Investigation of the stratospheric inorganic bromine budget for 1996-2000: balloon-borne measurements and model comparisons

I N A U G U R A L - DISSERTATION

Investigation of the Stratospheric Inorganic
Bromine Budget for 1996-2000: Balloon-Borne
Measurements and Model Comparisons

DISSERTATION
submitted to the
Combined Faculties for the Natural Sciences and for Mathematics
of the
Rupertus Carola University of Heidelberg, Germany

for the degree of
Doctor of Natural Sciences
Investigation of the Stratospheric Inorganic Bromine Budget for 1996-2000:
Balloon-Borne Measurements and Model Comparisons

presented by

Diplom-Physicist: Richard Fitzenberger

Heidelberg, 8.November 2000

Zusammenfassung
Anorganische Bromverbindungen spielen eine bedeutende Rolle in katalytischen Ozonabbauzyklen der Stratosphäre. Im Rahmen dieser Doktorarbeit wurden ballongetragene DOAS (Differentielle Optische Absorptions Spektroskopie) BrO Vertikalprofilmessungen mittels direkten Sonnenlichts mit bisher nicht erreichter Genauigkeit (+/-12%) durchgeführt. Die spektroskopischen Messungen fanden während acht erfolgreicher Flüge mit der gemeinsamen deutsch-französischen Nutzlast LPMA/DOAS unter sehr verschiedenen geophysikalischen Bedingungen in den Jahren 1996-2000 statt. Dabei ergaben sich völlig neue Einblicke in die Chemie und das Budget des stratosphärischen Broms. Die neuartigen Erkenntnisse umfassen (1) eine genaue und vollständige Erfassung des stratosphärischen Bromgehaltes der letzten vier Jahre, (2) ein verbessertes Verständnis der Chemie des stratosphärischen anorganischen Broms, und (3) den erstmaligen Nachweis und die Messung von Höhenprofilen der BrO Konzentration in der freien Troposphäre. Mit Hilfe der BrO Messungen konnte der stratosphärische Gehalt an anorganischem Brom zu Brⁱⁿ_y =(21.5+/-3) ppt in 5.6 Jahre alter Luft für 1999 bestimmt werden. Hingegen zeigt die erstmalige gleichzeitige Bestimmung des  stratosphärischen Bromgehaltes aus Messungen organischer Bromverbindungen nur Br^org_y =(18.4+/-1.8) ppt. Die Übereinstimmung des Gesamtbrommischungsverhältnisses ist befriedigend, jedoch deutet das konsistent größere Brⁱⁿ_y darauf hin, daß vermutlich anorganisches Brom (3.1+/-3.5 ppt) aus der Troposphäre in die Stratosphäre eingetragen wird. Diese Vermutung wird durch den Befund (3) und den kürzlich erfolgten Nachweis von im Aerosol der oberen Troposphäre gebundenem, anorganischem Brom von etwa 1 ppt qualitativ gestützt.

Als weiterer Teil der Arbeit wurde ein Algorithmus zur Auswertung der vom Ballon aus gemessenen Sonnenspektren für den Nachweis von Chlordioxid (OClO) neu entwickelt. OClO konnte in allen arktischen Winterflügen auch bei geringer Chloraktivierung detektiert werden. Überraschenderweise wurde OClO auch während eines Ballonfluges über Spanien im Herbst 1996 nachgewiesen (5-10 ppt in 20-30 km Höhe bei einem Sonnenzenitwinkel von 88-93o). Zur Interpretation der erhöhten OClO Werte wurden Ergebnisse des 3-D CTM Modells SLIMCAT und des dafür eigens entwickelten Lagrange Boxmodells LABMOS mit den Messungen verglichen. Dabei stellte sich heraus, daß sich die arktischen OClO Messungen im Winter gut verstehen lassen, während die in mittleren Breiten gemessenen OClO Konzentrationen deutlich höher sind als mit der bisherigen Theorie der stratosphärischen Chemie erklärt werden kann.

Summary
Inorganic bromine plays an important role in catalytic ozone depletion in the stratosphere at high and mid-latitudes. This study reports and discusses in detail stratospheric DOAS (Differential Optical Absorption Spectroscopy) BrO vertical profile measurements with direct sunlight, that were conducted aboard the french-german LPMA/DOAS (Laboratoire de Physique Moléculaire et Application) balloon gondola at mid and high northern latitudes during the years 1996 to 2000. The unprecedented high accuracy (+/-12%) of the BrO measurements together with 3-D CTM and Lagrangian photochemical modelling provided new insights into the budget and chemistry of atmospheric bromine. These include (1) a thorough investigation of the present and recent budget of stratospheric bromine, (2) an improved understanding of the chemistry of stratospheric bromine, and (3) the first measurements of free tropospheric BrO concentration profiles. The total inorganic bromine (in 5.6 year old air in 1999) determined from BrO measurements (Brin y =(21.5+/-3) ppt) was tested for the first time against the total stratospheric bromine inferred from the organic bromine method (Brorg y =(18.4+/-1.8) ppt). The agreement of total bromine found with both methods is good, however, the consistently larger Brin y suggests an influx (3.1+/-3.5 ppt) of inorganic bromine from the troposphere. This conclusion is supported by finding (3) and qualitatively by the recent detection of inorganic bromine at a 1 ppt level in the upper tropospheric aerosol. Thus this thesis could establish a more complete budget of stratospheric bromine. Also, a new DOAS retrieval algorithm was developed for the detection of chlorine dioxide (OClO) in the balloon-borne direct sun spectra. In all Arctic winter flights OClO could be detected - even at low chlorine activation. Surprisingly, OClO could also be measured in significant amounts (5-10 ppt at 20-30 km and a solar zenith angle (SZA) of 88-93o) during a fall 1996 mid-latitude balloon flight. The OClO measurements were inter compared with results from a 3-D CTM (SLIMCAT) and a Lagrangian trajectory box model (LABMOS), which was implemented for this purpose as a part of this study. While the OClO detected at high latitudes during winter corresponds well to the model predictions, the OClO detected at mid-latitudes cannot be explained by the presently known stratospheric photochemistry.

Contents

1 Introduction ... 1

2 Halogen species and their importance in atmospheric chemistry ... 5
2.1 Stratospheric Ozone ... 5
2.2 Tropospheric Ozone ... 7
2.3 Stratospheric gas phase chemistry related to ozone ... 9
2.3.1 Chapman Chemistry ... 9
2.3.2 Catalytic Cycles ... 10
2.3.3 Nitrogen chemistry in the stratosphere ... 11
2.3.4 Halogen chemistry in the stratosphere ... 13
2.4 Heterogeneous chemistry on PSCs leading to the Ozone Hole ... 17
2.4.1 Heterogeneous chemistry on sulphate aerosols ... 22
2.5 Fundamental Stratospheric Dynamics ... 24
2.6 The atmospheric halogen budget ... 27

3 Measurement Technique : Direct Sunlight Balloon Borne DOAS 
(Differential Optical Absorption Spectroscopy) ... 33
3.1 Solar Radiation and the Solar Spectrum ... 33
3.1.1 Interaction of light with matter ... 33
3.1.2 Lambert-Beer′s Law - Optical Absorption Spectroscopy ... 35
3.2 Differential Optical Absorption Spectroscopy (DOAS) ... 37
3.3 The DOAS double spectrograph for balloon-borne measurements ... 38
3.3.1 Noise sources of the measurements ... 42
3.4 The LPMA/DOAS balloon payload ... 43
3.4.1 The behaviour of the DOAS spectrograph during the balloon flights and its impact on the BrO evaluation ... 46
3.4.2 The BrO DOAS evaluation ... 50
3.4.3 The OClO DOAS evaluation ... 55
3.5 Determination of the SCD offset in the Fraunhofer reference - Langley Plot ... 55
3.6 Summary of the error sources of the bromine oxide SCD measurements ... 58
3.7 Profile Retrieval ... 58
3.7.1 Raytracing ... 59
3.7.2 AMF matrix inversion ... 59
3.7.3 Errors of the inversion technique ... 61
3.7.4 Differential Onion Peeling technique ... 61
3.8 Modelling of SCDs ... 62

4 Results and Discussion of the LPMA/DOAS balloon flights ... 65
4.1 The sunset flight at León on November 23, 1996 ... 65
4.1.1 BrO profile and SCD model comparison ... 72
4.1.2 OClO profile and SCD model comparison ... 75
4.2 The sunset flight at Kiruna on February 14, 1997 ... 80
4.2.1 BrO profile and SCD model comparison ... 81
4.2.2 OClO profile model comparison ... 87
4.2.3 O3 profile model comparison ... 93
4.2.4 NO2 profile model comparison ... 94
4.2.5 Summary of the model comparison ... 95
4.3 The sunrise flight at Gap on June 20, 1997 ... 95
4.3.1 BrO profile and SCD model comparison ... 97
4.4 The sunset flight at León on March 19, 1998 ... 100
4.4.1 BrO profile and VCD comparison with GOME ... 105
4.5 The sunset and sunrise flight at Kiruna on August 19/20, 1998 ... 109
4.5.1 BrO profile and SCD model comparison for the sunset ... 112
4.5.2 BrO SCD model comparison for the sunrise ... 121
4.6 The sunset flight at Kiruna on February 10, 1999 ... 121
4.6.1 BrO profile and SCD model comparison ... 124
4.6.2 OClO profile and SCD model comparison ... 133
4.7 The sunrise flight at Gap on June 25, 1999 ... 137
4.8 The sunset flight at Kiruna on February 18, 2000 ... 141
4.8.1 BrO profile and SCD model comparison ... 144
4.8.2 OClO profile and SCD model comparison ... 149
4.9 Summary of BrO measurements during the eight LPMA/DOAS balloon flights ... 155
4.9.1 BrO profile measurements ... 155
4.9.2 BrO VCD comparisons with satellite and ground-based instruments ... 156

5 The first measurement of a BrO profile in the free troposphere ... 159
5.1 Methodology and Measurements ... 160
5.2 Discussion of the free tropospheric BrO measurements ... 163

6 Comparison of the inorganic and organic bromine budget for the Arctic lower stratosphere in winter 1998/1999 ... 167
6.1 Methodology of the comparison ... 168
6.2 Discussion of the comparison ... 169
6.3 A recent history of total organic and inorganic stratospheric bromine ... 173

7 Lagrangian case studies for the interpretation of enhanced OClO  measurements at mid and high latitudes ... 175
7.1 The Lagrangian trajectory box model LABMOS ... 175
7.2 Case study of the in-vortex flight at Kiruna on February 10, 1999 ... 177
7.3 Case study of the out-of-vortex flight at León on November 23, 1996 ... 187

8 Conclusions and Outlook 197

A Appendix I
A.1 Meteorological Definitions ... I
A.2 Potential Temperature ... I
A.3 Ertl´s potential vorticity ... I
A.4 Geopotential Height ... II

B Chemical Reaction Rate Constants ... III 
B.1 Bimolecular Gas Phase Reactions ... III
B.2 Trimolecular Gas Phase Reactions ... V
B.3 Photochemical reactions ... VII
B.4 Heterogeneous reactions ... VIII
B.5 Concentration change because of changing volume along the trajectory ... IX

Bibliography xi

C Danksagung xxi

Chapter 1 

Introduction The chemical composition and dynamics of the atmosphere are vital for human, animal, and plant life on Earth. On a global scale, the system atmosphere-biosphere-ocean is in an equilibrium steady-state, where small short-term variations of the climate and the chemical composition of the different compartments are smoothed by the relatively long transport processes and therefore generally of small amplitude.

Since the beginning of industrialisation mankind has released large amounts of gases into the atmo- sphere and influenced thereby the natural cycles of material transport between the different reservoirs. One of the most important examples of this influence was the discovery of the Antarctic ozone hole by Farman et al. [1985], where nowadays at some altitude levels up to 100% of the stratospheric ozone can be depleted within a few weeks simultaneously reducing the total column of ozone by more than 70%. This observation was not expected and provoked a lot of concern among scientists and politicians, because the stratospheric ozone layer is protecting life on Earth against harmful UV radiation1. Research on this phenomenon led to the installation of a series of international political agreements on the ban of ozone depleting species like the CFCs, halons and halogenated hydrocarbons [WMO 1998].

Concern that some halogenated hydrocarbons may destroy ozone in the upper atmosphere was first raised in the 1970s [Molina and Rowland 1974; Stolarski and Cicerone 1974] predicting a global ozone reduction of 10-20% during the next 50-100 years. Acknowledging the realisation that these man-made chemicals are threatening the ozone layer, the Vienna Convention for the Protection of the Ozone layer was adopted by 28 countries in 1985. This committed signatories to a general obligation to take appropriate actions to protect the ozone layer and to co-operate on research. Shortly after the Vienna meeting, the ozone hole over the Antarctic was discovered, reinforcing the pressure to control potential ozone depleting substances. In 1987, the Montreal Protocol on Substances that Deplete the Ozone Layer was agreed and has since been ratified by over 160 countries. Initially, the Protocol imposed clear limits on the future production of CFCs and halons only and committed Parties to cutting down production by 50% by the year 2000. There were several amendments to the Montreal Protocol (London, 1990; Copenhagen, 1992; Montreal, 1997) thus strengthening the control of ozone depleting substances. The phase-out of substances already regulated was accelerated, and other chemicals found to cause ozone depletion were included, i.e. carbon tetrachloride, methyl chloroform, HCFCs and methyl bromide (CH3Br).

Because of the long lifetime of CFCs and halons (′organic′ F, Cl and Br) in the atmosphere, these species can reach the stratosphere where they are photolysed by sunlight (or attacked by chemical radicals like OH or O), thereby setting free active2 chlorine and bromine, which reacts nearly instantaneously with ozone. While the political actions taken world-wide do already show a decrease of the chlorine loading of the lower atmosphere [WMO 1998], the bromine containing halons are still increasing. As bromine and chlorine chemistry in the stratosphere are coupled together, the impact of bromine on the ozone budget is largest, where chlorine is activated - like in the Arctic and Antarctic winter. There have been a series of intensive measurement field campaigns during the last 10 years also studying the Arctic ozone layer and the occurrence of an Arctic ozone hole(e.g. EASOE, SESAME, THESEO, THESEO2000-EuroSolve)3 , which was observed during 3-5 of the Arctic winters in the 1990s [Goutail et al. 2000].

Although the abundance of active bromine is about 200 times less than the one of active chlorine, the efficiency of the bromine related catalytic cycles can be as high as the efficiency of the chlorine cycles. This is due to the lower stability of the bromine reservoir species in comparison with the chlorine reservoir species. During daytime inorganic stratospheric bromine is mainly present in the form of BrO, so that the balloon-borne DOAS4 measurement of BrO is an efficient method to investigate the inorganic bromine budget of the stratosphere. Using model studies to predict the behaviour of the measured species during the time of the measurements allows to derive the total amount of inorganic bromine [Fitzenberger et al. 2000; Fitzenberger et al. 2000; Harder et al. 2000; Pfeilsticker et al. 2000]. Thus the eight LPMA/DOAS5 balloon flights conducted between 1996 and 2000 provide the unique possibility to measure total inorganic bromine (Brin y ) directly and compare it to the increase of organic precursors in the troposphere and lower stratosphere.

There are different methods to measure BrO in the stratosphere. The in-situ chemical conversion resonance-fluorescence instruments measure BrO indirectly within a certain air sample, while the remote sensing DOAS instruments (Zenith Scattered Light (ZSL) [Platt et al. 1997], direct sunlight DOAS from ground, satellite or balloon platform [Ferlemann et al. 2000]) directly measure the absorption bands in the UV part in the sunlight associated to BrO. The in-situ resonance-fluorescence technique performs well only in the stratosphere, while under certain conditions direct sunlight and scattered light DOAS can be sensitive to the troposphere and lower stratosphere [Friess et al. 1999; Fitzenberger et al. 2000; Fitzenberger et al. 2000]. The remote sensing instruments using zenith-sky or other scattered sunlight apply the same absorption technique as balloon-borne direct sunlight DOAS, but suffer from a more complicated radiative transport in the atmosphere, which has to be well known to retrieve the vertical distribution of the measured species. Therefore balloon-borne DOAS with direct sunlight allows to measure accurately the profiles of chemical species and by SCD6 comparison it is also possible to check its photochemical variation.

The chemical species OClO is an indicator for chlorine activation as it is believed to be predom- inantly produced by the reaction of ClO and BrO in the stratosphere. During the measurements presented in this thesis it could be detected in the Arctic winter as well as at mid-latitudes in autumn under undisturbed conditions. Nevertheless, it was only possible to predict the activation seen in the Arctic by the models used in this thesis - the 3-D Chemical Transport Model(CTM) SLIMCAT [Chipperfield 1999], which is well established in the scientific community, and a Lagrangian box model on isentropic trajectories (LABMOS) especially implemented as a part of this thesis. The mid-latitude measurements cannot be explained with the standard gas phase and heterogeneous chemistry used in both models. Chapter 2 gives an overview about the chemistry of the stratosphere and especially highlight the importance of halogens in the atmosphere. Chapter 3 proceeds with a description of the direct sunlight balloon-borne DOAS measurement technique which was used to retrieve the concentrations of the chemical species of interest in this thesis. Then an overview of the results of the eight LPMA/DOAS balloon flights conducted so far is presented in chapter 4. Chapter 5 covers the additional results of first free tropospheric BrO profile measurements which were stimulated by the combination of different measurement platforms (balloon, satellite, and ground-based). In chapter 6, the first comparison of the total inorganic bromine method with the standard organic bromine method is shown. Then, chapter 7 outlines two case studies made to understand the observation of enhanced OClO amounts. Finally the thesis concludes with an outlook in chapter 8.

Chapter 2 

Halogen species and their importance in atmospheric chemistry Stratospheric ozone depletion through catalytic chemistry involving man-made chlorofluorocarbons is an area of focus in the study of geophysics and one of the global environmental issues of the twen- tieth and twenty-flrst century. It has been shown by several theoretical and experimental studies that catalytic cycles involving oxygen, hydrogen, nitrogen and last but not least halogens can affect fundamentally the abundance of ozone in the stratosphere, as well as in the troposphere. 2.1 Stratospheric Ozone Although it was already proposed in 1974 that stratospheric ozone could be depleted in chemical reactions involving the degradation products of chlorofluorocarbons(CFCs), it was not until 1985 that unequivocal evidence of ozone loss was reported in the scientific literature. In that year, scientists from the British Antarctic Survey [Farman et al. 1985] described the polar ozone depletion, now known as the ozone hole, in which during six weeks in the spring the total ozone column decreases by more than half. It was subsequently shown that in the lower stratosphere almost all the ozone is removed from a layer at altitudes between about 13 and 20 km. These observations attracted great public interest, and aroused considerable scientific debate. At that time, photochemical theory had predicted ozone loss in the upper stratosphere by catalytic cycles involving the chlorine monoxide radical(ClO) and oxygen atoms. As the Antarctic ozone loss occurs in the lower stratosphere, it cannot be explained by that mechanism. The unique role of ozone in absorbing certain wavelengths of incoming solar ultraviolet light was recognized in the latter part of the nineteenth century by Cornu [1879] and Hartley [1880]. Interest in ozone stems from the fact that such absorption of solar radiation is important in determining not only the thermal structure of the stratosphere but also the ecological framework for life on the Earth′s surface. Decreased ozone results in increased ultraviolet transmission, which can affect the health of humans, animals, and plants.

[...]