ISSN 0001-4338, Izvestiya, Atmospheric and Oceanic Physics, 2017, Vol. 53, No. 1, pp. 65–75. © Pleiades Publishing, Ltd., 2017. Original Russian Text © V.Yu. Ageyeva, A.N. Gruzdev, 2017, published in Izvestiya Rossiiskoi Akademii Nauk, Fizika Atmosfery i Okeana, 2017, Vol. 53, No. 1, pp. 74–85.
Seasonal Features of Quasi-Biennial Variations of NO2 Stratospheric Content Derived from Ground-Based Measurements V. Yu. Ageyeva† and A. N. Gruzdev* Obukhov Institute of Atmospheric Physics, Russian Academy of Sciences, Moscow, 119017 Russia *e-mail:
[email protected] Received June 19, 2015; in final form, October 8, 2015
Abstract⎯Seasonal and latitudinal distributions of amplitudes of quasi-biennial variations in total NO2 content (NO2 TC), total ozone content (TOC), and stratospheric temperature are obtained. NO2 TC data from ground-based spectrometric measurements within the Network for the Detection of Atmospheric Composition Change (NDACC), TOC data from satellite measurements, and stratospheric temperature data from ERA-Interim reanalysis are used for the analysis. The differences in the NO2 TC diurnal cycles are identified between the westerly and easterly phases of the quasi-biennial oscillations (QBO) of equatorial stratospheric wind. The QBO effects in the NO2 TC, TOC, and stratospheric temperature in the Northern (NH) and Southern (SH) hemispheres are most significant in the winter–spring periods, with essential differences between the NH and SH. The NO2 TC in the Antarctic is less for the westerly phase of the QBO than that for the easterly phase, and the NO2 TC quasi-biennial variations in the SH mid-latitudes are opposite of the variations in the Antarctic. In the NH, the winter values of the NO2 TC are generally less during the westerly QBO phase than during the easterly phase, whereas in spring, on the contrary, the values for the westerly QBO phase exceed those for the easterly phase. Along with NO2, the features of the quasi-biennial variations of TOC and stratospheric temperature are discussed. Possible mechanisms of the quasi-biennial variations of the analyzed parameters are considered for the different latitudinal zones. Keywords: nitrogen dioxide (NO2), ozone (O3), quasi-biennial oscillation (QBO) DOI: 10.1134/S0001433817010029
1. INTRODUCTION
maximum amplitude of ~20% was observed at an altitude of 28 km above the Equator. The authors of [6] associate the NO2 QBVs with the QBO effect on the vertical transport of the odd nitrogen species. According to [8], the QBVs of NO2 are the main cause of the QBVs of ozone above 30 km in the tropics. In [12], the analysis of data from measurements of vertical NO2 profiles with the GOMOS instrument in 2002–2008 (~3 quasi-biennial cycles) is carried out. The amplitude of the NO2 QBVs in a layer of maximum concentration was 18% above the equator, and the QBO effect on the NO2 content in mid-latitudes was negligible. It was found in [13] that the correlation coefficient between the NO2 content according to the GOMOS data and the velocity of the equatorial stratospheric wind changes with altitude. For a detailed analysis of QBO effects, long-term data are required. The longest series of observations of stratosphere NO2 are obtained in ground-based spectrometric measurements. The first estimations of the NO2 QBVs from such measurements were given in [14]. The QBVs of NO2 have been detected in all latitudinal zones, and the most significant of them are in the tropics and polar latitudes. To date, additional data
The quasi-biennial oscillations (QBOs) of the equatorial stratospheric wind are the main feature of the equatorial stratospheric circulation, which manifests itself in the alternation of the westerly and easterly winds with an average period of ~29 months [1, 2]. The QBO has an impact on atmospheric circulation, temperature, and content of atmospheric species not only in the tropics, but also in the extratropical latitudes of the Northern (NH) and the Southern (SH) hemispheres [1, 3–5]. The impact of QBOs on the total ozone content (TOC) and temperature in the stratosphere was studied most thoroughly [1, 6–13]. The QBO effects in the content of the other atmospheric species, including stratospheric NO2, are investigated insufficiently [6, 8, 12–14]. The QBO effect on the NO2 content was mainly studied for relatively short satellite data series and only in tropical latitudes [6, 8, 12, 13]. The quasi-biennial variations (QBVs) of NO2 were first discovered in the SAGE II data for 1984–1990 (~3 quasi-biennial cycles) in the latitudinal belt 30° S–30° N [6]. The † Deceased.
65
66
AGEYEVA, GRUZDEV
Stations and periods of NO2 TC observations Station
1. Ny-Ålesund 2. Scoresbysund 3. Kiruna 4. Sodankylä 5. Zhigansk 6. Harestua 7. Zvenogorod 8. Jungfraujoch 9. Issyk-Kul 10. Mauna Loa 11. Reunion Isl. 12. Bauru 13. Lauder 14. Kerguelen Isl. 15. Macquarie Isl. 16. Dumont d’Urville 17. Arrival Heights
Latitude Longitude
78.92° N 70.48° N 67.84° N 67.37° N 66.76° N 60.22° N 55.69° N 46.55° N 42.62° N 19.54° N 20.90° S 22.35° S 45.05° S 49.35° S 54.50° S 66.67° S 77.83° S
11.93° E 21.95° W 20.41° E 26.63° E 123.35° E 10.75° E 36.77° E 7.98° E 76.99° E 155.58° W 55.48° E 49.03° W 169.69° E 70.26° E 158.94° E 140.02° E 166.66° E
Observation period, years
1995–2008 1992–2013 1991–2010 1990–2013 1992–2012 1994–2013 1990–2013 1991–2012 1983–2012 1996–2008 1993–2013 1995–2013 1980–2013 1996–2013 1996–2012 1988–2013 1991–2012
from ground-based measurements of NO2 content are being accumulated. This makes it possible not only to refine estimations of QBO effects on NO2, but also to analyze their seasonal dependence, which has not been done yet. Considerable seasonal differences in the QBO effect on ozone were revealed in [7, 10]. It can be assumed that the QBO impact on NO2 must also depend on the season. The objective of this paper is to obtain and provide analysis of seasonally dependent estimations of the QBO effects on the stratospheric contents of NO2 and ozone and stratospheric temperature in different latitudinal zones of SH and NH. 2. DATA AND ANALYSIS METHOD Data from ground-based spectrometric measurements of the NO2 total content (NO2 TC) at stations of the Network for the Detection of Atmospheric Composition Change (NDACC) were used for the analysis. Coordinates of the stations and years of observation are indicated in the table. The measurements are carried out by zenith-scattered solar radiation within the visible spectral region during the morning and evening twilight. The data are available at http://ndacc.org. Stratospheric NO2 is a major contributor to its total content in the absence of pollution of the atmospheric boundary layer by nitrogen oxides. Most stations are located away from industrial centers (see table), so the tropospheric NO2 content is usually small there, and NO2 TC represents mainly the contribution of its
stratospheric part. The Zvenigorod station (scientific station of the Obukhov Institute of Atmospheric Physics, Russian Academy of Sciences) is located 50 km west to Moscow. However, the method of NO2 observations at this station allows retrieving vertical NO2 profiles and calculating directly its stratospheric content [15–17]. Further, by NO2 TC, we will mean NO2 content in a vertical column of the stratosphere. In this paper we also use the data of TOC from satellite measurements using the TOMS (1979–2005) and OMI (2004–2014) instruments and the ERAInterim reanalysis data on temperature at the isobaric pressure level of 50 hPa for 1979–2014 (http://apps. ecmwf.int/datasets/data/interim-full-daily/). The Giovanni internet service (http://gdata1-ts1.sci.gsfc. nasa.gov/daac-bin/G3/gui.cgi?instance_id=omi) was used to extract data on the TOC over the stations for NO2 observation. The annual variation was removed from the NO2 TC, TOC, and temperature according to the method of [18]. The NO2 TC values for 1992–1994 were excluded because of the significant impact of the Pinatubo eruption on the NO2 content [15, 19]. The obtained deviations were divided into two groups corresponding to the westerly and easterly phase of the QBO. The data of the Berlin Free University on the zonal velocity of stratospheric equatorial wind (http:// www.geo.fu-berlin.de/en/met/ag/strat/produkte/qbo/ index.html) were used as an index of the QBO. According to [3, 4, 13] and our own experience [20], the wind velocity at isobaric surface of 30 hPa was used for the SH, while the mean value of the wind velocities at surfaces of 40 and 50 hPa was used for the NH. To eliminate uncertainties associated with the change in wind direction, the following restrictions were applied: the QBO phase was considered westerly (easterly) if the velocity of the westerly (easterly) wind at that moment exceeded modulo the 30% of the value obtained by averaging the maxima of the velocity modulus of the westerly (easterly) wind. The monthly dependent QBO effects in the NO2 TC, TOC, and temperature were determined as the differences between monthly mean values of the deviations of these values from the annual variation in each month in the westerly QBO phase and the mean values of the deviations in the same month in the easterly QBO phase. The difference between the values obtained at different QBO phases characterizes the amplitude of the desired QBVs, and in the absence of gaps in data it is equivalent to differences between the average values of NO2 TC, TOC, and the temperature at the westerly and easterly QBO phases. Therefore, for brevity, the difference in deviations will further be called the difference in NO2 TC, TOC, and stratospheric temperature.
IZVESTIYA, ATMOSPHERIC AND OCEANIC PHYSICS
Vol. 53
No. 1
2017
Difference in NO2 TC, 1015 cm–2
SEASONAL FEATURES OF QUASI-BIENNIAL VARIATIONS
0.8
0.8
NO2
0.6
0.6
0.4
0.4
0.2
0.2
0
0
–0.2
–0.2
–0.4
–0.4
–0.6
–0.6
–0.8
–0.8
–1.0
1
2
3
4
5
20 10 0 –10 –20 –30 –40 –50 –60 –70 –80 –90 –100
6
7
8
9
10 11 12
Difference in TOC, DU
O3
Temperature difference, K
1 5 4 3 2 1 0 –1 –2 –3 –4 –5 –6 –7 –8 –9
2
3
4
5
6
7
8
9
10 11 12
T (50 hPa)
1
2
3
4
5
6 7 8 9 10 11 12 Month Arrival Heights Dumont d’Urville
67
NO2
–1.0 1
2
3
4
5
6
20 10 0 –10 –20 –30 –40 –50 –60 –70 –80 –90 –100
8
9
10 11 12
8
9
10 11 12
O3
1 5 4 3 2 1 0 –1 –2 –3 –4 –5 –6 –7 –8 –9
7
2
3
4
5
6
7
T (50 hPa)
1
2
3
4
5
6 7 8 9 10 11 12 Month Ny-Ålesund Scoresbysund Kiruna Zhigansk
Fig. 1. Monthly dependent differences in (upper panels) average NO2 TC from data of evening measurements, (middle panels) TOC, and (lower panels) temperature on 50 hPa isobaric surface between the westerly and easterly QBO phases (westerly minus easterly) for the Arrival Heights and Dumont d’Urville Antarctic stations in the SH (left), and the Ny-Ålesund, Scoresbysund, Kiruna, and Zhigansk Arctic stations in the NH (right). Vertical segments are 95% confidence intervals.
3. RESULTS 3.1. Seasonal Features of Quasi-Biennial Variations Figure 1 shows the difference between the values of NO2 TC, TOC, and temperature on the 50 hPa isoIZVESTIYA, ATMOSPHERIC AND OCEANIC PHYSICS
baric surface in the westerly and easterly QBO phases for the polar stations of the SH (left) and NH (right). First and foremost, let us note that in the both hemispheres the QBO effects on the NO2 TC, TOC, and temperature are most pronounced in winter and spring Vol. 53
No. 1
2017
68
AGEYEVA, GRUZDEV 2.0
ΔNO2, 1015 cm–2
1.5
1.0
0.5
0
1
2
3
4
5
6 7 8 9 10 11 12 Month Arrival Heights (westerly QBO phase) Arrival Heights (easterly QBO phase) Dumont d’Urville (westerly QBO phase) Dumont d’Urville (easterly QBO phase)
Fig. 2. Monthly dependent differences, ΔNO2 between evening and morning values of the NO2 TC (evening value minus morning value) during westerly (thick curves) and easterly (thin curves) QBO phases according to measurement at the Arrival Heights and Dumont d’Urville Antarctic stations. Vertical segments are 95% confidence intervals.
periods. The seasonal dependences of the QBV amplitudes for NO2 TC, TOC, and temperature in Antarctica are qualitatively similar (Fig. 1, left). The NO2 TC, TOC, and temperature are smaller in the westerly QBO phase in spring than in the easterly phase. This is caused, apparently, by different intensities of physical and chemical processes responsible for ozone deficiency in the Antarctic stratosphere. In the westerly QBO phase, the circumpolar stratospheric vortex and dynamic isolation of the polar stratosphere is stronger than in the easterly QBO phase [3], which contributes to a stronger cooling of the Antarctic stratosphere with a more active formation of polar stratospheric clouds. Heterogeneous reactions on the surface of their particles lead to denitrification of the polar stratosphere, followed by intensification of ozone depletion as a result of the increase in effectiveness of the chlorine cycle. The diurnal variation of NO2 has a daytime component, that is, an increase in NO2 content throughout the day due to photolysis of N2O5 [15, 21]. The total content of odd-nitrogen family species depends on the HNO3 content [21], and thus, the HNO3 content affects the daytime component of the diurnal NO2 TC variation. According to the results of satellite measurements in 2001–2009 using the SMR instrument, a significant early decrease in HNO3 in the Antarctic stratosphere is observed in the winter–spring period [22], which has been the result of denitrification of the Antarctic stratosphere. The impact of the denitrification on the diurnal variation of NO2 is highlighted in [23]. To estimate the QBO effect on these processes, we calculated the monthly mean differences between
morning and evening NO2 TC in the westerly and easterly QBO phases. Data for the winter-spring period of 2002 for the Antarctic stations were excluded because of the sudden stratospheric warming that took place that year, which is not typical for the SH. It was found that the difference between evening and morning NO2 TC values in Antarctica, in early spring, is smaller during the westerly QBO phase than during the easterly phase (Fig. 2). This can be explained by the stronger decrease in the HNO3 content over Antarctica in the westerly QBO phase due to the greater denitrification of the stratosphere. Unlike in the Antarctic, the seasonal dependences of QBO on the NO2 TC, TOC, and the temperature in the Arctic region vary more strongly from station to station in the winter–spring period (Fig. 1, right), but, on the whole, NO2 TC in the NH spring is higher in the westerly QBO phase than in the easterly phase. In the westerly QBO phase, the temperature is lower in winter and higher in spring than at the easterly QBO phase. Regional features of seasonal dependencies of QBO effects at the Arctic stations can be noted. In February, the NO2 TC, TOC, and temperature in the Ny-Ålesund station are smaller during the westerly QBO phase than during the easterly phase, as well as over the Antarctic in spring. At the European polar stations of Kiruna and Sodankylä (Sodankylä results are not shown in Fig. 1), one may note the similarity of seasonal dependencies of the QBO effects on NO2 TC and temperature and their difference from the seasonal dependence of the QBO effect on TOC. So, the NO2 TC and temperatures in winter are generally higher for the easterly QBO phase and in spring they are higher for the westerly QBO phase, while the TOC values, throughout the whole winter-spring period, are smaller in the westerly QBO phase than in the easterly phase. At the Zhigansk station, similar to the Scoresbysund station, the seasonal dependencies of QBO effect on NO2 TC, TOC, and temperature differ. Similar to the SH, on the polar station of the NH, the dependence of NO2 TC on the QBO phase manifests itself in the diurnal NO2 variation in the winter– spring period (Fig. 3), but the character of this effect is different. At the Scoresbysund station in February and the Kiruna station in April, the difference between evening and morning NO2 values are larger in the easterly QBO phase than in the westerly phase (Fig. 3a), which is typical for the Antarctic (Fig. 2). However, the difference between morning and evening NO2 values from February to April at Zhigansk in the easterly QBO phase is, on the contrary, smaller than in the westerly phase (Fig. 3b). In this regard, we should note the opposite behavior of the NO2 TC QBVs at the Kiruna and Zhigansk stations in April (Fig. 1). One of the additional factors affecting the Arctic stratosphere in winter–spring period are sudden stratospheric warmings (SSWs). According to [24, 25],
IZVESTIYA, ATMOSPHERIC AND OCEANIC PHYSICS
Vol. 53
No. 1
2017
SEASONAL FEATURES OF QUASI-BIENNIAL VARIATIONS
ΔNO2, 1015 cm–2
1.5
1.5
(а)
1.0
1.0
0.5
0.5
0 1
2
3
4
5
6 7 8 9 10 11 12 Month Scoresbysund (westerly QBO phase) Scoresbysund (easterly QBO phase) Kiruna (westerly QBO phase) Kiruna (easterly QBO phase)
0 1
69
(b)
2
3
4
5
6 7 8 9 10 11 12 Month Zhigansk (westerly QBO phase) Zhigansk (easterly QBO phase)
Fig. 3. Similar to Fig. 2, but for the Arctic stations: (a) Scoresbysund and Kiruna or (b) Zhigansk.
in the period from the 1989/1990 winter to the 1997/1998 winter (~4 quasi-biennial cycles), there were no SSWs, whereas in the period from the 1998/1999 winter to the 2009/2010 winter (~5 quasibiennial cycles) the SSWs occurred almost every year (except for the winter–spring period of 2004/2005). Analysis of the differences in NO2 TC between the westerly and easterly QBO phases in the NH, for the periods with and without SSWs, has shown that the main contribution to the revealed seasonal dependence of the QBO effect on NO2 TC in the winter– spring periods is caused by the years with the SSWs. The SSW effect on the species contents requires special analysis that is beyond the scope of this work. Let us turn to mid-latitudes now. Figure 4 shows the differences in the NO2 TC, TOC, and stratospheric temperature between the westerly and the easterly QBO phases for the mid-latitude and tropical stations of the SH (left) and NH (right). The seasonal course of the QBO effect in NO2 TC in the SH midlatitudes differs radically from the seasonal course in the Antarctic (Fig. 1). In the winter–spring period, the NO2 TC in the SH mid-latitudes is higher for the westerly QBO phase; therefore, the QBO-caused changes in NO2 TC over Antarctica and in the mid-latitudes of the SH during this period are opposite to each other. The character of the seasonal dependence of the QBO effect in the SH mid-latitude TOC is approximately the same as in the Antarctic. However, an important difference is that the QBVs of the TOC at mid-latitudes are significant not only in spring, but also in winter. The same feature can also be noted in the temperature effect of the QBO. In the winter–spring period, as follows from Fig. 4, the QBVs of the NO2 TC, on the one hand, and the QBVs of the TOC and IZVESTIYA, ATMOSPHERIC AND OCEANIC PHYSICS
temperature, on the other hand, in the SH mid-latitudes are opposite to each other. The QBVs of the TOC in the SH mid-latitudes are associated with the quasi-biennial modulation of the Brewer–Dobson circulation. According to [26], the circulation in the middle stratosphere of mid-latitudes is more intense during the easterly QBO phase, which leads to the increased TOC as compared to the westerly QBO phase. The same effect can explain the temperature dependence on the QBO phase in mid-latitudes. An additional factor affecting the temperature can be the radiative heating due to absorption of UV solar radiation by ozone. Quasi-biennial variations of NO2 in mid-latitudes are, apparently, also due to the effect of the QBO on the Brewer–Dobson circulation; however, the increase in the circulation in mid-latitudes during the easterly QBO phase leads to lower values of NO2 TC. This is because the mixing ratio of HNO3 (one of the source of NO2) is lower in the tropics than in middle latitudes [27]. An additional impact on NO2 comes from more intensive interlatitudinal exchange between the periphery and the inner areas of the circumpolar vortex in the easterly QBO phase due to the lower dynamic isolation of the Antarctic stratosphere, which also contribute the decrease in the NO2 TC in middle latitudes for the easterly QBO phase. Another difference in the seasonal dependencies of the amplitude of the NO2 TC QBVs between the middle and polar SH latitudes is that the signs of the effects in the middle latitudes are opposite in the summer and winter–spring season, while in the polar latitudes the QBO effects in NO2 in these seasons have the same sign (Figs. 1, 4). It should be noted in connection with this that if the wind velocities on isobaric surfaces 40 or 50 hPa are used to specify the QBO Vol. 53
No. 1
2017
70
AGEYEVA, GRUZDEV
Difference in NO2 TC, 1015 cm–2
0.8
0.8
NO2
0.6
0.6
0.4
0.4
0.2
0.2
0
0
–0.2
–0.2
–0.4
1
2
3
4
5
6
7
8
9
10 11 12
30
NO2
–0.4 30
1
2
3
4
5
6
Difference in TOC, DU
O3 20
10
10
0
0
–10
–10
–20
–20
–30 2
3
4
5
Temperature difference, K
4
6
7
8
9
10 11 12
T (50 hPa)
–30
3
2
2
1
1
0
0
–1
–1
–2
–2
–3
–3 1
2
3
4
5
1
9
10 11 12
2
3
4
5
6 7 8 T (50 hPa)
9
10 11 12
2
3
4
5
6 7 8 9 Month Zvenigorod Jungfraujoch Issyk-Kul Mauna Loa
10 11 12
4
3
–4
8
O3
20
1
7
6 7 8 9 10 11 12 –4 1 Month Macquarie Isl. Lauder Bauru Reunion Isl.
Fig. 4. Similar to Fig. 1, but for the mid-latitude and tropical stations: (left) Macquarie, Lauder, Bauru, and Reunion in the SH, and (right) Zvenigorod, Jungfraujoch, Issyk-Kul, and Mauna Loa in the NH.
phase for the QBO, the summer effect of the QBO on the NO2 TC becomes insignificant whereas the amplitude of the QBVs of the NO2 TC in the winter-spring period remains nearly the same. The QBO effect in the NO2 TC in the NH mid-latitudes is in antiphase with the effects in the TOC and temperature, similar to the SH mid-latitudes (Fig. 4).
In general, NO2 TC in the NH mid-latitudes is higher for the westerly phase of QBO in winter and for the easterly phase of QBO in spring, which is similar to the seasonal dependence of the QBO effects in NO2 TC in the NH polar latitudes. However, in contrast to the polar stations in the NH, seasonal dependencies of the QBO effects in the TOC and temperature in the mid-
IZVESTIYA, ATMOSPHERIC AND OCEANIC PHYSICS
Vol. 53
No. 1
2017
SEASONAL FEATURES OF QUASI-BIENNIAL VARIATIONS
2.5
(а)
2.0
ΔNO2, 1015 cm–2
2.0
71
(b)
1.5
1.5 1.0 1.0 0.5
0.5 0 1
2
3
4
5
6 7 8 9 10 11 12 Month Lauder (westerly QBO phase) Lauder (easterly QBO phase)
0 1
2
3
4
5
6 7 8 9 10 11 12 Month Zvenigorod (westerly QBO phase) Zvenigorod (easterly QBO phase) Issyk-Kul (westerly QBO phase) Issyk-Kul (easterly QBO phase)
Fig. 5. Similar to Fig. 2, but for the mid-latitude stations of (a) Lauder in the SH and (b) Zvenigorod and Issyk-Kul in the NH.
latitudes are similar to each other. Like in the case of the SH mid-latitudes, they are believed to be the result of the QBO effect on the Brewer–Dobson circulation. Similar seasonal dependencies of the QBO effects in NO2 TC during winter–spring periods in the middle and polar latitudes of NH may be due to a stronger interlatitudinal exchange in the NH compared to the SH. At the mid-latitude stations in the both hemispheres, small but statistically significant discrepancies are revealed in the differences between the evening and morning values of the NO2 TC for the westerly and easterly QBO phases (Fig. 5). For example, at the Lauder station in the SH (Fig. 5a), Zvenigorod, Issyk-Kul (Fig. 5b), and the Jungfraujoch in the NH, the differences between the evening and morning values of the NO2 TC in winter–spring periods for the westerly QBO phase are greater than for the easterly phase, which contrasts with the ratio of the differences in the polar regions (Figs. 2, 3a). According to [22] (Fig. 1), on the outer side of the Antarctic stratospheric vortex, in the winter and in the spring, there is a belt of increased HNO3 concentration in the lower and middle stratosphere. A similar feature is typical for the NH mid-latitudes. The QBO effect on the Brewer–Dobson circulation and interlatitudinal exchange in the area of the stratospheric circumpolar vortex is, apparently, affecting the HNO3 content in the mid-latitude stratosphere. This may cause a dependence of the difference between the evening and morning values of the NO2 TC in the middle latitudes of the NH and SH on the QBO phase. At the Bauru and Reunion tropical stations in the SH, as well as at the SH mid-latitude stations, the IZVESTIYA, ATMOSPHERIC AND OCEANIC PHYSICS
QBVs of the NO2 TC are opposite to the variations of the TOC and temperature (Fig. 4, left). One may notice the displacement of the maximum QBO effect in the NO2 TC and TOC from the spring in mid-latitudes to the winter in tropical latitudes. However, there are no qualitative differences in the QBO effects in theNO2 TC, TOC, and temperature between the middle and tropical SH latitudes. This similarity can be explained in terms of the QBO effect through modulation of the Brewer–Dobson circulation, assuming that the tropical region of the ascending branch of the circulation is somewhat shifted from the equator to the NH. Then, judging by the character of QBO effect on the TOC and stratospheric temperature, the Bauru and Reunion stations may “tend” towards the middle latitudes, unlike to the tropical Mauna Loa station in the NH, located in the region of the ascending branch of the Brewer–Dobson circulation. The opposite character of the QBVs of the TOC and NO2 TC in the southern tropics may be related to the destruction of stratospheric ozone in the nitrogen cycle in the tropical latitudes [8]. The QBO effect on the NO2 TC in a winter–spring period at the Mauna Loa station in the NH manifests itself in a similar way to the NH mid-latitudes; however, in contrast to the mid-latitudes, the tropical QBVs of the TOC are opposite to the variations of temperature (Fig. 4, right). The Brewer–Dobson circulation weakens during the westerly QBO phase [26], which results in higher TOC in its ascending branch. Lower efficiency of the nitrogen cycle of ozone destruction over the winter, due to lower NO2 TC in the westerly QBO phase, also contributes to higher TOC values in the westerly QBO phase. Vol. 53
No. 1
2017
72
AGEYEVA, GRUZDEV
(а)
1.5
(b)
2.0
ΔNO2, 1015 cm–2
1.5
1.0 1.0
0.5 0.5
0
1
2
3
4
5
6 7 8 9 10 11 12 Month Bauru (westerly QBO phase) Bauru (easterly QBO phase) Reunion (westerly QBO phase) Reunion (easterly QBO phase)
0
1
2
3
4
5
6 7 8 9 10 11 12 Month Mauna Loa (westerly QBO phase) Mauna Loa (easterly QBO phase)
Fig. 6. Similar to Fig. 2, but for the tropical stations: (a) Bauru and Reunion in the SH and (b) Mauna Loa in the NH.
The difference between the evening and morning values of the NO2 TC at the Bauru and Reunion tropical stations in the SH (Fig. 6a) is higher in the westerly QBO phase than in the easterly phase for almost the entire winter–spring period; at the Mauna Loa NH station (see Fig. 6b) it takes place only in late spring. In the same months, the NO2 TC in the tropics and SH is larger (while the TOC and stratospheric temperature are lower) in westerly QBO phase when compared to the easterly phase (Fig. 4). Estimates of the amplitudes of the NO2 TC QBVs and differences between evening and morning NO2 TC in westerly and easterly QBO phases indicate that the QBO effect in NO2 TC is stronger in the SH tropics than in the NH tropics. 3.2. Latitudinal Dependence of QBO Effects Using Figs. 1 and 4, one can trace the latitudinal dependence of the QBO effects in NO2 TC, TOC, and stratospheric temperature. For TOC, an increase in the amplitude of the maximum effect (regardless of the month) with the latitude takes place in both hemispheres, dramatically increasing in the polar regions, where it reaches 20% of the annual mean TOC. The amplitude of the maximum effect in temperature also achieves the highest values of 6–7 K near the poles. The character of the latitudinal dependence of the maximum QBO effect in NO2 TC is, in general, dissimilar. In the SH, some decrease in its amplitude can be noticed in a direction from the tropics to the mid-latitudes and an increase in the higher latitudes. Near the South Pole, the amplitude of the maximum effect in NO2 exceeds modulo 20% of the annual mean NO2 TC.
The annual mean values of QBO effects are also of interest. They were calculated on the same principle as the monthly means, but without separation for months. Latitudinal distributions of the annual mean effects are shown in Fig. 7. First and foremost, we note a significant interhemispheric asymmetry of the effects for all the studied variables. In the most general terms, the character of the latitudinal dependences of the temperature and ozone effects is the same: smaller (compared to the NH) variations of the amplitudes of these effects with latitude and well pronounced latitudinal dependence of the amplitude in the NH. There is a significant increase in the amplitude of the ozone effect (in absolute value) with latitude in the Antarctic, with an absolute maximum of ~12 DU near the South Pole. Latitudinal dependence of the annual mean QBO effects in the TOC and temperature in the NH is described by general decrease of modulo amplitudes in the direction from the polar region to the middle latitudes, change of the sign of the effects in the midlatitudes, and an increase in the amplitudes in the lower latitudes toward the tropics. The most striking feature in the latitudinal distribution of the annual mean NO2 TC QBVs is the opposite sign of the effect in the Antarctic and in out-ofpolar latitudes. Distribution of the amplitude modulo in the SH is characterized by a local minimum in the mid-latitudes and an increase in the direction towards the pole and tropics with a maximum value near the South Pole. The latitudinal feature of the annual mean QBO effect on the NO2 TC in Fig. 7 can be characterized by two groups of amplitude values. One group with relatively higher values includes the Ny-Ålesund and Scoresbysund arctic stations of the North Atlantic sec-
IZVESTIYA, ATMOSPHERIC AND OCEANIC PHYSICS
Vol. 53
No. 1
2017
Difference in NO2 TC, 1015 cm–2
SEASONAL FEATURES OF QUASI-BIENNIAL VARIATIONS
NO2
0.2
5
11 15
0.1
14
9 8 7
12
2
1
6
13
0 10
4
3
–0.1 16
–0.2
17
Difference in TOC, DU
–0.3 –80 –60 –40 –20 20 30 40 50 60 70 80 –70 –50 –30 4 O3 0 –4
–16 –80 –60 –40 –20 20 30 40 50 60 70 80 0.8 –70 –50 –30 T Temperature difference, K
amplitudes in the tropical belt. For the temperature effect, there is a significant (almost double) difference in its amplitudes between two similar in altitude tropical stations of the SH. Differences of the QBO effects in the TOC and NO2 TC manifest themselves in largely different amplitudes and different signs of these effects in the tropics of the NH and SH. These differences reflect zonal heterogeneity of the QBO effects in the TOC, NO2 TC, and the temperature in the tropics, because the Mauna Loa, Reunion, and Bauru tropical stations are very different by their longitude (see table). In this regard, we note the existence of the Walker zonal circulation [28] in the tropical stratosphere, and evidence that the QBO modulates intensity of its ascending branches in areas of deep convection [29]. The impact of the Walker circulation on zonal structure of stratospheric processes can be one of the reasons of zonal differences in QBO effects in the tropics identified in this work.
–8 –12
0.4 0
–0.4 –0.8 –1.2 –80 –60 –40 –20 20 30 40 50 60 70 80 –70 –50 –30 Latitude Fig. 7. Difference between the annual mean values of (upper panels) NO2 TC according to evening measurements, TOC (middle panels), and (bottom panels) temperature at 50 hPa isobaric surface, at the westerly and easterly QBO phases (westerly value minus value easterly), as function of a latitude. Vertical segments are 95% confidence intervals. The numbers on the upper plot are the station numbers according to the table.
tor and the East Siberian Zhigansk station. Amplitudes of the annual mean effect on the other stations (that make up the other group) are much lower. The value of the annual mean amplitude of the NO2 TC QBVs in this group increases with decreasing latitude. One common feature of all latitudinal distributions in Fig. 7 is the large difference between the QBVs IZVESTIYA, ATMOSPHERIC AND OCEANIC PHYSICS
73
4. CONCLUSIONS Using results of ground-based spectrometric measurements of the NO2 TC, satellite measurements of the TOC, and data of the ERA-Interim reanalysis, the seasonal dependencies and latitudinal distributions of amplitudes of the quasi-biennial variations of NO2 TC, TOC and stratospheric temperature are derived. The significant interhemispheric asymmetry of the QBO effects in all studied parameters is revealed. Along with this, dependence of the daily component of the NO2 diurnal variation on the QBO phase is revealed, which manifests itself in the difference between the evening and morning NO2 TC values. In the both hemispheres, the QBO effect in the NO2 TC, TOC, and stratospheric temperature was most noticeable in the winter–spring periods. Seasonal dependences of the amplitudes of the NO2 TC, TOC, and CCA and temperature QVBs over the Antarctic are similar. In the spring, the NO2 TC, TOC, and the temperature in the westerly QBO phase are lower than in the easterly phase, which is the result of different intensities of physical and chemical processes responsible for ozone deficit in the Antarctic stratosphere. These processes manifested themselves also in the daytime component of the NO2 TC diurnal variation. For example, the differences between evening and morning NO2 TC values in spring are smaller during the westerly QBO phase than during the easterly phase. Unlike at the Antarctic stations, the QBVs of the NO2 TC in the SH mid-latitudes is in antiphase with the TOC and temperature variations during the winter–spring period. In the westerly QBO phase, the NO2 TC values are higher than in the easterly phase, whereas the TOC and temperature values, on the contrary, are lower in the westerly QBO phase. Likely Vol. 53
No. 1
2017
74
AGEYEVA, GRUZDEV
mechanisms of the NO2 TC QBVs are the quasi-biennial modulation of the Brewer–Dobson circulation and changes in the degree of dynamic isolation of the Antarctic stratosphere (intensity of the polar stratospheric vortex). Weak exchange between the Antarctic and mid-latitude stratosphere with a stronger vortex during the westerly QBO phase promotes higher NO2 TC values in the SH mid-latitudes. It can be assumed that the QBO affects the HNO3 content in a similar way since the HNO3 content affects the NO2 diurnal variation. Seasonal features of the QBO effects in the NO2 TC, TOC, and temperature at the tropical and midlatitude stations of the SH are qualitatively similar. Winter values of the NO2 TC in the NH are usually smaller in the westerly QBO phase, whereas in spring, on the contrary, the NO2 TC in the westerly QBO phase is larger than in the easterly phase. In contrast to the SH, in the winter-spring period there is almost no qualitative differences in the NO2 TC QBVs between the middle and polar latitudes of the NH. This may be associated with a stronger interlatitudinal exchange between the polar and mid-latitude stratosphere in the NH. The impact of the QBO of the equatorial stratospheric wind on the NO2 TC in the polar regions and southern tropics, on the annual average, is much stronger than in the mid-latitudes. The average annual value of the amplitude of the NO2 TC QBVs in the SH have opposite signs in the Antarctic and nonpolar latitudes. Considerable zonal differences in the QBV of the NO2 TC, TOC and stratospheric temperatures in the tropical belt are revealed. ACKNOWLEDGMENTS Data on the NO2 TC used in this work are publicly available on the NDACC website, and the authors are grateful to everyone who contributed directly to the measurements as well as to data processing and storage. We are especially grateful to A.S. Elokhov, V.M. Dorokhov, V.P. Sinyakov, V.K. Semenov, F.V. Kashin†, V.N. Aref’ev, J.-P. Pommereau, F. Goutail, A. Pazmino, M. Van Roozendael, F. Hendrick, M. De Maziere, P.V. Johnston, K. Kreher, S.V. Wood, J.P. Burrows, K. Stebel, and T. Svendby. To obtain the TOC data, we use the Giovanni Internet service created and supported by NASA GES DISC. We also used temperature data of the ERA-Interim reanalysis of the European Center for Medium Range Weather Forecasts (ECMWF) and the data from the Berlin Free University on the velocity of the equatorial stratospheric wind. REFERENCES 1. M. P. Baldwin, L. J. Gray, T. J. Dunkerton, et al., “The quasi-biennial oscillation,” Rev. Geophys. 39 (2), 179– 229 (2001).
2. A. N. Gruzdev and V. A. Bezverkhny, “Two regimes of the quasi-biennial oscillation in the equatorial stratospheric wind,” J. Geophys. Res. 105 (D24), 29443– 29443 (2000). 3. M. P. Baldwin and T. J. Dunkerton, “Quasi-biennial modulation of the southern hemisphere stratospheric polar vortex,” Geophys. Res. Lett. 25 (17), 3343–3346 (1998). 4. J. R. Holton and H.-C. Tan, “The influence of the equatorial quasi-biennial oscillation on the global circulation at 50 mb,” J. Atmos. Sci. 37 (10), 2200–2208 (1980). 5. J. R. Holton and H.-C. Tan, “The quasi-biennial oscillation in Northern Hemisphere lower stratosphere,” J. Meteorol. Soc. Jpn. 60 (1), 140–147 (1982). 6. J. M. Zawodny and M. P. McCormick, “Stratospheric Aerosol and Gas Experiment II measurements of the quasi-biennial oscillations in ozone and nitrogen dioxide,” J. Geophys. Res. 96 (D5), 9371–9377 (1991). 7. A. N. Gruzdev and I. I. Mokhov, “Quasi-biennial oscillations in the total ozone global field from ground– based observations,” Izv., Atmos. Ocean. Phys. 28 (5), 358–366 (1992). 8. M. P. Chipperfield, L. J. Gray, J. S. Kinnersley, and J. M. Zawodny, “A two-dimensional model study of the QBO signal in SAGE II NO2 and O3,” Geophys. Res. Lett. 21 (7), 589–592 (1994). 9. S. A. Sitnov and A. N. Gruzdev, “Analysis of the quasibiennial oscillation in the vertical distribution of ozone and meteorological parameters from ozonesonde data,” Izv., Atmos. Ocean. Phys. 32 (3), 339–351 (1996). 10. A. N. Gruzdev and S. A. Sitnov, “Distinctions between vertical profiles of ozone and meteorological parameters in the phases of the quasi-biennial oscillation (from ozonesonde data),” Izv., Atmos. Ocean. Phys. 33 (1), 82–94 (1997). 11. F. T. Huang, H. G. Mayr, C. A. Reber, et al., “Stratospheric and mesospheric temperature variations for the quasi-biennial and semiannual (QBO and SAO) oscillations based on measurements from SABER (TIMED) and MLS (UARS),” Ann. Geophys. 24, 2131–2149 (2006). 12. E. Kyrölä, J. Tamminen, V. Sofieva, et al., “GOMOS O3, NO2, and NO3 observations in 2002–2008,” Atmos. Chem. Phys. 10 (16), 7723–7738 (2010). 13. A. Hauchecorne, J. L. Bertaux, F. Dalaudier, et al., “Response of tropical stratospheric O3, NO2 and NO3 to the equatorial quasi-biennial oscillation and to temperature as seen from GOMOS/ENVISAT,” Atmos. Chem. Phys. 10, 8873–8879 (2010). 14. A. N. Gruzdev, “Quasi-biennial variations in the total NO2 content,” Dokl. Earth Sci. 438 (2), 837–841 (2011). 15. A. S. Elokhov and A. N. Gruzdev, “Nitrogen dioxide column content and vertical profile measurements at the Zvenigorod research station,” Izv., Atmos. Ocean. Phys. 36 (6), 763–777 (2000). 16. A. N. Gruzdev and A. N. Elokhov, “Validating NO2 measurements in the vertical atmospheric column with the OMI instrument aboard the EOS Aura satellite against ground-based measurements at the Zvenigorod scientific station,” Izv., Atmos. Ocean. Phys. 45 (4), 444–455 (2009).
IZVESTIYA, ATMOSPHERIC AND OCEANIC PHYSICS
Vol. 53
No. 1
2017
SEASONAL FEATURES OF QUASI-BIENNIAL VARIATIONS 17. A. N. Gruzdev and A. S. Elokhov, “Variability of stratospheric and tropospheric nitrogen dioxide observed by the visible spectrophotometer at Zvenigorod, Russia,” Int. J. Remote Sens. 32 (11), 3115–3127 (2011). 18. A. N. Gruzdev, “Weekly cycle in the atmosphere,” Dokl. Earth Sci. 439 (1), 1034–1038 (2011). 19. A. N. Gruzdev, “Estimate of the effects of Pinatubo eruption in stratospheric O3 and NO2 contents taking into account the variations in the solar activity,” Atmos. Oceanic Opt. 27 (5), 403–411 (2014). 20. V. Yu. Ageyeva, A. N. Gruzdev, A. S. Elokhov, and M. V. Grishaev, “Winter–spring anomalies in the stratospheric content of NO2 from ground-based measurement results,” Izv., Atmos. Ocean. Phys. 51 (4), 397–404 (2015). 21. G. Brasseur and S. Solomon, Aeronomy of the Middle Atmosphere (D. Reidel, Dordrecht, 1986; Gidrometeoizdat, Leningrad, 1987). 22. J. Urban, M. Pommier, D. P. Murtagh, et al., “Nitric acid in the stratosphere based on odin observations from 2001 to 2009—Part 1: A global climatology,” Atmos. Chem. Phys. 9 (18), 7031–7044 (2009). 23. J.-P. Pommereau, F. Goutail, F. Lefèvre, et al., “Why unprecedented ozone loss in the Arctic in 2011? Is it related to climate change?,” Atmos. Chem. Phys. 13, 5299–5308 (2013). doi doi 10.5194/acp-13-5299-2013
IZVESTIYA, ATMOSPHERIC AND OCEANIC PHYSICS
75
24. A. J. Charlton and L. M. Polvani, “A new look at stratospheric sudden warmings. Part 1: Climatology and modeling benchmarks,” J. Clim. 20, 449–469 (2007). doi doi 10.1175/JCLI3996.1 25. J. Kuttippurath and G. Nikulin, “A comparative study of the major sudden stratospheric warmings in the Arctic winters 2003/2004–2009/2010,” Atmos. Chem. Phys. 12, 8115–8129 (2012). doi doi 10.5194/acp-128115-2012 26. T. Flury, D. L. Wu, and W. G. Read, “Variability in the speed of the Brewer–Dobson circulation as observed by Aura/MLS,” Atmos. Chem. Phys. 13, 4563–4575 (2013). doi doi 10.5194/acp-13-4563-2013 27. J. C. Gille, P. B. Bailey, and C. A. Craig, “Revised reference model for nitric acid,” Adv. Space Res. 18 (9– 10), 125–138 (1996). 28. Large-Scale Dynamical Processes in the Atmosphere, Ed. by B. Hoskins and R. Pearce (Academic, London, 1983; Mir, Moscow, 1988), pp. 261–305. 29. S. Liess and M. A. Geller, “On the relationship between QBO and distribution of tropical deep connection,” J. Geophys. Res 117, D03108 (2012). doi 10.1029/ 20117DO16317
Translated by I. Ptashnik
Vol. 53
No. 1
2017