STRATOSPH

ERIC OZONE

OF CONCEPTS

AND

DEPLETION'

A REVIEW

HISTORY

Susan Solomon

AeronomyLaboratory NationalOceanicand Atmospheric Administration Boulder, Colorado

Abstract. Stratosphericozone depletionthroughcatalytic chemistry involving man-made chlorofluorocarbons is an area of focusin the studyof geophysicsand one of the global environmentalissuesof the twentieth century.This review presentsa brief history of the science of ozone depletion and describesa conceptual frameworkto explainthe key processesinvolved,with a focuson chemistry.Observationsthat maybe considered as evidence(fingerprints)of ozone depletion due to chlorofluorocarbonsare explored, and the related gas phase and surfacechemistryis described.Observations of ozone and of chlorine-relatedtrace gasesnear 40 km provide evidence that gas phase chemistryhas indeed currentlydepletedabout 10% of the stratosphericozone there aspredicted,and the vertical and horizontal structures of this depletion are fingerprintsfor that process. More striking changesare observedeach austral spring in Antarctica, where about half of the total ozone col-

umn is depleted each September,forming the Antarctic ozone hole. Measurements of large amounts of C10, a key ozone destruction catalyst,are among the finger-

1.

prints showingthat human releasesof chlorofluorocarbons are the primary cause of this change. Enhanced ozone depletion in the Antarctic and Arctic regions is linked to heterogeneous chlorine chemistry that occurs on the surfaces of polar stratospheric clouds at cold temperatures. Observationsalso show that some of the same heterogeneous chemistry occurs on the surfaces of particles present at midlatitudes as well, and the abundancesof these particles are enhanced following explosivevolcanic eruptions. The partitioning of chlorine between active forms that destroy ozone and inert reservoirsthat sequesterit is a central part of the framework for our understanding of the 40-km ozone decline, the Antarctic ozone hole, the recent Arctic ozone lossesin particularly cold years, and the observation

of record

midlatitude

ozone

de-

pletion after the major eruption of Mount Pinatubo in the early 1990s.As human use of chlorofluorocarbons continues to decrease, these changesthroughout the ozone layer are expected to gradually reverse during the twenty-first century.

INTRODUCTION

vealed that the total ozone abundancesover many regionsof the globehave decreasedmarkedlysinceabout The unique role of ozone in absorbingcertain wave- 1980, as is illustratedin the data presentedin Figure 1. lengthsof incomingsolarultravioletlight wasrecognized Indeed, the depletion of the global ozone layer has in the latter part of the nineteenth century by Cornu emergedas one of the major global scientificand envi[1879] and Hartley [1880]. Interest in ozone stemsfrom ronmental issuesof the twentieth century. Downward trends are evident in the time series of the fact that suchabsorptionof solarradiation is important in determiningnot only the thermal structureof the spatially or time-averagedspring column ozone obserstratosphere [e.g.,Andrewset al., 1987]but alsothe eco- vations shownin Figure 1. Ozone varies from year to logicalframeworkfor life on the Earth'ssurface.(Terms year at all locations, but the behavior seen in recent in italic type are defined in the glossaryfollowing the decadesin Antarctic springlies very far outsideof the main text.) Decreasedozoneresultsin increasedultra- historicalvariability.The longestavailablehigh-quality violet transmission, which can affect the health of hu-

record is that of Arosa, Switzerland, which dates back to

mans,animals,and plants [e.g.,van derLeun et al., 1995, and referencestherein]. Observationsof the total integrated column ozone based on ultraviolet absorptionbegan in the first few decadesof the twentieth century [e.g., Fabry and Buisson, 1913;Dobson, 1968, and referencestherein; Dtitsch, 1974]. Systematicmeasurementsof this type have re-

the 1920s[Staehelinet al., 1998a,b]. The record at this siteagreeswell with the larger-scalechangesobservedby satellite since 1979. Figure 1 showsthat the observed ozone changesin the 1990scomparedwith earlier decades are large enough that sophisticatedstatistical treatmentsare not neededto discernthem, not only over Antarctica

Copyright1999 by the AmericanGeophysicalUnion.

but also in the Arctic

and at midlatitudes.

For

Reviewsof Geophysics, 37, 3 / August1999 pages275-316

8755-1209/99/1999

RG900008515.00

Papernumber1999RG900008 ß 275

ß

276 ß Solomon' STRATOSPHERIC OZONE DEPLETION

500

500j" i /

1

i

37, 3 /REVIEWS OF GEOPHYSICS

i

t

13 55NNimbus TOMS

I

F o 55NSBUV/SBUV2 ! ...... 55NSBUV/SBUV2 3yr.•rr•n 450 I- -e- Aros•Switzerbnd

450

L

•,,.running mean ,,m ,. . 1.

•' 'l

_l 1

-I

!

/

1550

Arctic 63-90øN

MarchAverage

Figure 1. Observationsof total ozone at various lo-

cations.The AntarcticdataarefromHalley[Farmanet al., 1985;Jonesand Shanklin,1995]and updatedcourtesy of J. Shanklin.The Arctic data are from satellite observationsdescribedby Newman et al. [1997], updated courtesyof P. Newman. The Arosa, Switzerland datasetis the longestrunningin the world [Staehelinet al., 1998a, b]. Satellite observationsfrom a slightly higher midlatitude region are shownfor comparison [Hollandsworthet al., 1995], updated courtesyof R. Nagatani. The satellite data are zonally and monthly averaged,while the ground-baseddata at each site have alsobeen averagedover time asindicatedin each

30() K

19'PO

1980 Year 1990/ 2.000 1_.••1940 19•0 1980 2.000 Yeer

/ •7/

ß •,,•-:• .

,

400

case. -

--

-

.

' October Average

]00 ..... •

I

1950

1960

J

i

•

I97q

I

1980

•

ß• • I

I990

•

2000

Year

detailed discussion• of data quality and ozone trend trated in Figure 1 is the focusof sections3-6. Section3 detectionapproaches,seethe recentreviewsby Harris et describeshow the discoveryand explanationof the Antal. [1998] and J. Staehelinet al. (Observationsof ozone arcticozoneholeradicallyalteredthe gasphasechemical trends, submittedto Reviewsof Geophysics,1998, here- picture by revealingthe key role playedby reactionsof inafter referred to as Staehelin et al., submitted manu- chlorinecompoundson and within surfaces(heterogescript,1998). neouschemistry),particularlyunderverycoldconditions The aim of this review is to describe a framework for in polar regions.The chemicalnature of stratospheric a conceptualand historicalunderstandingof the pro- surfacescapableof drivingsuchchemistryis the focusof cessescontrollingstratosphericozone depletion,partic- section 4, where it is shown that water ice, nitric acid ularly the role playedby human use of chlorofiuorocar- hydrates,and liquid sulfuricacid/watersurfacesall must bons (CFCs). Key historical and illustrative recent be considered.Laboratorystudies,observationsof midreferenceswill be cited.Sucha reviewis by designlimited latitude ozone trends, and measurements of stratoin scopeandis intendedto be accessible to the nonspecialspheric chemical composition have underscoredthe ist. It focusesstrictlyon ozonedepletionprocesses rather need to consider both gas phase and heterogeneous thanon the broaderaspects of the current,highlydetailed chemistrynot only under extreme cold but also under understanding of stratospheric ozonechemistry,radiative transfer,dynamics,and meteorology.For recentin-depth relativelywarm conditions,as is discussedin sections3, treatmentsof thosetopics,see,for example,WorldMeteo- 4, and 5. Recent changesin Arctic ozone have further rologicalOrganization/United Nations EnvironmentPro- illustrated the strong couplingbetween heterogeneous gramme(WMO/UNEP) [1994,1999,andreferences there- chemistryand extensivepolar ozonedepletionand have raised important questions regarding meteorological in],Andrewset al. [1987],andHoltonet al. [1995]. trends (section6). A key frameworkfor understanding Section2 of this paper briefly discusses the general theoretical understandingof the vulnerabilityof ozone that is emphasizedthroughoutthis reviewis the concept to chemicalchange,particularlythe depletionsthat were of chemicalpartitioningof chlorine between forms that predictedto occurin the distantfuture basedupon gas are inert with regard to ozone (HC1, C1ONO2) and phasechlorineandbrominechemistry[MolinaandRow- othersthat can destroyit (C1,C10). Major conclusions land, 1974]. The transformationof this theory to the are briefly summarizedin section7. A glossaryof terms remarkablereality depictedby the ozone decline illus- used follows the main text.

37, 3 / REVIEWSOF GEOPHYSICS 2.

GAS PHASE CHEMISTRY

AND

Solomon:STRATOSPHERICOZONE DEPLETION ß 277 RELATED

CONSIDERATIONS

2.1. CatalyticCyclesand ChemicalFamilies

drogen may arise through human modifications of sourcegasessuch as H20 and CH4, while natural odd nitrogen can be perturbed through direct emissionsof high-flyingaircraft,by nuclearexplosions,or by changes in its primarysourcegas,N20. Heath et al. [1977] demonstrated that large solar proton events can lead to transientperturbationsin upperstratospheric ozonedue to a natural modulationof odd nitrogen chemistry.This verified the nitrogen-catalyzedozone destructioncycle in a dramaticfashion (see also the review by Jackman and McPeters[1987]). Refinementsto measuredlabora-

A photochemicaltheory for formation and destruction of ozonebasedon an oxygen-onlychemicalscheme was first proposedby Chapman [1930]. An updated version of this framework is shownby the first seven reactionspresentedin Table 1. Perhapsmost importantly, Chapman noted that ozone and atomic oxygen rapidlyinterchange with eachother,whilethe sumof the modelestimates two is linked to much slower chemicalprocesses.This torykineticratesallowedthe numerical work laid the foundationfor the understandingof "odd of the impactsof suchperturbationsupon ozone to be oxygen"chemistry.Such a conceptualpicture allowsa gradually improved over a period of several decades clear distinction to be drawn between net and gross [e.g., Ko and Sze, 1983]. While the study of possible productionand lossof ozone over a chosentimescale ozonedepletiondue to hydrogenandnitrogenchemistry whichwill be briefly summarizedhere (see the seminal remains an area of active research(see WMO/UNEP review by Johnstonand Podolske[1978] for further de- [1998] for current studiesand Johnston[1992] for an historicalreview),the weightof evidenceshowsthat the tails). Ozone photolysisbelow ---50 km representsa gross bulk of the observedrecentdepletiondepictedin Figure but not net lossprocessover timescalesof the order of 1 is due to other processes,particularlythe chlorineminutes or more, sincenearly all of the atomic oxygen related chemistrythat is the primary subject of this thus producedreforms ozone (throughthe reaction of review. In 1974 it was shownthat chlorine could also engage O + 02 with a third body,M; see(R2) in Table 1) in just a few secondsor less.Ozone and atomicoxygenthereby in a catalyticcycleresultingin ozone destruction[Stocyclevery rapidly betweenone another in the strato- larskiand Cicerone,1974]. Of particularimportancewas as sphere.A very small fraction of the oxygenatomspro- the identification of man-made chlorofluorocarbons stratospheric chloducedfrom ozonephotolysiscanreactwith ozone(O + the major sourceof ozone-destroying 0 3 --->202) , yieldinga net lossof the sumof the two over rine [Molina and Rowland,1974].Like the nitrogenand extendedtimescales.Hence it is conceptuallyuseful to hydrogenoxides,chlorinecan destroyozonein catalytic consideratomic oxygenand ozone together as an odd cyclessuchasthoseshownin Table 1. Wofsyet al. [1975]; oxygenfamily distinctfrom the much longer-livedform Yunget al. [1980], Tunget al. [1986], and McElroy et al. of "even oxygen,"0 2 (for further discussionsee, e.g., [1986] showedthat bromocarbonscould alsocontribute to ozonedepletion,particularlythroughthe couplingof Brasseurand Solomon,1986]). In the 40 yearsfollowingChapman'sgroundbreaking bromine and chlorinechemistry.Collectively,the deplepaper, it became clear that stratosphericozone was tion of ozoneby chlorine,bromine, and the interactions chemicallydestroyednot solelyby reactionwith atomic between them will be referred to herein as halogen oxygen,but also by hydrogen[Batesand Nicolet, 1950; chemistry. Hampson,1964] and nitrogenoxidechemistry[Crutzen, 1970, 1971;Johnston,1971]. Each of thesespeciesmay 2.2. Processes ControllingChlorocarbonLifetimes also be consideredin terms of their own odd hydrogen Molina and Rowland[1974] and Rowlandand Molina hypothand odd nitrogen families, the membersof which can [1975]pointedout that the chlorofluorocarbons interchangechemicallywith one another[see,e.g.,Bras- esized as ozone depletorshave very long atmospheric seurand Solomon,1986].Table 1 illustratesthe fact that residencetimes,so that if thesegaseswere to be deplethydrogenand nitrogenoxidescandestroyodd oxygenin ing stratosphericozone, they would continue to do so a catalyticfashionwherein the initiating active species well into the twenty-first century. This critical point (e.g., OH, NO) are regenerated,so that even small merits a brief elucidation(expert readersmay wish to amountsof these gasescan influencethe much greater skip to section2.3). Figure 2 is a schematicdiagramof ozone abundances.Table 1 alsopresentssomeillustra- the key processesthat contributeto and control chlotive reactionsthat coupleone family of gasesto another rofluorocarbonlifetimes in the Earth's atmosphere.As (suchas the formationof C1ONO2 throughreactionof was emphasizedby Molina and Rowland [1974], the C10 with NO2; C1ONO2 is thus a member of both the chlorofluorocarbonsare not significantlysolublein waodd chlorineand odd nitrogenfamilies) and processes ter; nor do they reactwith oceanor soil surfacesor with that form relativelylong-livedreservoirs (HC1, C1ONO2, any chemicalspeciespresentin the lower atmosphere Their chemical HNO3) , whichcan stronglyinfluencethe abundances of (below ---12-15 km, the troposphere). the ozone-destroying gases(e.g., C10, NO2), as is dis- destructiondependsupon the ultravioletlight found in cussed further below. the upper atmosphere(between---12-15and 50 km, the Perturbationsto the natural abundancesof odd hy- stratosphere).This radiationbreaksup the chlorofluo-

278 ß Solomon' STRATOSPHERICOZONE DEPLETION

37, 3 / REVIEWSOF GEOPHYSICS

TABLE 1. Key Chemical Processesand Catalytic Cycles Reaction

Chemical

Process

Number

ChapmanChemistry a 02 + hv -• 20 O + 0 2 + M-•O

R1

3+ M

R2

0 3 +hv-•O 2 + O(•D) O(•D) + M-•O + M

R3

0 3 + hv -• 0 2 + O O+O+ M-->O 2 + M 0 + 03 --> 202

R5

R4 R6 R7

Illustrative OddHydrogen Catalytic Cycles b O + OH -• 0 2 + H H + 0 2 + M --> HO 2 q- M O q- HO 2 --> 0 2 q- OH Net Cyclel:O + O +M-•O2 OH + 03 -• HO2 + 02 HO 2 + 0 3 --> OH + 20 2 Net Cycle 2:20 3 -• 302

R8 R9

R10 R6 Rll

+M

R12

R13

IllustrativeOdd NitrogenCatalyticCyclec R14

NO + O 3 --> NO 2 q- 02 O q- NO 2 --> NO + 02 Net Cycle 3' O + O3 -• 02

R15 R7

Illustrative OddChlorineCatalytic Cycles d R16

C1 + O3-•C10 + 0 2 C10 + O-• C1 + 02 Net Cycle 4: O + O3 -• 02 C1 + O3-•C10 + 0 2 C1 + O3-•C10 + 02 C10 + C10 + M -• C1202 q- M C1202 + hv-• C1 + CIO 2 C102 + M -• C1 + 02 + M Net Cycle 5:2 0 3 -• 30 2

R17 R7 R16

R16 R18 R19

R20 R13

IllustrativeCl-Br CatalyticCycle½ R16

CI+ O3-•C10 + 02 Br + 03-->BrO + 02 BrO + C10-• Br + C102 C102 q- M -• C1 + 02 + M Net Cycle 6:2 03 -o 302

R21

R22

R20 R13

SomeImportantCouplingand ReservoirReactions R23

C10 + NO-• C1 + NO 2 C1 + CH 4 --> HC1 + CH 3 HO 2 q- C10-• HOC1 + 0 2 C10 + NO 2 + M -• C1ONO2 + M OH + NO 2 q- M--> HNO 3 + M

R24

R25 R26 R27

Key Heterogeneous Reactions

HC1 + C1ONO2 --> HNO 3 + C12 N20 5 + H20 --> 2HNO 3 C1ONO2 + H20--> HNO 3 + HOC1 HC1 + HOC1 -• H20 q- C12 BrONO 2 + H20--> HNO 3 + HOBr HC1 + BrONO 2 --> HNO 3 + BrC1 HC1 + HOBr -• H20 q- BrC1

R28 R29

R30 R31 R32 R33 R34

aChapman[1930].

bBates andNicolet[1950];Hampson [1964]. CCrutzen [1970];Johnston[1971].

dStolarski andCicerone [1974];MolinaandMolina[1987]. eMcElroyet al. [1986];Tunget al. [1986].

rocarbonmolecules,yieldingC1atomsthat can go on to destroyozone in catalyticcyclessuchas thoseshownin Table 1 as they move throughthe stratosphere. A reviewof the fluid mechanicalprinciplesunderlying the dynamicsandmeteorologythat is responsible for the

movement of air from the troposphereto the stratosphereis provided,for example,byAndrewset al. [1987] andHolton et al. [1995].It is interestingto note that long before the fluid dynamical underpinningsof stratospherictransportwere fully established, a broadconcep-

37 3 / REVIEWS OF GEOPHYSICS

/

Stratosphere = I0%

of the

mass of the

atmosphere

/

Solomon: STRATOSPHERIC OZONE

DEPLETION

ß 279

lower atmospheredistributesthe CFC-11 throughout the troposphere. Observationsof chlorofluorocarbons 03 -• Brewer-Dobson I .................... } • CircuMion from surfacestationsas far apart as the SouthPole and Colorado as indicated in Figure 2 (data taken from +....... 'O, NO" -"-• Montzka et al. [1996]) show that the mixing ratios of / \ CFCs in the Southern Hemisphere lag those of the Northern Hemisphereby about a year. However, the

//CFC+h•,--•'-Cl CiO ,-,,_•

.•, o'o _x

•/,/•Rapid Mixing •""•"•

I

fact that the abundances of chlorofluorocarbons

are so

large at a remote site like the South Pole, far removed

510

from their emissionin the industrializedparts of the Northern Hemisphere,atteststo the fact that their destructionin the tropospheremust be extremelyslow or nonexistent.Key factors are the near-insolubilityof CFCs in water (whichmakesthem resistantto the rainout andwashoutprocesses that removesomeother gases emitted by industrial activities,such as those that form the local pollution of acid rain) and their chemically

_• 500

5OO

inert character.

E 490

49O

Tro•sphere = •% ,-., >

540

NorthernHemisphere

of t• m•s of t• ••re SouthernHemisphece

54O

53O

.• 5•o o• 5•0

520

48O

r• 470

470

199t

1992

igg3

YeQr

•

igg5

Ig•

Year

Figure 2. Schematicdiagram illustratingthe breakdownof CFCs and catalytic destructionof ozone in the middle and upper stratosphere.Becausethe stratospherecontainsonly 10% of the massof the total atmosphere, the atmospheremust turn over many times to destroy all of the CFCs present, resultingin long atmosphericresidencetimes for thesegases. The simplifiedcartoon illustratesonly the key net processes that transport CFCs and other gasesin a zonally averaged sense.The Brewer-Dobsoncirculationillustratesa typical average flow pattern. Waves mix trace gaseswhen they break down,particularlyin the winter hemisphere(SouthernHemisphere in this illustration). The long CFC lifetimes are reflected in the surface observationsof CFC-12 at stations,such

as SouthPole, that are far removedfrom the emissionregions in the industrializedNorthern Hemisphere. tual framework

had been deduced from chemical

obser-

vations that remains basically intact today. Dobson [1930] inferred the existenceof a large-scalestratosphericcirculationcell characterizedby risingmotion in the tropics and descendingmotion at mid and high latitudes

on the basis of his observations

of the latitude

gradientsin ozone. He pointed out that greater ozone columnabundancesobservedat higherlatitudesmustbe the resultof downward,polewardmotion.Brewer[1949] reached a similar conclusionbased upon an elegant analysisof early measurements of water vapor. Recent studieshave, for example, used observationsof very long-livedgaseswith knowntropospherictrendssuchas CO2 [e.g., Schmidt and Khedim, 1991; Boeringet al., 1996] to showthat the timescalefor the overturningof this"Brewer-Dobson"circulationcell is -5 years.About 90% of the total atmospheric massresidesin the troposphere,and -10% residesin the stratosphere. Consider the fate of 1 kg of CFC-11 released in today'satmosphere,usingFigure2. Rapid mixingin the

A fraction of the troposphericmassentersthe stratosphereand is slowlytransportedupward,poleward,and backto the troposphere.Rapid horizontalmixingin the troposphere,coupled with the fact that the primary point of entry to the stratosphereis in the tropics(as sketchedin simplifiedform in Figure 2; seeHolton et al. [1995] for a more detailed picture), implies that the chlorine content of stratosphericair will not depend substantiallyupon proximity to local sources.Ozone depletion is therefore a global phenomenon,sincethe amount of total chlorine (also called chlorineloading) both at and abovethe SouthPole is nearly the sameas that aboveindustrializedregions.The observedspatial and temporalvariationsin ozonelossare closelytied to chemicalprocesses that partition thistotal chlorineloading amongitsvariousformsand therebymodulateozone destructionin spaceand time, as is discussedbelow. Within -5 years,air will havecycledthroughthe mid to upper stratosphere.Most of the CFC-11 containedin thisair breaksdownin the upperstratosphere to release itschlorine(whichin turn destroysozone)andreturnsto the tropospherelargelyin the form of hydrochloricacid (which ultimately rains out and removes the chlorine from the system).Since only -10% of the massof the troposphereexchangeswith the upper stratospherein each5-yearperiod,the processwill haveto be repeated approximately10 timesto destroythe bulk of chlorofluorocarbon initially released.In the case of CFC-11, this leadsto a lifetime of -50 years.For someof the other chlorofluorocarbons, stratospheric photodissociation destroysa smallerfraction of the parent compoundwithin a singlecircuit through the Brewer-Dobsoncirculation, extendingthe lifetimesconsiderably(in the caseof CFC115,for example,the lifetime is about500 years[WMO/ UNEP, 1994]). If all emissionsof thesechlorofluorocarbonswere to ceaseimmediately,these gaseswould be slowlyremovedfrom the atmosphereon suchtimescales accordingto the processesdepictedin Figure 2. A concisereviewof globalemissionsand future projectionsis providedby Pratheret al. [1996].

280 ß Solomon: STRATOSPHERICOZONE DEPLETION

37, 3 / REVIEWSOF GEOPHYSICS

Stolarskiand Rundel, 1975; Sze, 1978]. The amount of atomicfluorineand FO availableto participatein ozonedestroying catalyticcycles(or, in chemicalterms, the F + 03'•- FO+02 O+ ClO---•Cl+ 02 catalyticchainlength[seeJohnston andPodolske,1978]) Fluorine has a short chain NO2 lengthbecauseHF doesnot Net: O+ 03---->202 is hence extremelyshort, and fluorine has a negligible react with OH h•/ impacton ozone(seethe recentanalysisby D. J. Lary et CIONO 2• al. (Atmosphericfluorine photochemistry, submittedto Journalof Geophysical Research,1998)). Chlorineforms Br +03 --•-Br•)+O 2 I Chlorine chain length strongly both HC1 [Stolarskiand Cicerone,1974] and C1ONO2 C1+03 -"•C10+02 and CION02/CIO reservoirs[Rowlandet al., 1976].These gasescan, however, be reconvertedto chlorine atoms by gas phase / Net: 20'•'-•'30•I -chainlength because both chemistry(i.e., by reaction with OH and photolysis, / I BrON02 Ii-I Bromine has a long catalytic HBr and BrON02 photolyze respectively).The amount of C1 and C10 availableto andreact rapidlywith OH participate in ozone-destroying catalytic chemistry Figure 3. Key contrastsbetweenF, C1,and Br for ozoneloss therefore is criticallydependenton the partitioningof

CH4 •"•'•,•OH Endothermic

BrO+CIO--•-Br*CI+02 IN02depends upon partitioning HCI/CIO

are linked to their gasphasepartitioningprocesses illustrated chlorine between these "active" chlorine radicals and the here.

non-ozone-destroying "reservoirs,"HC1 and C1ONO2. The rates of chemical

The challengesfacing geoengineeringstrategiesto mitigate ozone lossesby, for example, making more ozone, have been recognizedfor decades.The energy required to break the 02 bond in order to make two ozonemoleculesis about5.1 eV, sothat the powerinput

required to produce theozonelayerisabout2 x 1013 W

formation

and destruction

of the

reservoirs control this partitioning. Bromine is less tightlyboundthan chlorine,sothat relativelylittle of the brominereleasedfrom bromocarbonsis tied up in HBr and BrONO2, rendering this atom quite effective for ozone loss [Wofsyet al., 1975; Yung et al., 1980; Lary, 1996], especiallyin combinationwith chlorine [Tunget al., 1986;McElroyet al., 1986].Althoughthere are significanthuman sourcesof bromine, the contemporary abundances of total stratosphericbromineare about200 times smallerthan those of chlorine [e.g., Schauffieret al., 1993; Wamsleyet al., 1998]. Iodine may alsoparticipate in ozone-destroyingcatalyticcycleswith bromine and chlorine[Solomonet al., 1994, 1997] but its stratosphericabundanceis believedto be much smallerthan thoseof bromine and chlorine,and its primary sources are believedto be natural rather than largelyor partly

[Hunten,1977].This power is providedon a natural and continuingbasisby the Sun but was estimated at timesmankind'stotal artificialpowergenerationin 1970 [Hunten, 1977]. Making enoughozone to artificiallyreplace even a small fraction of the global burden would still be an extremelyexpensivepropositiontoday.Alternative schemes involving interference with chlorine chemistryhave also been shownto be impractical[see, e.g., Viggianoet al., 1995].Hence the reductionof global emissionsand the resulting gradual removal of atmo- man-made as in the case of fluorine, chlorine, and brosphericchlorineis the only knownpracticalapproachto mine. future recoveryof the ozone layer. While many natural processesproduce chlorine at groundlevel (includingfor example,sea salt and volca2.3. ChemicalPartitioning,ChlorineSources,and nic emissionsof HC1), these compoundsare efficiently Gas PhaseChemistryFingerprints removedin precipitation(rain and snow)owingto high The foregoingdiscussionand the referencestherein solubility.The removalof HC1 emitted,for example,by outline the dominant catalytic processesthat control volcanoesis extremelyefficient[see,e.g., Tabazadehand ozonechemistryand describein generaltermswhy com- Turco,1993],renderingeventhe mostexplosivevolcanic poundssuchas chlorofluorocarbons releasedat ground plumesineffectiveat providingsignificantinputsof chlolevel residein the global atmosphereover timescalesof rine to the stratosphere(as was demonstratedin direct decadesto centuries.Along with these catalyticcycles, observations of volcanicplumesby Mankin et al. [1992] chemical partitioning processesplay a major role in and Wallaceand Livingston[1992]). In contrast, airborne observations of the suite of ozone destructionthat is dramaticallyillustratedby the contrastsbetween F, C1, and Br gas phase chemistry chlorofluorocarbons at the base of the tropical stratoshownin simplifiedform in Figure3. Briefly,the halogen sphere [see, e.g., Schauffieret al., 1993] showthat the atoms releasedin the stratospherefrom chlorocarbon, total chlorine content in air entering the lowermost bromocarbon,and fluorocarbonsourcegasescan form stratospheredue to chlorofluorocarbonsin 1992 was acids (through abstractionof a hydrogenatom) and about3.0ppbv,comparedwith only ---0.1-0.2 ppbvfrom nitrates (through reaction with NO2). In the case of concurrentmeasurementsof HC1 and ---0.5-0.6 ppbv fluorine, the acid HF is quicklyformed and so tightly from CH3C1, which is the sole stratosphericchlorocarboundthat essentiallyall fluorine releasedfrom fluorine bon that has significantnatural sources.Observations sourcegasesin the stratosphereis irreversiblyand rap- suchas thosein Figure 2 have confirmedthat the temidly "neutralized" as HF [Rowlandand Molina, 1975; poral trendsin global surfacelevel abundancesof chlo-

37, 3 / REVIEWSOF GEOPHYSICS Chlorine P(•rtitioning

Solomon: STRATOSPHERICOZONE DEPLETION ß 281 Ozone Trend (•t NorthernMid-L(•titudes

54• ...............

Observed

50]• Zander et al' (1996 Figure 4. (left) Observations of chlorinepartitioning as a function

of altitude

from an instrument

on board

the spaceshuttle[Zanderet al., 1996].(right) Observed vertical profile of the ozone trend at northern midlatitudes [Harris et al., 1998], together with a current model estimate[from Solomonet al., 1997].

.01

.I

I -12

Fr(]ctionof Av(]il(]ble Cly

-I0

-8

-6

-4

-2:

0

OzoneTrend(%/dec(]de)

(Cly=HCl+ClONO 2 + Cl+ ClO+HOCl...)

rofluorocarbons

are consistent with the known industrial

emissions[e.g.,Montzka et al., 1996;Prinn et al., 1995; WMO, 1985; WMO/UNEP, 1991, 1994, 1999], both in termsof the buildup of thesegasesin past decadesand the slower accumulationin the 1990sfollowing reductionsin globaluse (see alsoPlate 5 below). Observationsof HC1, C10, C1ONO2, and other chlorine-bearinggasesby infrared spectroscopy onboardthe spaceshuttle[Michelsenet al., 1996;Zander et al., 1996] or from satellites[Dessleret al., 1995, 1998] allow study of how chlorineis chemicallypartitionedin the middle and upper stratospherein somedetail, as shownin the left-handsideof Figure4. In the uppermoststratosphere above ---45 km, nearly all of the chlorine releasedfrom sourcemoleculessuchas CFCs (hereinafterreferred to

region [e.g., Russellet al., 1993; Zander et al., 1996; Montzka et al., 1996; Schauf-fier et al., 1993] provide direct evidence that the chlorine content of the contem-

porary stratospherehas been greatly perturbed, with about 85% of the 1992 stratosphericchlorine burden attributable

to human

activities.

Crutzen[1974] and Crutzenet al. [1978] carried out some of the first detailed

chemical

models

of ozone

depletion, building upon the studiesof Stolarskiand Cicerone[1974], Molina and Rowland [1974] and Rowland and Molina [1975] and includingthe chemicalunderstandingoutlined above.Crutzen[1974] predicteda relativemaximumin C10 near 40 km, whichwasbroadly confirmeda few years later by observationsof C10 by Andersonet al. [1977],Parrishet al. [1981], and Waterset asCly)issequestered in theHC1reservoir, owinglargely al. [1981]. Largelybecauseof this relative maximumin to the efficacyof the reaction of C1 + CH 4 at warm C10, Crutzen [1974] predicted a maximum in ozone temperaturesand high C1/C10ratiosthere. Recentmod- depletion in the same region (although other factors els [Michelsenet al., 1996], stratosphericobservations suchas the availabilityof atomicoxygenalsocontribute [Stachniket al., 1992;Chandraet al., 1993], and labora- to the verticalprofile of ozonedepletion).The left-hand tory measurements [Lipsonet al., 1997] (seeJetPropul- side of Figure 4 showsthat current observationsof the sionLaboratory(JPL) [1997]) suggestthat a smallyield C10/Cly profile in themiddle andupperstratosphere of HC1 in the reaction of C10 with OH also affects the agreewith thoseearly predictionsand observations. HC1/C10 partitioning in this region. While there havebeen indicationsof ozonedepletion Becausenearly all of the chlorine and fluorine re- in the upper stratospherefor more than a decade[e.g., leased from chlorofluorocarbons resides as HC1 and HF OzoneTrendsPanel,1988],onlywithin the pastfew years in the stratosphere near 50 km, observations of thesetwo hasit beenwell quantified[e.g.,Miller et al., 1995;Harris gasesin this region provide key verification of their et al., 1998]. Sample observationsof the northern midattribution to CFC sources.Recent global data by Rus- latitude ozone profile trendsas shownin the right-hand sell et al. [1996] displayabundancesand trendsin both side of Figure 4 displaya maximum near 40 km, just as HC1 and HF near 50 km that are quantitativelyconsis- predicted more than 2 decadesago. The close agreetent with observations of the chlorofluorocarbons at ment between the vertical shapesof the observedand ground level; these observationstherefore confirm that predictedozonechangesin the upper stratosphereproCFCs are the key sourcesfor stratosphericchlorineand vides strong evidencefor gas phase chlorine-catalyzed fluorine. Taken together, the global measurementsof ozonedepletionchemistry.Further, the correspondence HC1, HF, other chlorine compounds,and the CFC between the shapesof the vertical profiles of the obsourcegasesboth at the surfaceand in the tropopause served ozonedepletionandof the C10/Clyratioattests

282 ß Solomon: STRATOSPHERICOZONE DEPLETION

37, 3 / REVIEWS OF GEOPHYSICS

to lhe role of partitioningprocessesin modulatingthis chemistry.Indeed, this is the first of several "fingerprints" that can be usedto establishthe role of chlorine in ozone depletion. Plate 1 presentsanother fingerprint illustrating the role of gasphasechemistryand chlorinepartitioningin ozone depletion, namely, the latitudinal gradients in upper stratosphericC10 and ozone depletion. In the early 1980s,globalmeasurementsof methaneby satellite illustratedthat the strongupwellingof the Brewer-Dobson circulationin the tropical upper stratosphere(see Figure 2) givesrise to a maximumin methane there [Jonesand Pyle, 1984]. The enhancedmethane,in turn, was predicted to lead to a tropical minimum in C10 through its dominant role in C10/HC1 partitioningand hence in ozone depletion [Solomonand Garcia, 1984]. Current global satellite observationsof C10 [Waterset al., 1993; Waterset al., 1999], as shownin the left panel of Plate 1, indeed displaya stronglatitudinal gradient with a pronouncedminimum in the tropics. Satellite observations

of the latitudinal

variation

of the ozone

trends over the past 15 years (right panel of Plate 1) reveal a similarspatialpattern in the upper stratosphere as predicted.Thus not only the vertical profile but also the latitudinal structure of the ozone depletion above -25 km parallels the patterns observedin C10. These spatial variationsin ozone depletion point towardsgas phasechlorinechemistryand highlightthe role of chemical partitioning in modulating ozone depletion. For further discussionof other factors influencing upper stratospheric ozone(including,for example,the rolesof temperature,water vapor, and other factors)seeMtiller et al. [1999] in the WMO/UNEP [1999] ozone assessment. Plate 1 also showsvery high C10 abundancesin the lowerstratosphere (below30 km) overthe Antarctic in australspring,where gasphasechemicalpartitioning would not predict it. These are a focus of the next section.

3.

HETEROGENEOUS

CONDITIONS:

CHEMISTRY

THE ANTARCTIC

UNDER

OZONE

COLD

HOLE

3.1. Discoveryand Verificationof the Ozone Hole Measurableozonedepletionwasfirst documentedin the Antarctic springat the BritishAntarctic Surveystation at Halley [Farman et al., 1985]. Farman et al. showed that the ozone hole is confined to particular seasons (i.e., spring)and to southpolar latitudes.These pioneering findingswere quickly confirmed by spacebased measurements[Stolarskiet al., 1986] and by observationsat other Antarctic sites [e.g., Komhyr et al., 1986].Observationsof total columnozoneusinginfrared [Farmeret al., 1987] and visiblespectroscopy [Mount et al., 1987] providedfurther supportfor the seasonaldepletion of springtimeozoneusingindependentmethods. As the satellitemeasurements confirmedthat the depletion extended over roughly the entire continent, the

phenomenonbecame known as the Antarctic ozone "hole."

While the Antarctic ozone hole is not a true hole, in the sense that some column

ozone remains

even in the

most extreme depletions observed in the mid 1990s (when October ozone minima were near 100 Dobson units(DU) over the South Pole, or depletionof about two thirds of the historicallevels [see Hofmann et al., 1997]), the descriptorcapturesthe fact that the peak depletionis sharplylimited to Antarctic latitudes.Dobson[1968and referencestherein]notedthat there is less ozone naturally present over Antarctica than over the Arctic in winter and spring,but this climatologicaldifferencebetweenthe natural ozonelevelsover the poles of the two hemispheresshouldnot be confusedwith the abrupt decline that began near the mid-1970sas depicted in Figure 1. Newman [1994] discusses theseand other historical

measurements

of total ozone and shows

that the Antarctic ozone hole began in the last few decades.

The latitudinal gradientsin Antarctic ozonedepletion are related to the dynamicalstructureof the polar winter stratosphere,whosecirculationcanbe viewedas a vortex [see, e.g., Schoeberlet al., 1992a; Holton et al., 1995]. Briefly, the absenceof solarilluminationin high-latitude winter leadsto coolingover the polesand hencea large temperaturegradient near the polar terminator. This thermal gradient implies rapid zonal (west-east)flow characterizingthe "jet" at the edge of the vortex,while the air within the vortexis relativelyisolatedin comparison with surrounding midlatitude regions, allowing deep depletion to develop.Differencesin the pre-1970s ozoneabundances in the two polar vorticesfirstnotedby Dobson[1968] are related to differencesin atmospheric wavesand circulationpatterns,which are in turn driven by factorsrelatingto surfacetopography(e.g., distribution of mountains,oceans,and continents).In brief, the north polar vortex is generallymore disturbedby atmosphericwavesforced from beneathby flow over a more variablesurfacetopography.Theselead to greatermixing and faster downwardmotion, which both increases the natural wintertime Arctic ozone abundances(by bringingdownozone-richair from above)andwarmsthe lower stratosphere(through adiabatic compression). Temperaturesin the Antarctic vortex are both colder and lessvariablethan thoseof the Arctic,whichstrongly influencesthe polar ozone depletion in the two hemispheres(sections3.3 and 6 below). Plate 2 showsmeasurementsof the seasonalcycleof ozoneat Halley in historicaland recentdata,whichshow that the depletionoccursonly over a limited portion of the year. These observationsdemonstratethat contemporary observationsof ozone at Halley in late August (end of australwinter) are near historicallevels,while the bulk of the ozonelossthere occursrapidlyduringthe month of September.The ozonesondedata [e.g.,Hofmann et al., 1987] further underscorethis point. Farman et al. [1985] presented evidence for large

37, 3 / REVIEWSOF GEOPHYSICS

Solomon' STRATOSPHERICOZONE DEPLETION ß 283 trends in October

Antarctic

total ozone

that were com-

pletely unanticipatedat the time. They suggestedchlorofluorocarbonsas the likely cause.This assertionwas remarkable because the observed depletion was far larger than was ever anticipatedup to that point. Scientific understandingof the behavior of the ozone layer prior to Farman et al.'s discoverysuggestedthat trends of a few percent in total ozone might begin to become observablesometimein the twenty-firstcenturyif chlorofluorocarbonemissionscontinued[see,e.g., Wuebbles et al., 1983].The reasonthat the predictedchangeswere relatively small and far in the future is reflected in the discussionof gasphasechemistryoutlined in section2. Figures 3 and 4 show that a gas phase chemical understanding predicts that chlorine's greatest impact on ozone occurs in the upper stratospherenear 40 km. Sincethe bulk of the total ozone column lies in a layer at much lower altitudes near 10-30 km, the integrated impact on the total ozone column is small with such a depletionprofile. Further, this frameworkdoesnot identify Antarctica as a site of particular significancefor ozone depletion. As Figure 1 illustrates,the Antarctic ozone depletion was the earliest to be observedand

( tuq) lq61aH

remains

•

the most extreme

on Earth

at the time

of this

writing [seeJonesand Shanklin,1995]. Measurementsof the vertical profile of the depletion within the ozone hole were first presentedby Chubacht [1984] and were rapidly followedby other data suchas those of Hofmann et al. [1987], Gatdiner [1988], and Iwasakaand Kondoh[1987],and more recentlyin satellite studies such as that of Bevilacquaet al. [1997]. Hofmann et al. [1997] presenteda detailed analysisof many years of ozonesondemeasurementsat the South Pole. Figure 5 summarizesa key result of that study, showingthat the depletion of the Antarctic ozone column is largelyconfinedto altitudesfrom ---12to 25 km, far below the altitudeswhere gasphasechlorine chemistry would predict major changes.Figure 5 also illustrates the shapeof the unperturbedozone "layer" ob-

o

o

served in historical ozone data, as well as the near-total

removal of ozone in the heart of the layer in a typical contemporarysondingfor early October. Finally, Figure 5 showsthe close correspondencebetween the region where most of the ozone is depleted and a vertical profile of a typicalpolar stratospheric cloud (PSC) observedat the SouthPole [Collinset al., 1993];the critical

o

to

role of these clouds in formation o

•

•

o

o

o

(Odq) a,•nsse•d o

,q

,e'

o

c5

•

o•

c5

cS

-

o

0

•:

•

0 N

u• q•

o

•

hole is

3.2. Solarand DynamicalTheoriesof the Origin of the Ozone

•

of the ozone

discussed in section 3.3.

o

i•

(wq) eprq!tl•alow!x0••

•

--

Hole

Observationsnot just of Antarctic ozone,but also of the factorsthat affect it, suchas chemicalspecies(e.g., NO 2 and C10) and meteorologicalproperties,were extremely limited at the time of the discoveryof this dramatic and unanticipatedozone loss. As a result, a variety of different theorieswere advancedas plausible explanations.The conflictingtheorieswere reviewedby

SeasonalCycleof Ozoneat Halley .....

400

'

i

..... •

I

ß

I

'

'"1 ...... ' "'

..

I ....

'

I

ß

A

76/77 87/88 77/78

350

96/97

300

250

. ß ß••0 •0 •00•

..

0

200 o

o

150

Sept 5O 210

2:40

Oct

270

Nov 300

Dec

5•

dan 360

Feb :390

Mar 420

450

Time sinceJan l (days) Plate 2. Observations of the full seasonal cycleof dailyozoneat Halley,Antarctica,in theyearsbeforethe ozonehole(1976-1977and 1977-1978),andin 1987-1988and 1996-1997,courtesy of J. Shanklin.Note the rapid Septemberdrop in total ozonein the ozoneholeyears.

__

260

I

_

I

.

.I

.

!

.

I ....

!

I

.

I

ß ...I

. ,

I

ß

,

,

Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr Moy Jun Jon Feb Mar Apr May Jun Jul

Aug Sep Oct Nov Dec SH month

88/89

240

96/97 Arctic

,.-.

NH month

Arctic

/ x

•

x!• :v:";"'

220

2O0

82 ø Latitude, zonally averaged 7Orebar (near 18kin) 180

I

80

'

I

240

'

'!'

'

I

;500

'

I "' '

!

360

' ' I

''

!

420

'

'l'

'

!

480

'

!

'

540

Day Number Plate 3. Observations of the zonallyaveragedtemperaturesin the Arctic for 82øNnear 18 km, from the NationalCenterfor EnvironmentalPrediction(NCEP) meteorological databasefor two illustrativerecent

years,togetherwithAntarcticobservations at 82øSin 1997(shiftedby 6 monthsfor comparison).

37, 3 / REVIEWSOF GEOPHYSICS

Solomon: STRATOSPHERICOZONE DEPLETION ß 285

30

0

_

25

'"',.,• South Pole Ozone '•...•.• October Average ';•

1967-1971

I I I

5

Typical Polar

0

Stratospheric

•_.;• ...._• •DU

Cloud

E 20 I

.,.'•

I I

•

5

_ •-;•

I0

_

0

5

O-

I

I

I

0

5

I0

15

0

2:0

OzonePartial Pressure(mPa)

0

I

I

I

I

I

I

I

I

2:

3

4

5

6

7

Backscatter Ratio

Figure 5. (left) Observations of the verticalprofile of ozoneobservedat the SouthPole duringOctobersin the late 1960sand early1970s,contrastedwith thoseof 1986and 1997.Total ozone(DU) is indicatedfor each profile,from Hofmannet al. [1997].(right) Typicalpolar stratospheric cloudobservedat the SouthPole from the observations of Collinset al. [1993].

Solomon[1988];the theoriesthat were not supportedby observationswill be briefly discussedin this section. The primary dynamical theory of ozone depletion rested upon the notion that illumination of the cold polar lower stratosphereat the end of winter couldgive rise to heatingand net upwardmotion [Tunget al., 1986; Mahlman and Fels, 1986]. As is illustratedin Figure 5, there is far less ozone in the troposphere than the stratosphere,sothat upwardtransportof ozone-poorair from the troposphereto the stratospherecould locally decrease the Antarctic ozone column. However, observations of conservative

tracers that serve to illustrate

the

directionof dynamicalflow suchasaerosols[Hofmannet al., 1987],nitrousoxide[Parrishet al., 1988;Loewenstein et al., 1989], and other long-livedgases[Jaramilloet al., 1989; Toon et al., 1989] quicklydemonstratedthat the ozone hole was not caused by such upward motion. Indeed, much as Brewer[1949] deducedthe nature of global transportfrom observationsof water vapor as a tracer of atmosphericdynamics,so have observationsof a wide varietyof chemicaltracersshownthat transportis directed downwardwithin the Antarctic stratospherein springrather than upward,althoughthere is still debate about the strengthof this fluid flow and the degree of exchangeof air between lower latitudes and polar regions[e.g.,Hartmannet al., 1989;Tuck,1989;Tucket al., 1997;Schoeberlet al., 1990, 1992a, 1995;Manneyet al., 1995a,b]. Enhanced nitrogen oxides from high solar activity that occurredin the early 1980swere alsoproposedas a causeof the ozone hole [Callis and Natarajan, 1986], drawingupon the well-knowncatalyticchemistryof NOx and its enhancementby processessuchas solar proton events as discussed in section 1. While

chemical

in char-

3.3). The solar theory proved to be in conflict with observations. Measurementsof the nitrogendioxidecolumn byNoxon[1978]andMcKenzieandJohnston[1984] displayedreducedrather than enhancednitrogenoxides over the southpolar regions.After the discoveryof the ozonehole, similarmeasurementsconfirmedthoseearly data usingboth infrared and visiblespectroscopy methods [Coffeyet al., 1989;Farmeret al., 1987;Mount et al., 1987]. Airborne measurementsof the latitudinal gradient of nitric oxide at 20 km (in the heart of the south polar ozonedestructionregionas shownin Figure5) by Faheyet al. [1989a] usinga chemiluminescence method are comparedwith the data of Noxon[1978]in Figure 6. Both data sets show that the southern high-latitude winter-springstratospherecontainsa minimum in nitrogen oxidesrather than a maximum as required by the solar theory of Antarctic ozone depletion.The differencesin the location of the steepgradientbetweenthe two data setslikely reflect differencesin seasonand local motion of the polar vortex. The observedshapeof the profileof ozonedepletionis alsoin conflictwith the solar theory,whichwouldpredictgreaterozonelossesat higher altitudesrather than removalonly in a narrow range of altitudefrom -12 to 25 km as shownin Figure 5. It is interestingto note that the observationsof Noxon [1978, 1979] revealed strong evidencefor a "cliff" in NO 2 in polar regionsseveralyearsbefore the Antarctic ozonehole was discovered.The chemistryof this anomaly was not understood,and it was called out as one of the challengesto scientific understandingof stratosphericchemistryof the time [see,e.g.,WMO, 1985].We now knowthat the chemistrythat producedNoxon'scliff is tied to that of the ozone hole; the relationshipbetween NO 2 and C10 will be discussedfurther below. Arguably, the ozone hole might have been predicted

acter,thistheoryis diametricallyoppositeto the chlorine theory, which requires that nitrogen oxide abundances be suppressed so that chlorineoxidesare not tied up in before it was observed had the the chlorinenitrate reservoir(see Figure 3 and section understoodin the early 1980s.

Noxon

cliff been better

286 ß Solomon' STRATOSPHERICOZONE DEPLETION

ß' ' ' ' I ' ' ' ' I i , , , I '

_

'E --0

E

Noxon, 1978 x

/

oJ,,,,,,,, ,,J 40 45 ..... 5'0 55 Latitude

120 -4Sept.,1987 ..,., •Fahey et al., 1989 1 F"'I'"'"I'"• 8o 40

0

I

•

•

58

60

62

64

66

68

70

72

Latitude (øS)

37, 3 /REVIEWS OF GEOPHYSICS

(R28) in Table l). The C12formed would photolyze rapidly in sunlit air and rapidly form C10. They also pointed out that this and related heterogeneousreactionswould suppressthe concentrationof NO 2 by forming HNO3, so that the C10 thereby releasedcould not readily reform the C1ONO2 reservoir.Thus it was recognized that rapid ozone loss via chlorine chemistry would require (1) the heterogeneous"activation"of chlorinefrom both the HC1 and C1ONO2 reservoirsand (2) the suppressionof NO2, an essentialelement in keepingthe chlorineactive.The productionof C12in this processimplies that sunlightwould be required to releaseC1,so that the ozonedepletionwould occurwhen air wasnot onlycoldbut alsosunlit(i.e., largelyin spring as observed,rather than in winterwhen the polar cap is continuouslydark or summerwhen it is warm). Some ozonelosscan take place evenin polar winter, however, due to atmosphericwavesthat movepolar air out to the sunlitatmospherefor brief periods[Tuck, 1989;Sanders et al., 1993;Roscoeet al., 1997]. Observationsof PSCs, low NO 2 amountsin polar regions(Figure 6), enhancedpolar HNO3 [Murcrayet al., 1975;Williamset al., 1982]and the verticalprofile of the ozone depletionbasedupon the Japanesemeasurements [Chubachi,1984]were cited in supportof heterogeneouschemistry as the primary processinitiating Antarctic ozone depletion.Sucha mechanismwould be

Figure 6. Observationsof the "cliff" in NO 2 reported by Noxon [1978]. The solidand dashedlinesrepresentNorthern Hemispheremeasurements, while the solidcirclesand crosses showSouthernHemisphereeveningand morningtwilightdata, respectively. The NO measurements of Faheyet al. [1989a]are shownfor comparison.The two moleculesinterchangerapidly with one anotherin the sunlitatmosphereand henceprovidea measure of NOx. Both data sets show very low NOx in the most effective high-latitudestratosphere.

in the Antarctic

because of colder

tem-

peraturesand greaterPSC frequenciesthere than in the correspondingseasonsin the Arctic [McCormicket al., 1982],a point discussed further below. 3.3. Heterogeneous ChlorineChemistryand As in the discussionof gas phasechemistry,a comAntarcticOzone Depletion:EarlyTheoreticalStudies plete understandingof ozonedepletionrequiresconsidThe light was especially good today; the sun was eration not only of how much C10 is present (i.e., directly reflected by a single twisted iridescent cloud in the North, a brilliant and most beautiful object.

C10/Cly)but alsoof the catalytic cyclesin whichC10

This quotationfrom one of the firstexplorersto stand at the South Pole documentsthe fact that polar stratosphericcloudswere presentin the Antarcticlongbefore the advent of the ozone hole. The term "polar stratosphericclouds"was coinedby McCormicket al. [1982], who firstpresentedsatelliteobservations of high-altitude clouds in the Antarctic and Arctic stratospheres.The data showedthat the Antarctic cloudswere presentfrom June to late September,that they were associatedwith coldtemperaturesbelow -200 K, andthat theyoccurred between ---12 and 25 km. Three quartersof a century after their observationduring Scott's expedition and severalyears after the first satellite observations,it be-

may engage.Solomonet al. [1986] emphasizedthe catalytic ozone destructioninitiated by the reaction between HO 2 and C10. However, this processcannot destroyenoughozoneearly enoughin the springseasonto be consistentwith the detailed seasonalityof the ozone lossprocessas shownabovein Plate 2. Molina and Molina [1987] showedthat very rapid ozone depletioncan occurthrougha previouslyunrecognizedcatalyticcycleinvolvingformationand photolysis of a ClO dimer, C1202.Following a period of some uncertaintyregardingthe kineticsandphotochemistry of the dimer, laboratory studiesconfirmed its importance [e.g.,Sanderet al., 1989] (seeRodriguezet al. [1990]for model calculationsand JPL [1997] for a detailed summary of laboratorydata). This cycleis now well recognized as the primary catalyticprocessresponsiblefor

came clear that PSCs are a critical

about 75% of the ozone removal

Robert Falcon Scott, diary entry for August 1, 1911 [Scott, 1996, p. 264]

factor in the ozone

hole.

Solomon et al. [1986] suggestedthat HC1 and C1ONO2might react on the surfacesof PSCs,perturbing gas phasechlorine partitioning in a manner that could greatly accelerate ozone loss in the Antarctic lower stratosphere(HC1 + CIONO 2 --> HNO 3 + CI2; see

in the ozone hole.

McElroy et al. [1986] and Tunget al. [1986] emphasizedthe role of bromine chemistryin ozone hole formation (in particular,its couplingto chlorinethrough the reactionbetweenC10 and BrO); this cycleis now known to contribute

about 20% to the annual formation

of the Antarcticozonehole [e.g.,Anderson et al., 1989].

37, 3 / REVIEWS OF GEOPHYSICS

Both McElroy et al. [1986] and Tung et al. [1986] also emphasizedthe need for reducedNO 2 in order for C10 to remainactive(notingthe linksto the Noxoncliff), and McElroy et al. [1986] also emphasizedthe Japanese ozonesondeobservations, particularlythe observationof ozone lossat low altitudes,where bromine can be very

Solomon: STRATOSPHERICOZONE DEPLETION ß 287

importance of atmosphericwaves in modulating the temperaturesand sunlightthat influencethe ozone loss process[e.g.,Joneset al., 1989], while othershave underscoredthe role of interannualvariabilityin dynamical conditionsin determining not only the temperaturedependentchemistry[Tie et al., 1997] but also the reeffective for ozone destruction. supplyof ozone into the depletedregion, hence moduPartly becauseof limited analysisof Antarctic strato- lating the chemistry[Knightet al., 1998]. Schoeberland spheric temperatures,early studiessuch as those de- Hartmann [1991] and Schoeberlet al. [1996] have used scribedabovewere not specificabout the type of parti- dynamicaltracerssuchas N20 to carefully identify the clesof which the observedPSCswere composed.It was edge of the vortex and showits links to both dynamics generallyassumedthat the particleswere mainly water and chemistry.In particular,they have simulatedchemice [Steeleet al., 1983]. Stratosphericice clouds are ical perturbationsat the outer fringesof the Antarctic frequentlyopticallythick andbrilliant in color,like those vortex, including high C1ONO2 abundancesobserved, observedby Captain Scott.Suchcloudsform when tem- for example,by Toon et al. [1989]. peratures drop below the frost point and are now reSince ozone providesthe primary sourceof heat to ferred to as type2 PSCs.However, more sensitivesatel- the stratospherethroughits absorptionof UV radiation, lite measurements[McCormicket al., 1982] suggested Shine [1986] noted that the ozone hole shouldbe exthat opticallythinner PSCswere alsopresentat warmer pectedto lead to a stratosphericcooling,which in turn temperatures. could make heterogeneouschemistryeven more effecToon et al. [1986] and Crutzenand Arnold [1986] tive. The seasonal increase in Antarctic ozone observed pointedout that the PSCsparticlesmight be composed after October as shown in Plate 2 is associated with the not only of water ice but also of solid nitric acid trihy- seasonalwarming and breakdownof the Antarctic vordrate (NAT). Both studiesnoted that suchcomposition tex, which allowsozone-richair to flow into the region. could affect the impact on ozone in two ways:(1) by Its delayin recentyearscomparedwith historicaldata is reducingthe amount of nitrogen oxide that could be evident in Plate 2 and suggeststhat such a positive present(i.e., not only by formingnitric acid but alsoby feedbackmechanismhas indeed modified not only the removingit from the gasphase)and (2) by raisingthe October-Novembertemperatures[NewmanandRandel, temperatureat whichcloudscouldform, sincethermo- 1988] but also the meteorologicalcharacteristicsof the dynamicanalysessuggested that NAT couldcondenseat Antarcticstratosphere[seeJonesand Shanklin,1995]in temperatureswell above the frost point. These clouds a fashionthat prolongsthe ozonehole. It is importantto cameto be known astype1 PSCs.In addition, Toonet al. note, however,that analysesof winter temperaturesin [1986] suggestedthat sedimentationof large particles the Antarcticstratospherereveallittle or no evidencefor could result in denitrificationof the stratosphere.The coolingbeforethe ozonedepletionoccursin September removalof nitric acid not only from the gasphasebut [Newmanand Randel, 1988; Trenberthand Olson,1989; from the stratospherealtogetherwould have a potential Jonesand Shanklin,1995],confirmingthat the meteoroto further reduce NO 2 concentrationsand hence en- logical changesare primarily a consequenceand not a hance C10/CIONO 2 ratios and attendant chlorine-cata- cause of the ozone hole. lyzed ozone loss.McElroy et al. [1986] also considered the possibilityof nitric acid-water particles,suggesting 3.4. Heterogeneous ChlorineChemistryand the that nitric acidmonohydrate(NAM) waslikely to form. Ozone Hole: Fieldand LaboratoryObservations Table 1 includesa list of the major heterogeneous Foremostamongthe data that establishedthe cause processesof importancein the stratosphere.Through of the ozone hole are observations of active chlorine theseheterogeneous reactions,the chemicalpartitioning species,particularlyC10. De Zafra et al. [1987, 1989] of chlorinein the Antarcticlower stratospherein spring presentedground-basedmicrowaveemissionmeasurecan be greatlyperturbedin comparisonwith gasphase ments at McMurdo Station, Antarctica, showingevichemistry,making chlorine (and its couplingwith bro- dence for greatly enhancedC10 in the lower stratomine) far more damagingto ozonethan it wouldbe in a sphere.Near 20 km the observationssuggestedmixing gasphaseframework.A broad rangeof models,includ- ratiosof ---1ppbvin September,about 100 timesgreater ing two-dimensional[e.g., Isaksenand Stordal, 1986; than the 10 pptv predictedby gasphasephotochemical Chipperfieldand Pyle, 1988; Brasseurand Hitchman, theory.Andersonet al. [1989] presentedin situ airborne 1988; Ko et al., 1989; Rodriguezet al., 1989; Tie et al., measurementsof C10 using a resonancefluorescence 1997],three-dimensional[e.g.,Cariolleet al., 1990;Aus- method. The latitudinal coverageof the airborne data tin et al., 1992;Brasseuret al., 1997;Knightet al., 1998], taken flyingsouthfrom Chile near 20 km showeda very and trajectoryand Lagrangianstudies[e.g.,Joneset al., steepgradient in C10 as the airplane crossedinto cold 1989; Schoeberlet al., 1996] have probed this basic regionswithin the Antarcticvortex,increasingto about1 frameworkand presentednumericalanalysesof calcu- ppbv in mid-Septemberas shown in Figure 7. Hence lated ozonetrends.Severalauthorshaveemphasizedthe within a few yearsafter the discoveryof the ozonehole,

288 ß Solomon'STRATOSPHERIC OZONE DEPLETION I

I

I

I

I

!

I

I

I

',

observations asdepictedin Plate1 haveallowedstudyof the full globaldistributionof C10 [Waterset al., 1993, 1999]usingmicrowaveemissionmethodsthat further tie the ozonedestruction regionwith the spatialdistribution of enhancementsin C10 [Manneyet al., 1995b;MacKenzie et al., 1996]. The front cover of this issue of Reviewsof Geophysics presentssatelliteobservationsof lower stratosphericozoneand C10 on August30, 1996, over Antarctica.The data showthat the region of reducedozoneextendsover an area larger than the continentbeneathandillustratethe closespatialcorrespondencebetweenthe regionsof depletedozoneand those of enhancedC10 as first emphasizedbyAndersonet al. [1989]. Observationsof chlorinedioxide(OC10) via visible spectroscopy wasanotherindependentmethodof probing the chlorine chemistrythat also revealed hundred-

Southern Hemisphere

1.2

-20

o

I

37, 3 / REVIEWSOF GEOPHYSICS

km

0.4

h:---:i:'i:--i'" I

0.0 12

1

22 Sept., 1987 I I I

56

58

60

62

I

I

I

I

I

64

66

68

70

72

fold enhancements

of active chlorine in the Antarctic

74

vortex [Solomonet al., 1987; Wahneret al., 1989;Kreher Latitude (øS) et al., 1996].This techniquealso allowedstudyof the seasonalchangesin Antarctic chlorine activationand its Figure 7. Observationsof the latitude gradients in C10, links to PSC chemistry.Theseshoweda seasonaldecline NOy, andH20 on a flightof the ER-2 aircraftin September of OC10 betweenlate Augustand early October,asso1987,showingevidencefor extremelyhighC10 in the Antarctic togetherwith substantialdenitrificationand dehydration ciatedwith increasingtemperaturesand the cessationof heterogeneous chemistry[Solomonet al., 1987]. (removalof NOyandH20) associated withPSCs. Observations of the HC1 column,and in particularits ratio to the HF column,stronglysuggested that HC1 had two independentmethods confirmedremarkablyele- indeed been converted to active chlorine in the Antarctic vated C10 abundances in the ozonehole region,which spring[Farmeretal., 1987;Toonet al., 1989;Coffeyetal., are possibleonly if chlorineis releasedfrom both of the 1989].The recent global satellitedata by Russellet al. reservoirgases,HC1 and C1ONO2. [1993]furtherdemonstrate thisbehavioron largerspaFigure 8 showsboth ground-basedand airbornemea- tial scales. The first in situ measurements of HC1 showsurementsof C10 in Septemberover Antarctica from ing evidencefor conversionto active chlorinewere ob1987,and comparesthem with gasphaseand heteroge- tained in Arctic studiesby Websteret al. [1993]; see neous photochemicaltheory. More recently, satellite section6.1. Concurrentglobal HC1 and C1ONO2 data from the UARS 700

I

i

i

i

Observations

X McMurdo, :>0-24 Sept., 1987 g Aircraft 2:> Sept., 1987 o Aircraft 21 Sept., 1987 '•

600

24• o _.

Gas- phase photochemistry

E

500 • x

•

-20

.u_ -•

400

the simultaneous

Antarctic in situ observationsof both C10 and HC1,

whichdramaticallyillustraterapid activationat temperaturesbelowabout195K [Kawaet al., 1997](seeearlier studiesby Tooheyetal. [1993]andSchoeberl etal. [1993a, b]) andprovidea key demonstration of rapidheteroge-

-16

neouschemistryunder cold conditions.

-12

C10, HC1,OC10, NO, NO2, andothergasesby a variety

The Modelincludingheterogeneous chemistry

satellite illustrated

chemicalconversionof both specieswherepolar stratosphericcloud surfaceswere also present[Gelleret al., 1995;Yudinet al., 1997].A detailedviewof the temperaturedependence of chlorineactivationisprovidedfrom

combination

of simultaneous

observations

of

of independentchemicalmethodsdemonstratesthat the springtimeAntarcticstratosphere is indeedheavilyperClO (pptv) turbedcomparedto expectations from gasphasechemistry, and in a manner consistentwith heterogeneous Figure 8. Observationsof the verticalprofilesof C10 in the reactions on PSC particles.Throughthe resultingdraAntarcticstratosphere in September1987from both ground300

0

I000

2000

basedmicrowaveremote sensing[de Zafra et al., 1989] and aircraft resonancefluorescencetechniques[Andersonet al., 1989].Thesedata are comparedwith a gasphasephotochemistry model and with the heterogeneous chemistrymodel of Joneset al. [1989],whichaccountsfor air parceltrajectories.

maticenhancements in the C10/Clyratio,chlorine's ef-

fectivenessfor ozone destructionis greatly enhanced. For C10 abundances near 1-1.3 ppbvas observedsince 1986-1987 (and BrO abundances near 7-10 pptv [see Bruneet al., 1989]),Antarcticozoneis destroyednear20

37, 3 / REVIEWS OF GEOPHYSICS

Solomon: STRATOSPHERICOZONE DEPLETION ß 289

kmin September at a rateof about0.06-0.1ppmvd-1, lower stratosphereinclude those of Santeeet al. [1998, so that within -40-60 daysvirtually all of the ozone at this level can be depletedunlessrapid dynamicalresupply occurs,broadlyconsistentwith Figure 5 [seeAnderson et al., 1989;Murphy, 1991;MacKenzieet al., 1996]. The cold temperaturesobservedin the Antarctic during September in most years suggestthat net downward motion (which would tend to warm the air through adiabaticcompression)and horizontal mixing is relatively limited at that time. This generalpicture of relative dynamicalisolationin the heart of the ozone depletion region is supported by a number of dynamical studies[e.g.,Hartmann et al., 1989;Manneyet al., 1994b, 1995a,b; Schoeberlet al., 1995, 1996]. The observationsoffer several different spatial and temporal fingerprintsthat stronglysupportthe identification of chlorine chemistryand its perturbationsby heterogeneousprocessesas the principal cause of the ozonehole. The measurementsshownin Figure 8 reveal

1999], Voemelet al. [1995] and Pierceet al. [1994]. Many laboratory studieshave confirmed that rapid heterogeneous processes do indeed take place on the kind of surfacespresentin polar regions.The fundamental principles of surface chemistryare outlined in the excellent book by Somorjai [1994], which illustrates many of the factors that allow surfaces to facilitate processesthat do not happen, or happen only very slowly,in the gasphase.Fairbrotheret al. [1997] review thermodynamicprinciplesbehind stratosphericheterogeneouschemistry.The first laboratorystudiesof stratospheric reactions on ice by Molina et al. [1987] and Tolbertet al. [1987]showedthat the reactionof HC1with C1ONO2 indeed takes place readily on water ice films. Hansonand Ravishankara[1994] showedthat a portion of the reaction is due to HOC1 + HC1 --• C12q- H20 , followingformation of HOC1 through the surfacereaction C1ONO2+ H•O-• HOC1 + HNO3. Prather[1992a] that the enhanced C10 occurs over about the 12- to discussedthe implications of a surface HOC1 + HC1 25-km range, the region where PSCs are observedand reaction for ozone depletion. Numerous laboratory inwhere the ozone is depleted as shownin Figure 5. The vestigationshave shownthat HC1 reactswith C1ONO• airborne data of Andersonand colleagues,as depicted, on nitric acidtrihydrateice surfacesaswell, althoughthe for example,in Figure 7, demonstratethe steeplatiturate dependson factorssuchas the HC1 partial pressure dinal gradient in C10, consistentwith the connectionof and on the water contentof the NAT surface[Leu, 1988; the ozone hole to cold Antarctic latitudes;a fully threeAbbatt et al., 1992; Hanson and Ravishankara, 1993; dimensionalview of the samebehaviorbasedupon satPeter,1997;Carslawand Peter,1997].Someauthorshave ellite data [Waterset al., 1993,1999]is illustratedin Plate noted that heterogeneousreactions between bromine 2 and the cover of this issue. Seasonal observations of OC10, C10, HC1, NO2, and other gaseshave been used to show that the large temporal changesin the abundancesof thesechemicalspeciesare consistentwith the time evolution of the ozone hole and with heterogeneous chemistry.In short, the vertical, latitudinal, and

and chlorine can also contribute to chlorine activation,

through,for example,reactionbetweenHOBr and HC1 [Hansonand Ravishankara,1995; Danilin and McConnell, 1995]. For a review of recent laboratorystudiesof these and other heterogeneous processes,see JPL [1997]. seasonal behavior observed in active chlorine and a host Both the nature of stratosphericsurfacesand the of related speciesall provideindependentevidencecondetailed reaction mechanisms that can occur within firming the basicprocessesthat control the occurrence them have been the subjectsof many studies.C1ONO• of the ozone hole. Antarctic field measurementsalso allowed study of hydrolysison ice may proceedvia nucleophilicattack at the formation, composition,and seasonalbehavior of C1 by a lattice water molecule in concert with proton polar stratosphericclouds.Fahey et al. [1989b, 1990b], transfer [Blanco and Hynes, 1998]. It is possiblethat Gandrudet al. [1989], and Pueschelet al. [1989] carried proton transfer from HC1 to water forms C1- on ice out the first observationsof the compositionof polar surfaces,allowing reaction with C1ONO• through an stratosphericcloudsand demonstratedthat the particles ion-assistedand hence efficient mechanism[Van Doren do indeed contain nitric acid as had been predicted. et al., 1994].FundamentalsurfacechemistrymodelssugLaboratory studies [Hanson and Mauersberger,1988] gestthat HC1 forms a bilayer on ice, allowingionization confirmedthe thermodynamicstabilityof NAT at tem- and subsequentsurfacechemistry[Gertnetand Hynes, peratureswell abovethe frost point. Observationsdem- 1996]. On the other hand, Matereret al. [1997] suggest onstrated thatgasphasereactive nitrogen orNOy[Fahey that a quasi-liquidlayer displayingless order than the et al., 1989b] and water vapor [Kelly et al., 1989] are bulk maybe presentat an ice surfaceunder stratospheric stronglydepleted in the Antarctic stratosphere,as was conditions,so that uptake of HC1 into stratosphericice predictedby Toonet al. [1986] on the basisof sedimen- particlesmay occurthrougha processakin to solutionat tation of large type 2 PSC particles. Figure 7 shows a quasi-liquidinterface.In additionto theseinteresting evidence for NOyandH20 removalin thesameregion questionsregardingthe fundamentalphysicalchemistry displayingenhancedC10 in the Antarctic stratosphere of heterogeneousreactionsin the stratosphere,the un(poleward of -64øS in that particular transect) from derstandingof the compositionof polar stratospheric airborne studies.Later measurementsdocumentingthe cloudshas alsobeen a subjectof intensestudyand is the "denitrification" and "dehydration" of the Antarctic subjectof the next section.

290 ß Solomon: STRATOSPHERIC OZONE DEPLETION

37, 3 /REVIEWS OF GEOPHYSICS

freezingof these"background"sulfateaerosolparticles followed by uptake of nitric acid and water when the RETHINKING PSCS NAT condensation temperaturewasreached[Dyeet al., 1992;Molina et al., 1993] (see Peter [1997] and referThe foregoingdiscussion illustratesthe rapidprogress ences therein). Recent observationsand theoretical made toward understandingthe key role of heteroge- studiesraise many questionsabout this picture, as is neouschlorinechemistryon polar stratosphericclouds discussed below. in the formation of the Antarctic ozone hole in the last ObservationsbyDye et al. [1992] and their analysisby half of the decadeof the 1980s.However, at presentour Carslawet al. [1994],Drdlaet al. [1994]and Tabazadeh et conceptualpicture of the compositionand chemistryof al. [1994]have demonstratedthat somePSCsare probPSCshasevolvedconsiderably from the relativelysimple ablycomposedof supercooled ternaryliquid solutionsof one that prevaileda decadeago.Detailed recentreviews HNO3-H2SO4-H20. Briefly, the observationsshoweda of the microphysics, thermodynamics, and heterogenous smooth growth in particulatevolume with decreasing chemistryof stratosphericparticulatesare providedby temperaturerather than a "stepfunction"in growthat Peter [1997] and Carslawet al. [1997a], and excellent the stability point for NAT. The dependenceof the shortsummariesof laboratorywork are givenby Tolbert growth of particulatevolume with temperatureclosely [1994, 1996].Only the key pointswill be reviewedhere. followedmodel predictionsbasedon the thermodynamBoth Arctic and Antarctic observationsthat jointly con- icsof ternaryliquid solutions(seeCarslawet al. [1997a] tribute to this understandingwill be describedin this for an in-depthreview).Laboratorystudiessupportthis section. picture by demonstratingthat realistic solutionsand particlescontainingsulfuricacid, water, and nitric acid 4.1. What Are PSCs Made of? remain liquid even at very cold temperatures,as low as Toon et al. [1990]were the first to presentevidence 188 K [e.g.,Beyeret al., 1994;Koop et al., 1995, 1997; that PSCs may be composedof liquid as well as solid Anthony et al., 1997; Clapp et al., 1997; Bertram and particles,drawinguponlidar measurements byBrowellet Sloan, 1998]. Indeed, the resultsof theserecent laboraal. [1990]. Briefly, the observationsby Browell et al. tory studiesshowthat it is extremelydifficult to make [1990],andlater work [e.g.,Beyerle et al., 1994;Adrianiet such particlesfreeze at temperaturesabove the frost al., 1995; Steffanuttiet al., 1995; Gobbi et al., 1998] point, evenwhen they are kept cold for many hours. Thus recent field and laboratory observations,and showedevidencefor high backscatter(hencethe detection of clouds).However,while somedatarevealedhigh related modelingstudies,have substantiallyaltered the accompanying depolarizationas expectedfor aspherical conceptualunderstandingof PSCs.The data showthat solid particles, other measurementsshowedvery low PSCs are liquid much of the time. While there is no depolarizationof the backscattersignal,suggestingliq- difficultyunderstandingthe formation of frozen type 2 uid rather than solid particles.The two distinctcloud water ice cloudsthat form below the frost point [see typesresultedin a further subdivisionof PSCsinto type Poole and McCormick, 1988a, b; MacKenzie et al., 1995; la (depolarizingsolid) and type lb (nondepolarizing Peter,1997],the mechanismwherebysolidtype 1 PSCs liquid). sometimes(but not always)are presentat temperatures In situ and satellite measurements also showed eviabovethe frostpoint [e.g.,Pooleet al., 1988]is not clear. dencefor shortcomings in our understandingby demon- The studiesby Tabazadehet al. [1996], Santeeet al. stratingthat while PSCsform at temperaturesabovethe [1998a],and Larsenet al. [1997] suggestthat temperafrost point and do containnitric acid,the detailedrela- ture historiesaswell as local temperaturesare likely to tionshipsto temperatureare often difficultto reconcile be importantin determiningwhen and if freezingoccurs. with NAT thermodynamics[Kawa et al., 1990, 1992; Mesoscaletemperaturefluctuations(i.e., transientrapid Rosen et al., 1989; Arnold, 1992; Dye et al., 1996; De! coolingin atmosphericwaves)couldlead to nonequilibNegroet al., 1997;Santeeet al., 1998a].Toonand Tolbert rium conditions under which NAT or the nitric acid [1995] further showedthat infrared spectraof type 1 dihydrate,NAD (emphasizedby Worsnopet al. [1993]) PSCs observed over Antarctica were inconsistent with could freeze [Meilingeret al., 1995; Tsiaset al., 1997]. On the other hand, Tabazadehet al. [1994]and Tabathoseexpectedfor NAT. ' Arnold [1992] showedevidencefor uptake of HNO3 zadeh and Toon [1996] have emphasizedamorphous into PSCsthat could not be reconciledwith NAT parti- solid solutionsand water-richhydratesas possibleprecles and suggestedthat ternary liquid HNO3-H2SO4- cursorsto freezing,while MacKenzieet al. [1995] and H20 particles might be responsiblefor the observed Drdla et al. [1994] suggestedthat trace impuritiesin anomalies.The introductionof H2SO4 in this discussion stratospheric particles(suchas meteoriticmaterial, ormerits a brief descriptionof its origins. It has been ganics,or soot;seeobservations byMurphyet al. [1998]) known sincethe pioneeringwork of Jungeet al. [1961] may play a key role in the freezingprocess.However, that particlescomposedof sulfuricacid and water form Iraci et al. [1998]recentlyshowedthat solidsulfuricacid throughoutthe stratosphere.The mechanismresponsi- tetrahydrate (SAT) particlesmay form in the laboratory ble for growth of PSCswas originallythoughtto involve undercertainconditions,then (in a perhapsironictwist) 4.

FORMATION, COMPOSITION,

ROLE OF POLAR

STRATOSPHERIC

AND CHEMICAL

CLOUDS:

37, 3 / REVIEWS OF GEOPHYSICS

Solomon: STRATOSPHERICOZONE DEPLETION ß 291

hat

HCl + ClONO2•HN032i+ Cl2

Range, •

IOøI' ,/solid 'I'-•,LI ,,,,,,,,,'

•' water •/Liquid sulfuric :>0 IJ_\\

•

c •

Upper es•ma•,

Lower estimam,

•

nitricacid/water

•eF 5ppmv H20 ',,

•

; • •Han•n md•

ß Tolbertetal.

•

•

ß ,. 1•4• TTT• ] T J• Ti T, •!, • , 0

I0

20

30

•

50

•

15

I0-3

-I0

-

c••e •acid/water -- nitric acid/water • •'•', •

• •o-•

Liqui dsulfurJc -acid/water

ICj

60

•

% H2SO 4

-

•

•

•

13•

•

; H_O •-

•

80

-

I1• 190

•

195

200

25ppmv '

• "2- •._ 205

Figure 9. (left) Laboratorydata on the efficiency of the reaction between HC1 and C1ONO2 for ice, nitric acid-water solid surfaces,and liquid sulfuricacid-water solutions. (right) Altitude variationof the temperatureat which the efficiencyof this reactionon liquid sulfuricacid-water solutionsbecomesgreater than 0.3 for water vapormixingratiostypically observedin the lower stratosphere.

210

Approximate Temperature at which Y>0.3

take up HNO 3 not in a coolingbut in a melting phase, which couldbe followednot by completemelting of the particle but rather by crystallizationof NAT. Thus the solidSAT couldform a corethat allowsuptakeof HNO 3 into a liquid melting surface,followed by freezing. 4.2.

What is the Effect of Different

Surfaces on

Ozone DepletionChemistry? It is extremely important to consider whether the phase and microphysicalmechanismsunderlying PSC formation are important for ozone depletion. The uptake of condensable vaporsenhancessurfaceareaswhen PSCs are present.This increasesthe gas-particlecollision frequency and hence can enhance the rates of heterogeneous reactions. Background stratospheric aerosolsgrow into type la and/or type lb particles as they cool, then further grow into type 2 water ice PSCs if temperaturesfall below the freezingpoint. The impact of enhanced surface areas for chemistrydepends not only on the frequency with which gasesstrike these surfacesbut alsoon the reactivityof the surfacesand the availabilityof those gases. Let us first consider

how reactive

the different

sur-

faces are and where they are to be found in the stratosphere.Laboratory studieshave shownthat water ice, NAT, and liquid ternary solutionsare all effective for activatingchlorine heterogeneously,but with differing efficienciesand with different dependencieson temperature, water vapor abundance, and pressure [e.g., Carslawet al., 1997a;JPL, 1997,and referencestherein]. These dependenciesare related to the thermodynamics of the different surfaces,which control not only their

surfaceareasbut alsotheir composition(especiallythe uptakeof HC1 onto/intothe particles). Figure 9 summarizesa number of laboratory measurementsof the efficiencyof the key heterogeneous surfacereactionHC1 + C1ONO2 -->C12+ HNO 3 (where

1 indicatesreactionon everycollisionof C1ONO2with a surface,0.1 indicates1 reaction in 10 collisions,etc.). Water ice is believed to be highly reactivewherever it can form, but the thermodynamicsof ice condensation implythat rather coldtemperaturesare requiredto form it in the stratosphere(e.g.,below ---188K near 20 km). There is currently debate about the efficiencyof NAT for this reaction[Carslawand Peter,1997;JPL, 1997] as well as the conditions under which NAT can form, but

from a thermodynamicviewpoint,NAT may form near 20 km when temperaturesdrop below ---195 K, thus allowing reaction on a solid surface at temperatures abovethe frost point. In the caseof liquid solutions,the efficiencyof reactiondependsstronglyon the fractionof water in the particle [e.g., Tolbertet al., 1988;Hansonet al., 1994]. The HC1 + C1ONO2 reactionbecomesmore efficient for lower percentagesof sulfuric acid (and higher water in the liquid particles, which greatly increasesthe solubilityof HC1; seeRobinsonet al. [1998] for a detailedrecentanalysis).This reactiontakesplace in liquid solutionswith an efficiencygreaterthan 1 in 100 (0.01 asshownin Figure9) for temperaturescolderthan ---197 K at 20 km, and an efficiency of 1 in 10 for temperaturesbelow ---195K. Ravishankaraand Hanson [1996] have emphasizedthat liquid PSCscan be comparable to or more effective than solid PSCs for many surfacereactionsat temperaturesbelow ---195 K at 20 km, a point also illustratedby Cox et al. [1994], Borrmann et al. [1997a] and Del Negroet al. [1997]. Becauseliquid aerosolsare present throughout the global stratosphereand becausethe water vapor pressures available

to condense

into

them

increases

with

increasingtotal pressure,the temperaturesat which effective reactions may occur in liquid particles are higherfor lower altitudes[Hofmannand Oltman, 1992], as is also shownin Figure 9. This is a critical issuefor both the polar and midlatitudelower stratospheres. Fig-

292 ß Solomon: STRATOSPHERICOZONE DEPLETION

ure 9 suggeststhat both liquid and solid surfacescan activatechlorine efficientlynear the tropopause[Borrmann et al., 1996, 1997b;Bregmanet al., 1997;Solomon et al., 1997]. Observationsof enhancedC10 and reduced NO closeto the tropopausefor relativelywet (15 ppmv of H20 ) conditionsprovide evidencefor such liquid surfacechemistry[Keimet al., 1996].Note that Figure 9 is baseduponthe thermodynamicmodelof Carslawet al. [1997a];its extrapolationto very highwater vaporpressures(e.g.,>5 ppmvat 200mbar)isuncertainat present and requireslaboratorystudiesfor thoseconditions[see Robinsonet al., 1998]. It is useful to note that the HC1 + C1ONO2 reaction competes with H20 + C1ONO2 for the available C1ONO2 on liquid aerosol surfaces.Thus if HC1 has been depleted,the rate of the latter reactionincreases, so that effectiveheterogeneousactivationof chlorineis not dependent upon both HC1 and C1ONO2 being present.Further, the reactionsof HC1 with HOC1 and HOBr are also quite efficient on liquid aerosolunder moderately cold and/or wet conditions[Ravishankara andHanson,1996;JPL, 1997],providingadditionalpathwaysfor chlorine activation. A key conclusionof Figure 9 is that while there are differencesand uncertaintiesin the reactivityof various surfaces,rather effective chlorine activatingreactions can occurirrespectiveof particlephasebelow---198K at 20 km and below 200-210

K near 12-14

km. As an air

parcel cools and particle reactivitiesincrease,liquid chemistrywill occurfirst. This may be followedby reactions on NAT and ice, dependingon factorsincluding microphysics,the minimum temperaturereached,and whether or not all of the chlorineactivationhas already occurred[Turcoand Hamill, 1992]. This latter point is critical.For example,if effectivechemicalprocessing on liquid surfaceshas depleted all of the available HC1 and/or C1ONO2within an air parcel,then further lowering of temperature and formation of, for example NAT, may have a limited effect on ozone depletion. Moreover, an increased rate of reaction and/or an in-

37, 3 / REVIEWS OF GEOPHYSICS

and explosivevolcanic eruptions that inject SO2 gas directly into the stratosphere[e.g., McCormicket al., 1995] which subsequentlyforms liquid sulfate aerosols. Observationsof PSC extinction show that the major eruptionsof E1 Chich6nin 1981and Mount Pinatuboin 1991 led to large increasesin particle surfaceareasin polar regions[e.g.,McCormicket al., 1995;Deshleret al., 1992;Thomasonet al., 1997].Hofmannet al. [1992,1997] andHofmannand Oltmans[1993]showedthat enhanced aerosol surface areas due to Pinatubo expandedthe altitude range of significantAntarctic ozone depletion into one of its margins,downto lower,warmer altitudes (i.e., 10-14 km) where solidPSCsdo not form. Observationsof OC10 in the Antarctic fall season(MarchApril) at temperaturesabove200 K in the year immediately following Pinatubo also suggest significant activationof chlorinethroughsulfateaerosolprocessing [Solomonet al., 1993]. Hence both ozoneand trace gas observations from the Antarcticprovidesupportfor the role of temperature-dependent heterogeneous chemistry on liquidaerosols. Portmannetal. [1996]showedthat volcanicallyenhanced PSC surface areas were likely responsiblefor the sharponsetof the ozonehole in the early 1980sfollowingthe E1 Chich6n eruption, and for the very deep ozone holesobservedin the early 1990s followingthe Pinatuboeruption. It may be usefulto pausefor a brief summaryof the conceptualpicture for Antarctic ozone depletionthat emerged in the late 1980s and describe how it has changed.Initial observationsof Antarctic chemistryas discussedabove showed evidence both for heterogeneous chlorine

activation

on PSCs and for denitrifica-

tion. An understandingemergedthat chlorine-activating reactionstook place on solidPSCsin Antarcticwinter, accompaniedby denitrificationthat allowedthe depletion to persistin spring,even in the absenceof further PSC formation. This picture was simple and easy to explainin chemicalterms.However,the currentunderstandingsuggests that denitrificationcan increaseozone destructionsomewhatbut is not required for polar

creasedsurfacearea (through,for example,formationof NAT type 1 PSCs or type 2 PSCs) may not enhance ozone losses. This is because chlorine activation can ozone depletionin a time averagedsense.If, for exam- continueto occur on liquid aerosolsin spring,keeping ple, reactionson sulfate aerosolsare sufficientto acti- the chlorine active in sunlit air whether denitrified or not vate all of the available chlorine within a day, ozone [Portmannet al., 1996;ChipperfieMand Pyle, 1998].Obdepletionwill not be increasedif insteadreactionson ice servationsfollowing the eruption of Mount Pinatubo activate all of the chlorine in an hour, since the ozone supportthe view that liquid aerosolchemistryhas been depletionis a processthat occursover a much longer a key factor in determiningthe depth of the ozonehole period (weeks)followingthe activation.Hence the de- after major eruptions.More generally,the expansionof tails of the reactivitiesand the microphysics that control heterogeneouschemistryfrom ice, to NAT, to liquid particle surface areas, while playing a role to some sulfateaerosolshas lessenedthe expecteddependence degree,are not critical to formation of the ozone hole of the ozone losson extreme cold to one of relative cold, [e.g.,Portmannet al., 1996;Carslawet al., 1997b].They therebyexpandingthe height,time, and latitude ranges are likely to be more importantat the margins,particu- whereozonedepletionmaybe expectedto be enhanced larly regionswhere temperaturesare cold but not ex- by heterogeneousprocessesthat affect chlorinepartitremely cold. tioning. The next two sectionsexplore these issuesby Primary sourcesof sulfur to the stratosphereare illustratingthe important role of heterogeneousprocarbonylsulfide [Crutzen,1976; Chin and Davis, 1995] cessesfor both mid-latitudeand Arctic ozonedepletion.

37, 3 / REVIEWSOF GEOPHYSICS 5.

MIDLATITUDE

OZONE

DEPLETION

Solomon: STRATOSPHERICOZONE DEPLETION ß 293

dependenceon the particlecomposition,specifically,the water contentof the particle (henceon the temperature The discoveryof the Antarctic ozone hole naturally and water vapor pressure). raisedthe questionof whetherother latitudesmight also The laboratoryinvestigationsof Tolbertet al. [1988] displaygreaterozonedepletionthan expected.Within a andMozurkiewicz and Calvert[1988]promptedHofmann few years after the ozone hole was discovered,statisti- and Solomon[1989]to studythe role of N20 5 hydrolysis cally significanttrends in ozonewere found at northern and chlorine activation on sulfuric acid aerosols at midmidlatitudesas well [Ozone TrendsPanel, 1988, and latitudes, particularly under volcanic conditionswhen referencestherein]. By the 1990s,significanttrendshad suchprocesses would be enhanced.They suggestedthat been established for both northern and southern midthis chemistrycould be significantfor both background latitudes, not only in winter and spring but also in and volcanically perturbed conditions, and that the summer[WMO/UNEP, 1991, 1994;Stolarskiet al., 1991; ozonereductionsnoted by severalauthors[e.g.,,4driani McPeterset al., 1996a, b; Harris et al., 1997; Staehelin et et al., 1987]followingthe eruptionof E1 Chich6nin 1981 al., submittedmanuscript,1998]. Midlatitude ozonecol- might be linked to heterogeneousreactionssimilar to umn trends as of the 1990s are of the order of 5-10%, thoseoccurringin Antarctica, albeit with reducedrates. muchsmallerthan thoseof the Antarctic (Figure 1) but Observationsof marked reductionsin NO 2 over New far greater than gas phase model predictions.As in Zealand after the E1 Chich6n eruption [Johnstonand Antarctica, recent analyseshave shownthat the bulk of McKenzie, 1989] provided some of the first chemical the northern midlatitude ozone decline is occurringin evidencethat suchprocessescouldbe important at midthe lower stratosphere(near 12-20 km [seeMcCormick latitudes.Severalmodelingstudies[e.g.,Rodriguezet al., et al., 1992;Miller et al., 1995;WMO/UNEP, 1994;Bojkov 1991, 1994; Brasseurand Granier, 1992; Prather, 1992b; McElroy et al., 1992; Pitari and Rizi, 1993; Toumi et al., and Fioletov,1997;Harris et al., 1998]). 1993] further probedthe role of this chemistryin determining global ozone trends and related questionsof 5.1. Heterogeneous Chemistryand Midlatitude chemicalpartitioningand odd oxygendestructioncycles. Ozone Depletion One mechanism that could affect midlatitude ozone Prather[1992b]pointed out that the hydrolysisof N20 s depletionis heterogeneouschemistry.It had long been saturatesbeyonda certainaerosolload at whichN20 5 is suspectedthat some heterogeneousprocessinvolving convertedto HNO 3 as fast as it can be formed, so that N20 5 might be responsiblefor the Noxon 'cliff' and for further increasesin aerosol do not affect NOx abunanomalouslyhigh HNO 3 abundancesin middle to high dancesthrough this process. latitudesof the Northern Hemisphere[e.g.,Wofsy,1978; Rodriguezet al. [1991] and McElroy et al. [1992] Noxon, 1979;Austinet al., 1986]but it was not until the pointed out that hydrolysisof N20 5 would have the late 1980sthat laboratorystudiesshowedthat N20 5 can effect of dramaticallyaltering the competitionbetween hydrolyzerapidly (reaction efficiencyof about 0.1) on the various catalytic cyclesin the lower stratosphere, sulfuricacid-waterfilms [Tolbertet al., 1988] and parti- enhancingthe roles of the odd hydrogenand odd chlocles [Mozurkiewicz and Calvert,1988]. Hence the possi- rine/bromine destruction mechanisms, even for backbility of heterogeneouschemistryon the liquid sulfate groundaerosolconditions.Direct observations by Wennlayer that is pervasivethroughoutthe stratospherebegan berget al. [1994] later confirmedthis view by providing to be consideredin earnest(but see also Cadle et al. simultaneousmeasurementsof a suite of key radicals [1975]for an early and interestingexploratorypaper). including OH, HO2, NO, and C10 near 20 km; these The hydrolysisof N20 5 reducesNOx and its impact data can be related in hindsightto the anomalouslylow on ozone in the lower stratosphere,and indirectly en- NO mixingratiosreportedin the midlatitudelowermost hances the effect of C10 through its control of the stratosphereby Ridleyet al. [1987]. Cohenet al. [1994] C1ONO2/C10 ratio, as was discussedearlier. Recent presenta detailedchemicalargumentdemonstratingthe studies have examined the dependence of the N20 5 dominanceof HOx chemistryin the natural lower stratohydrolysisreaction on temperatureand pressure[Rob- spheric ozone balance based on observations.Taken insonet al., 1997] and haveprobedreactionconditionsin together,this improvedunderstandingof the balanceof extensivedetail [JPL, 1997]. This key reaction and the terms among chemicalcyclesof ozone destructionis a related hydrolysisof bromine nitrate [Hanson et al., key buildingblock for attemptsto evaluateozone loss, 1996] both take place rather rapidly at virtually all which is tied to the competitionof chlorine- and brostratosphericconditions, making their influence ex- mine-catalyzeddestructioncomparedwith other chemand to transport.In short,slowerrates of tremelywidespread(and, as is shownbelow, extremely ical processes important). In addition to these indirect effects,how- ozone loss through other processes(especiallyNOx ever, there is evidence for direct activation of chlorine chemistry)result in a larger relative role for humanon liquid sulfate aerosolsas well.. As noted above in induced perturbations at midlatitude due to chlorine connectionwith polar chemistry,Tolbert et al. [1988] and/or bromine increases. From about 1988 to the early 1990s the scientific suggestedthat C1ONO2couldreactwith water and with HC1 on sulfuric acid-water surfaces,but with a strong understandingof midlatitude ozone depletion evolved

294 ß Solomon' STRATOSPHERICOZONE DEPLETION

37, 3 / REVIEWSOF GEOPHYSICS

ObservedChangesinChemicalPartitioning DueTothe Eruptionof Mt.Pinatubo 0.07

0

Z

0.14• [ [[[[ 0.12: o

-

0.10

-

c•008

O

-

'

Z +

oø.ø61

-

z

'-• 0.04

0.0?_/ 0

Mid-Latitude,

near 2_0 km I I 5

I0

I 15

I 20

I 25

]

[

[

0.06 -

,A

0.04 -

A,

øo.o•• 0.02

0.01 30

•

0

I

I

I

I

I

5

I0

15

2:0

25

Surface Area(pm2/cm •)

30

Surface Area (/zm% m:•)

ClO

h:/

Figure 10. Observationsof the changesin chemicalpartitioningas a function of aerosolload after the eruptionof Mount Pinatuboat 20 km at midlatitudes[Faheyet al., 1993].The primarychemistryresponsible for the observedbehavioris depictedby the schematicdiagram.

from a gas phase picture into the expectationof enhancedozone depletionat leastvia the N20 5 hydrolysis process,not only for volcanicallyperturbedconditions but alsofor backgroundaerosolloading.The eruptionof Mount Pinatuboin June 1991wasthe largestthusfar in the twentieth centuryand occurrednear the peak loading of atmosphericchlorine (see section7). This geophysicaleventprovidednumerouslinesof evidencesuggestingthat heterogeneouschemicalreactionson sulfate aerosolsplay a key role in ozone chemistryand its depletion.Gleasonet al. [1993] were the first to report

Toohey[1995] and WMO/UNEP [1994, 1998]. Because the stratosphericBrewer-Dobson circulation (as depictedin Figure2) transportsmaterialupwardandpoleward, major volcaniceruptionsthat inject material into the tropicalstratospherecanhavethe greatestand longest impactson globalozone,while volcanicinjectionsat higherlatitudesare removedby downwardmotion. Both E1 Chich6n and Pinatuboare tropicalvolcanoes.

record

curring in the ozone hole region. Observationsfrom New Zealand showedboth reducedNO 2 and enhanced HNO 3 columnabundances[Johnston et al., 1992;Koike et al., 1994]. Aircraft, ground-based,and balloon- and shuttle-borne experiments revealed similar large

low northern

midlatitude

ozone

abundances

in

the followingyear. Hofmann et al. [1994] and McGee et al. [1994] demonstratedthat substantialozone losses occurredin the lower stratospherefollowingthe Pinatubo eruption, particularly in winter and spring, with peak local depletionsnear 20 km at 40ø-50øNas large as about 25%. Randel and Cobb [1994] showed that changesin temperaturesrelating to aerosolheating can provide an important means of distinguishingozone lossesdue to volcaniceruptionsfrom those relating to the quasi-biennialoscillation(QBO), E1 Nifio, or other perturbationsin statisticalanalysesof ozone data (see also Jiggerand Wege [1990], Bojkov et al. [1993] and Zerefoset al. [1994]).The high aerosolload presentjust after the eruption in mid-1991 changedstratospheric heating and hence reducedtropical ozone through dynamical effects [Brasseurand Granier, 1992], but this lastedonly a few monthsand waslargelyconfinedto the tropics[seeSchoeberlet al., 1992b;Tie et al., 1994].For reviewsof the many studiesestablishingthe large and persistentmidlatitudeozonechangesafter Pinatubo,see

Chemical

measurements

after Pinatubo

have identi-

fied many signaturesof heterogeneousreactionson sulfate aerosols at midlatitudes

that are akin to those oc-

changes in NOx/NOypartitioningassociated with the roughlythirty-fold increasesin aerosolsurfaceobserved [e.g.,Rinslandet al., 1994; Websteret al., 1994;Faheyet al., 1993;Mills et al., 1993;Coffeyand Mankin, 1993;Sen et al., 1998]. Figure 10 showsdirect observationsof

perturbations in NOx/NOyandC10/Clyat 20 km from Faheyet al. [1993], associated with the buildupof Pina-

tuboaerosols atmidlatitudes. NOx/NOydecreases follow the behaviorbroadlypredictedby Prather [1992b] and expectedfrom the dominanceof N20 5 hydrolysis.Randeniyaet al. [1997] and Slusseret al. [1997] usedsummer polar observations of NO 2 to show evidence for BrONO2 hydrolysison sulfate aerosolsas well. Observationsof enhancedOH at sunrisefurther suggestthat the latter processis significantnot only in reducingNOx via heterogeneouschemistrybut alsoas a sourceof OH

37, 3 / REVIEWSOF GEOPHYSICS

[Hansonand Ravishankara,1995; Hanson et al., 1996; Salawitche! al., 1994;Lary e! al., 1996]. Turning to the key chlorine-relatedspecies,Avallone et al. [1993a, b] and Wilsonet al. [1993] showedpostPinatubo C10 observationssuggestingheterogeneous perturbationsin midlatitude air. HC1 observationsby Websteret al. [1998]provideevidencethat C10 not only is enhancedby high volcanicloading at midlatitudesas an indirect effect through shifts in NO2, but also is directlyaffectedby chlorineactivation(as reflectedin reducedHC1). Debate on the magnitudeof the latter effect in someregionshas focusedboth on the rates of chlorine activation in liquid aerosolsfor midlatitude

Solomon: STRATOSPHERICOZONE DEPLETION ß 295

chlorine(e.g., in the nineteenthcenturyafter the eruption of Krakatoa) would likely be a slight column increaseas a result of suppressionof NOx-catalyzeddestruction as depicted in Figure 10 rather than the observed

decreases

obtained

for current

chlorine

loads

[Solomonet al., 1996; Tie and Brasseur,1995]. It is also useful to note that observations

such as those at Arosa in

Figure 1 showno noticeabledepletionafter a seriesof large eruptions in the 1960s, most notably the major tropical eruption of Agung in 1963. Observationsfrom many other ground-basedsites confirm that the enhancements

in aerosol of the 1960s had little

effect on

ozone [e.g.,Bojkovet al., 1995]. Only sinceabout 1980 conditions andonmassbalanceamongClyspecies [see have chlorine levels become sufficientlyelevated that Dessleret al., 1996, 1997, 1998; Stimpfieet al., 1994]. volcanic perturbations to C10/C• suchasthoseshown in While chlorine-activatingreactions on liquid sulfate Figure 10 result in significantozone loss. Hence the aerosolsare thought to be relativelyslow at 20 km for evidencesuggests that volcanicparticlesat midlatitudes the averagetemperaturesthat prevail at midlatitudes, exacerbatehalogen-inducedozonedepletionin the conthe stronglynonlineardependenceof thesereactionson temporary stratosphere(much as PSCs do for polar temperaturesimplies that the reaction rate averaged regions,andwith somesimilarchemistry)but cannoton over the actual temperaturesincludingcold fluctuations their own significantlydestroystratosphericozone. associatedwith wave motions will substantiallyexceed Lary et al. [1997] were the first to suggestthat soot the rate computedfor the averagetemperature[Murphy may also affect Northern Hemisphere midlatitude andRavishhnkara, 1994].In otherwords,briefexposure ozone,mainlythroughpossiblereactionsinvolvingreacto coldtemperatures mayalterC10/Clypartitioning and tive nitrogenspecies[seeRogaskietal., 1997].A studyby hence enhanceozone depletion at midlatitudes,espeBekki [1997]further probedthischemistryin somedetail cially under high aerosol loads [Websteret al., 1998; and arguedfor significantimpactson ozonetrendsin the Solomonet al., 1998]. vicinity of the tropopause.There is currently debate Figure 10 showsthat even rather modestchangesin aboutthe surfacearea of soot availableat stratospheric

aerosolabundances cansubstantially affectthe C10/Cly altitudes, the extent to which it can remain active for

partitioningnear 20 km. Indeed, Figure 10 suggests that aerosolsurfacearea increasesof a factor of only about 5 chemistryor be quickly"poisoned,"and whether or not (as observed,for example,in somelocationsfollowing chemicaldata support such perturbations[Gao et al., the relatively minor Mount St. Helens eruption [see 1998]. The observationof large midlatitudeozonedepletion Thomason et al., 1997]couldincrease C10/Clyby 50%, following Pinatubo and E1 Chich6n, substantialrelated thus greatly enhancingthe chlorine-drivenlocal ozone changesin chemicalspecies,and a wide range of moddestruction reactions. As hasbeen emphasizedthroughoutthis paper, pro- elingstudies[e.g.,Hofmannand Solomon,1989;Brasseur cesses thatenhance C10 relativeto Clyareat the heart and Granier, 1992; Michelangeliet al., 1989; Pitari and of ozonedepletion.A point of usefulcomparisonmaybe Rizi, 1993;Bekki and Pyle, 1994; Tie et al., 1997;Solomon drawnbynotingthatif C10/Clyhadbeenconstant from et al., 1996;Jackmanet al., 1996]providestrongevidence 1980 to 1990, then C10 would have been expectedto that heterogeneoussulfate aerosol chemistryplays a increaseby about 50% over this decade(owing to the major role togetherwith man-madechlorinein the proroughly 50%increase in Clyfromthegradualincrease in cessescontrolling midlatitude ozone trends. The obchlorofluorocarbons duringthat period). However, Fig- served C10/ClyandNOx/NOydependencies uponvolcaure 10 demonstratesthat much larger changesin C10 nic aerosolamoufftsas shown,for example,in Figure 10 can be rapidly induced by volcanic aerosol increases may be considereda chemical fingerprint underlying throughtheir effectson chemicalpartitioning.Solomon these effects,like the observationsof greatly enhanced et al. [1996,1998]showedthat both the long-termozone C10 in the ozone hole region. Another parallel with trend at northern midlatitudesand its year-to-yearvari- Antarctic ozone depletion is the observationof a close ations over the past 20 years are highly likely to be correspondencein altitude between the region of encloselytied to volcanic-aerosol-driven changesin C10/ hancedPinatuboaerosolabundancesand ozone depleClypartitioning (seePlate6 below). Jackman etal. [1996] tion [e.g.,McGee et al., 1994;Hofmann et al., 1994]. A andZerefoset al. [1997]reachedsimilarconclusions with third fingerprintis the onsetand slowrelaxationof the their models,and showedthat solar cyclecontributions ozonedepletionafter Pinatuboobservedat midlatitudes to interannualozone depletion are much smaller. Sev- [see, e.g., Solomonet al., 1996, 1998; Jackman et al., eral authors have shown that the ozone responseto 1996] over a period of a few years,mirroring in a slower volcanicaerosolsbeforehumansperturbedstratospheric manner the seasonaldepletion of the ozone hole.

296 ß Solomon: STRATOSPHERICOZONE DEPLETION

37, 3 / REVIEWSOF GEOPHYSICS

5.2. DynamicalProcesses and MidlatitudeOzone

sphericdynamicsthemselvescould have contributedto

Trends

the observed

Although the focusof this review is on ozone chemistry,other mechanismsthat could contributeto midlatitude ozone depletionwill be briefly summarizedhere. Severalstudiesexaminedthe extent to which dynamical processesmight spreadthe influenceof the ozonehole, either through a one-time "dilution" at the end of the winter when stratosphericwarmingsbreak up the polar vortex or throughvortex "processing"wherebyflow of air throughthe vortex(and hencechemicalactivationof chlorine) might be transportedto lower latitudes[e.g., Tuck, 1989; Tuck et al., 1992; Waughet al., 1994, 1997; Waubenet al., 1997; Tuck and Proffitt, 1997]. The amount of ozone depletion observed at both northern and southern midlatitudes is considerably greater than that implied by a one-time end-of-winter dilution process[see,e.g., Szeet al., 1989;Pratheret al., 1990;Pitari et al., 1992]. For the SouthernHemisphere, suchone-time dilution likely providesan averagemidlatitude column ozone depletion of--•1-2%. Locally larger but transient dilution effects following the breakupof the Antarctic ozonehole in late springhave

is providedby Ravishankara et al. [1999](seereferences therein). In brief, some studies[e.g., Hood and Zaff,

midlatitude

ozone trends. A recent review

1995; McCormack and Hood, 1997; Hood et al., 1997;

Fuscoand Salby,1999] have arguedfor a componentof purely dynamicalchangein midlatitude ozone relating, for example,to changesin the transportof ozone. It is well known that dynamicalprocessesstronglyinfluence ozone from year to year, particularlyin Januaryin the Northern Hemisphere [Fuscoand Salby, 1999]. However, evaluation of trends requires long records and analysisof low-frequencytrends(i.e., timescalesof the order of a decade)rather than higher-frequencyvariations. While

some contribution

to the observed

trends

from dynamicalprocessesthat could changeover long time intervals(decadal)cannotbe ruled out, the evidence cited above and in Plate 6 below demonstrates

that chlorinechemistryhasplayedan importantandvery likely dominant role in the observedtrends in midlatitude ozone over the past 2 decades.

been documented in observations over New Zealand,

Australia, Brazil, and Chile [Atkinsonet al., 1989;Lehmann et al., 1992;Kirchhoffet al., 1996, 1997a].The city of Punta Arenas, Chile, at 53øS occasionallylies just beneath the tip of the Antarctic ozone hole itself for brief periods in October when wave disturbancespush the vortex overhead[Kirchhoffet al., 1997a,b]. Because of greater dynamical activity, the northern vortex is likely to be subjectto a greaterdegreeof processing, and many studiesconcludethat there is ample evidencefor the spreadof polar "filaments"to midlatitudesat times [Tuck et al., 1992;Gerberand Kampfer,1994;Pyleet al., 1995;Lutman et al., 1997].However,dynamicalanalyses and tracer studiessuggestthat the transportfrom polar regions alone cannot account for the observedozone losses in midlatitudes [e.g., Schoeberlet al., 1992b; Waugh et al., 1994; Manney et al., 1994b; Jonesand MacKenzie,1995; ChipperfieMet al., 1996; Waubenet al., 1997; Greweet al., 1998]. This subjectwill be discussed further below in the section relating to Arctic ozone depletion. In addition to vortex processingas describedabove, the notion of PSC processinghas also been suggested (wherein PSCsforming outsidethe vortex providethe sitesfor heterogeneousreactions),particularlyin association with locally cold temperaturesthat may be related to mountain lee wavesand hence of quite small

spatialand temporalscale[e.g., Godinet al., 1994; Carslawet al., 1998].All of theseprocessing mechanisms dependupon heterogeneouschlorine-relatedchemistry in some fashion

and hence connect

midlatitude

ozone

depletion to chlorine trends,but with important differencesin the degreeof nonlocal(i.e., transport-related) linkages. A few authors have argued that changesin strato-

6.

CHEMICAL/DYNAMICAL

ARCTIC

OZONE

COUPLING:

DEPLETION

Perhaps ironically in view of the extremelyremote nature of the Antarctic, ozone depletion was more readilyobservedthere than in the Arctic. This wasdue in part to the fact that no corresponding"hole" developed in the Arctic stratospherein the early 1980s,but alsoto the paucityof ground-basedlong-termmeasurementsin the high Arctic and to the greater local variability of Arctic ozone associatedwith atmosphericwaves,as discussedabove [see Reed, 1950]. As the mechanismfor Antarcticozonedepletionbeganto be elucidatedin the latter half of the 1980s, it was understood in general terms that Arctic ozone depletion would likely be smaller on account of warmer temperatures(hence fewer PSCsas documentedby McCormicket al. [1982]) and the associateddynamicaldifferences(i.e., a less isolatedvortex). The top panel of Figure 11 illustrates the climatologicaldifferencesin the seasonalcyclesof temperaturefor 65øNand 65øS,the edge regionsof the Arctic and Antarctic. Colder temperaturesare typically found at higher latitudes, but this region is shown in order to illustrate accompanyingsatellite total ozone data(whichare availableonlyin the sunlitatmosphere). Perhaps most importantly, the typical springtime increasein stratospherictemperaturesoccursin association with much earlier stratosphericwarmings in the north than in the south [e.g., Andrews et al., 1987], suggestingthat the overlap betweencold temperatures and sunlightwould be limited and the Arctic ozone depletionhencelesssevere(see,e.g.,the reviewby Pyle et al. [1992]).However,not all yearsare typical(a point discussed below).

37, 3 / REVIEWSOF GEOPHYSICS 2•0

'

I

'

I

'

I

'

I

Climatological

2•0

temperature at 83 mbar

Solomon:STRATaSPHERIC OZONE DEPLETIONß 297 '

I'

'

I

---o-- - 65S

0.8

Arctic • !

0.0 0.5

2OO

,

I

500

'

I

450

0 ©

400

$

o

o

o

I

I

I

I

,

I

,

I

J

I

'

I

'

I

'

I

Spring Maximum o 1996

and 1997 ß 1979 and 1980

o o

'tO. 2 .. ß

0.1

'

I

'

I

'

I

,

ß

o

350 o

ß o

o

Arctic Total Ozone at 65N

200

i

I

,

I

450

i

I

i

I

13 Feb., 1992 , [

I

[

I

Spring Maximum

ß

n n

,

56

I

60

,

I

64

,

I

68

Latitude (øN)

Figure 12. Observations of the chemicalcomposition of the Arcticstratosphere fromthe ER-2 aircraftin February1992.

350

I

I

52

n 1996 and 1997 ß 1979 and 1980

4O0

3OO

.

0.2

•,.

! Antarctic

25O

• ,•:

.•.,..,•.\ ,.N,'•. ß

ß ;.=

•0.4

I

300

-20 km

0.6

..0"'/

220

NorthernHemisphere

....o..... 65N

/[3- ..D.

B B

o

n

ß

ß

ß

The data showhigh C10 abundances associated with reduced

HC1 abundances as wouldbe expected from heterogeneous conversion. H20 doesnotdisplayevidence for dehydration on

250

this Arctic transect. Antarctic Total Ozone at 65S

200

,

•

,

•

•

JL A

S

0

N

I

•

I

I

I

J F M A M d JL A •; 0 • O (Arctic) D J Month

F

M

A

M

d (Antarctic)

measurements of HC1 were particularlyimportantin sharpeningthe link betweenenhancedCiO and converFigure11. (top) Observations of the averagetemperatures sion from HC1 [Websteret al., 1993;Michelsenet al., at 65øSand65øNfromtheFleminget al. [1990]CosparInter- 1999],asshownfor examplein Figure12.Measurements nationalReference Atmosphere (CIRA) climatology, withsat- of C1ONO2alsodisplayed evidencefor heterogeneous ellite measurements of the annual cyclesof total ozone at processingon PSCs [e.g., yon Clarmann et al., 1993; (middle)65øNand(bottom)65øSin the late 1970sandin 1996 Rocheet al., 1994;Oelhafet al., 1994;Adrianet al., 1994; and1997(fromSBUV/SBUV2,courtesy of R. Nagatani). Gelleret al., 1995;Yudinet al., 1997].Concurrentin situ andspace-based observations of C10 andPSCstogether with trajectorystudiesfurther linked the activatedchlo6.1. Chemical Processes in the Arctic rine to heterogeneous chemistry[Joneset al., 1990b; As in the Antarctic, direct observations of a broad

Yudin et al., 1997; Dessleret al., 1998, and references

rangeof chemicalspecieshave shownthat heteroge- therein].Thusthe samegeneralfingerprints of heteroneouschemistry greatlyperturbsthe composition of the geneouschemistrythat were first observedin the AntArctic vortex. Evidence for effective winter activation of arctic were not only apparent in, but also further chlorinewas providedby measurements of enhanced strengthenedby Arctic data. OC10 [Solomonet al., 1988;Schilleret al., 1990;PomObservations of NOyandwatervapordisplayed signs mereauandPiquard,1994a;Pernetet al., 1994;Pfeilstic- of sporadicandlimiteddenitrification and dehydration ker and Platt, 1994]and C10 [Bruneet al., 1990, 1991; (compareFigures12 and7), in markedcontrast withthe Tooheyet al., 1993; Waterset al., 1993; Crewellet al., pervasivecharacteristicsof these chemicalconditionsin 1994;Bellet al., 1994;deZafra et al., 1994;Shindellet al., theAntarctic[Kawaetal., 1990;Faheyetal., 1990b](see 1994;Donovanet al., 1997].DecreasedNO and NO2 later work by Kondoet al., 1994;Oelhafet al., 1994; were alsoobservedwith severalindependentmethods Khattatovet al., 1994;Rinslandet al., 1996;Santeeet al., [Faheyet al., 1990a;Noxon, 1978; Toon et al., 1994; 1998, 1999]. In somecases,denitrificationwas observed Mankin et al., 1990;Wahneret al., 1990;Pommereauand without accompanying dehydration,raisingnew chalPiquard,1994b;Goutailetal., 1994;Pfeilsticker andPlatt, lengesregardingthe mechanism underlying the micro1994;Van Roozendael et al., 1994].The columnabun- physicsof the denitrificationprocessthat still are not dancesof HC1 andHF supportedthe viewthat chlorine completelyresolved[see,e.g.,Toonetal., 1990;Gandrud activation onPSCsmustbe effectivein theArctic[Toon et al., 1990;Salawitch et al., 1989;Koopet al., 1995]. etal., 1994;Mankinetal., 1990;Traubetal., 1994].In situ On the basisof C10 observations [Bruneet al., 1990] ß

298 ß Solomon: STRATOSPHERICOZONE DEPLETION

37, 3 / REVIEWSOF GEOPHYSICS

and related model calculations, observed and calculated

rates of ozone lossin February 1989were shownto be of

theorderof 20ppbvday-• near20km[Schoeberl etal., 1990;Salawitchet al., 1990;McKenna et al., 1990]. Further, the BrO observationsof Tooheyet al. [1990] revealed that the C10-BrO catalyticcyclewas probablyof particularimportancefor the Arctic, sinceC10 enhancements

were

smaller

there

than

in the Antarctic

and

hencethe efficiencyof the C10 dimer cyclewasreduced (note that the rate of the latter dependson the squareof C10 density[e.g.,Salawitchet al., 1990,1993]).However, the early warming observedin February 1989 as illustrated in Plate 3 preventedextensivetotal ozonelossin that year. Several studiessuggestedthat the lessextensive denitrification

of the Arctic would limit ozone losses

there [Bruneet al., 1991;Salawitchet al., 1993] through lesseffective NOx reduction in sunlit air and hence an early cutoff of the depletionprocessin spring;the precedingsectionillustratesthat new understandingof liquid aerosolchemistryhas affectedthis picture. Thermal decompositionof the C1202dimer (which cuts off the C10 dimer ozone losscycle) also affectsthe degreeof ozone loss as air warms in spring even if denitrified [McKennaet al., 1990;MacKenzieet al., 1996].As will be discussedfurther below, recent observationsof large Arctic ozone depletions(see Figure 1) have not been associated with extensive

denitrification.

in the amount of ozoneobservedfor a givenamountof conservedtracer such as N20 or CH 4 [Proffittet al., 1993; Mtiller et al., 1996, 1997a, b] provide a useful (albeit imperfect)diagnosticfor ozonelossbasedupon understanding of ozone-inert tracer relationships [Plumb and Ko, 1992] and their spatial distributions. Using satellitedata for CH4, 03, and HC1, for example, Mtiller et al. [1996] suggestthat 60 DU of total Arctic ozonewas depletedon constantCH 4 surfacesin a manner inconsistent with transportfrom any other regionof the stratospherein the Arctic winter of 1991-1992. The reduced ozone was associatedwith pronouncedHC1 depletionobservedin the sameair, aswas expectedon the basisof heterogeneouschemistryon PSCs. Another method of quantifying ozone destruction involvesthe useof trajectoryanalysesof airflowtogether with multiple ozonesondesto find "matches"wherein the air observedat one site is observedagainsomedays later. Changesin the observedozone then provide a measureof ozone loss [Von der Gaathenet al., 1995]. This approachhas provided strongevidencefor extensive Arctic ozone depletion that is closelytied to cold temperaturesnear 195 K [Rexet al., 1997, 1998].Understandingcan be further tested by comparingthe observed depletion derived from such "matches" with chemistrycalculationsalongthe sametrajectories.These studieshave shown good agreement in February and March, but some evidence for midwinter ozone loss that

6.2. QuantifyingArctic Ozone Depletion The more complexdynamicsof the Arctic vortex as comparedwith the Antarcticdemandsthe applicationof sophisticated toolsfor analysisof ozonedestruction.The greater wave activity of the Northern Hemisphere can enhanceozone losseseven in winter by increasingthe exposureof polar air to sunlightin the distortionscaused by atmosphericwaves, as comparedwith the Southern Hemisphere[see,e.g.,Joneset al., 1990a].However,the samewave activitycan warm the air and perhapseven distort it sufficientlyto mix with its surroundings,thus reducingozone depletion.Detailed methodshave been developed[Schoeberl et al., 1990;Manneyet al., 1994b, 1995a, b, 1996] to evaluate the air parcel trajectories along which ozone and other trace gases are transported. These help to quantify the amount of ozone chemicallydestroyedby revealing that while the time evolution of inert tracers such as N20 can be well simulatedin the Arctic usingsuchapproaches,the evolution of ozone showslarge departuresfrom conservation that likely reflect chemical loss [Manney et al., 1994a, 1995a,b, c, 1996, 1997]. Further, the regionsof apparent ozone depletion identified in this manner occur in regionsof enhancedC10 revealedby concurrent satelliteobservations[e.g.,Waterset al., 1993;Manneyet al., 1994a, 1995c; Lutman et al., 1994a, b; MacKenzie et al., 1996]. Tracer-ozone

correlations

are another

method

used

to provideinsightsinto polar ozoneloss.Briefly,changes

exceedsphotochemicaltheory has recently been suggested[Rexet al., 1998;Beckeret al., 1998]. Fully three-dimensionalmodels driven in somecases by the meteorologicaldata for specificyears have also been used to probe the Arctic ozone lossesand test photochemicalunderstanding.These modelshave succeededin explainingmuchof the observedozonedepletion, documentingits connectionsto chemicalprocesses, and even reproducingmuch of the observedvariability seenfrom one year to anotheras depicted,for example, in Figure 1 [see, e.g., ChipperfieMet al., 1994, 1996; Deniel et al., 1998;Douglasset al., 1995]. Taken together,thesecombinedapproachesto transport analysesusingtracers,matches,chemicaltransport models,or Lagrangiancalculationstogetherwith ozone and trace constituentobservationsprovide strong evidencefor a chemicallydrivenArctic ozoneloss(order of 60-120 DU) in severalrecentyears.Each approachis subjectto different sourcesof quantitative error and uncertainty, such as inaccuraciesin temperature data usedas input in observationally-based transportstudies, incomplete understandingof the factors influencing tracer-tracer correlations, and small scale dynamical processesthat are not well represented in modelling studies(e.g., mountainwaves).In spite of theseshortcomings and in contrast with the Antarctic, there is substantialevidence for a dynamical contribution to recent trends as well. These are discussed in section 6.3.

Solomon: STRATOSPHERIC

37, 3 / REVIEWS OF GEOPHYSICS

6.3. Variabilityof ArcticTemperatures Since Antarctic ozone depletion occurs mainly in Septemberunder cold conditions,it is natural to consider whether comparableconditionsare ever attained in the Arctic in the analogousmonth of March. Nagatani et al. [1990] pointed out that while such conditions appear to be quite rare basedon the availablerecord (which extendsback to about the 1950s),they are not unknown. For example, during the Arctic winter of 1975-1976, March temperatureswere close to those typicallyseenin the Antarcticin September,but chlorine loadingwas small in 1975, and no discernibleAntarctic ozonehole was observedat that point (see Figure 1). Nagatani et al. [1990] noted that extensiveArctic ozone lossmight be expectedif suchmeteorologicalconditions were to be realizedin an atmospherewith currentchlorine loadings. There havebeen severalunusuallycoldArctic winters since 1990, with correspondinglylarge Arctic ozone losses[seeNewman et al., 1997; Coy et al., 1997] illustrated in Figure 1. Not all the yearssince1990havebeen cold, asis reflectedfor examplein the high springozone observedin 1998 for example.Enhancedvolcanicaerosolfrom the Mount Pinatuboeruptionprobablycontributed to the very low ozone observedin 1992 and 1993, but the continuingdepletionsin, for example,1996 and 1997 suggesta strong effect of temperature. Plate 3 illustrates the full seasonalbehavior of temperatures observedin some recent cold years, and contraststheir behavior with the Antarctic. As has already been emphasized,the warmer temperaturesgenerallyobserved in the winter Arctic stratosphereas comparedwith the Antarctic reflect adiabaticheatingassociatedwith faster downwardmotion, which also leads to a rapid wintertime increasein Arctic total ozone(from valuesof ---300 DU in Septemberto as much as 450 DU at the spring maximum in March in 1979 and 1980, for example, as shown in Figure 11). In the much colder Antarctic, pre-ozone hole total ozonedid not showsucha winter increase(remaining instead near 250-300 DU from March throughSeptember;seePlate 2 and Figure 11), suggestiveof an isolated and less dynamicvortex. In today'sAntarctic atmosphere,an abrupt drop in ozone occursin Septemberdeepin the vortexasshownin Plate 2 and even earlier on the edge of the vortex as shownin Figure 11, reflectingrapid chemicalremovalin sunlitair with limited dynamicalresupplyas discussedearlier. In the relativelystagnantAntarctic vortex the total ozone actuallydecreasesin springto form a "hole" compared with the surroundingmidlatitude air. In the more dynamicArctic, transportreplacesa substantialportion of the ozone lost, even in recent cold years [see, e.g., Manneyet al., 1997]. Indeed, Figure 11 showsthat even in the very cold years 1996 and 1997, Arctic ozone continuedto increaseat 65øNduringspring;it simplydid not do so as rapidly as it had in 1979 or 1980. Hence large chemicalozonelossesof the order of 60-120 DU occurred,but no Arctic hole formed (see the detailed

OZONE

DEPLETION

ß 299

analysesby Manneyet al. [1997] andM•iller et al. [1997a, b]). The formationof an Arctic ozonehole may require not only cold March temperaturesbut alsocold temperatures throughout the winter, both in order to cause activation

of chlorine

in sunlit

air and to inhibit

the

buildup of ozone through downward transport. Such conditionswere not satisfiedevenin the very cold Arctic winter/spring seasonsof recent years [see Coy et al., 1997;Zurek et al., 1996]. It is important to note that denitrificationwas observed but was rather limited in degree in the Arctic springsof 1993,1996,and 1997 [Santeeet al., 1995,1996, 1997, 1999], so that the observationsof the order of 60-120 DU of ozonedepletionin eachof theseyearsare not associated with denitrification. Rather, as in the

Antarctic and consistentwith current understandingof liquid aerosolchemistry,the evidencesuggests that heterogeneous reactions in the sunlit atmosphere are mainlyresponsiblefor maintainingthe high C10 [Santee et al., 1997]that depletedthe Arctic ozonein thoseyears [Manneyet al., 1997]. Plate 4 showsobservedchangesin the verticalprofile of Arctic ozone at Sodankyl•i,Finland (67øN), in 1996 that illustratethisgeneralpicture.Plate 4 is intendedfor the purposeof illustration.A detailed analysiswould be needed to quantify dynamical and chemical contributionsto ozonelossesas in the studiesof Rexet al. [1998] and Manney et al. [1997]; the points sketchedhere are consistentwith thosepapers.The ozone observedin late March 1996 lies well below the climatology for this location, much as the South Pole ozone in September 1986lay belowits climatology(Figure 5); other datesin March displaysimilar behavior. The ozone at ---15-20 km at South Pole was depleted in 1986, similar to the layer of reducedozone observedover Finland at nearly the samealtitudesin March 1996(and at thosealtitudes where PSCsare frequentlyobserved).However, at the South Pole, the historical and current ozone profiles displaynearly the same values above the depleted region, showinglittle evidence for large changesin the amount of ozone brought down from above. The data from Finland present an interestingcontrast,with reducedozone above20 km not only in March but alsoin February,likely reflectingreduceddynamicaltransport from above. This is not surprising,since the cold temperatures observedin that year must reflect reduced downwardmotion. Hence particularlycold Arctic winters must be associated with less downward

motion

and

a componentof dynamicimpact on total ozone. The shapeof the profile is suggestive of chemicalremovalin the broad layer near 15-20 km. In summary, there is abundant evidence for some chemical perturbations and ozone destruction in the Arctic even in relativelywarm years, but the degree of ozone depletion depends upon cold temperatures in sunlit conditions,just as in the Antarctic. An unprecedented number of cold yearshave occurredin the Arctic since 1990. Each of these is reflected

in low ozone in the

300 ß Solomon'

STRATOSPHERIC

OZONE

DEPLETION

37

3 / REVIEWS OF GEOPHYSICS

30

'• 25

Ozone Arctic

67 øN Sod(•nkyl(•,

':... -'..,,'",•,

Finl(•nd

,

20

15

I0

'-

.....

Fe bru•r•'•iil•tology[ M•rchClimotology I

2/14/96

I

..... I

• 0

5

I0

15

20

Ozone Prtiel Pressure (hP) Plate 4. Arctic ozonesondedata from Sodankyl•i,Finland. The climatologiesfor February and March representthe averagesof all data for 1988-1997. Sampleprofilesobservedin February and March 1996 are shown for comparison.

Arctic record as shown in Figure 1. Five of the years from 1991 to 1998 have been significantlycolder than average[Coy et al., 1998;Zurek et al., 1996]. This series of unusuallycold years raisesthe key questionof cause. Randeland Wu [1999]arguethat the coolingobservedin both

the Arctic

and the Antarctic

is due to the ozone

depletion itself; hence they proposea feedbackmechanism,followingShine[1986],wherein ozone losseslead to colder temperaturesand hence even greater depletion. The studyby Thompsonand Wallace [1998] suggeststhat changesin the dynamicsof the north polar vortex are linked to the underlying troposphericwave field,particularlythe North Atlantic Oscillation(NAO). These authors thus suggesta wave-driven systematic linkagebetweentropospheric wavesandstratospheric temperature,whichcouldreflectozonechanges. Hartleyet al. [1998]arguefor a similarlinkageinvolvingthe modification of stratospheric dynamicsdue to the ozonechanges, with troposphericpropagationas a key element. It has long been known that the "greenhouseeffect" due to increasesin CO2 and other gaseswarm the planet surfacebut cool the stratosphere[e.g.,Felset al., 1980], with attendant effectson temperature-dependentozone chemistry[Haigh and Pyle, 1979]. While this effect is

predictedto be small(onlya few tenthsof a degreein

observedin the Arctic), dynamicalamplificationof such changesis also possible,as was noted above.A number of studieshave suggestedthat increasedCO2 and other greenhousegasescouldsubstantiallyaffect Arctic ozone [e.g.,Austin et al., 1992;Shindellet al., 1998]. The work of Shindell et al. arguesfor a key role for sucha feedback both in the 1990sand perhapsin future years,with the peak Arctic ozone lossesbeing predicted to occur near 2010, well after the expected peak of chlorine loading (see also Dameriset al. [1998]). However, at presentthe possibilitythat the recent colder Arctic temperatures are part of a natural low-frequencycycle that could, for example,induce a seriesof colder years every 50 yearsor socannotbe ruled out giventhe shortrecord of existingglobal stratospherictemperaturedata. Hence while it is clear that there has been significantchemical ozone depletion associatedwith the cold Arctic winterspringseasonsof recent years,the fundamentalreason for thosecold temperaturesremainsa topic of research.

7.

SUMMARY

OF THE PAST AND

A LOOK

TO THE

FUTURE

This paper has outlined the history and conceptual today's atmosphere,far less than the recent coolings understandingof the processesresponsiblefor ozone

37, 3 /REVIEWS OF GEOPHYSICS

Solomon' STRATOSPHERIC OZONE DEPLETION ß 301

ii iii.......

O

302 ß Solomon:

STRATOSPHERIC

OZONE

DEPLETION

37

3 / REVIEWS OF GEOPHYSICS

Late 1990speak•3.5ppbv Ozone depletionbegan • 2.2 ppbv

Return to !ate-1970s level around 2040

Total Tropospheric Chlorine I

1900

ß

1920

I

ß

1940

1

I

1960

1980

,

I

•

2000

6

I

ß

2020

I

.

2040

I

I

2060

2080

.

2100

._

5-year runningmeans o

4

Observed- Switzerland

...........

Model-with

Model-

volcanoes

no volcanoes

0 (D

O N

-2

o

-4

Northern Mid-Latitude Total Ozone

-6

-' -8

•

1900

1920

ß

,

1940

ß

•

1960

.

i

.

1980

,

.

2000

,

2020

,

I

2040

,

I

,i

2060

I

2080

,

*

2100

Time (years) Plate 6. (top) Total troposphericchlorine content estimatedfrom the baselinescenarioof WMO/UNEP [1999]; this is basedon a gas-by-gas analysislike thoseshownin Plate 5. (bottom) Changesin the 5-year runningmeanozoneobservedover Switzcrland[Staehelinet al., 1998a,b] comparedwith a modelcalculation for 45øNapplyingthe sametime averaging,with and withoutconsideringthe effectsof volcanicenhancements in aerosolchemistry(from the model of Solomonet al. [1996, 1998]). The major eruptionssince 1980were those of E1 Chich6n

in 1982 and Pinatubo

in 1991.

depletionby chlorofiuorocarbons in the stratosphere.In brief, the long lifetimes of chlorofiuorocarbonsare reflected in their observed worldwide

accumulation

in the

atmosphere.Their role in stratosphericozone depletion dependscritically on partitioning processesthat follow release of halogen atoms;indeed, the marked contrasts betweenfluorine (whichdoesnot depletestratospheric ozone),chlorine,and bromineillustratethe centralrole of partitioningchemistry.Table 2 summarizesa seriesof spatial and temporal fingerprintsthat connectchlorine chemistryto ozone depletion. Observationalevidence for gas phase chlorine chemistryimpacts on ozone is

provided,for example,by observations of the C10/Cly and ozone trend profilesaboveabout 25 km at midlatitudesand by the similaritiesin their observedlatitudinal distributions.

The

cold

conditions

of

the

Antarctic

winter

and

spring stratospherelead to formation of polar stratosphericclouds.Heterogeneouschemistryinvolvingmanmade chlorinetakesplace on thesesurfacesand results in the dramaticand unanticipatedAntarctic ozonehole. The heterogeneousactivationof chlorinefrom both its HCI and C1ONO2reservoirsand the suppression of the NO 2 (that wouldotherwisereformC1ONO2)alterschlorine partitioningand allowseffectiveozone lossin cold sunlit air. The close correspondencebetween observed enhancementsin C10 and depleted Antarctic ozone through independent observationalmethods as functions of altitude, latitude, and longitude illustratesthe key role of these chemical partitioning processesin producingthe ozone hole. A broad range of chemical observationsof HC1, HNO3, NO 2, OC10, and other

37, 3 / REVIEWS OF GEOPHYSICS

Solomon: STRATOSPHERIC OZONE DEPLETION ß 303

TABLE 2. Summaryof Key Fingerprintsof OzoneDepletion Observation

Profile shapesof upper stratosphericozone

Method

Latitude

Altitude

satelliteand ground-based Northern Hemisphere 30-50 km midlatitude

depletion andC10/Cly

Primary Chemistry Linkage

Gas phasechlorine chemistry, particularly partitioning processes

Latitudinalstructureof upper stratosphericozone depletionand C10

satellite

Latitudinal structure of C10, HC1, NO, NO2, and ozone

airborne

Polar, midlatitude,

30-50 km

Gas phasechlorine chemistry, particularlyHC1/

20 km

Heterogeneous

and tropical

C10

and satellite

50ø-85øS, 50-85øN

chlorine activation

lossin polar regions Vertical

structure of seasonal

ozone loss and PSCs in

polar regions Seasonalchangesin PSCs, ozone depletion,O'C10, C10, HC1, and C1ONO2 Post-Pinatubo(---1992-1995) ozone depletionand recovery;contrastwith post-Agung(---1964-1968)

balloon-borne, lidar, and

90øS, 50ø-85øS, 50-

satellite

12-24 km

85øN

balloon-borne, lidar,

50ø-90øSin both

ground-based, and

polar regions

and

NO 2 suppression Heterogeneous chlorine activation

---12-24 km and

column

Heterogeneous chemistry

satellite

ground-basedand some

midlatitudesand

satellite

polar regions

near 20 km and

column

Heterogeneous liquid surface chemistry relating to chlorine

Post-Pinatubochangesin stratosphericchemical state and aerosol content

airborne,ground-based,

midlatitudesand

near 20 km, near

somepolar

tropopause,

balloon-borne, and satellite

and column

(NOx, C•O, I-INO•, OC•O)

Heterogeneous chemistry, particularly N20 5 hydrolysis and some chlorine

activation

speciessupport and extend this picture. Quantitative numericalmodelingstudiesthat includedetailedanalysesof transportand chemistryfurther connectthe enhanced C10 producedby heterogeneouschemistryto

firm the impact of liquid aerosolsurfaceson chlorine and nitrogenpartitioningchemistry.Observationsand laboratorystudieshave demonstratedthe efficacyof heterogeneous processeson suchsurfaces(both at 20 the formation of the Antarctic ozone hole. km and at loweraltitudes,wherehighwatervaporpresScientificunderstanding of PSCsand heterogeneous suresenhancechlorineactivationchemistry).As in the chemistryhasevolvedconsiderably in recentyears.The Antarctic,concurrentobservations of a broad rangeof detailed microphysicalmechanismsresponsiblefor chemical speciesshow evidencefor surface reactions freezing of PSC particles and for denitrification are associatedwith particles,which work to enhanceC10/ subjectto debateat present,but theseprocesses appear Clypartitioning at midlatitudes. Dilutionandprocessing to be less critical to ozone depletion than was once of the polar ozonelossesalsocontributeto midlatitude thought.There is evidencefrom field, laboratory,and ozonedepletion.While somestudiessuggesta role for a modelingstudiesthat PSCscanbe composed not onlyof purelydynamicaltrend in midlatitudeozonedepletion, solidwaterice andnitricacidhydratesbut alsoof liquid thesehavenot yet succeeded in quantifyinga significant solutions of water, sulfuric acid, and nitric acid. The

contribution.

chemistryassociated with thesevaryingsurfacesdisplays important differencesin detail but has the common feature that all can suppressNO 2 and activatechlorine from the reservoirspecies,makingthe ozonedepletion processmorecontinuousin temperatureand lessdependent upon the abrupt temperaturethresholdsthat are associated with formationof solidsthan waspreviously thought.

There is abundantevidencefor heterogeneous perturbationsto Arctic chemistrythroughobservationsof C10, OC10, HC1, and many other key gases.Arctic ozone has reachedrecord low valuesin many yearsin the 1990s,linkednot onlywith heterogeneous chemistry on Pinatuboaerosolsbut alsowith unusuallycoldspring temperatures.A chemicalcontribution to these low val-

ues has been documentedwith a variety of methods including trajectory "matches," chemistry transport ozonedepletionfollowingthe eruptionof Pinatubocon- modeling, and tracer correlation studies. The fundaObservations

of enhanced Antarctic

and midlatitude

304 ß Solomon: STRATOSPHERIC OZONE DEPLETION

37, 3 / REVIEWS OF GEOPHYSICS

This reviewhasemphasizedmanyspatialand tempomental questionof the causeor causesof record low temperatures in many of the Arctic winter-spring ral "fingerprints"that illustratethe role of chlorinein stratosphere (Taseasonsof the 1990sremains a topic of debate and a depletingozonein the contemporary ble 2). Indeed, it is the structureof the ozonelossin key issue. In closing,Plates5 and 6 are presentedto showthe space(e.g., in the 40-km region) and time (e.g., in the impactof changesin globalemissions of chlorofluoro- Antarcticspringand in midlatitudesin the yearsfollowcarbonsand the likelyfuture of the ozonelayerbasedon ing Pinatubo)that testsand confirmsscientificunderthe conceptualpicturedevelopedin this review.Plate 5 standing,illustratinghow gasphaseand heterogeneous displayssurfaceobservationsof CFC-11 and methyl modulation of C10/Clypartitioning affectsozonedeplechloroform(CH 3 CC13).The latter gasis the onlyshort- tion. Through the impactsof this chemistry,the stratolived industrial chlorofluorocarbonproduced in large sphericozonelayer in the twenty-firstcenturywill conamounts in the 1970s and 1980s. Becauseof its 5-year tinue to reflect the impact of the changesin chlorine lifetime, the abundancesof methyl chloroformhave al- enacted in the twentieth. ready begun to decline,as a result of reducedglobal emissions.Those of CFC-11 are just passingtheir peak and are projectedto declineslowlyin comingdecades, GLOSSARY reflectingits 50-year lifetime. Active chlorine: chlorine compoundsthat destroy Plate 6 (top) showsthe pastandfutureprojectionsof the total troposphericchlorinecontent(whichleadsthe ozone and interchangerapidly with one another in the stratosphereby 3-5 years). It is anticipatedthat the sunlit atmosphere(mainly C1, C10, C1202,OC10, and combinedeffect of all CFCs will lead to a peak strato- HOC1);chlorinethat is not tied up in the reservoirgases sphericchlorineloadingin the late 1990s.By about2040, (HC1 and C1ONO2). the chlorine will return to levels close to those of the late Chlorine loading: Abundanceof total chlorinein 1970s, when ozone depletion was first apparent. All all forms (includingCFCs) at a givenlocation. Chlorofluorocarbons (CFCs): Chemicals, used in a other thingsbeingequal,the Antarcticozonehole and midlatitudeozonedepletionwill likelydisappeararound varietyof industrialapplications,that are the dominant this time. However, the key role of temperature and sourceof chlorineto the present-daystratosphere. aerosolsin modulatingozone depletion must also be CIr: Thesumof all chlorine gases liberated bydeC10, HC1, C1ONO2, considered.The unusuallycoldArctic winter-springsea- compositionof CFCs,including.C1, sonsof recentyearsstandat the time of thiswriting as a HOC1, C1202,and other trace species. CIO dimer: C1202,a key intermediatein the formacriticalchallengeto our understanding that couldaffect the future of polar ozonedepletionin bothhemispheres. tion of the Antarctic ozone hole. See the catalyticcycle For example,if the majority of future Arctic winters involvingthis gasillustratedin Table 1. Denitrification' Removal of reactive nitrogen were to be colder than average,then the Arctic ozone depletionwouldlikely be prolonged.The bottompanel (NOy)fromthestratosphere through sedimentation of of Plate 6 showsthe long-runningArosa, Switzerland, largeparticlescontainingnitric acid. Dehydration: Removal of water vapor from the ozonerecordillustratingthe onsetof midlatitudeozone depletion,its linksto heterogeneous chemistry,and its stratospherethrough sedimentationof large particles simulationwith a currentstratospheric chemistrymodel containingwater. Dobson Unit (DU): Unit of measurement of total includingthe processesdescribedin this review. The changesin ozone observedover Arosa are in good ozone column abundance,named for G. M. B. Dobson, agreementwith the zonallyaveragedglobalsatellitedata a pioneerin measurementof ozone.One Dobsonunit to 2.6x 10•6molecules cm-2 of totaloverdiscussed earlier, and the time-averagedtrendsobtained corresponds there are representativeof northernmidlatitudes.Plate head column ozone. 6 illustratesthat the future of midlatitude ozone depleFrostpoint: The temperatureat whichwater contion is likely to be linkednot onlyto chlorinebut alsoin denses to form solid ice. NAI: Nitric acid trihydrate, or HNO3' (H20)3). part to volcanoesfor at least severaldecades.If there cloudsare probablycomposed were to be an extremelylarge volcaniceruptionsuchas Somepolar stratospheric that of Tambora (whose1815 eruptionis estimatedat of solid NAT particles. NOx: NO + NO2, two reactiveforms of nitrogen about 3 timesthe stratospheric impactof Pinatubo)in comingdecades,it is likely that midlatitudeozonede- that interchangevery rapidly with each other in the pletion would be increasedeven though the chlorine sunlitatmosphere.The amountof NOx is linkedto NO 2 contentof the stratosphereis expectedto be lower than and hence to formation of the C1ONO2 reservoir. it is today. This illustratesthe connectionbetweenthe NOr: Thesumof therelatively reactive totalnitroaccumulationof chlorine in today'satmospheredue to gengases,includingN, NO, NO2, C1ONO2,NO3, N20 s, human activitiesof the industrialera and the unpredict- BrONO 2, HNO3, and other trace species. Ozone hole: Widespreadremovalof total ozonein able timescalesof geologicphenomenathat coupleinto this altered chemical state. Antarcticspring.The hole is reflectedin both the steep

37, 3 / REVIEWS OF GEOPHYSICS

Solomon: STRATOSPHERICOZONE DEPLETION ß 305

latitudinal gradients in the observedozone depletion sondedata. Antarctic and Arctic ozone data were generously furnished by J. Shanklin and P. Newman. The author also and in its temporal evolutionsincethe mid-1970s.

Partitioning: Distributionof chlorinebetweenac- wishesto thank S. Montzka, J. Elkins, D. Fahey, M. Santee, tive compoundsthat destroyozone and reservoirsthat are inert toward

ozone.

andJ. Watersfor generouslyprovidingdata usedin key figures

here.

Observations

available

on the World

Wide

Web

for

ozone at Halley and Arosa, and the CMDL and ALE/GAGE ppbv,pptv: Partsper billionbyvolumeor partsper measurementsof CFCs are also deeply appreciated.The astrillion by volume, indicating relative abundanceof a sistanceof SusanHovde and Maria Neary in preparationof the givengas(i.e., 1 ppbv = 1 moleculeper billion total air figuresis alsogratefullyacknowledged.Helpful commentsand molecules). suggestions by P. Crutzen,R. de Zafra, V. Dvortsov,D. Fahey, Processing: Generalterm describing conversionof R. Garcia,J. Holton, J. Lynch,M. MacFarland,G. Manney,H. chlorine to active forms. Chemical processingrefers to Michelsen,D. Murphy, M. Santee,R. Salawitch,J. Staehelin, in situ chemistry.Vortex processingrefers to flow of air M. Tolbert, J. Waters, C. Webster and the reviewersgreatly aided in the author in preparationof this review. to midlatitudesfrom the vortex, while PSC processing Michael Coffeywas the Editor responsiblefor this paper.

refersto flow of air throughPSCsassociatedwith locally He thanks M. Schoeberl and J. Waters for the technical cold temperatures. viewsand S. Islam for the cross-disciplinary review.

re-

Polarstratospheric clouds(PSCs): Cloudsthat are observedto form at cold temperatures(below ---200K) in the polar stratospheresof both hemispheres. REFERENCES Reservoir: Long-livedcompoundcapableof storing NOx or activechlorinein a relativelyinert form (mainly Abbatt, J.P., L. T. Molina, and M. J. Molina, Heterogeneous interactionsof C1ONO2and HC1 on nitric acidtrihydrateat HNO 3, HC1, and C1ONO2). 202 K, J. Phys. Chem., 96, 7674-7679, 1992. Stratosphere: The region of the atmospherebeAdrian, G. P., et al., First resultsof ground-basedFTIR meatween ---12and 50 krn (a few kilometerslower in polar surementsof atmospherictrace gasesin north Swedenand regionsand higher in the tropics)in which heatingby Greenland during EASOE, Geophys.Res. Lett., 21, 13431346, 1994. ozone leadsto increasingtemperatureswith increasing Adriani, A., G. Fiocco, G. P. Gobbi, and F. Congeduti,Cor-

altitude.

Sulfuricacid tetrahydrate(SAI): A solidform of sulfuric

acid and water

that

can form

under

certain

related

behavior

of aerosol

and ozone

contents

of the

stratosphereafter the E1 Chich6n eruption, J. Geophys. Res., 92, 8365-8372, 1987.

thermodynamicconditions. Adriani, A., T. Deshler, G. Di Donfrancesco, and G. P. Gobbi, Polar stratosphericclouds and volcanic aerosol during lracer: Long-livedchemicalcompoundthat can be spring 1992 over McMurdo Station, Antarctica: Lidar and usedto trace the atmosphericairflow. particle countercomparisons, J. Geophys.Res.,100, 25,877lropopause: The transitionregion, in which tem25,897, 1995. peratures reach a minimum, between the troposphere Anderson, J. G., J. J. Margitan, and D. H. Stedman,Atomic chlorine and the chlorine monoxide radical in the stratoand stratosphere.

lroposphere: The region of the atmospherebetween the surface and the stratosphere,in which temperaturesdecreasewith increasingaltitude. lype 1: Polar stratospheric cloudsthat form at temperaturesabove the frost point.

sphere:Three in-situ observations,Science,198, 501-503, 1977.

Anderson, J. G., W. H. Brune, and M. H. Proffitt, Ozone

destructionby chlorineradicalswithin the Antarcticvortex: The spatialand temporal evolutionof C10-O3 anticorrelation based on in situ ER-2 data, J. Geophys.Res., 94,

lype 1a (1b): Solid (liquid) polar stratospheric 11,465-11,479, 1989. Andrews,D. G., J. R. Holton, and C. B. Leovy,MiddleAtmocloudsat temperaturesabovethe frost point. sphereDynamics,Academic,San Diego, Calif., 1987. lype 2: Solid water ice polar stratosphericclouds Anthony,S. E., T. B. Onasch,R. T. Tisdale,R. S. Disselkamp,

that form when temperaturesdrop belowthe frost point. M. A. Tolbert, and J. C. Wilson, Laboratory studiesof Vortex: Dynamical structureof the stratospherein ternaryH2SO4/HNO3f[-I20particles:Implicationsfor polar stratospheric cloud formation, J. Geophys. Res., 102, polar winter causedby the absenceof solarillumination, 10,777-10,784, 1997. which leads to a cooling over the poles and a large Arnold, F., Stratosphericaerosolincreasesand ozonedestructemperaturegradientrelativeto midlatitudes.This temtion: Implicationsfrom massspectrometermeasurements, perature gradient implies rapid zonal (east-west)flow Ber. BunsenGes.Phys. Chem., 96, 339-350, 1992. characterizingthe "jet" at the edge of the vortex, while Atkinson, R. J., W. A. Matthews, P. A. Newman, and R. A. Plumb,Evidenceof mid-latitudeimpactof Antarcticozone the air within the vortex is relativelyisolatedin compardepletion,Nature, 340, 290-294, 1989. ison with surroundingregions. Austin, J., R. R. Garcia, J. M. Russell III, S. Solomon, and

ACKNOW!.I:DGMENIS. The author is very grateful to the manycolleagueswho sentreprints,made suggestions, and supplieddata for inclusionhere. R. Zander and E. Mathieu providedATMOS observationsof chlorinepartitioning,while T. Turunen and E. Kyro kindly providedthe Finnish ozone-

A. F. Tuck, On the atmosphericphotochemistryof nitric acid, J. Geophys.Res., 91, 5477-5485, 1986. Austin,J., N. Butchart,and K. P. Shine,Possibilityof an Arctic ozonehole in a doubled-CO2climate,Nature,360, 221-225, 1992.

Availone, L. M., D. W. Toohey,W. H. Brune, R. J. Salawitch, A. E. Dessler, and J. G. Anderson, Balloon-borne in situ

306 ß Solomon: STRATOSPHERIC OZONE DEPLETION

37, 3 / REVIEWS OF GEOPHYSICS

measurementsof C10 and ozone: Implicationsfor hetero- Brasseur, G., and C. Granier, Mount Pinatubo aerosols,chlorofluorocarbons,and ozone depletion,Science,257, 1239geneouschemistryand mid-latitudeozone loss,Geophys. Res. Lett., 20, 1795-1798, 1993a.

Availone, L. M., D. W. Toohey, M. H. Proffitt, J. J. Margitan, K. R. Chan, and J. G. Anderson, In situ measurements of C10 at mid-latitudes: Is there an effect from Mt. Pinatubo?,

1242, 1992.

Brasseur,G., and M. H. Hitchman, Stratosphericresponseto tracegasperturbations:Changesin ozoneand temperature distributions, Science,240, 634-637, 1988.

Geophys.Res.Lett., 20, 2519-2522, 1993b. Bates, D. R., and M. Nicolet, Atmospherichydrogen,Publ.

Brasseur,G., and S. Solomon,Aeronomyof the Middle Atmosphere,2nd ed., D. Reidel, Norwell, Mass., 1986.

Astron. Soc. Pac., 62, 106-110, 1950. Becker, G., R. Miiller, D. S. McKenna, M. Rex, and K. S. Carslaw,Ozone lossrates in the Arctic stratospherein the

Brasseur, G. P., X. X. Tie, P. J. Rasch, and F. Lefevre, A

winter 1991/92:Model calculationscomparedwith Match results,Geophys.Res.Lett., 23, 4325-4328, 1998. Bekki, S., On the possiblerole of aircraft-generated sootin the middle latitude ozone depletion, J. Geophys.Res., 102, 10,751-10,758, 1997.

Bekki, S., and J. A. Pyle,A two-dimensionalmodelingstudyof the volcaniceruption of Mount Pinatubo,J. Geophys.Res., 99, 18,861-18,869, 1994. Bell, W., N. A. Martin, T. D. Gardiner, N. R. Swann, P. T.

Woods, P. F. Fogal, and J. W. Waters, Column measure-

ments of stratospheric tracespecies over•re, Sweden, in

three-dimensional

simulation

of the Antarctic

ozone hole:

Impact of anthropogenicchlorineon the lower stratosphere and upper troposphere,J. Geophys.Res., 102, 8909-8930, 1997.

Bregman,A., M. van den Broek, K. S. Carslaw,R. Miiller, T. Peter, M.P. Scheele,and J. Lelieveld, Ozone depletion in the late winter lower Arctic stratosphere:Observationsand model results,J. Geophys.Res., 102, 10,815-10,828,1997. Brewer, A. W., Evidencefor a world circulationprovidedby the measurementsof helium and water vapour distribution in the stratosphere,Q. J. R. Meteorol.Soc., 75, 351-363, 1949.

Browell, E. V., C. F. Butler, S. Ismail, P. A. Robinette, A. F.

the winter of 1991-1992,Geophys. Res.Lett., 21, 1347-1350,

Carter, N. S. Higdon, O. B. Toon, M. R. Schoeberl,and

1994.

A. F. Tuck, Airborne lidar observations in the wintertime

Arctic stratosphere:Polar stratosphericclouds,Geophys. Res. Lett., 17, 385-388, 1990. and freezing mechanismof nitric acid trihydrate aerosol under stratosphericconditions, J. Geophys.Res., 103, Brune, W. H., J. G. Anderson, and K. R. Chan, In situ

Bertram, A. K., and J. J. Sloan, The nucleation rate constants

13,261-13,265, 1998.

Bevilacqua,R. M., et al., POAM II ozoneobservations in the Antarctic ozone hole in 1994, 1995, and 1996, J. Geophys. Res., 102, 23,643-23,657, 1997.

Beyer,K. D., S. W. Seago,H. Y. Chang,and M. J. Molina, Compositionand freezingof aqueousH2SO4-HNO3 solutions under polar stratosphericconditions,Geophys.Res. Lett., 21,871-874,

1994.

Beyerle, G., R. Neuber, and O. Schrems,Multiwavelength lidar measurementsof stratosphericaerosolsabove Spitsbergenduringwinter 1992/93,Geophys. Res.Lett., 21, 5760, 1994.

Bianco, R., and J. T. Hynes, Ab initio model study of the mechanismof chlorine nitrate hydrolysison ice, J. Phys. Chem., 102, 309 -314, 1998.

Boering,K. A., S.C. Wofsy,B.C. Daube,H. R. Schneider,M. Loewenstein,J. R. Podolske, and T. J. Conway, Stratosphericmeanagesand transportratesfrom observations of carbon dioxide and nitrous oxide, Science,274, 1340-1343, 1996.

Bojkov, R. D., and V. E. Fioletov, Changesof the lower stratospheric ozoneover Europe and Canada,J. Geophys. Res., 102, 1337-1347, 1997.

Bojkov, R. D., C. S. Zerefos, D. S. Balis, I. C. Ziomas, and Z. F. Bais,Record low total ozoneduringnorthernwinters of 1992-1993, Geophys.Res.Lett., 20, 1351-1354, 1993. Bojkov, R. D., L. Bishop, and V. E. Fioletov, Total ozone trendsfrom quality-controlled ground-based data (19641994),J. Geophys.Res.,100, 25,867-25,876,1995. Borrmann,S., S. Solomon,J. E. Dye, andB. Luo, The potential of cirruscloudsfor heterogenouschlorineactivation,Geophys.Res.Lett., 23, 2133-2136, 1996. Borrmann, S., S. Solomon,J. E. Dye, D. Baumgardner,K. K. Kelly, and K. R. Chan, Heterogeneousreactionson stratosphericbackgroundaerosols, volcanicsulfuricaciddroplets, and type I polar stratosphericclouds:Effects of temperature fluctuationsand differencesin particle phase,J. Geophys.Res., 102, 3639-3648, 1997a. Borrmann, S., S. Solomon, L. Availone, D. Toohey, and D. Baumgardner,On the occurrenceof C10 in cirrusclouds and volcanic aerosol in the tropopauseregion, Geophys. Res. Lett., 24, 2011-2014, 1997b.

observations

of BrO

over Antarctica:

ER-2

aircraft

results

from 54øSto 72øSlatitude, J. Geophys.Res., 94, 16,63916,647, 1989.

Brune,W. H., D. W. Toohey,J. G. Anderson,and K. R. Chan, In situobservations of C10 in the Arctic stratosphere: ER-2 aircraft resultsfrom 59øN to 80øN latitude, Geophys.Res. Lett., 17, 505-508, 1990.

Brune, W. H., J. G. Anderson, D. W. Toohey, D. W. Fahey, S. R. Kawa, R. L. Jones, D. S. McKenna, and L. R. Poole,

The potentialfor ozonedepletionin the Arctic polar stratosphere,Science,252, 1260-1266, 1991. Cadle,R. D., P. Crutzen, and D. Ehhalt, Heterogeneouschemical reactions in the stratosphere,J. Geophys.Res., 80, 3381-3385, 1975.

Callis,L. B., and M. Natarajan,The Antarcticozoneminimum: Relationshipto odd nitrogen,odd chlorine,the final warming, and the 11-yearsolarcycle,J. Geophys. Res.,91, 10,77110,798, 1986.

Cariolle, D., A. Lasserre-Bigorry,and J.-F. Royer, A general circulationmodel simulation of the springtimeAntarctic ozone decreaseand its impacton midlatitudes,J. Geophys. Res., 95, 1883-1898, 1990.

Carslaw,K. S., and T. Peter, Uncertaintiesin reactiveuptake coefficientsfor solid stratosphericparticles, 1, Surface chemistry,Geophys.Res.Lett., 24, 1743-1746, 1997. Carslaw,K. S., B. P. Luo, S. L. Clegg, T. Peter, P. Brimblecombe,and P. J. Crutzen,Stratosphericaerosolgrowthand HNO 3 gasphasedepletionfrom coupledHNO 3 and water uptake by liquid particles, Geophys.Res. Lett., 21, 24792482, 1994.

Carslaw,K. S., T. Peter, and S. L. Clegg, Modeling the compositionof liquid stratosphericaerosols,Rev. Geophys.,35, 125-154, 1997a. Carslaw, K. S., T. Peter, and R. Miiller, Uncertainties in

reactiveuptakecoefficientsfor solidstratosphericparticles, 2, Effect on ozone depletion,Geophys.Res.Lett., 24, 17471750, 1997b.

Carslaw,K. S., et al., Increasedstratosphericozone depletion due to mountain-inducedatmosphericwaves,Nature, 391, 675-678, 1998.

Chandra, S., C. H. Jackman,A. R. Douglass,E. L. Fleming, and D. B. Considine, Chlorine catalyzed destructionof

37, 3 / REVIEWSOF GEOPHYSICS

ozone: Implications for ozone variability in the upper stratosphere,Geophys.Res.Lett., 20, 351-354, 1993. Chapman, S., On ozone and atomic oxygen in the upper atmosphere,Philos.Mag., 10, 369-383, 1930. Chin, M., and D. D. Davis,A reanalysisof carbonylsulfideas a source of stratosphericbackgroundsulfur aerosols,J. Geophys.Res., 100, 8993-9005, 1995. Chipperfield,M.P., and J. A. Pyle, Two-dimensionalmodelling of the Antarcticlower stratosphere,Geophys. Res.Lett.,

Solomon: STRATOSPHERICOZONE DEPLETION ß 307

chlorocarbonindustryon the ozone layer,J. Geophys.Res., 83, 345-363, 1978. Dameris, M., V. Grewe, R. Hein, C. Schnadt, C. Brfihl, and B.

Steil, Assessmentof the future developmentof the ozone layer, Geophys.Res.Lett., 25, 3579-3582, 1998. Danilin, M. Y., and J. C. McConnell, Stratosphericeffectsof bromine activation on/in sulfate aerosol,J. Geophys.Res., 100, 11,237-11,243, 1995.

Del Negro, L. A., et al., Evaluatingthe role of NAT, NAD, and 15, 875-878, 1988. liquid H2SO4/H20/HNO3 solutions in Antarctic polar stratosphericcloud aerosol:Observationsand implications, Chipperfield,M.P., and J. A. Pyle,Model sensitivitystudiesof J. Geophys.Res.,102, 13,255-13,282,1997. Arctic ozone depletion, J. Geophys.Res., 103, 28,38928,403, 1998. Deniel, C., R. M. Bevilacqua,J.P. Pommereau,and F. Lefevre, Arctic chemicalozonedepletionduringthe 1994-1995 Chipperfield,M.P., D. Cariolle,and P. Simon,A 3D transport winter deduced from POAM II satellite observations and model study of chlorine activationduring EASOE, Geothe REPROBUS three-dimensional model, J. Geophys. phys.Res.Lett., 21, 1467-1470, 1994. Res., 103, 19,231-19,244, 1998. Chipperfield,M.P., A.M. Lee, and J. A. Pyle, Model calculations of ozone depletion in the Arctic polar vortex for Deshler, T., A. Adriani, G. P. Gobbi, D. J. Hofmann, G. Di Donfrancesco, and B. J. Johnson, Volcanic aerosol and 1991/92to 1994/95, Geophys.Res.Lett., 23, 559-562, 1996. Chubachi, S., Preliminary result of ozone observationsat ozone depletionwithin the Antarctic polar vortex during SyowaStation from February, 1982 to January,1983,Mem. the austral springof 1991, Geophys.Res. Lett., 19, 18191822, 1992. Natl. Inst. Polar Res.Jpn. Spec.Issue,34, 13-20, 1984. Clapp, M. L., R. F. Niedziela, L. J. Richwine,T. Dransfield, Dessler, A. E., et al., Correlated observations of HC1 and R. E. Miller, and D. R. Worsnop,Infrared spectroscopy of C1ONO2 from UARS and implicationsfor stratospheric sulfuric acid/water aerosols: Freezing characteristics,J. chlorine partitioning, Geophys.Res. Lett., 22, 1721-1724, 1995. Geophys.Res., 102, 8899-8907, 1997. Coffey, M. T., and W. G. Mankin, Observationsof the lossof Dessler,A. E., S. R. Kawa, A. R. Douglass,D. B. Considine, J. B. Kumer, A. E. Roche,J. L. Mergenthaler,J. W. Waters, stratosphericNO 2 followingvolcaniceruptions,Geophys. Res. Lett., 20, 2873-2876, 1993. J. M. RussellIII, and J. C. Gille, A test of the partitioning betweenC10 and C1ONO2 usingsimultaneousUARS meaCoffey, M. T., W. G. Mankin, and A. Goldman, Airborne surementsof C10; NO2 and C1ONO2,J. Geophys.Res.,101, measurements of stratosphericconstituentsoverAntarctica 12,515-12,521, 1996. in the austral spring,1987, 2, Halogen and nitrogen trace Dessler, A. E., D. B. Considine, J. E. Rosenfield, S. R. Kawa, gases,J. Geophys.Res.,94, 16,597-16,613,1989. A. R. Douglass,and J. M. RussellIII, Lower stratospheric Cohen, R. C., et al., Are modelsof catalyticremoval of 03 by chlorinepartitioningduring the decayof the Mt. Pinatubo HO x accurate?,Constraintsfrom in situ measurementsof aerosolcloud, Geophys.Res.Lett., 24, 1623-1626, 1997. the OH to HO 2 ratio, Geophys.Res. Lett., 21, 2539-2542, 1994. Dessler,A. E., M.D. Burrage,J.-U. Grooss,J. R. Holton, J. L. Collins, R. L., K. P. Bownan, and C. S. Gardner, Polar stratoLean, S. T. Massie, M. R. Schoeberl,A. R. Bouglass,and C. H. Jackman,Selectedsciencehighlightsfrom the first 5 sphericcloudsat the SouthPole in 1990:Lidar observations and analysis,J. Geophys.Res., 98, 1001-1010, 1993. yearsof the Upper AtmosphereResearchSatellite(UARS) Cornu, A., Observation de la limite ultraviolette du spectre program,Rev. Geophys.,36, 183-210, 1998. solaire h diversesaltitudes, C. R. Hebd. Seances,,Icad. Sci., 89, 808, 1879. Cox, R. A., A. R. MacKenzie, R. H. Mtiller, T. Peter, and P. J.

Crutzen,Activationof stratospheric chlorineby reactionsin liquid sulphuricacid, Geophys.Res. Lett., 21, 1439-1442, 1994.

Coy, L., E. R. Nash, and P. A. Newman,Meteorologyof the polar vortex: Spring 1997, Geophys.Res. Lett., 24, 26932696, 1997.

Crewell, S., K. Ktinzi, H. Nett, T. Wehr, and P. Hartogh, Aircraft measurementsof C10 and HC1 during EASOE 1991/92, Geophys.Res.Lett., 21, 1267-1270, 1994. Crutzen, P. J., The influenceof nitrogen oxide on the atmosphericozonecontent,Q. J. R. MeteoroLSoc.,96, 320-327, 1970.

de Zafra, R. L., M. Jaramillo, A. Parrish, P. Solomon, B. Connor, and J. Barnett, High concentrationsof chlorine

monoxide at low altitudes in the Antarctic spring stratosphere:Diurnal variation,Nature, 328, 408-411, 1987. de Zafra, R. L., M. Jaramillo, J. Barrett, L. K. Emmons, P.M.

Solomon, and A. Parrish, New observationsof a large concentrationof C10 in the springtimelower stratosphere over Antarctica and its implicationsfor ozone-depleting chemistry,J. Geophys.Res., 94, 11,423-11,428, 1989. de Zafra, R. L., L. K. Emmons, J. M. Reeves, and D. T.

Shindell, An overview of millimeter-wave spectroscopic measurements of chlorine monoxide at Thule, Greenland,

February-March, 1992:Vertical profiles,diurnal variation, and longer-termtrends,Geophys.Res.Lett., 21, 1271-1274, 1994.

Crutzen,P. J., Ozoneproductionratesin an oxygen-hydrogen- Dobson, G. M. B., Observationsof the amount of ozone in the nitrogenoxideatmosphere,J. Geophys. Res.,76, 7311-7327, Earth's atmosphereand its relation to other geophysical 1971. conditions,Proc. R. Soc. London, Ser.A., 129, 411, 1930. Crutzen, P. J., Estimatesof possiblefuture ozone reductions Dobson,G. M. B., Forty years'researchon atmosphericozone at Oxford: A history,Appl. Opt., 7, 387-405, 1968. from continueduse of fluoro-chloro-methanes (CF2C12, Donovan, D. P., et al., Ozone, column C10, and PSC measureCFC13),Geophys. Res.Lett., 1, 205-208, 1974. Crutzen,P. J., The possibleimportanceof OCS for the sulfate mentsmadeat the NDSC Eurekaobservatory (80øN,86øW) during the spring of 1997, Geophys.Res. Lett., 24, 2709layer of the atmosphere,Geophys. Res.Lett., 3, 73-76, 1976. 2712, 1997. Crutzen, P. J., and F. Arnold, Nitric acid cloud formation in the cold Antarctic stratosphere:A major cause for the Douglass,A. R., M. R. Schoeberl,R. S. Stolarski,J. W. Waters, J. M. Russell III, A. E. Roche, and S. T. Massie, Interhemispringtime'ozone hole,' Nature, 324, 651-655, 1986. Crutzen, P. J., I. S. A. Isaksen,and J. R. McAfee, Impact of sphericdifferencesin springtimeproductionof HC1 and

308 ß Solomon: STRATOSPHERICOZONE DEPLETION

C1ONO2 in the polar vortices,J. Geophys. Res.,100, 13,96713,978, 1995. Drdla, K., A. Tabazadeh, R. P. Turco, M. Z. Jacobsen,J. E.

Dye, C. Twohy, and D. Baumgardner,Analysisof the physical state of one Arctic polar stratosphericcloud basedon observations,Geophys.Res.Lett., 21, 2475-2478, 1994. Dtitsch, H. V., Ozone distribution in the atmosphere,Can.

37, 3 / REVIEWS OF GEOPHYSICS

ozone and their relationshipto variationsof planetarywave activity,J. Clim., 12, 1619-1629, 1999. Gandrud, B. W., P.S. Sperry, L. Sanford, K. K. Kelly, G. V. Ferry, and K. R. Chan, Filter measurementresultsfrom the Airborne Antarctic Ozone Experiment,J. Geophys.Res.,94, 11,285-11,297, 1989.

Gandrud, B. W., J. E. Dye, D. Baumgardner,G. V. Ferry, M. J. Chem., 52, 1491-1504, 1974. Loewenstein,K. R. Chan, L. Sanford,B. Gary, and K. Kelly, Dye, J. E., D. Baumgardner,B. W. Gandrud,S. R. Kawa, K. K. The January 30, 1989, Arctic polar stratosphericclouds Kelly, M. Loewenstein,G. V. Ferry, K. R. Chan, and B. L. (PSC) event: Evidence for a mechanismof dehydration, Geophys.Res.Lett., 17, 457-460, 1990. Gray, Particle size distributionsin Arctic polar stratosphericclouds,growthand freezingof sulfuricaciddroplets Gao, R. S., B. Kaercher, E. R. Keim, and D. W. Fahey, and implicationsfor cloud formation,J. Geophys.Res.,97, Constrainingthe heterogneneouslossof 03 on sootparti8015-8034, 1992. cleswith observationsin jet engine exhaustplumes, Geophys.Res.Lett., 25, 3323-3326, 1998. Dye, J. E., et al., In-situ observationsof an Antarctic polar stratosphericcloud: Similaritieswith Arctic observations, Gardiner, B.C., Comparative morphology of the vertical Geophys.Res.Lett., 23, 1913-1916, 1996. ozone profile in the Antarctic spring,Geophys.Res. Lett., 15, 901-904, 1988. Elkins, J. W., T. M. Thompson,T. H. Swanson,J. H. Butler, B. D. Hall, S. O. Cummings,D. A. Fisher,and A. G. Raffo, Geller, M. A., V. Yudin, A. R. Douglass,J. W. Waters, L. S. Elson, A. E. Roche, and J. M. Russell III, UARS PSC, Decreasein the growthrates of atmosphericchlorofluorocarbons 11 and 12, Nature, 364, 780-783, 1993. C1ONO2,HC1 and C10 measurementsin earlywinter: Additionalverificationof the paradigmfor chlorineactivation, Fabry, C., and M. Buisson,L'absorptionde l'ultraviolet par Geophys.Res.Lett., 22, 2937-2940, 1995. l'ozone et la limite du spectresolaire,J. Phys.,3, 196-206, 1913. Gerber, L., and N. K/impfer, Millimeter-wave measurements of chlorine monoxideat the JungfraujochAlpine Station, Fahey,D. W., D. M. Murphy, K. K. Kelly, M. K. W. Ko, M. H. Geophys.Res.Lett., 21, 1279-1282, 1994. Proffitt, C. S. Eubank, G. V. Ferry, M. Loewenstein,and K. R. Chan, Measurements of nitric oxide and total reactive Gertner, B. J., and J. T. Hynes,Molecular dynamicssimulation of hydrochloricacid ionization at the surface of stratonitrogen in the Antarctic stratosphere:Observationsand sphericice, Science,271, 1563-1566, 1996. chemicalimplications,J. Geophys.Res.,94, 16,665-16,681, 1989a. Gleason,J. F., et al., Recordlow globalozonein 1992,Science, 260, 523-526, 1993. Fahey, D. W., K. K. Kelly, G. V. Ferry, L. R. Poole, J. C. Wilson,D. M. Murphy, M. Loewenstein,and K. R. Chan,In Gobbi, G. P., G. D. Donfrancesco,and A. Adriani, Physical situ measurementsof total reactive nitrogen, total water propertiesof stratosphericcloudsduringthe Antarcticwinter of 1995,J. Geophys.Res., 103, 10,859-10,873, 1998. and aerosolin a polar stratosphericcloudin the Antarctic, J. Geophys.Res., 94, 11,299-11,315, 1989b. Godin, S., G. M6gie, C. David, D. Haner, C. Flesia, and Y. Emery, Airborne lidar observationsof mountain-wave-inFahey, D. W., S. R. Kawa, and K. R. Chan, Nitric oxide measurementsin the Arctic winter stratosphere,Geophys. ducedpolar stratosphericcloudsduring EASOE, Geophys. Res. Lett., 17, 489-492, 1990a.

Fahey, D. W., K. K. Kelly, S. R. Kawa, A. F. Tuck, M. Loewenstein, K. R. Chan, and L. E. Heidt, Observationsof

denitrificationand dehydrationin the winter polar stratospheres,Nature, 344, 321-324, 1990b. Fahey,D. W., et al., In situmeasurementsconstrainingthe role of sulfate aerosolsin mid-latitude ozone depletion,Nature, 363, 509-514, 1993. Fairbrother, D. H., D. J. D. Sullivan, and H. S. Jonston, Global

thermodynamicatmosphericmodeling:Searchfor new heterogeneousreactions,J. Phys.Chem.A., 101, 7350-7358, 1997.

Farman, J. C., B. G. Gardiner, and J. D. Shanklin,Large losses of total ozone in Antarctica reveal seasonal C1Ox/NO• interaction, Nature, 315, 207-210, 1985.

Farmer, C. B., G. C. Toon, P. W. Shaper, J. F. Blavier, and L. L. Lowes, Stratospherictrace gasesin the spring 1986 Antarctic atmosphere,Nature, 329, 126-130, 1987. Fels, S. B., J. D. Mahlman, M.D. Schwarzkopf,and R. W. Sinclair,Stratosphericsensitivityto perturbationsin ozone and carbon dioxide radiative and dynamical response,J.

Res. Lett., 21, 1335-1338, 1994. Goutail, F., J.P. Pommereau,A. Sarkissian,E. Kyro, and V. Dorokhov,Total nitrogendioxideat the Arctic polar circle since 1990, Geophys.Res. Lett., 21, 1371-1374, 1994. Grewe, V., M. Dameris, and R. Sausen,Impact of stratosphericdynamicsand chemistryon Northern Hemisphere midlatitude ozoneloss,J. Geophys.Res.,103, 25,417-25,433, 1998.

Haigh, J. D., and J. A. Pyle, A two-dimensionalcalculation including atmosphericcarbon dioxide and stratospheric ozone, Nature, 279, 222-224, 1979.

Hampson,J., Photochemicalbehaviorof the ozonelayer, Tech. Note TN 1627/64, Can. Armament Res. and Dev. Estab., Valcartier, Que., Canada, 1964.

Hanson, D. R., and K. Mauersberger,The vapor pressuresof HNO3-H20 solutions,J. Phys.Chem.,92, 6167-6170, 1988. Hanson, D. R., and A. R. Ravishankara,Reaction of C1ONO2 with HC1 on NAT, NAD,

and frozen.sulfuric acid and

hydrolysisof N20 s and C1ONO2 on frozen sulfuricacid,J. Geophys.Res.,98, 22,931-22,936, 1993. Hanson,D. R., and A. R. Ravishankara,Reactiveuptake of Atmos. Sci., 37, 2265-2297, 1980. C1ONO2 onto sulfuric acid due to reaction with HC1 and H20 , J. Phys.Chem., 98, 5728-5735, 1994. Fleming,E. L., S. Chandra,J. J. Barnett,andM. Corney,Zonal mean temperatures,pressures,zonal winds, and geopote- Hanson,D. R., andA. R. Ravishankara,Heterogeneouschemnial heights as a function of latitude, Adv. SpaceRes., istryof brominespeciesin sulfuricacidunder stratospheric conditions,Geophys.Res.Lett., 22, 385-388, 1995. 10(12), 11-59, 1990. Fried, A., B. E. Henry, J. G. Calvert, and M. Mozurkewich, Hanson, D. R., A. R. Ravishankara, and S. Solomon, Heterogeneousreactionsin sulfuricacidaerosols:A frameworkfor The reactionprobabilityof N20 s with sulfuricacid aerosols model calculations,J. Geophys.Res., 99, 3615-3629, 1994. at stratospherictemperaturesand compositions, J. Geophys. Res., 99, 3517-3532, 1994. Hanson, D. R., A. R. Ravishankara,and E. R. Lovejoy,Reactions of BrONO2 and H20 on submicron sulfuric acid Fusco,A. C., and M. L. Salby,Interannual variationsof total

37, 3 / REVIEWS OF GEOPHYSICS

aerosoland the implicationsfor the lower stratosphere,J. Geophys.Res., 101, 9063-9069, 1996. Harris, N. R. P., et al., Trends in stratosphericand free troposphericozone,J. Geophys.Res.,102, 1571-1590, 1997. Harris, N., R. Hudson,and C. Phillips(Eds.), Assessment of trendsin the vertical distributionof ozone, SPARC Rep. 1, Ozone Res. and Monit. Proj. Rep. 43, World Meteorol. Organ., Geneva, May 1998. Hartley, D. E., J. T. Villarin, R. X. Black, and C. A. Davis, A new perspectiveon the dynamicallink betweenthe stratosphereand troposphere,Nature, 391,471-474, 1998. Hartley, W. N., On the probableabsorptionof solarradiation by atmosphericozone, Chem.News,42, 268, 1880. Hartmann, D. L., L. E. Heidt, M. Loewenstein, J. R. Podolske,

J. Vedder, W. L. Starr, and S. E. Strahan,Transportinto the south polar vortex in early spring,J. Geophys.Res., 94, 16,779-16,795, 1989. Heath, D. F., A. J. Krueger, and P. J. Crutzen, Solar proton event:Influenceon stratosphericozone,Science,197, 886889, 1977.

Solomon: STRATOSPHERIC OZONE DEPLETION ß 309

Isaksen,I. S. A., and F. Stordal, Antarctic ozone depletion: 2-D model studies,Geophys.Res.Lett., 13, 1327-1330, 1986. Iwasaka, Y., and K. Kondoh, Depletion of Antarctic ozone: Height of ozone loss region and its temporal changes, Geophys.Res.Lett., 14, 87-90, 1987. Jackman,C. H., and R. D. McPeters, Solar proton eventsas tests for the fidelity of middle atmospheremodels,Phys. Scr., T18, 309-316, 1987.

Jackman,C. H., E. L. Fleming, S. Chandra,D. B. Considine, and J. E. Rosenfield, Past, present, and future modeled ozone trendswith comparisonsto observedtrends,J. Geophys.Res., 101, 28,753-28,767, 1996. Jfiger, H., and K. Wege, Stratosphericozone depletion at northern midlatitudes after major volcanic eruptions, J. Atmos. Chem., 10, 273-287, 1990. Jaramillo, M., R. L. de Zafra, J. Barrett, L. K. Emmons, P.M.

Solomon, and A. Parrish, Measurementsof stratospheric hydrogencyanideat McMurdo Station,Antarctica:Further evidence of winter stratosphericsusidence?,J. Geophys. Res., 94, 16,773-16,777, 1989.

Hofmann,D. J., and S. J. Oltmans,The effect of stratospheric Johnston,H. S., Reductionof stratosphericozoneby nitrogen oxide catalystsfrom supersonictransportexhaust,Science, water vapor on the heterogeneousreaction rate for 173, 517-522, 1971. C1ONO2 and H20 for sulfuricacid aerosol,Geophys.Res. Lett., 19, 2211-2214, 1992. Johnston,H. S., Atmosphericozone,Annu. Rev. Phys.Chem., Hofmann, D. J., and S. J. Oltmans, Anomalous Antarctic

43, 1-32, 1992.

ozone during 1992:Evidencefor Pinatubovolcanicaerosol Johnston, H. S., and J. Podolske, Interpretation of stratosphericphotochemistry, Rev. Geophys.,16, 491-519, 1978. effects,J. Geophys.Res.,98, 18,555-18,561, 1993. Hofmann, D. J., and S. Solomon,Ozone destructionthrough Johnston,P. V., and R. L. McKenzie, NO2 observationsat 45øS duringthe decreasingphaseof solarcycle21, from 1980 to heterogeneouschemistryfollowingthe eruption of E1 Chi1987,J. Geophys.Res., 94, 3473-3486, 1989. ch6n, J. Geophys.Res., 94, 5029-5041, 1989. Hofmann, D. J., J. W. Harder, S. R. Rolf, and J. M. Rosen, Johnston,P. V., R. L. McKenzie, J. G. Keys, and W. A. Matthews,Observationsof depletedstratospheric NO2 folBalloon-borneobservations of the developmentand vertilowingthe Pinatubovolcaniceruption, Geophys.Res.Lett., cal structure of the Antarctic ozone hole in 1986, Nature, 326, 59-62, 1987. Hofmann, D. J., S. J. Oltmans, J. M. Harris, S. Solomon, T.

Deschler, and B. J. Johnson,Observationsand possible causesof new ozone depletion in Antarctica in 1991, Nature, 359, 283-287, 1992.

Hofmann, D. J., S. J. Oltmans,W. D. Komhyr, J. M. Harris, J. A. Lathtop, A. O. Langford,T. Deshler, B. J. Johnson,A.

19, 211-213, 1992. Jones, A. E., and J. D. Shanklin, Continued decline of total

ozone over Halley, Antarctica, since 1985, Nature, 376, 409-411, 1995. Jones, R. L., and A. R. MacKenzie, Observational studies on

the role of polar regionsin midlatitudeozoneloss,Geophys. Res. Lett., 22, 3485-3488, 1995.

Jones,R. L., and 2. A. Pyle, Observationsof CH 4 and N20 by the NIMBUS 7 SAMS: A comparisonwith in situ data and stratosphereoverthe United Statesin 1992-1993:Evidence two dimensionalnumericalmodel calculations,J. Geophys. for heterogeneouschemistryon the Pinatuboaerosol,GeoRes., 89, 5263-5279, 1984. phys.Res.Lett., 21, 65-68, 1994. Jones,R. L., et al., LagrangianphotochemicalmodelingstudHofmann, D. J., S. J. Oltmans, J. M. Harris, B. J. Johnson,and ies of the 1987Antarcticspringvortex,1, Comparisonwith J. A. Lathrop, Ten years of ozonesondemeasurementsat AAOE observations,J. Geophys.Res., 94, 11,529-11,558, the South Pole: Implications for recovery of springtime 1989. Antarctic ozone, J. Geophys.Res., 102, 8931-8943, 1997. Jones, R. L., SuSolomon, D. S. McKenna, L. R. Poole, W. H. Hollandsworth,S. M., R. D. McPeters,L. E. Flynn,W. Planet, Brune, D. W. Toohey, J. G. Anderson,and D. W. Fahey, A. J. Miller, and S. Chandra, Ozone trends deduced from The polar stratosphericcloud eventof January24, 2, Phocombined Nimbus 7 SBUV and NOAA 11 SBUV/2 data, tochemistry,Geophys.Res.Lett., 17, 541-544, 1990a. Geophys.Res.Lett., 22, 905-908, 1995. Jones, R. L., D. S. McKenna, L. R. Poole, and S. Solomon, On Holton, J. R., P.'H. Haynes,M. E. Mcintyre, A. R. Douglass, the influence of polar stratosphericcloud formation on R. B.' Rood, and L. Pfister, Stratosphere-troposphere exchemical compositionduring the 1988/89 Arctic winter, change,Rev. Geophys.,33, 403-439, 1995. Geophys.Res.Lett., 17, 545-548, 1990b. Hood, L• L., and D. A. Zaff, Lower stratosphericstationary Jet PropulsionLaboratory(JPL), ChemicalKineticsand Phowaves and the longitude dependenceof ozone trends in tochemicalData for Use in StratosphericModelling, Evalwinter, J. Geophys.Res.,100, 25,791-25,800,1995. uation number 12 of the NASA panel for data evaluation, JPL Publ. 97-4, 1997. Hood, L. L., J.P. McCormack, and K. Labitzke, An investigation of dynamicalcontributionsto midlatitudeozonetrends Junge,C. E., C. W. Changnon,and J. E. Manson,Stratospheric aerosols,J. Meteorol., 18, 81-108, 1961. in winter, J.'Geophys.Res., 102, 13,079-13,093, 1997. Hunten, D. M., Ozone, in The UpperAtmosphereand Magne- Kawa, S. R., D. W. Fahey, L. C. Anderson, M. Loewenstein, tosphere,ppo156-168, Natl. Acad. Sci.,Washington,D.C., and K. R. Chan, Measurementsof total reactive nitrogen 1977. duringthe Airborne Arctic StratosphericExpedition,GeoIraci, L. T., T. J. Fortin, and M. A. Tolbert, Dissolution of phys.Res.Lett., 17, 485-488, 1990. sulfuricacid tetrahydrateat low temperaturesand subse- Kawa, S. R., D. W. Fahey, K. K. Kelly, J. E. Dye, D. Baumgardner,B. W. Gandrud, M. Loewenstein,G. V. Ferry, and quentgrowthof nitric acidtrihydrate,J. Geophys.Res.,103, 8491-8498, 1998. K. R. Chan, The Arctic polar stratosphericcloud aerosol: Torres, and W. A. Matthews, Ozone loss in the lower

310 ß Solomon: STRATOSPHERIC

OZONE

DEPLETION

37, 3 / REVIEWS OF GEOPHYSICS

and J. B. Renard, Carbon aerosolsand atmosphericphotochemistry,J. Geophys.Res.,102, 3671-3682, 1997. Lehmann,P., D. J. Karoly, P. A. Newman, T. S. Clarkson,and W. A. Matthews,An investigationinto the causesof strato3921-3933, 1997. sphericozonelossin the southernaustralasian region,Geophys.Res.Lett., 19, 1463-1466, 1992. Keim, E. R., et al., Observationsof large reductionsin the NO/NOyratio near the mid-latitudetropopause and the Leu, M.-T., Laboratory studiesof sticking coefficientsand heterogeneousreactionsimportant in the Antarctic stratorole of heterogeneouschemistry,Geophys.Res. Lett., 23, 3223-3226, 1996. sphere,Geophys.Res.Lett., 15, 17-20, 1988. Kelly, K. K., et al., Dehydrationin the lower Antarctic strato- Lipson,J. B., M. J. Elrod, T. W. Beiderhase,L. T. Molina, and M. J. Molina, Temperaturedependenceof the rate constant sphereduringlate winter and earlyspring,1987,J. Geophys. Res., 94, 11,317-11,357, 1989. and branchingratio for the OH + C10 reaction,J. Chem. Soc.Faraday Trans.,93, 2665-2673, 1997. Khattatov,V., V. Yushkov,M. Khaplanov,I. Zaitzev,J. Rosen, and N. Kjome, Some resultsof water vapor, ozone and Loewenstein, M., J. R. Podolske, K. R. Chan, and S. E. aerosolballoon-bornemeasurementsduringEASOE, GeoStrahan, Nitrous oxide as a dynamicaltracer in the 1987 phys.Res.Lett., 21, 1299-1302, 1994. AirborneAntarcticOzoneExperiment,J. Geophys. Res.,94, Aircraft measurementsof reactive nitrogen, total water, and particles,J. Geophys.Res.,97, 7925-7938, 1992. Kawa, S. R., et al., Activation of chlorine in sulfate aerosol as inferred from aircraft observations,J. Geophys.Res., 102,

Kirchhoff, V. W. J. H., N.J. Schuch,K. K. Pinheiro, and J. M.

Harris, Evidence for an ozone hole perturbation at 30ø South, Atmos. Environ., 30, 1481-1488, 1996. Kirchhoff, V. W. J. H., C. A. R. Casiccia,S. Zamorano B., and F. Zamorano B., The ozone hole over Punta Arenas, Chile,

11,589-11,598, 1989. Lutman, E. R., R. Toumi, R. L. Jones,D. J. Lary, and J. A. Pyle, Boxmodel studiesof C1Oxdeactivationand ozoneloss duringthe 1991/92Northern Hemispherewinter, Geophys. Res. Lett., 21, 1415-1418, 1994a.

Lutman, E. R., J. A. Pyle, R. L. Jones, D. J. Lary, A. R. MacKenzie, I. Kilbane-Dawe, N. Larsen, and B. Knudsen, Kirchhoff, V. W. J. H., Y. Sahai, C. A. R. Casiccia S., F. Zamorano B., and V. Valderrama V., Observations of the Trajectory model studies of C1Ox activation during the 1995 ozonehole over Punta Arenas, Chile, J. Geophys.Res., 1991/92Northern Hemispherewinter, Geophys.Res.Lett., 21, 1419-1422, 1994b. 102, 16,109-16,120, 1997b. Knight, J. R., J. Austin, R. G. Grainger, and A. Lambert, A Lutman, E. R., J. A. Pyle, M.P. Chipperfield,D. J. Lary, I. Kilbane-Dawe, J. W. Waters, and N. Larsen, Three-dimenthree-dimensionalmodel simulationof the impact of Mt. Pinatubo aerosol on the Antarctic ozone hole, Q. J. R. sionalstudiesof the 1991/1992Northern HemispherewinMeteor. Soc., 124, 1527-1558, 1998. ter usingdomain-fillingtrajectorieswith chemistry,J. GeoKo, M. K. W., and N. D. Sze, Effect of recent rate data phys.Res., 102, 1479-1488, 1997. revisionson stratospheric modeling,Geophys. Res.Lett., 10, MacKenzie, A. R., M. Kulmala, A. Laaksonen, and T. Vesala, 341-344, 1983. On the theoriesof type I polar stratosphericcloud formation, J. Geophys.Res., 100, 11,275-11,288, 1995. Ko, M. K. W., J. M. Rodrigues,N. D. Sze,M. H. Proffitt, W. L. Starr, A. Krueger, E. V. Browell, and M.P. McCormick, MacKenzie, I. A., R. S. Harwood, L. Froidevaux, W. G. Read, and J. W. Waters, Chemical loss of polar vortex ozone Implicationsof AAOE observations for proposedchemical inferred from UARS MLS measurementsof C10 duringthe explanationsof the seasonaland interannualbehavior of Arctic and Antarctic late winters of 1993,J. Geophys.Res., Antarctic ozone,J. Geophys.Res.,94, 16,705-16,715,1989. 101, 14,505-14,518, 1996. Koike, M., N. B. Jones, W. A. Matthews, P. V. Johnston,R. L. McKenzie, D. Kinnison, and J. Rodriguez,Impact of Pina- Mahlman, J. D., and S. B. Fels, Antarctic ozone decreases:A dynamicalcause?,Geophys. Res.Lett., 13, 1316-1319, 1986. tubo aerosolson the partitioningbetweenNO 2 and HNO3, Mankin, W. G., M. T. Coffey, A. Goldman, M. R. Schoeberl, Geophys.Res.Lett., 21,597-600, 1994. L. R. Lait, and P. A. Newman, Airborne measurementsof Komhyr, W. D., R. D. Grass,and R. K. Leonard, Total ozone decreaseat South Pole, Antarctica, 1964-1985, Geophys. stratosphericconstituentsover the Arctic in the winter of Res. Lett., 13, 1248-1251, 1986. 1989, Geophys.Res.Lett., 17, 473-476, 1990. Kondo, Y., U. Schmidt,T. Sugita,P. Aimendieu, M. Koike, H. Mankin, W. G., M. T. Coffey, and A. Goldman, Airborne Ziereis, and Y. Iwasaka,Total reactivenitrogen, N20 and observationsof SO2, HC1, and 03 in the stratospheric ozonein the winter Arctic stratosphere, Geophys. Res.Lett., plume of the Pinatubovolcanoin July 1991, Geophys. Res.

J. Geophys.Res.,102, 8945-8953, 1997a.

Lett., 19, 179-182, 1992.

21, 1247-1250, 1994.

Koop, T., U. M. Biermann,W. Raber, B. P. Luo, P. J. Crutzen, Manney, G. L., et al., Chemical depletion of ozone in the Arctic lower stratosphereduring winter 1992-93, Nature, andT. Peter,Do stratospheric aerosolsfreezeabovethe ice 370, 429-434, 1994a. frost point?, Geophys.Res.Lett., 22, 917-920, 1995. Koop,T., K. S. Carslaw,andT. Peter,Thermodynamic stability Manney, G. L., R. W. Zurek, A. O'Neill, and R. Swinbank,On the motion of air throughthe stratosphericpolar vortex,J. and phasetransitionsof PSC particles,Geophys. Res.Lett., Atmos. Sci., 51, 2973-2994, 1994b.

24, 2199-2202, 1997.

Kreher, K., J. G. Keys, P. V. Jonston,U. Platt, and X. Liu, Ground-based

measurements

of OC10

and HC1 in austral

spring 1993 at Arrival Heights, Antarctica, Geophys.Res. Lett., 23, 1545-1548, 1996.

Larsen, N., B. M. Knudsen, J. M. Rosen, N. T. Kjome, R. Neuber, and E. Kryo, Temperaturehistoriesin liquid and solidpolar stratosphericcloudformation,J. Geophys. Res.,

Manney, G. L., et al., Lagrangiantransportcalculationsusing UARS data, I, Passivetracers,J. Atmos. Sci., 52, 3049-3068, 1995a.

Manney, G. L., R. W. Zurek, L. Froidevaux,J. W. Waters, A. O'Neill, and R. Swinbank,Lagrangian transport calculationsusingUARS data, II, Ozone,J. Atmos.Sci.,52, 30693081, 1995b.

Manney, G. L., R. W. Zurek, L. Foidevaux,and J. W. Waters, Lary, D. J., Gas phasebromine photochemistry, J. Geophys. Evidencefor Arctic ozone depletionin late Februaryand Res., 101, 1505-1516, 1996. early March 1994,Geophys.Res.Lett., 22, 2941-2944, 1995c. Lary, D. J., M.P. Chipperfield,R. Toumi, and T. Lenton, Manney, G. L., M. L. Santee,L. Froidevaux,J. W. Waters, and R. W. Zurek, Polar vortex conditionsduring the 1995-96 Heterogeneousatmosphericbrominechemistry, J. Geophys. Res., 101, 1489-1504, 1996. Arctic winter: Meteorologyand MLS ozone, Geophys.Res. Lett., 23, 3203-3206, 1996. Lary, D. J., A.M. Lee, R. Toumi, M. J. Newchurch,M. Pirre, 102, 23,505-23,517, 1997.

37, 3 / REVIEWS OF GEOPHYSICS

Manney, G. L., L. Froidevaux,M. L. Santee,R. W. Zurek, and

Solomon: STRATOSPHERIC OZONE DEPLETION ß 311

Miller, M. H. Proffitt, R. W. Sanders, and S. Solomon, On

J. W. Waters, MLS observations of Arctic ozone loss in

the relationshipbetween stratosphericaerosolsand nitrogen dioxide, Geophys.Res.Lett., 20, 1187-1190, 1993. Materer, N., U. Starke, A. Barbieri, M. A. Van Hove, G. A. Molina, L. T., and M. J. Molina, Productionof C1202from the Somojai, G. J. Kroes, and C. Minot, Molecular surface self-reactionof the C10 radical, J. Phys. Chem., 91, 433436, 1987. structureof Ice(0001): Dynamicallow energyelectrondiffraction total-energycalculationsand molecular dynamics Molina, M. J., and F. S. Rowland, Stratosphericsink for simulations,Su• Sci., 381, 190-210, 1997. chlorofluoromethanes: Chlorineatom catalyzeddestruction of ozone, Nature, 249, 820-812, 1974. McCormack, J.P., and L. L. Hood, Modeling the spatial distributionof total ozonein Northern Hemispherewinter: Molina, M. J., T-L. Tso, L. T. Molina, and F. C.-Y. Wang, 1979-1991, J. Geophys.Res., 102, 13,711-13,717, 1997. Antarcticstratosphericchemistryof chlorinenitrate, hydroMcCormick, M.P., H. M. Steele, P. Hamill, W. P. Chu, and gen chloride, and ice: Release of active chlorine, Science, 238, 1253-1257, 1987. T. J. Swissler,Polar stratosphericcloud sightingsby SAM II, J. ,4tmos. Sci., 39, 1387-1397, 1982. Molina, M. J., R. Zhang, P. J. Wooldridge,J. R. McMahon, McCormick,M.P., R. E. Veiga, and W. P. Chu, Stratospheric J. E. Kim, H. Y. Chang,and K. D. Beyer,Physicalchemistry ozone profile and total ozone trends derived from the of the H2SO4/HNO3/H20 system:Implications for polar SAGE I and SAGE II data, Geophys.Res. Lett., 19, 269stratosphericclouds,Science,261, 1418-1423, 1993. 272, 1992. Montzka, S. A., J. H. Butler, R. C. Myers, T. M. Thompson, T. H. Swanson,A.D. Clarke, L. T. Lock, and J. W. Elkins, McCormick, M.P., L. W. Thomason,and C. R. Trepte, Atmosphericeffects of the Mt. Pinatubo eruption, Nature, Decline in the troposphericabundanceof halogen from 373, 399-404, 1995. halocarbons:Implicationsfor stratosphericozone depletion, Science,272, 1318-1322, 1996. McElroy, M. B., R. J. Salawitch,S.C. Wofsy, and J. A. Logan, Reductionsof Antarctic ozone due to synergisticinterac- Mount, G. H., R. W. Sanders,A. L. Schmeltekopf,and S. tions of chlorine and bromine, Nature, 321,759-762, 1986. Solomon,Visible spectroscopy at McMurdo Station, AntMcElroy, M. B., R. J. Salawitch, and K. Minschwaner, The arctica, 1, Overview and daily variations of NO 2 and 03 changing stratosphere,Planet. Space Sci., 40, 373-401, duringaustralspring,1986,J. Geophys. Res.,92, 8320-8328, 1996-97, Geophys.Res.Lett., 24, 2697-2700, 1997.

1992.

McGee, T. J., P. Newman, M. Gross,U. Singh,S. Godin, A.-M. Lacoste,and G. M6gie, Correlation of ozone losswith the presenceof volcanicaerosols,Geophys.Res.Lett., 21, 28012804, 1994. McKenna, D. S., R. L. Jones, L. R. Poole, S. Solomon, W. H.

Brune, D. W. Fahey, K. K. Kelly, M. Loewenstein, and K. R. Chan, Calculationsof ozone destructionduring the 1988/89 Arctic winter, Geophys.Res. Lett., 17, 553-556, 1990.

McKenzie, R. L., and P. V. Johnston,Springtimestratospheric NO 2 in Antarctica, Geophys.Res.Lett., 11, 73-75, 1984. McPeters, R. D., and G. J. Labow, An assessment of the

1987.

Mozurkewich,M., and J. G. Calvert, Reaction probability of N205 on aqueousaerosols,J. Geophys.Res., 93, 15,88915,896, 1988. M/iller, R., P. J. Crutzen, J-U. Grooss, C. Br/ihl, J. M. Russell

III, and A. F. Tuck, Chlorine activationand ozone depletion in the Arctic vortex: Observationsby the Halogen Occultation Experiment on the Upper Atmosphere ResearchSatellite,J. Geophys.Res., 101, 12,531-12,554, 1996. M/iller, R., P. J. Crutzen, J-U. Grooss, C. Br/ihl, J. M. Russell III, H. Gernandt, D. S. McKenna, and A. F. Tuck, Severe

chemical ozone loss in the Arctic during the winter of 1995-96, Nature, 389, 709-712, 1997a.

accuracyof 14.5 yearsof Nimbus 7 TOMS version7 ozone M/iller, R., J-U. Grooss, D. S. McKenna, P. J. Crutzen, C. Br/ihl, J. M. Russell III, and A. F. Tuck, HALOE observadata by comparisonwith the Dobson network, Geophys. Res. Lett., 23, 3695-3698, 1996a. tions of the vertical structureof chemicalozone depletion McPeters, R. D., S. M. Hollandsworth, L. E. Flynn, J. R. in the Arctic vortex duringwinter and early spring1996Herman, and C. J. Seftor, Long-termozonetrendsderived 1997, Geophys.Res.Lett., 24, 2717-2720, 1997b. M/iller, R., R. J. Salawitch, P. J. Crutzen, W. A. Lahoz, G. L. from the 16-year combined Nimbus 7/Meteor 3 TOMS version7 record, Geophys.Res.Lett., 23, 3699-3702, 1996b. Manney, and R. Tuomi, Upper stratosphericprocesses,in Meilinger, S. K., T. Koop, B. P. Luo, T. Huthwelker, K. S. ScientificAssessment of OzoneDepletion:1998,Rep. 44, pp. Carslaw, U. Krieger, P. J. Crutzen, and T. Peter, Size6.1-6.4, World Meteorol. Organ./U. N. Environ. Prodependent stratosphericdroplet compositionin leewave gramme, Geneva, 1999. temperature fluctuationsand their potential role in PSC Murcray, D. G., D. B. Barker, J. N. Brooks,A. Goldman, and W. J. Williams, Seasonal and latitudinal variation of the freezing,Geophys.Res.Lett., 22, 3031-3034, 1995. Michelangeli,D. V., M. Allen, and Y. L. Yung, E1 Chich6n stratospheric concentrationof HNO3, Geophys. Res.Lett., 2, 223-225, 1975. volcanicaerosols:Impact of radiative, thermal, and chemicalperturbations,J. Geophys. Res.,94, 18,429-18,443,1989. Murphy, D. M., Ozone lossratescalculatedalongER-2 flight Michelsen, H. A., et al., Stratosphericchlorine partitioning: tracks,J. Geophys.Res., 96, 5045-5053, 1991. Constraintsfrom shuttle-bornemeasurementsof [HC1], Murphy, D. M., and A. R. Ravishankara,Temperature aver[C1NO3] and [C10], Geophys.Res. Lett., 23, 2361-2364, ages and rates of stratosphericreactions, Geophys.Res. 1996.

Michelsen,H. A., C. M. Spivakovsky,and S.C. Wofsy, Aero-

sol-mediated partitioning of stratospheric ClyandNOy at temperatures above200'K, Geophys. Res.Lett.,26, 299-302, 1999.

Miller, A. J., G. C. Tiao, G. C. Reinsel, D. Wuebbles, L.

Bishop, J. Kerr, R. M. Nagatani, J. J. DeLuisi, and C. L. Mateer, Comparisonsof observed ozone trends in the stratospherethroughexaminationsof Umkehr and balloon ozonesondedata, J. Geophys.Res., 100, 11,209-11,217, 1995.

Mills, M. J., A. O. Langford, T. J. O'Leary, K. Arpag, H. L.

Lett., 21, 2471-2474, 1994.

Murphy, D. M., D. S. Thomson,and M. J. Mahoney, In situ measurementsof organics,meteoritic material, mercury, and other elements in aerosols at 5 to 19 kilometers, Science, 282, 1664-1669, 1998.

Nagatani, R. M., A. J. Miller, M. E. Gelman, and P. A. Newman,A comparisonof Arctic lower stratosphericwinter temperaturesfor 1988-89 with temperaturessince1964, Geophys.Res.Lett., 17, 333-336, 1990. Newman, P. A., Antarctic total ozone in 1958, Science,264, 543-546, 1994.

Newman,P. A., and W. J. Randel, Coherentozone-dynamical

312 ß Solomon: STRATOSPHERIC OZONE DEPLETION

37, 3 / REVIEWS OF GEOPHYSICS

tionsin anthropogenic ozonedepletionin polar regions,J. Geophys.Res., 101, 22,991-23,006, 1996. Newman, P. A., J. G. Gleason, R. D. McPeters, and R. S. Prather, M., More rapid polar ozone depletionthroughthe reactionof HOC1 with HC1 on polar statosphericclouds, Solarski,Anomalouslylow ozoneover the Arctic, Geophys. changesduring the Southern Hemisphere spring, 19791986,J. Geophys.Res.,93, 12,585-12,606,1988.

Res. Lett., 24, 2689-2692, 1997.

Noxon, J. F., StratosphericNO2 in the Antarcticwinter, Geophys.Res.Lett., 5, 1021-1022, 1978. Noxon, J. F., StratosphericNO2, 2, Global behavior,J. Geophys.Res.,84, 5067-5076, 1979.

Nature, 355, 534-537, 1992a.

Prather,M., Catastrophiclossof stratospheric ozonein dense volcanicclouds,J. Geophys.Res.,97, 10,187-10,191,1992b. Prather, M., M. M. Garcia, R. Suozzo, and D. Rind, Global

impact of the Antarctic ozone hole: Dynamicaldilution with a three-dimensional chemicaltransportmodel,J. Geophys.Res.,95, 3449-3471, 1990. V61ker, StratosphericC1ONO2 and HNO 3 profilesinside Prather, M., P. Midley, F. S. Rowland,and R. Stolarski,The ozone layer: The road not taken, Nature, 381, 551-554, the Arctic vortex from MIPAS-B limb emission spectra 1996. obtained during EASOE, Geophys.Res. Lett., 21, 12631266, 1994. Prinn, R. G., R. F. Weiss,B. R. Miller, J. Huang, F. N. Alyea, D. M. Cunnold, P. J. Fraser, D. E. Hartley, and P. G. OzoneTrendsPanel(OTP), Report of the internationalozone Simmonds,Atmospherictrends and lifetime of CH3CC13 trendspanel, WMO Rep. 18, World Meteorol. Organ.,Geneva, 1988. and globalOH concentrations, Science,269, 187-192, 1995. Parrish, A., R. L. de Zafra, P.M. Solomon, J. W. Barrett, and Proffitt, M. H., K. Aikin, J. J. Margitan, M. Loewenstein,J. R. Podolske,A. Weaver, K. R. Chart, H. Fast, and J. W. Elkins, E. R. Carlson, Chlorine oxide in the stratosphericozone Ozone loss inside the northern polar vortex during the layer: Ground-baseddetectionand measurement,Science, Oelhaf, H., T. von Clarmann, H. Fischer, F. Friedl-Vallon, C. Fritzsche, A. Linden, C. Piesch, M. Seefeldner, and W.

211, 1158-1161, 1981. Parrish, A., R. L. de Zafra, M. Jaramillo, B. Connor, P.M.

Solomon, and J. W. Barrett, Extremely low N20 concentrationsin the springtimestratosphere at McMurdo Station,

1991-1992 winter, Science,261, 1150-1154, 1993. Pueschel,R. F., et al., Condensednitrate, sulfate, and chloride

in Antarctic statosphericaerosols,J. Geophys.Res., 94, 11,271-11,284, 1989.

Pyle,J. A., G. Carver,J. L. Grenfell, J. A. Kettleborough,and D. J. Lary, Ozone lossin Antarctica:The implicationsfor Perrier, D.• A. Roth, and T. Kl•ipfel, Groundbasedmeasureglobal change,Philos. Trans.R. Soc.London, Ser. B, 338, ments of stratosphericOC10, NO2 and 03 at S0ndre 219-226, 1992. Str0mfjordin winter 1991/92,Geophys. Res.Lett., 21, 1367Pyle, J. A., M.P. Chipperfield,I. Kilbane-Dawe,A.M. Lee, 1370, 1994. R. M. Stimpfie,D. Kohn, W. Renger, and J. W. Waters, Peter,T., Microphysics and heterogeneous chemistryof polar Early modelling resultsfrom the SESAME and ASHOE stratospheric clouds,Annu. Rev.Phys.Chem.,48, 785-822, campaigns,Faraday Discuss.Chem. Soc., 100, 371-387, 1997. 1995. Pfeilsticker,K., and U. Platt, Airborne measurementsduring Randel, W. J., and J. B. Cobb, Coherentvariationsof monthly the Arctic StratosphericExperiment:Observationsof 03 mean total ozone and lower stratospherictemperature,J. and NO•., Geophys.Res.Lett., 21, 1375-1378, 1994. Geophys.Res.,99, 5433-5447, 1994. Pierce, R. B., W. L. Grose, J. M. Russell III, A. F. Tuck, R. Randel, W. J., and F. Wu, Coolingof the Arctic and Antarctic Swinbank,and A. O'Neill, Springdehydrationin the Antpolar stratospheres due to ozonedepletion,J. Climate,12, arctic stratosphericvortex observedby HALOE, J. Atmos. 1467-1479, 1999. Antarctica, Nature, 332, 53-55, 1988.

Sci., 51, 2931-2941, 1994. Pitari, G., and V. Rizi, An estimate of the chemical and

radiativeperturbationof stratospheric ozonefollowingthe eruption of Mt. Pinatubo,J. Atmos. Sci., 50, 3260-3276, 1993.

Pitari, G., G. Visconti, and M. Verdecchia, Global ozone

Randeniya,L. K., P. F. Vohralik, I. C. Plumb, and K. R. Ryan, HeterogeneousBrONO 2 hydrolysis:Effect on NO 2 columns and ozone at high latitudesin summer,J. Geophys. Res., 102, 23,543-23,557, 1997. Ravishankara, A. R., and D. R. Hanson, Differences in the

reactivity of type 1 PSC's dependingon their phase,J. Geophys.Res., 101, 3885-3890, 1996. 8075-8082, 1992. Ravishankara,A. R., et al., Lower stratosphericprocesses,in Plumb, R. A., and M. K. W. Ko, Interrelationshipsbetween ScientificAssessment of OzoneDepletion:1998, Rep. 44, pp. mixing ratios of long-lived stratosphericconstituents,J. 7.1-7.76, World Meteorol. Organ./U. N. Environ. ProGeophys.Res., 97, 10,145-10,156, 1992. gramme, Geneva, 1999. Pommereau,J-P., and J. Piquard,Observations of the vertical Reed, R. J., The role of vertical motions in ozone-weather distributionof stratospheric OC10, Geophys. Res.Lett., 21, relationships, J. Meteorol.,7, 263-267, 1950. 1231-1234, 1994a. Rex, M., et al., Prolonged stratosphericozone loss in the Pommereau,J-P., and J. Piquard,Ozone and nitrogendioxide 1995-96 Arctic winter, Nature, 389, 835-838, 1997. vertical distributionsby UV-visible solar occultationfrom Rex, M., et al., In situ measurementsof stratosphericozone balloons,Geophys.Res.Lett., 2l, 1227-1230, 1994b. depletionratesin the Arctic winter 1991/1992:A LagrangPoole,L. R., and M.P. McCormick,Polar stratospheric clouds ian approach,J. Geophys.Res.,103, 5843-5853, 1998. and the Antarctic ozone hole, J. Geophys.Res., 93, 8423- Ridley, B. A., M. McFarland, A. L. Schmeltekopf,M. H. 8430, 1988a. Proffitt, D. L. Albritton, R. H. Winkler, and T. L. Thomp-

depletionandthe Antarcticozonehole,J. Geophys. Res.,97,

Poole, L. R., and M.P. McCormick, Airborne lidar observa-

tionsof Arctic polar stratospheric clouds:Indicationsof two distinctgrowthstages,Geophys. Res.Lett., 15, 21-23, 1988b. Poole, L. R., M. T. Osborn, and W. H. Hunt, Lidar observa-

tionsof Arcticpolarstratospheric clouds,1988:Signatureof small, solid particlesabovethe frost point, Geophys.Res. Lett., 15, 867-870, 1988. Portmann, R. W., S. Solomon, R. R. Garcia, L. W. Thomason, L. R. Poole, and M.P. McCormick, Role of aerosol varia-

son, Seasonal differences in the vertical distributions of NO,

NO2, and 03 in the stratospherenear 50øN, J. Geophys. Res., 92, 11,919 -11,929, 1987. Rinsland, C. P., M. R. Gunson, M. C. Abrams, L. L. Lowes, R. Zander, E. Mahieu, A. Goldman, M. K. W. Ko, J. M.

Rodriguez, and N. D. Sze, Heterogeneousconversionof N205 to HNO 3 in the post-Mount Pinatubo eruption stratosphere, J. Geophys.Res.,99, 8213-8219, 1994. Rinsland, C. P., et al., ATMOS measurementsof H20 + 2CH 4

37, 3 / REVIEWSOF GEOPHYSICS

Solomon: STRAIOSPHERIC OZONE DEPLETION ß 313

and total reactivenitrogenin the November 1994 Antarctic stratosphere: Dehydrationand denitrificationin the vortex, Geophys.Res.Lett., 23, 2397-2400, 1996. Robinson,G. N., D. R. Worsnop,J. T. Jayne,C. E. Kolb, and P. Davidovits,Heterogeneousuptake of C1ONO2 and N20 5 by sulfuricacid solutions,J. Geophys.Res.,102, 3583-3601, 1997.

Robinson,G. N., D. R. Worsnop, J. T. Jayne, C. E. Kolb, E. Swartz,and P. Davidovits,Heterogeneousuptakeof HC1 by sulfuricacid solutions,J. Geophys.Res.,103, 25,371-25,381,

Miller, J. G. Keys, and A. L. Schmeltekopf,Visible and near-ultravioletspectroscopy at McMurdo Station,Antarctica, 9, Observationsof OC10 from April to October 1991, J. Geophys.Res., 98, 7219-7228, 1993. Santee, M. L., W. G. Read, J. W. Waters, L. Froidevaux, G. L. Manney, D. A. Flower, R. F. Jarnot, R. S. Harwood, and

G. E. Peckham, Interhemispheric differences in polar stratosphericHNO3, H20, C10, and 03, Science,267, 849852, 1995.

Santee, M. L., G. L. Manney, W. G. Read, L. Foidevaux,and 1998. J. W. Waters, Polar vortex conditionsduring the 1995-96 Arctic winter: MLS C10 and HNO3, Geophys.Res.Lett., 23, Roche, A. E., et al., Observations of lower-stratospheric 22, 3207-3210, 1996. C1ONO2,HNO3, and aerosolby the UARS, CLAES experiment betweenJanuary 1992 and April 1993,J. Atmos. Sci., Santee,M. L., G. L. Manney, L. Froidevaux,R. W. Zurek, and 51, 2877-2902, 1994. J. W. Waters, MLS observationsof C10 and HNO3 in the 1996-97 Arctic polar vortex, Geophys.Res.Lett., 24, 2713Rodriguez, J. M., et al., Nitrogen and chlorine speciesin the 2716, 1997. springAntarctic stratosphere:Comparisonof modelswith Airborne Antarctic Ozone Experiment observations,J. Santee,M. L., A. Tabazadeh,G. L. Manney, R. J. Salawitch,L. Froidevaux, W. G. Read, and J. W. Waters, UARS MicroGeophys.Res., 94, 16,683-16,703, 1989. wave Limb Sounder HNO 3 observations:Implicationsfor Rodriguez,J. M., M. K. W. Ko, and N. D. Sze, The role of Antarctic polar stratosphericclouds,J. Geophys.Res.,103, chlorinechemistryin Antarctic ozone loss:Implicationsof 13,285-13,313, 1998. new kinetic data, Geophys.Res.Lett., 17, 255-258, 1990. Rodriguez, J. M., M. K. W. Ko, and N. D. Sze, Role of Santee, M. L., G. L. Manney, L. Froidevaux,W. G. Read, and heterogeneousconversionof N205 on sulphateaerosolsin J. W. Waters, Six years of UARS microwavelimb sounder global ozone losses,Nature, 352, 134-137, 1991. HNO 3 observations: Seasonal,interhemispheric,and interRodriguez,J. M., M. K. W. Ko, N. D. Sze, C. W. Helsey,G. K. annual variations in the lower stratosphere,J. Geophys. Yue, and M.P. McCormick, Ozone responseto enhanced Res., in press,1999. heterogeneousprocessingafter the eruption of Mt. Pina- Schauffler,S. M., L. E. Heidt, W. H. Pollock,T. M. Gilpin, J. F. tubo, Geophys.Res.Lett., 21,209-212, 1994. Vedder, S. Solomon, R. A. Lueb, E. L. Atlas, MeasureRogaski,C. A., D. M. Golden, and L. R. Williams, Reactive ments of halogenatedorganiccompoundsnear the tropical uptake and hydration experimentson amorphouscarbon tropopause,Geophys.Res.Lett., 20, 2567-2570, 1993. treated with NO2, SO2, 03, HNO3, and H2SO4, Geophys. Schiller, C., A. Wahner, U. Platt, H-P. Dom, J. Callies, and Res. Lett., 24, 381-384, 1997. D. H. Ehhalt, Near UV atmosphericabsorptionmeasureRoscoe, H. K., A. E. Jones, and A.M. Lee, Midwinter start to ments of column abundances during Airborne Arctic Antarcticozonedepletion:Evidencefrom observations and StratosphericExpedition,January-February1989,2, OC10 models, Science, 278, 93-96, 1997. observations,Geophys.Res.Lett., 17, 501-504, 1990. Rosen, J. M., S. J. Oltmans, and W. F. Evans, Balloon borne Schmidt, U., and A. Khedim, In situ measurements of carbon observationsof PSCs,frost point, ozone and nitric acid in dioxide in the winter Arctic vortex and at mid-latitudes: An the north polar vortex, Geophys.Res. Lett., 16, 791-794, indicatorof the age of stratosphericair, Geophys.Res.Lett., 1989.

Rowland, F. S., and M. J. Molina, Chlorofluoromethanes in

18, 763-766, 1991.

Schoeberl,M. R., and D. L. Hartmann, The dynamicsof the the environment,Rev. Geophys.,13, 1-35, 1975. polar vortex, Science,251, 46-52, 1991. Rowland,F. S., J. E. Spencer,and M. J. Molina, Stratospheric Schoeberl,M. R., L. R. Lait, M. H. Proffitt, K. K. Kelly, M. formationand photolysisof chlorinenitrate,J. Phys.Chem., Loewenstein, J. R. Podolske, S. E. Strahan, P. A. Newman,

80, 2711-2713, 1976.

Russell,J. M., III, A. F. Tuck, L. L. Gordley, J. H. Park, S. R. Drayson,J. E. Harries, R. J. Cicerone, and P. J. Crutzen, HALOE Antarctic observationsin the springof 1991, Geophys.Res.Lett., 20, 719-722, 1993. Russell, J. M., III, M. Luo, R. J. Cicerone, and L. E. Deaver, Satellite

confirmation

of the dominance

of chlorofluorocar-

bons in the global stratosphericchlorine budget, Nature, 379, 526-529, 1996.

Salawitch,R. J., G. P. Gobbi, S.C. Wofsy, and M. B. McElroy, Denitrification in the Antarctic stratosphere,Nature, 339, 525-527, 1989. Salawitch, R. J., et al., Loss of ozone in the Arctic vortex for

the winter of 1989, Geophys.Res.Lett., 17, 561-564, 1990.

and J. E. Rosenfield,Stratosphericconstituenttrendsfrom ER-2 profile data, Geophys.Res.Lett., 17, 469-472, 1990. Schoeberl, M. R., L. R. Lait, P. A. Newman, and J. E. Rosen-

field, The structureof the polar vortex,J. Geophys.Res.,97, 7859-7882, 1992a.

Schoeberl,M. R., P. K. Bhartia, and J. R. Herman, Tropical ozonelossfollowingthe eruptionof Mt. Pinatubo,Geophys. Res. Lett., 20, 29-32, 1992b.

Schoeberl,M. R., A. R. Douglass,R. S. Stolarski,D. Toohey, L. Availone, J. G. Anderson,W. Brune, D. W. Fahey, and K. Kelly, The evolution of C10 and NO along air parcel trajectories,Geophys.Res.Lett., 20, 2511-2514, 1993a. Schoeberl, M. R., R. S. Stolarski, A. R. Douglass, P. A.

Salawitch, R. J., et al., Chemical loss of ozone in the Arctic

Newman, L. R. Lait, J. W. Waters, L. Froidevaux, and

polarvortexin the winter of 1991-1992,Science,261, 1146-

W. G. Ready, MLS C10 observationsand the Arctic polar vortex temperatures, Geophys.Res. Lett., 20, 2861-2864,

1149, 1993.

Salawitch, R. J., et al., The diurnal variation of hydrogen, nitrogen, and chlorine radicals:Implicationsfor the heterogeneousproduction of HNO2, Geophys.Res. Lett., 21, 2551-2554, 1994.

Sander,S. P., R. R. Friedl, and Y. L. Yung, Rate of formation of the C10 dimer in the polar stratosphere:Implicationsfor ozone loss, Science, 245, 1095-1098, 1989. Sanders, R. W., S. Solomon, J.P. Smith, L. Perliski, H. L.

1993b.

Schoeberl,M. R., M. Luo, and J. E. Rosenfield,An analysisof the Antarctic Halogen Occultation Experiment trace gas observations,J. Geophys.Res., 100, 5159-5172, 1995. Schoeberl,M. R., A. R. Douglass,S. R. Kawa, A. E. Dessler, P. A. Newman, R. S. Stolarski, A. E. Roche, J. W. Waters,

and J. M. RussellIII, Developmentof the Antarctic ozone hole, J. Geophys.Res., 101, 20,909-20,924, 1996.

314 ß Solomon:STRATOSPHERICOZONE DEPLETION Scott, R. F., Scott'sLast Expedition:The Journals,Carroll and Graf, New York, 1996. Sen, B., G. C. Toon, G. B. Osterman, J.-F. Blavier, J. J.

Margitan, R. J. Salawitch,and G. K. Yue, Measurementsof reactivenitrogenin the stratosphere, J. Geophys. Res.,103, 3571-3585, 1998. Shindell, D. T., J. M. Reeves, L. K. Emmons, and R. L. de

37, 3 / REVIEWSOF GEOPHYSICS

the homogenized total ozoneseriesof Aroso(Switzerland), 1926-1996, J. Geophys.Res.,103, 8389-8399, 1998b. Steele, H. M., P. Hamill, M.P. McCormick, and T. J. Swissler,

The formationof polar stratospheric clouds,J. Atmos.Sci., 40, 2055-2067, 1983. Stefanutti, L., M. Morandi, M. Del Guasta, S. Godin, and C.

David, Unusual PSC's observedby LIDAR in Antarctica, Zafra, Arctic chlorinemonoxideobservations duringspring Geophys.Res.Lett., 22, 2377-2380, 1995. 1993 over Thule, Greenland, and implicationsfor ozone Stimpfle,R. M., et al., The responseof C10 radical concendepletion,J. Geophys.Res.,99, 25,697-25,704,1994. trations to variations in NO 2 radical concentrationsin the Shindell, D. T., D. Rind, and P. Lonergan,Increasedpolar lower stratosphere,Geophys.Res. Lett., 21, 2543-2546, 1994. stratosphericozone lossesand delayed eventual recovery owingto increasinggreenhouse-gas concentrations, Nature, Stolarski,R. S., and R. J. Cicerone,Stratosphericchlorine:A 392, 589 -592, 1998. possiblesinkfor ozone,Can.J. Chem.,52, 1610-1615, 1974. Shine, K. P., On the modelled thermal responseof the Ant- Stolarski,R. S., and R. D. Rundel, Fluorine chemistryin the arctic stratosphereto a depletion of ozone, Geophys.Res. stratosphere,Geophys.Res.Lett., 2, 443-446, 1975. Lett., 13, 1331-1334, 1986. Stolarski,R. S., A. J. Krueger, M. R. Schoeberl,R. D. McSlusser,J. R., D. J. Fish, E. K. Strong,and R. L. Jones,Five Peters,P. A. Newman, and J. C. Alpert, Nimbus7 satellite years of NO 2 vertical column measurementsat Faraday measurementsof the springtimeAntarctic ozonedecrease, Nature, 322, 808-811, 1986. (65øS):Evidencefor the hydrolysis of BrONO2 on Pinatubo Stolarski, R. S., P. Bloomfield, R. D. McPeters, and J. R. aerosols,J. Geophys.Res.,102, 12,987-12,993,1997. Herman, Total ozone trends deduced from Nimbus 7 Solomon,S., The mysteryof the Antarctic ozone hole, Rev. Geophys.,26, 131-148, 1988. TOMS data, Geophys.Res.Lett., 18, 1015-1018, 1991. Solomon,S., and R. R. Garcia, On the distributionsof long- Sze, N. D., Stratosphericfluorine: A comparisonbetween livedtracersand chlorinespeciesin the middleatmosphere, theory and measurements, Geophys. Res.Lett., 5, 781-783, 1978. J. Geophys.Res.,89, 11,633-11,644,1984. Solomon, S., R. R. Garcia, F. S. Rowland, and D. J. Wuebbles, Sze,N. D., M. K. W. Ko, D. K. Weisenstein,J. M. Rodriguez, R. S. Stolarski, and M. R. Schoeberl, Antarctic ozone hole: On the depletion of Antarctic ozone,Nature, 321,755-758, 1986. Possibleimplicationsfor ozone trends in the Southern Solomon, S., G. H. Mount, R. W. Sanders, and A. L. SchmeltHemisphere,J. Geophys.Res.,94, 11,521-11,528,1989. ekopf, Visible spectroscopy at McMurdo station, Antarc- Tabazadeh,A., and O. B. Toon, Freezingbehaviorof stratotica, 2, Observationsof OC10, J. Geophys.Res.,92, 8329sphericsulfate aerosolsinferred from trajectorystudies, 8338, 1987. Geophys.Res.Lett., 22, 1725-1728, 1995. Solomon, S., G. H. Mount, R. W. Sanders, R. O. Jakoubek, Tabazadeh,A., and O. B. Toon, The presenceof metastable and A. L. Schmeltekopf,Observationsof the nighttime HNO3/H20 solidphasesin the stratosphereinferred from abundance of OC10 in the winter stratosphereabove ER-2 data,J. Geophys. Res.,101, 9071-9078,1996. Thule, Greenland, Science,242, 550-555, 1988. Tabazadeh,A., and R. P. Turco, Stratospheric chlorineinjecSolomon,S., R. W. Sanders,R. R. Garcia, and J. G. Keys, tion by volcaniceruptions:HC1 scavenging and implications for ozone, Science,260, 1082-1086, 1993. Increasedchlorine dioxide over Antarctica causedby volcanic aerosolsfrom Mount Pinatubo, Nature, 363, 245-248,

Tabazadeh, A., R. P. Turco, K. Drdla, M. Z. Jacobson,and

1993.

O. B. Toon, A studyof type I polar stratosphericcloud formation, Geophys.Res.Lett., 21, 1619-1622, 1994. Tabazadeh,A., O. B. Toon, B. L. Gary, J. T. Bacmeister,and

Solomon, S., R. R. Garcia, and A. R. Ravishankara, On the

role of iodine in ozone depletion, J. Geophys.Res., 99, 20,491-20,499, 1994. Solomon, S., R. W. Portmann, R. R. Garcia, L. W. Thomason, L. R. Poole, and M.P. McCormick, The role of aerosol

variations in anthropogenicozone depletion at northern midlatitudes,J. Geophys.Res.,101, 6713-6727, 1996. Solomon, S., S. Borrmann, R. R. Garcia, R. Portmann, L. Thomason, L. R. Poole, D. Winker, and M.P. McCormick,

Heterogeneouschlorine chemistryin the tropopauseregion,J. Geophys.Res.,102, 21,411-21,429,1997.

M. R. Schoeberl, Observational constraints on the forma-

tion of type Ia polar stratosphericclouds,Geophys.Res. Lett., 23, 2109-2112, 1996.

Thomason,L. W., L. R. Poole, and T. Deshler, A global climatologyof stratosphericaerosol surface area density deducedfrom StratosphericAerosol and Gas Experiment II measurements:1984-1994, J. Geophys.Res.,102, 89678976, 1997.

Thompson,D. W. J., and J. M. Wallace,The Arctic oscillation Solomon, S., R. W. Portmann, R. R. Garcia, W. Randel, F. signaturein the wintertime geopotentialheight and temWu, R. Nagatani, J. Gleason,L. Thomason,L. R. Poole, perature fields,Geophys.Res.Lett., 25, 1297-1300, 1998. and M.P. McCormick, Ozone depletion at mid-latitudes: Tie, X. X., and G. Brasseur,The responseof stratospheric Couplingof volcanicaerosolsand temperaturevariabilityto ozone to volcanic eruptions: Sensitivityto atmospheric anthropogenicchlorine,Geophys.Res.Lett., 25, 1871-1874, chlorineloading,Geophys.Res.Lett., 22, 3035-3038, 1995. 1998. Tie, X. X., G. P. Brasseur,B. Briegleb, and C. Granier, Two-dimensional simulation of Pinatubo aerosol and its Sornorjai,G. A., Introductionto SurfaceChemistryand Catalysis,pp. 1-667, John Wiley, New York, 1994. effect on stratosphericozone,J. Geophys. Res.,99, 20,54520,562, 1994. Stachnik,R. A., J. C. Hardy, J. A. Tarsala,J. W. Waters, and N. R. Erickson, Submillimeterwave heterogynemeasure- Tie, X. X., C. Granier, W. Randel, and G. P. Brasseur,Effects ments of stratosphericC10, HC1, 03, and HO2: First reof interannualvariation of temperatureon heterogeneous sults,Geophys.Res.Lett., 19, 1931-1934, 1992. reactionsand stratosphericozone, J. Geophys.Res., 102, Staehelin, J., A. Renaud, J. Bader, R. McPeters, P. Viatte, B.

Hoegger, V. Bugnion, M. Giroud, and H. Schill, Total ozone seriesat Arosa (Switzerland):Homogenizationand data comparison,J. Geophys.Res.,103, 5827-5841, 1998a. Staehelin,J., R. Kegel, and N. R. P. Harris, Trend analysisof

23,519-23,527, 1997.

Tolbert, M. A., Sulfateaerosolsand polar stratospheric cloud formation, Science,264, 527-528, 1994. Tolbert, M. A., Polar cloudsand sulfate aerosols,Science,272, 1597, 1996.

37, 3 / REVIEWSOF GEOPHYSICS Tolbert, M. A., M. J. Rossi, R. Malhotra, and D. M. Golden,

Reaction of chlorine nitrate with hydrogenchloride and water at Antarctic stratospherictemperatures,Science,238,

Solomon: STRATOSPHERICOZONE DEPLETION ß 315

Van der Leun, J., X. Tang, and M. Tevini, Environmental effectsof ozonedepletion:1994assessment, Ambio, 14, 138, 1995.

1258-1260, 1987.

Van Doren, J. M., A. A. Viggano, and R. A. Morris, Rate enhancementof the reaction of HC1 with C1ONO2 by ions: Tolbert, M. A., M. J. Rossi,and D. M. Golden, Heterogeneous interactionsof chlorinenitrate hydrogenchlorideand nitric Implicationsfor the mechanisms of stratospherically imporacid with sulfuric acid surfacesat stratospherictemperatant heterogeneous reactions,J.Am. Chem.Soc.,116, 69576958, 1994. tures, Geophys.Res.Lett., 15, 847-850, 1988. Toohey, D. W., A critical review of stratosphericchemistry Van Roozendael,M., C. Fayt, D. Bols•e, P. C. Simon, M. Gil, researchin the U.S.: 1991-1994, U.S. Natl. Rep. Int. Union M. Yela, and J. Cacho, Ground-basedstratosphericNO 2 Geod. Geophys.1991-1994, Rev. Geophys.,33, 759-773, monitoringat Keflavik (Iceland) duringEASOE, Geophys. 1995.

Res. Lett., 21, 1379-1382, 1994.

Toohey,D. W., J. G. Anderson,W. H. Brune, and K. R. Chan, Viggiano, A. A., R. A. Morris, K. Gollinger, and F. Arnold, In situ measurementsof BrO in the Arctic stratosphere, Ozone destructionby chlorine:The impracticalityof mitiGeophys.Res.Lett., 17, 513-516, 1990. gationthroughion chemistry,Science,267, 82-84, 1995. Toohey, D. W., L. M. Availone, L. R. Lait, P. A. Newman, Voemel, H., S. J. Oltmans, D. J. Hofmann, T. Deshler, and J. M. Rosen, The evolution of the dehydrationin the M. R. Schoeberl,D. W. Fahey,E. L. Woodbridge,and J. G. Anderson, The seasonal evolution of reactive chlorine in Antarctic stratospheric vortex, J. Geophys. Res., 100, 13,919-13,926, 1995. the Northern Hemispherestratosphere,Science,261, 11341136, 1993.

Toon, G. C., C. B. Farmer, L. L. Lowes,P. W. Schaper,J.-F. Blavier, and R. H. Norton, Infrared aircraft measurements

von Clarmann, T., H. Fischer, F. Friedl-Vallon, A. Linden, H. Oelhaf, C. Piesch, M. Seefeldner, and W. V61ker, Retrieval

of stratosphericcompositionover Antarctica during September 1987,J. Geophys.Res., 94, 16,571-16,596, 1989.

of stratospheric03, HNO3 and C1ONO2 profilesfrom 1992 MIPAS-B limb emissionspectra:Method, results,and error analysis,J. Geophys.Res.,98, 20,495-20,506,1993.

Toon, G. C., J.-F. Blavier, and J. T. Szeto, Latitude variations

Von der Gaathen, P., et al., Observational evidence for chem-

of stratospherictrace gases,Geophys.Res.Lett., 21, 2599-

ical ozone depletion over the Arctic in winter 1991-2,

Nature, 375, 131-134, 1995. Toon, O. B., and M. A. Tolbert, Spectroscopicevidence Wahner, A., R. O. Jakoubek, G. H. Mount, A. R. Ravishan2602, 1994.

againstnitric acid trihydratein polar stratosphericclouds, Nature, 375, 218-221, 1995. Toon, O. B., P. Hamill, R. P. Turco, and J. Pinto, Condensa-

kara, andA. L. Schmeltekopf,Remotesensingobservations of nighttime OC10 column during the Airborne Arctic Ozone Experiment,September8, 1987,J. Geophys. Res.,94,

11,405-11,411, 1989. tion of HNO 3 and HC1 in the winter polar stratosphere, Wahner, A., J. Callies, H.-P. Dorn, U. Platt, and C. Schiller, Geophys.Res. Lett., 13, 1284-1287, 1986. Toon, O. B., E. V. Browell, S. Kinne, and J. Jordan, An Near UV atmosphericabsorptionmeasurementsof column analysisof lidar observationsof polar stratosphericclouds, abundancesduring Airborne Arctic StratosphericExpediGeophys.Res.Lett., 17, 393-396, 1990. tions, January-February1989, 1, Technique and NO 2 observations,Geophys.Res.Lett., 17, 497-500, 1990. Toumi, R., S. Bekki, and R. Cox, A model studyof ATMOS observationsand the heterogeneousloss of N20 5 by the Wallace, L., and W. Livingston,The effect of the Pinatubo sulphateaerosollayer,J. Atmos. Chem.,16, 135-144, 1993. cloud on hydrogenchloride and hydrogenfluoride, GeoTraub, W. A., K. W. Jucks,D. G. Johnson, and K. V. Chance, phys.Res.Lett., 19, 1209-1212, 1992. Chemical change in the Arctic vortex during AASE II, Wamsley,P. R., et al., Distributionof halon-1211in the upper Geophys.Res. Lett., 21, 2595-2598, 1994. troposphereand lower stratosphereand the 1994 total Trenberth,K. E., and J. G. Olson,Temperaturetrendsat the bromine budget,J. Geophys.Res.,103, 1513-1526, 1998. South Pole and McMurdo Sound, J. Clim., 2, 1196-1206, Waters, J. W., J. C. Hardy, R. F. Jarnot, and H. M. Pickett, 1989. Chlorine monoxideradical, ozone,and hydrogenperoxide: Tsias, A., A. J. Prenni, K. S. Carslaw, T. P. Onasch, B. P. Luo, Stratosphericmeasurementsby microwavelimb sounding, Science,214, 61-64, 1981. M. A. Tolbert, andT. Peter,Freezingof polar stratospheric clouds in orographicallyinduced strong warming events, Waters, J. W., L. Froidevaux,W. G. Read, G. L. Manney, L. S. Elson, D. A. Flower, R. F. Jarnot, and R. S. Harwood, Geophys.Res.Lett., 24, 2303-2306, 1997. Tuck, A. F., Synopticand chemicalevolutionof the Antarctic StratosphericC10 and ozone from the Microwave Limb vortexin late winter and early spring1987,J. Geophys.Res., Sounder on the Upper Atmosphere Research Satellite, 94, 11,687-11,737, 1989.

Tuck, A. F., and M. H. Proffitt, Commenton "On the magnitude of transport out of the Antarctic polar vortex" by W. M. F. Wauben et al., J. Geophys.Res., 102, 28,21528,218, 1997.

Nature, 362, 597-602, 1993.

Waters,J. W., G. L. Manney, W. G. Read, L. Froidevaux,D. A. Flower, and R. F. Jarnot, UARS MLS observationsof lower

stratosphericC10 in the 1992-1993 and 1993-94 Arctic winter vortices,Geophys.Res.Lett., 22, 823-826, 1995.

Tuck, A. F., et al., Polar stratosphericcloudprocessedair and Waters, J. W., et al., The UARS and EOS Microwave Limb Sounder (MLS) experiments,J. Atm. Sci., 56, 194-218, potential vorticity in the Northern Hemisphere lower 1999. stratosphereat midlatitudesduringwinter,J. Geophys. Res., 97, 7883-7904, 1992. Wauben,W. M. F., R. Bintanja,P. F. J. van Velthoven,and H. Tuck, A. F., et al., The Brewer-Dobsoncirculationin the light Kelder, On the magnitudeof transportout of the Antarctic of high altitude in-situaircraft observations, Q. J. R. Meteopolar vortex,J. Geophys.Res., 102, 1229-1238, 1997. rol. Soc., 123, 1-69, 1997. Waugh,D. W., et al., Transportout of the lower stratospheric Tung, K. K., M. K. W. Ko, J. M. Rodriguez,and N. D. Sze,Are Arctic vortexby Rossbywavebreaking,J. Geophys. Res.,99, 1071-1088, 1994. Antarctic ozone variationsa manifestationof dynamicsor chemistry?,Nature, 333, 811-814, 1986. Waugh, D. W., et al., Mixing of polar vortex air into middle latitudes as revealed by tracer-tracer scatterplots,J. GeoTurco,R. P., and P. Hamill, Supercooledsulfuricaciddroplets: phys.Res., 102, 13,119-13,134, 1997. Perturbedstratospheric chemistryin earlywinter,Bet. Bunsen Ges.Phys.Chem., 96, 323-334, 1992. Webster, C. R., et al., Chlorine chemistryon polar strato-

316 ß Solomon:STRATOSPHERIC OZONE DEPLETION sphericcloud particlesin the Arctic winter, Science,261, 1130-1133, 1993.

Webster, C. R., R. D. May, M. Allen, L. Jaeg16,and M.P. McCormick, Balloon profiles of stratosphericNO 2 and HNO3 for testingthe heterogeneous hydrolysis of N20 5 on sulfate aerosols,Geophys.Res.Lett., 21, 53-56, 1994.

37, 3 /REVIEWS OF GEOPHYSICS Worsnop,D. R., L. E. Fox, M. S. Zahniser, and S.C. Wofsy, Vapor pressuresof solidhydratesof nitric acid:Implications for polar stratosphericclouds,Science,259, 71-74, 1993.

Wuebbles, D. J., F. M. Luther, and J. E. Penner, Effect of

coupled anthropogenicperturbationson stratospheric

Webster, C. R., et al., Evolution of HC1 concentrationsin the

ozone,J. Geophys.Res.,88, 1444-1456, 1983. Yudin, V. A., M. A. Geller, B. V. Khattatov,A. R. Douglass, M. C. Cerniglia,J. W. Waters,L. S. Elson,A. E. Roche,and J. M. RussellIII, A UARS studyof lower stratospheric polar processing in the early stagesof northernand southand BrO, Science,266, 398-404, 1994. ern winters,J. Geophys.Res., 102, 19,137-19,148, 1997. Williams,W. J., J. J. Kosters,and D. G. Murcray,Nitric acid Yung, Y. L., J.P. Pinto, R. T. Watson, and S. P. Sander, column densities over Antarctica, J. Geophys.Res., 87, Atmosphericbromineandozoneperturbations in the lower 8976-8989, 1982. stratosphere,J. Atmos. Sci., 37, 339-353, 1980. Wilson, J. C., et al., In situ observationsof aerosol and chlorine Zander, R. E., et al., The 1994northernmidlatitudebudgetof monoxideafter the 1991 eruptionof Mount Pinatubo:Efstratosphericchlorinederivedfrom ATMOS/ATLAS-3 obfect of reactions on sulfate aerosol, Science,261, 1140servations,Geophys.Res.Lett., 23, 2357-2360, 1996. 1143, 1993. Zerefos,C. S., K. Tourpali,andA. F. Bais,Further studieson Wofsy, S.C., Temporal and latitudinalvariationsof stratopossiblevolcanicsignalto the ozonelayer,J. Geophys. Res., 99, 25,741-25,746, 1994. spherictrace gases:A critical comparisonbetweentheory and experiment,J. Geophys.Res.,83, 364-378, 1978. Zerefos, D. S., K. Tourpali, B. R. Bojkov, D. S. Balis, B. Wofsy,S.C., M. B. McElroy,andY. L. Yung,The chemistryof Rognerund,and I. S. A. Isaksen,Solar activity-totalcolumn atmospheric bromine,Geophys. Res.Lett.,2, 215-218, 1975. ozone relationships:Observationsand model studieswith World MeteorologicalOrganization(WMO), Atmospheric heterogeneouschemistry,J. Geophys.Res.,102, 1561-1569,

lower stratosphere from 1991 to 1996 followingthe eruption of Mt. Pinatubo,Geophys. Res.Lett., 25, 995-998, 1998. Wennberg,P. O., et al., Removalof stratospheric 0 3 by radicals:In situ measurementsof OH, HO2, NO, NO2, C10,

ozone, Rep. 16, Geneva, 1985. World MeteorologicalOrganization/UnitedNations Environment Programme(WMO/UNEP), ScientificAssessment of OzoneDepletion:1991, Rep. 25, Geneva, 1991. World MeteorologicalOrganization/UnitedNations Environment Programme(WMO/UNEP), ScientificAssessment of OzoneDepletion:1994, Rep. 37, Geneva, 1994. World MeteorologicalOrganization/UnitedNations Environment Programme(WMO/UNEP), Scientific Assessment of OzoneDepletion:1998, Rep. 44, in press,Geneva,1999.

1997.

Zurek, R. W., G. L. Manney,A. J. Miller, M. E. Gelman, and R. M. Nagatani,Interannualvariabilityof the north polar vortexin the lower stratosphereduringthe UARS mission, Geophys.Res.Lett., 23, 289-292, 1996.

S. Solomon,AeronomyLaboratory,NOAA R/E/AL8,National Oceanicand Atmospheric Administration, 325 South Broadway,Boulder,CO 80303. ([email protected])