event. Analyzing Figure 4a and 4b, it

Statistical analyses: atmospheric dynamics 295
can be observed that for -3 days (-3d) the variation of potential vorticity over the region remains stable, without variation in the content of APV on the south of Brazil, with APV values between 40 and 60PVU. 300 Already from 2-days before (-2d) the event, Figure 4.c, we can observe a slight increase in APV values over the study region, mainly between Argentina, Uruguay and Southern Brazil with APV values from 60 to 80 PVU. From one day before (-1d) the event, the increase of APV over the study region becomes more important, with APV values between 100 and 140 PVU. For the days after the event, +1d (Figure 4.e) and +2d (Figure 4.f), air masses with APV higher than 100 PVU bound 305 mid-latitude region in Chile, Argentina, Uruguay and Southern Brazil, with values up to 160 PVU. From the third day after the event, we found a decrease in APV values similar to the -2d situation (not shown). These results indicate that during the 37 identified secondary effect events due to the AOH development, low-ozone air masses are transported from polar region to mid-latitudes and covering a wide region over North of Chile, Argentina, Uruguay and Southern Brazil. In average, such low-ozone event may last and affect that sub-region during at least 4 days. This is agreement with previous works published 310 by Peres (2016).
Plots on Figure 5 show the average monthly distributions of potential vorticity anomalies on the 700 K isentropic level, averaged for August, September, October and November over the study period, 2006-2017. The PV anomaly fields show the predominance of positive potential vorticity anomalies in southern Brazil (values around 35 to 55 PVU). Figure 5.a shows 315 that in August there is a predominance of positive anomalies in southern Brazil, according to the number of events identified this month (7 events, see Table 2) in southern Brazil. November has the lowest number of low-ozone AOH events identified in the region (5 events, see Table 2) and also shows the predominance of a positive anomaly in southern Brazil, with possible vorticity anomalies between 10 and 30 PVU. For the months of September and October, positive PV anomalies were very evident in the 12 years of data. 320 Significant increases in positive anomalies (between 10 and 50 PVU for September, and from 30 to 60 PVU) are concomitant and consistent with the large number of low-ozone AOH events recorded during these months (12 events in September and 13 events in October). Physically, it is possible to confirm the importance of these months for the analysis, due to the greater number of AOH influencing events that affect southern Brazil, as observed by Bittencourt et al., 2018, 325 which is explained by polar filaments that release from the Ozone Hole region and then bring O 3 to mid-latitude regions.
For a better understanding of the tropospheric dynamics during the 37 events identified, medium fields were made for the horizontal and vertical cuts of the atmosphere. Figure 6 shows the average field for the horizontal cut (jet at 250 hPa and Omega at 500 hPa). In the mean of the 37 AOH influence events identified in this study (Table 2), the presence of the jet 330 stream (subtropical or polar) is observed in practically all identified events, Figure 6 confirms this where the presence of the jet stream is observed mainly on the southern region of Brazil. However, there is a predominance of a center with negative values of Omega at 500 hPa, indicating surface convergence, which explains the majority of events identified after the passage of frontal systems over the southern region of Brazil. Therefore, the importance of the jet stream for the vertical SANTOS, 2016) on the southern region of Brazil.
Finally, Figure 7 presents the average for the 37 AOH influence events identified on the Southern Brazil of the vertical cut of the atmosphere between 1000 and 30 hPa. Similar to Figure 6, the presence of the jet stream with an intense nucleus (~ 45 to 50 m/s) near the latitude and longitude of the study region, besides the presence of a jet near 30 hPa, indicating the probable 340 presence of the polar jet current in the average of the events. However, it is confirmed that the jet stream (subtropical and / or polar, depending on the case) was also present at higher levels of the atmosphere. Therefore, analyzing the average tropospheric dynamics of the 37 events of influence of the AOH on the southern region of Brazil, the presence of the polar jet stream, at higher levels of the atmosphere, as well as the presence of the subtropical jet stream probably explaining the transport of O 3 -poor air masses from polar regions to mid-latitude regions, like the southern Brazil. 345

Conclusions
In this work, we analyzed daily TOC measured by the Brewer Spectrophotometer operational at the SSO site in the South of Brazil, and by OMI instrument from 2006 to 2017. Analysis of TOC datasets revealed 37 low-ozone events that have occurred and extended during the austral spring period (August-September-October-November) over the SSO site.
Moreover, examination of potential vorticity fields in the stratosphere (on the 700 K isentropic level) and of back-trajectories 350 obtained by the Lagrangian HYSPLIT model showed that the 37 low-ozone events resulted from the transport of air masses from polar regions to mid-latitudes, and correspond therefore to the secondary effect of the AOH. In addition, it has been shown from PV anomaly fields that the detected events have spread over a large region, covering northern Chile, Argentina, Uruguay and southern Brazil, and can last and affect this sub region for at least 4 days. In accordance with the period of development of the (AOH) and with previous published works, we found that most of the events took place in September 355 (35%) and October (39%), while 17.6% of them were identified in August and 12.7% in November.
The analysis of tropospheric dynamics confirmed the importance of jet as the main synoptic system that assists in the exchange of air masses between the stratosphere and the troposphere. Of the 37 events, about 92% of the cases identified the presence of the jet stream (subtropical and/or polar), in the remaining 8% no action of the jet stream was identified, or it was 360 weak, not assisting in the exchange of air masses. In addition, on the surface, events were identified in 70% of cases after the passage of frontal systems in southern Brazil, where together with the performance of a high-pressure system characterized by downward stabilization of the atmosphere, explains the arrival of ozone-depleting air masses from the Antarctic region that can reach the mid-latitude regions. Regarding the statistical analyzes of the tropospheric fields, confirmation of the importance of the jet stream was obtained. The vertical cut of the atmosphere showed the presence of the two jet streams 365 (polar and subtropical jet) at higher levels of the atmosphere, besides the current lines converge to regions close to 30ºS, southern region of Brazil. The average fields of the 37 events identified in the region, show the presence of the jet stream in relation to the horizontal cut (jet 250 hPa) and vertical cut (1000 and 30 hPa).
The results found here highlight the importance of the jet stream actuation as the main synoptic system that supports the 370 exchange of masses of ozone-deficient air from the stratosphere to the troposphere. It is evident that the two jet streams (subtropical and polar) act together in this exchange mechanism, possibly becoming a "connection" between the two atmospheric layers during the occurrence of events of side effect of the (AOH) on the southern region of Brazil.