Stratospheric Nitrogen Dioxide in Antarctic Regions from Ground Based and Satellite Observation During 2001 Daniele ~ortoli*".~, Giorgio ~iovanelli~, Fabrizio ~ a v e ~ n a nIvan i ~ ,~ostadinov~, Andrea petrholib, Francescopiero ~ a l z o l a r iMaria ~ , Joao Costaa,and Ana Maria Silvaa a CGE-UE, University of Evora, Portugal ISAC-CNR, Via Gobetti, 101,40129 Bologna, Italy ABSTRACT The application of Differential Optical Absorption Spectroscopy (DOAS) methodology to the zenith scattered light data collected with the GASCOD spectrometer developed at the ISAC Institute allow for the detection of stratospheric trace gases involved in the ozone cycle such as NO2, OC10, BrO. The instrument was installed in December 1995 in the Italian Antarctic station at Terra Nova Bay (74"26'S, 164'03E', Ross Sea), after several tests both in laboratory and in Antarctic region, for unattended and continuous measurement in extreme high-latitude environment. The GASCOD is still working and producing very interesting data for the study of the denitrification processes during the formation of the so-called ozone hole over the Antarctic region. For the continuous NO2 monitoring for whole the year, also during winter when the station is unmanned, the [407 - 4601 nm spectral region is investigated. The results for Nitrogen Dioxide, obtained by application of DOAS algorithms to the data recorded during the year 2001, are presented. ERS-2 was launched in April 1995 into a near-polar sun-synchronous orbit at a mean altitude of 795 km. The descending node crosses the equator at 10:30 local time. GOME is a nadir-scanning double monochromator covering the 237 nm to 794 nm wavelength range with a spectral resolution of 0.17-0.33 nm. The spectrum is split into four spectral channels, each recorded quasi-simultaneously by a 1024-pixel photodiode array. The global spatial coverage is obtained within 3 days at the equator by a 960 km across-track swath (4.5 S forward scan, 1.5 S back scan). The ground pixel size of the measurements is 320 X 40 km'. A comparison of GASCOD and GOME results for NO2 total column is performed. Keywords: Ozone, nitrogen dioxide, stratosphere, cross section, satellites, Antarctica.
1. INTRODUCTION When the possibility of stratospheric ozone depletion was first predicted in the 1970s1", it was not expected to occur in the lower stratosphere4.Now we know that it does'-7. We also now know this region to be the major contributor to the observed column ozone changes. This is true both for mid latitudes and for Polar Regions, where the springtime ozone holes are confined to altitudes below 30 km. However. the extent of the observed depletion and the confidence with which it can be quantified vary significantly with location and season. In Polar Regions, ozone depletion is seasonal and well outside the natural variability. It is always associated with cold temperatures, a lack of local ozone production, and a rapid loss of ozone following exposure of polar air parcels to sunlight. Natural variability is however manifested in year-to-year variability in late winterlspring ozone depletion, especially in the Arctic. In contrast, ozone depletion at mid-latitudes can only be identified as a statistical trend after removing the seasonal cycle. In this region one needs to understand the natural variability before quantifying the depletion and, especially, attributing the depletion to known (or unknown) forcing. The lower stratosphere has a number of unique features that are relevant to its ozone abundance: (A) It is far removed from the major ozone production region, the more so the closer one gets to the poles. (B) It is somewhat removed from the regions where chlorine and nitrogen oxides are released from their precursor gases. Thus, transport plays a major role in bringing catalysts, as well as ozone itself, to the lower stratosphere. (C) It is the region that contains most of the sulphate aerosol and is vulnerable to enhanced sulphate loading following volcanic eruptions. (D) It contains the coldest parts of the stratosphere, to the extent that the sulphate aerosol can become highly reactive. Thus, heterogeneous reactions and multiphase reactions play a major role in the chemistry of this region. The high degree of non-linearity in the rates of these reactions with * Further authors information: Daniele Bortoli, ISAC-CNR via Gobetti 101,40129 Bologna, Italy; Tel.: +39-051-6399593;Fax: +39-05 1-
6399652; E-mail: [email protected]
Remote Sensing of Clouds and the Atmosphere VII, Klaus Schafer, Olga Lado-Bordowsky, Adolfo Comeron, Richard H. Picard, Editors, Proceedings of SPlE Vol. 4882 (2003) 0 2003 SPlE 0277-786)(/03/$15.00
temperature also makes temperature, and temperature fluctuations, very important. (E) In the extra-polar upper stratosphere, outside the polar night, a balance between chemical production and destruction of ozone is established much faster than air can be transported from one region to another. However, in the lower mid-latitude stratosphere, there is no clear separation of the time scales for chemistry and transport. (F) In the Polar Regions, radiative cooling together with the dynamics of air motions sets up a vortex during winter and early spring. This polar vortex and its immediate outer surroundings provide very cold temperatures to produce condensed matter (solid and liquid polar stratospheric clouds (PSCs)) that is generally not possible in other parts of the stratosphere. At the same time - and for related reasons - the vortex also inhibits the transport of air from the inside to the outside of the vortex, and vice versa. Thus, the coldest parts of the stratosphere can sustain large concentrations of active species. Some of them that, trough active reactions with other tracers, primarily drive the ozone chemical cycle are nitrogen compounds. Chemical interactions of odd nitrogen constituents with other trace species are important in oxidation processes throughout the Earth's atmosphere. Active nitrogen constituents, primarily NO and NOz, are necessary for the photochemical production of O3 in the troposphere. In contrast, in the stratosphere, the emphasis of odd nitrogen chemistry is on its ability to destroy O3 in so-called catalytic cycles. The combination of these destruction cycles with those involving odd chlorine and bromine, odd oxygen, and odd hydrogen determines the photochemical balance of stratospheric 03. Since observed and very large depletions of O3 in austral spring over Antarctica have been of headline importance, and the current downward trend of stratospheric O3 globally is on record, it is clearly important to understand in detail the chemical interactions of odd nitrogen with other trace species. NO2 can be oxidized by O3 to form No3, a strong atmospheric oxidant and a precursor to the formation of dinitrogen pentoxide, N2O5. Because NO3 is rapidly photolyzed at visible wavelengths, both its daytime concentration and chemistry are of relatively minor importance in the lower and middle stratosphere. In contrast, N2O5can have morning concentrations in the lower stratosphere comparable to NO, (NO+NOd even though its production requires the formation of NO3
The explanation results primarily from the difference in photolysis rates of N2O5and NO3. During the night N2OSbuilds at the expense of NO, or more precisely at the expense of NO? After sunrise, N205can be photolyzed to essentially two NO, molecules
but the photolysis time constant is dependent on both the usual factors (zenith angle, altitude, albedo, etc.) and the ambient temperature. In summer typical photolysis times range from 7-24 h in the 20-30 km region. The formation of N205is an example of formation of a long-lived odd-nitrogen reservoir that occurs at night or in Polar winter. Other important reservoir constituents are bromine nitrate (BrON02) and the chlorine analog (ClONO*), nitric acid (HN03) and peroxynitric acid (H02N02).The homogeneous formation of the reservoir species requires the molecular chain termination reactions (1.6-1.9).
The occurrence of these reactions not only curtails O3destruction through temporary removal of active NO,, but they also sequester HO,. and CIO,, radicals and thus decrease the contribution of their catalytic cycles to O3 destruction. Alternatively, any other stratospheric process that diminishes NO,, directly or indirectly will enhance the abundance of active HO,. and CIO,. by decreasing the rates of (1.7) (1.9) and therefore enhance their ozone destruction efficiency. Instruments on board satellitess. (MLS on UARS, GOME on ERS2 etc.), have extended the in situ (aircraft and balloon) and remote sensing (ground based equipment) observations1° of tracers in the stratosphere, but this last kind of measurements, even if more limited in space, are really useful to verify the results obtained for the same atmosphere from another point of view. A relatively new addition to atmospheric remote sensing is GOME, launched on ERS-2 in April 1995. The GOME instrument is an ultraviolet-visible-near-infrared spectrometer similar in heritage to TOMS, but with superior spectral Proc. of SPlE Vol. 4882
resolution. The greater spectral resolution has allowed retrieval of several products not previously observed using spaceborne in~trumentation'~. Since deployment, GOME has proven suitable for the generation of long-term datasets of trace atmospheric constituents, a series soon to be augmented with the launches of SCIAMACHY on ENVISAT and GOME-2 on METOP-1. (SCIAMACHY is a GOME-type instrument with an increased spectral range into the IR spectral region.) In this work we analyse observations of the light scattered from the zenith-sky that were obtained with a ground-based WVisible system (GASCOD-Gas Analyser Spectrometer Correlating Optical ~ifferences)'~, developed at ISAC - CNR and installed in December 1995 in the Italian Station at Terra Nova Bay -TNB- (74"26'S, 164"03E', Ross Sea). By application of DOAS technique to the solar spectra it was possible to extract the tracer contents along the optical path depending on sun elevation - usually called slant column (sc) - for ozone and other gases involved in the ozone cycle. Some of the results obtained during the 2001 will be presented. In addition a comparison of GOME and GASCOD results is performed, regarding the column values of N02.
2. ANALYSIS The ratio of the signal obtained by the spectrometer during a measurement and the "exposure" time" '(the Flux Index' - FI) was calculated for the whole period of measurements (Figure 1). The calculated values shown that during the year many cloudy days were encountered. In fact on a cloudy day the FI can increase up to 4-5 times caused by the multiple scattering mainly due to tropospheric cloud^'^. No episodes of snow accumulation on the quartz window placed to protect the GASCOD radiation input are evident. In this case F1 values lower than expected should be present.
W 1 1/01
Date Figurel. Flux Index obtained at local noon during 2001 at Terra Nova Bay station.
The flow index has been much profit in order to check the presence of cloudy days, snow accumulation and technical problems occurred to the equipment during the winter when the station is closed.
Proc. of SPlE Vol. 4882
TNB local solar time Figure2. FI on 18/03/2001 and 19/03/2001.A 'one hour shift' is found in the measurements header time.
A complete explanation of Differential Optical Absorption Spectroscopy (DOAS) technique can be found in references19.20. In this work a software package developed by the authors" is employed for the application of DOAS algorithms. The use of advanced mathematical techniques as Singular Value Decomposition (SVD), Marquardt method and ANalysis Of VAriance (ANOVA) tables, allowed for a very fast computation (in average 30 times faster than the previous software package used by the group). The analysis and interpretation of the results will be the subject of the next paragraph.
3. OBSERVATIONS AND DISCUSSION All the following results are presented as vertical or total column (VC).The NO2 VC are obtained applying to the measured column (slant column) an Air Mass Factor (AMF) computed with the AMEFCO~' RT model. In order to avoid bias due to the seasonal dependency of the vertical profile and hence in the AMF, different boundary condition are applied to the model with the aim to distinguish summertime and wintertime air masses. The value at 90 degrees of Solar Zenith Angle (SZA) is calculated with a cubic interpolation of the data, because the GASCOD is not able to take measurements at a fixed angle. The errors associated to the presented results are in the range of 5-8% for the VC at 90". Values with larger errors, due to very low signal intensity, are rejected. In the plots the errors bars are omitted for clearness. A clear seasonal trend emerges from the results obtained for the whole period of activity of the GASCOD spectrometer at TNB station (Figure 3). During the austral autumn the NO2 VC values decrease from a value of about 5.OE1.5 mol/cm2 to less than 1.OE15 mol/cm2. At the end of the polar night the nitrogen dioxide total columns increase with a different slope starting from more or less 1.OE14 mol/cm2. Both features are due to the know processes of the lower stratospheric denoxification (conversion of NO and NO2 into HN03) in the winter season. In other words: the length of the daylight strongly affects the NO2 total columns that are also influenced by the stratospheric temperature. The small pre- and post-winter column amounts are the result of gas phase reactions converting NO2 in N205, heterogeneous reactions converting N2O5 and ClONO? into HN03, and possible denitrification (that is sedimentation of HN03 particles leading to reduced NO,). In summer months the lifetime of HN03 is reduced by photolysis and reactions with OH, both releasing N02.The continuos sunlight in summer inhibits the formation of N2O5,thus reducing the diurnal variation. In winter there is no daylight to photolyze N2Osrwhich similarly reduces the diurnal variation.
Proc. of SPlE Vol. 4882 307
For the austral spring it is also possible to note some inversions in AM and PM values and more scattered results, due to the TNB geographical location that cause the station to be inside the vortex in some periods, while in other periods it is outside. In this situation the dynamics effects hide the usual photochemistry
- - - GASCOD VC 90 AM-NO2
Figure 3. NO2 vertical column amounts measured by GASCOD at Terra Nova Bay station (1996-2001),during morning and evening
twilight. The evening measurements are largest in spring and autumn, when morning and evening twilight are well separated. The correlation between morning - evening twilight NO2 VC and stratospheric temperature is quite remarkable (Figure 4), to confirm the strong relation between temperature and nitrogen dioxide total amount. . During the austral autumn the NO2 VC amounts follow the temperature trend lying below it, while this behaviour is opposite in the spring also for the lowest temperatures that occur in the second part of the year. Moreover, it is during the austral spring that the stronger effects of ozone depletion occur. Considering this last sentence it is also possible that the variations in NO2 are related to polar stratospheric clouds (PSC), whose abundance and variability are closely tied to temperature fluctuations. Also the dynamical processes play an important role (see local transport of air masses depleted in NO2 to Terra Nova Bay). Only with Lagrangian observations, which follow an air parcel as a function of time, the detailed information's about the time rate of change of NO2 can be obtained. However, the Eulerian measurements presented here can provide some indications. The analysis of meteorological data as temperature. potential vorticity and geopotential height, suggests that TNB station was located inside the polar vortex on days 09/07 - 05/08, 15/08 - 20/08 and 30108 - 09/09. In these days the temperature at 20 hPa reached values lower than -80°C and minimum values for NO2 VC are obtained. We have to note that after the lothof September, NO2 starts its recovery to the summer levels in a good correlation with the variation in the stratospheric temperatures. For the 6-9 August, the remarkable increase and inversion in the AM and PM values is imputable to the contemporaneous exit of TNB from the Polar Vortex (as also suggested by the Potential Vorticity analysis) and to an air parcel transport from the lower latitudes. The same for the 'jump' in the sunset NO2 VC value, located on the 27 of August, when the total column amount is 6.4E14 mol/cm2, but without the dynamic contribution of air masses transport. 308
Proc. of SPlE Vol. 4882
--+NO2 PM VC, SZA = 90"
, h , 8
Date Figure 4. Evening and morning twilight observation of NO2 along with 20 hPa temperatures at Terra Nova Bay station during 2001.
NO2 AM VC, SZA = 90" NO2 PM VC, SZA = 90"
Date Figure 5. Potential Vorticity at 500 K and twilight NO2 VC
Proc. of SPlE Vol. 4882
The Potential Vorticity is a very good indicator of the presence of the polar vortex at a fixed altitude. Figure 5 shows the correlation between the Potential Vorticity at 500 K and twilight NO2 VC.We can note that for some days the NO2 AM values follow the PV shape better than the PM values, while in the following period there is the inverse behaviour. This aspect can be explained considering phenomena of injection of air parcel from different altitude.
-o- Photochemical Model
7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70 73 76 79 82 85
Day of year 2001 Figurd. R (NO2 A W M ratio) calculated and measured at Terra Nova Bay Station during the austral autumn 2001.
During austral fall the NO2 diurnal variation at a given altitude is mainly driven by temperature, ozone concentration and night length as the variations take place from N205photodissociation created at night trough reactions involving ozone. A simple photochemical model [Gil et al., 19921 can reproduce the ratio R (AMIPM) by the formula:
R = exp ( K1  At ) with K1= 1.2 x10-l3exp(-2450ff). Where T is temperature at NO2 bulk altitude,  ozone concentration at the same level, assumed in this paper from climatologic mean, and At the night duration (Figure 6). The discrepancies between model output and experimental results can be attributed mainly to the variations in not accounted for by the climatologic mean. At Terra Nova Bay station R is comprised between 0.5 (during the equinoxes) and 1 (close to the solstices). The small diurnal variation during the solstices reflects the short time between sunset and sunrise. Ground-based measurements collocated in time and space with the satellite ones are of great importance in estimating the real accuracy of satellite and ground based NO2 data. Comparison of GOME NO2 total content measurements in 2001 with the simultaneous ground-based observations at TNB station was accomplished (Figure 7). The satellites and the ground-based spectrometers capture major stratospheric features similarly. This is truer in the Antarctic regions where low tropospheric NO2 contents occur. Although it is difficult to evaluate precisely the accuracy of the NO2 total column due to various problems such as the diurnal variation of NO2 and the profile shape effect on the Air Mass Factor (AMF), the overall accuracy jn areas of low tropospheric NO2 is estimated to fall within the 5% to 20%
Proc. of SPlE Vol. 4882
-.a.. GASCOD VC 90 AM-NO2
VC 90 PM-NO2
Figure 7. T i e series of GASCOD and GOME NO2 VC values for 2001 at Terra Nova Bay station
GOME NO2 VC
U GASCOD NO2 VC
2E+15 3E+15 4E+15 GOME NO2VC [moleculelcm2]
Figure %Scatter plot of GASCOD and GOME NO2 VC values Proc. of SPIE Vol. 4882
range. GOME GDP total NO2 is affected by larger errors under particular circumstances, e.g., over polluted areas or during midnight sun as evinced from the last days of measurements. The results in our ground-based measurements are in good agreement with the GOME total nitrogen dioxide especially during the autumn when the number of cloudy days is very low as we can see with the Flux-Index (Figure 1). Figure 8 shows the good correlation of the 2 data set. A positive correlation of r = 0.97 has been found between GOME NO2 VC and GASCOD NO2 VC. It is meaningful that the two series of data are in good agreement also for rather low values of concentration (of about 1.OE15 molec/cm2)
4. CONCLUSIONS The preliminary results for year 2001 obtained by DOAS processing of data carried out by GASCOD spectrometer installed since 1995 at TNB station are presented. No evident problems occur to the optical equipment during the Antarctic winter. The seasonal trend of NO2 VC values is reported and it shows the expected behaviour: maximum values at the end and at the beginning of the summer period while the minimum occur in the winter season. For year 2001, the NO2 VC fit very well both the stratospheric temperatures and the PV values at 500 K. The application of a simple photochemical model confirms that the NO2 AMPM ratio can be modelled with the temperature at NO2 bulk altitude, the ozone concentration at the same level and the night length. The comparison of the ground-based measurements with the GOME total nitrogen dioxide highlight a good agreement of the 2 data set, especially during the autumn The good correlation is remarked by a r = 0.97 that has been found between GOME NO2 VC and GASCOD NO2 VC,also including low values of concentration close to instrumental detection limit.
ACKNOWLEDGMENTS The National Antarctic Research Program (PNRA) supported this research. The author Daniele Bortoli was financially supported by the Subprograma CiCncia e Tecnologia do 3" Quadro Comunitfirio de Apoio. PV data are provided by NILU (database), temperature data from NCEPNCAR and GOME data from Steffen Beirle of the Heidelberg University.
REFERENCE 1. Crutzen P.J. 'The influence of nitrogen oxides in the atmospheric ozone content", J. Roy. Met. Soc. 96, pp. 320325,1970. 2. Johnston J.S. "Reduction of stratospheric ozone by nitrogen oxide catalysis from supersonic transport exhaust", Science 173, p.517, 1971. 3. Nicolet, M. "Stratospheric Ozone: an introduction to its study", Rev. Geophys. Space Phys., 13, p. 593, 1975 4. "Scientific assessment of ozone depletion:1995", World Meteorological Organization , 1996. 5. Cunnold D.M., H.Wang, W.Chu and L.Froidevaux "Comparison between SAGEII and MLS ozone measurements and abasing of SAGE I1 ozone trend in the lower stratosphere" J. Geophys. Res. 101, pp. 10061-10065, 1996. 6. M.L. Chanin "Stratospheric Processes and their Role in Climate newsletter" 13, Ed. SPARC Office, Meteo-France, July 1999 7. "Scientific assessment of ozone depletion: 1998", World Meteorological Organisation, 1999. 8. T.Wagner, C.Leue, K. Pfeilsticker and U. Platt: "Monitoring of stratospheric chlorine activation by global Ozone Monitoring Experiment (GOME) OclO measurements in the austral and boreal winters 1995 through 1999" J. Geophys. Res. 106, pp. 4971-4986,2001. 9. Santee, M.L., W.G. Read, J.W.Waters, L.Froidevaux, G.L.Manney, D.A. Flower, R.F. Janot, R.S. Harwood and G.E.Peckharn "Interhemispheric differences in Polar stratospheric HN03, H20, C10 and 0 3 from UARS MLS" Science 267,849, 1995 10. R.W. Sanders, S. Solomon. K. Kreher and P.V. Johnston, "An Intercomparison of NOz and OClO Measurements at Arrival Heights, Antarctica during Austral Spring 1996", J. of Atmospheric Chemistry 33, pp. 283-298, 1999. 11. Otten, C., F. Ferlemann, U. Platt, T. Wagner and K. Pfeilsicker. Ground based DOAS UVNisible measurements at Kiruna (Sweden) during the SESAME winters 1993194 and 1994195. J. ofAtmospheric Chemistry 30, pp. 141-162, 1998. 12. Sarkissian, A., G. Vaughan, H. K. Roscoe, L. M. Barlett, F. M. O'Connor, D. G. Drew, P. A. Huges, and D. M.Moore: "Accuracy of measurements of total ozone by a SAOZ ground-based zenith sky visible spectrometer", J. Geophys. Res. 102, pp. 1379-1390,1997. 312
Proc. of SPlE Vol. 4882
13. Oelhaf, H., G.Wetze1, T. von Clarmann, M. Schmidt, "Corelative balloon measurements of the vertical distribution of N20, NO, NO?, NO3, HN03, N205, ClON02 and total reactive NO, inside the polar vortex during SESAME", Polar Stratospheric Ozone, pp.187-192, Ed. J.A.Pyle, N.R.P. Harris, and G.T. Arnanatidis, Schliersee, Bavaria, Germany, 1995. 14. Pfeilsticker, K. and U. Plat: "Airborne measurements during the Artic stratospheric experiment: Observation of 0 3 and NOT, Geophys.Res.Lett. 21, pp. 1375-1378, 1994 15. Burrows J.P., M. Weber, M. Buchwitz, V.V. Rozanov, A. Ladstiidter-Weissenmayer, A. Richter, R. de Beek, R. Hoogen, K. Bramstedt, K.U. Eichmann, M. Eisinger und D. Perner, The Global Ozone Monitoring Experiment (GOME): Mission Concept and First Scientific Results, J. Atm. Sciences, 56, 151-175, 1999. 16. Evangelisti F., A. Baroncelli, P. Bonasoni. G. Giovanelli and F. Ravegnani. "Differential optical absorption spectrometer for measurement of tropospheric pollutants", App. Opt. 34, 1995. 17. Bortoli, D., F. Ravegnani, G. Giovanelli, Iv. Kostadinov, A. Petritoli. "Stratospheric Nitrogen Dioxide observations at mid-and high latitude performed with ground-based spectrometers", SPIE Conf. Proc. 4168, 297-308,2000 18. Erle, F. K. Pfeilsicker and U. Platt, "On the influence of stratospheric clouds on zenith-scattered-light measurements of stratospheric species", Geophys.Res.Lett. 22, pp. 2725-2728, 1995. 19. Noxon J.F., "Stratospheric NO2, Observational Method and behaviour at mid latitude", Geophys. Res. Left. 84, pp. 5047-5065,1979 20. U.Platt "Differential Optical Absorption Spectroscopy (DOAS)" in Air Monitoring by spectroscopic Techniques, Ed W. Sigrist, Chemical Analysis Series, 127, 22-85, 1994 21. Bortoli, D. "Spettroscopia ad Assorbimento Ottico Differenziale. Analisi di misure effettuate a1 Circolo Polare Artico" Graduation Thesis in Physics, Bologna University, 1998 22. Petritoli, G. Giovanelli, P. Bonasoni, T. Colombo, F. Evangelisti, U.Bonafe, D. Bortoli, Iv. Kostadinov and F. Ravegnani. "Ground Based NO2 and O3 Analysis at Mt. Cimone Station during 1995-1996: a case study for spring 1995 NO2 concentration profile", in Spectroscopic Atmospheric Monitoring Techniques, K.Schafer, ed., Proc. EUROPTO 3867, pp. 280-289,1999.
Proc. of SPIE Vol. 4882