1 Ozonejuno/HsuPrather_JGR2008.pdfHSU AND PRATHER: STRATOSPHERIC VARIABILITY AND TROPOSPHERIC OZONE...

JOURNAL OF GEOPHYSICAL RESEARCH, VOL. ???, NO. , PAGES 1–28,

Stratospheric Variability and Tropospheric1

Ozone2

Juno Hsu and Michael J. Prather

Juno Hsu, Earth System Science, University of California, Irvine. 92697. Email: [email protected]

1Earth System Science, University of

California, Irvine, California, USA

D R A F T September 3, 2008, 12:07pm D R A F T

2 HSU AND PRATHER: STRATOSPHERIC VARIABILITY AND TROPOSPHERIC OZONE

Abstract.

Changes in the stratosphere-troposphere exchange (STE) of ozone are ex-

pected to alter the tropospheric ozone abundance, both over the last decades

of stratospheric ozone depletion and the coming century of climate change.

Combining an updated linearized stratospheric ozone chemistry (Linoz v2)

with parameterized PSCs and a five-year sequence of EC meteorology we ex-

amine the variations in STE O3 flux and how they perturb tropospheric O3.

Our best estimate for the current STE ozone flux is 290 Tg/yr in the NH

and 225 Tg/yr in the SH with interannual variability over years 2001-2005

of ± 25 Tg/year and ± 30 Tg/year, respectively. The STE flux alone drives

a large seasonal change in the tropospheric ozone burden with NH mid-latitude

peak-to-peak changes of about 8 DU, mimicking summertime photochem-

ical production but with half the amplitude. The model matches the quasi-

biennial oscillation (QBO) in column ozone, and the STE shows negative anoma-

lies over the mid-latitudes during the easterly phases of the QBO and vice

versa. The QBO-induced circulations over mid-latitudes during the easterly

phase create conditions that reduce STE. The tropospheric burden of this

O3 of stratospheric origin is indeed linear with STE. When the observed col-

umn ozone depletion from 1979 to 2004 is modeled with Linoz v2, we pre-

dict STE reductions of at most 10 % in NH, corresponding to about 1 ppb

decrease in hemispheric tropospheric O3.


HSU AND PRATHER: STRATOSPHERIC VARIABILITY AND TROPOSPHERIC OZONE 3

1. Introduction

Scientific e!orts to understand the trends and variations in ozone observed over the past3

few decades has demonstrated the role of both photochemical and meteorological factors4

in driving stratospheric ozone change (e.g., Randel and Wu, 2007; Stolarski et al., 2006,5

Salawitch et al., 2005). It has been proposed that these stratospheric changes have altered6

the tropospheric ozone burden over the past few decades (Fusco and Logan, 2003) and7

will continue to a!ect it in the future (Sudo et al., 2003). This paper presents a series of8

highly constrained modeling experiments that capture the observed trends and variations9

in stratospheric ozone and diagnose the corresponding changes in the stratosphere-to-10

troposphere flux of ozone. We are thus able to better understand the seasonal, interannual,11

and decadal trends in tropospheric ozone and the oxidative capacity of the atmosphere12

that are driven by the stratosphere.13

The coupling of stratospheric and tropospheric ozone with chemistry models or with14

chemistry-climate models is occurring across the community (Eyring et al., 2005). These15

full models include a nearly complete set of chemical species and reactions that a!ect16

ozone, but are costly to run, and are often di"cult to diagnose as to the causative fac-17

tors of variability. We approach the problem with a simplified chemical model that is18

focused on simulating the stratosphere-to-troposphere exchange (STE) of ozone: a lin-19

earized ozone chemistry (Linoz version 1: McLinden et al., 2000) combined with unique20

transport diagnostics that quantify the STE flux as a function of time and place (Hsu21

et al., 2005). The Linoz model is revised (Section 2) to use an updated climatology22

for the background stratospheric composition and current photochemical data (IUPAC,23

2004; Sander et al., 2006). Stratospheric ozone simulated with the new Linoz version 224



driven by Oslo ECMWF-IFS meteorology data is tested against observed ozone climatol-25

ogy (Section 3). The seasonal and interannual relationship between stratospheric ozone,26

STE flux, and tropospheric ozone is derived for a continuous sequence of meteorological27

fields from January 2000 to December 2005 (Section 4). The extent of the halogen-driven28

ozone STE decrease since 1979 is derived (Section 5), and we discuss the overall role of29

the stratosphere in driving tropospheric ozone change (Section 6).30

We find that stratosphere alone produces a peak-to-peak seasonal variation in tropo-31

spheric column ozone of about 8 DU at northern mid-latitudes that mimics tropospheric32

photo-chemistry. For longer time scales, the Quasi-Biennial Oscillation (QBO) signature33

in total column ozone is roughly matched to the observed and the QBO signals in ozone34

STE flux have maximum amplitudes in midlatitudes that are opposite in phase to its35

midlatitude QBO signal in total ozone. The observed, post-1979 ozone depletion for the36

NH can be best simulated in our chemistry-transport model (CTM) with a 4K higher37

threshold activation temperature than the typical 195K threshold for PSC formation.38

Enhanced background bromine levels are found to have negligible e!ect on ozone deple-39

tion but our PSC chemistry is parameterized for fixed bromine and so only our gas-phase40

chemistry responds to enhanced Bry. The maximum simulated decrease in the STE flux41

for post-1979 ozone depletion is about 10% in the northern hemisphere (NH) and 22% in42

the southern hemisphere (SH). Furthermore, the latitude-season pattern of STE decrease43

due to ozone depletion is distinctly di!erent from the change in total column ozone.44

2. A linearized stratospheric Ozone Chemistry – Linoz version 2

The Linoz tables are derived using the photochemistry box model of Prather (1992) with45

background atmospheric composition specified as a monthly, latitude-height climatology46



from observations. The local net photochemical production (P-L) is represented by a47

first-order Taylor Series expansion using only three independent variables: local ozone48

mixing ratio (f), temperature (T), and overhead ozone column (c),49

df

dt= (P ! L)o +

!(P ! L)

!f

!!!!o

(f ! f o) +!(P ! L)

!T

!!!!o

(T ! T o) +!(P ! L)

!c

!!!!o

(c ! co). (1)

The photochemical tendency for the climatology is denoted by (P-L)o ,and the clima-50

tology values for the independent variables are denoted by fo , co , and To , respectively.51

Including these four climatology values and the three partial derivatives, Linoz is defined52

by seven tables. Each table is specified by 216 atmospheric profiles: 12 months by 1853

latitudes (85S to 85N). For each profile, the net production for the climatology, (P-L)o ,54

and the three derivatives are evaluated at every 2 km in pressure altitude from z* = 1055

to 58 km (z* = 16 km log10 (1000/p)). These tables are automatically remapped onto56

any CTM grid with the mean vertical properties for each CTM level calculated as the57

mass-weighted average of the interpolated Linoz profiles.58

We adopt the ozone climatology compiled by McPeters et al. (2007), which has im-59

proved profiles over the tropics and the SH as compared with Linoz v1. The temperature60

climatology is unchanged (Nagatani and Rosenfield, 1993). The remaining chemical com-61

position is specified as a climatology scaled to the tropospheric abundance of the long-lived62

source gases (i.e., N2O, CH4, and the halocarbons) so that it can be changed to reflect63

a changing atmosphere. This includes climatologies for three chemical families (NOy =64

NO + NO2 + HNO3 + ...; Cly = ClO + HOCl + ClONO2 + HCl + ...; Bry = Br +65

BrO + BrONO2 + ...). We use N2O as the primary measure of stratospheric composition66

and tracer-tracer relations to define the other trace gases. A monthly, latitude-by-height67



N2O climatology above 22 km is based on CLAES satellite measurements (Oct 1991- May68

1993, Randel et al, 1994) and below 22 km is constructed from the compact correlation69

with O3 from the NASA ER-2 in-situ measurements in the lower stratosphere (Strahan70

et al., 1999). Minor smoothing is applied to the transition region. The CH4 and NOy71

distributions are obtained using the polynomial fit with respect to N2O from ATMOS mea-72

surements (Michaelsen et al., 1998a; 1998b). The Cly climatology assumes conservation73

of halogens, thus increasing in the stratosphere as the organic source gases (e.g., CFCs,74

CH3CCl3, CCl4) are photochemically destroyed (Woodbridge et al., 1995). The Bry cli-75

matology likewise assumes increasing Bry as the tropospheric bromine source gases (e.g.,76

CH3Br, CF2BrCl, CF3Br) are destroyed (Wamsley et al., 1998). For Bry, we consider a77

sensitivity case where the tropopause value is increased to 6 ppt to include the relatively78

large amounts of inorganic bromine (Bry) that may cross the tropopause (Salawitch et79

al., 2005). Both families are keyed to the N2O distribution. Water vapor adopts a lower80

boundary fixed at 3.65 ppm (Nassar et al., 2005) and conserves of total hydrogen (H2O +81

2xCH4). The tracer-tracer correlations are applied with N2O scaled to the year of their82

observed correlations. Normalized distribution patterns for N2O, CH4, H2O, NOy and83

Bry in January are shown in Fig. 1. Compared to v1, these Linoz v2 climatologies for84

background stratospheric composition more accurately match observations.85

Ozone and temperature climatologies, to first order, determine stratospheric photolysis86

rates. We adopt a surface reflectivity of 0.3 as an average cloud cover. The photochem-87

ical box model is initialized with an approximate balance of species within each of the88

chemical families and integrated for 30 days to reach an approximate, diurnally repeat-89

ing steady state, whereby the initialization of species within the families is, for the most90



part, forgotten. During this integration, the abundance of ozone and long-lived gases91

are fixed, and the chemical families are conserved. The net ozone production and three92

partial derivatives are evaluated numerically by perturbing the local ozone by +5%, the93

column ozone by +5%, and the temperature by +4 K. Regarding the nonlinearity in the94

derivatives, see Fig. 3 of McLinden et al (2000).95

For Linoz version 2, the photochemistry has been updated from the 1997 -vintage version96

1 to current rate coe"cients (Sander et al., 2006) and cross sections (IUPAC, 2004). New97

solar fluxes are taken from the average solar irradiance reference spectra derived by the98

SUSIM team for two di!erent levels of of solar activity (Thuillier et al., 2004). Compared99

to Linoz v1, these are 15-20 % larger at wavelengths 177-200 nm, 5-10% larger at 200-100

300 nm, but relatively unchanged long-ward of 300 nm. Other notable updates a!ecting101

photolysis rates include the quantum yield of O(1D) from O3 photolysis and the NO2 cross102

sections.103

As an example of how the stratospheric chemistry model has evolved since v1, we follow104

the chemistry updates using a standard ATMOS profile (May 31, 30N) from previous105

models and measurements studies (Prather and Remsberg, 1993). The height profiles of106

net ozone production (P-L) and its derivatives with respect to ozone, temperature, and107

column ozone are shown in Fig. 2. A sequence of six model calculations are shown with108

successive updates tracking the change in chemistry from Linoz v1 to v2. Values generated109

with the JPL-1997 kinetics rates and cross sections and with the old solar flux data used110

for generating Linoz v1 are shown for comparison (JPL97-S3). Updating the quantum111

yields and cross sections only (JPL97-S2) has no e!ect on the three derivatives and a112

barely noticeable e!ect on net production, i.e., a small increase near 40 km. Updating113



the solar fluxes in addition (JPL97-S1) also has no e!ect on the derivatives but causes a114

large increase in net production throughout the stratosphere above 25 km. The update115

to JPL 2000 kinetics (JPL00-S1) causes a notable decrease in the temperature derivative116

between 34 and 48 km with an increase in net production from 34 to 44 km and a decrease117

below 34 km. Updating the kinetics to JPL 2002 (JPL02-S1) and JPL 2006 (JPL06-S1)118

has minor e!ects, with the latter causing a small decrease in net production about 30 km.119

The largest and most extensive changes in the chemical model occur in the net ozone120

production and not in the derivatives. The largest change in updating from JPL 1997121

to JPL 2000 is caused by the addition of a new branch for the reaction, OH + ClO "122

HCl +O2. This new pathway weakens the Cly-catalyzed ozone loss and thus results in123

an increase in net production peaking around 38 km. The other major change with JPL124

2000 kinetics was a stronger NOy-catalyzed ozone loss from the increased kinetic rate for125

the reaction, NO2 + O " NO + O2, and decreased kinetic rate for the reaction, NO2 +126

OH " HNO3. Changes to chemical reaction rates after JPL 2000 have relatively minor127

e!ects on the ozone chemistry (outside of PSC conditions).128

Linoz v1 considered only gas-phase photochemistry and did not include chlorine acti-129

vation by Polar Stratospheric Clouds (PSCs). Thus, in v1 there was no Antarctic ozone130

hole and no enhanced Arctic loss during cold winters. In v2, we incorporate the PSC131

parameterization scheme of Cariolle et al. (1990) when the temperature falls below 195132

K and the sun is above the horizon at stratospheric altitudes. The O3 loss scales as the133

squared stratospheric chlorine loading (normalized by the 1980 level threshold). In this134

formulation PSC activation invokes a rapid e-fold of O3 based on a photochemical model,135

but only when the temperature stays below the PSC threshold. It does not consider that136



the activated chlorine continues to destroy ozone for several days after encountering a137

PSC (Schoeberl et al., 1993). Recently, Cariolle and Teyssedre (2007) added a cold-tracer138

to account for this e!ect. Their new parameterization, which is not used here, will be139

more important in the Arctic where PSCs are not sustained throughout the winter. In140

view of this process and the evidence of Cly activation on ternary aerosols at warmer tem-141

peratures (Thornton et al., 2005), we test another version using a higher PSC-activation142

temperature of 199K.143

Linoz chemical tendencies are applied only in the stratosphere, defined here as to CTM144

grid points for which the O3 abundance is greater than 100 ppb. These simulations do145

not include realistic tropospheric ozone chemistry, but instead invoke a parameterized146

sink that restores O3 to 20 ppb in the lowest 600 m of the troposphere with an e-folding147

time-scale of 2 days (Hsu et al., 2005). The choice of 20 ppb was made to imitate a more148

realistic chemistry and produce reasonable tropospheric column O3. This tropospheric149

pseudo-chemistry is uniform, and thus variations in tropospheric O3 calculated here are150

driven entirely by the STE flux. When combining Linoz with a full tropospheric chemistry151

model, we simulate a separate Linoz tracer (O3s) and use it every time step in each grid152

box to determine if the tropospheric chemistry is invoked (e.g., O3s <100 ppb) or if the153

Linoz net chemical tendencies are used (>100 ppb).154

Using the normalized, monthly 2-D climatologies for stratospheric composition, we cal-155

culate five sets of Linoz v2 tables (see Table 1). Linoz-1979 uses the 1979 mean abundances156

from REF 1 of Eyring et al. (2005) and represents a stratosphere prior to significant157

ozone depletion. With Cly levels below the chlorine-loading threshold, PSC-induced loss158

is never invoked with Linoz-1979. Linoz-2004 uses year-2004 mean tropospheric abun-159



dances (WMO, 2006, Table 1-2) and generates an Antarctic ozone hole. A second pair of160

Linoz tables, -1979Br and -2004Br, assume a 6ppt greater background of Bry throughout161

the stratosphere (see Salawitch et al., 2005). We also use the Linoz-2004Br tables with162

a warmer PSC threshold of 199K and denote this case as Linoz-2004BrT. Note that the163

Linoz tables assume only gas phase chemistry plus some sulfate reactions using a SAGE164

climatology for the aerosol surface area. Thus ClO levels in the lower stratosphere are165

always low and the enhanced Bry does not notably enhance ozone loss.166

3. Evaluating Column Ozone with Linoz

To assess the impact of the updated Linoz v2 on stratospheric ozone, we repeat the167

Linoz v1 simulations of Hsu et al. (2005) with Linoz-2004 and the Oslo/EC meteorology168

for year 1997 derived from the European Centre for Medium-Range Weather Forecasts169

(ECMWF) Integrated Forecasting System (IFS) Cycle 23r4. Linoz v1 is known to be170

biased low in column O3 in the tropics and high in high latitudes (see Fig. 1 of Wild171

et al., 2003). With Linoz-2004 this bias is mostly eliminated: tropical ozone columns172

increase by 5-20% for all but the northern winter, and outside of the tropics ozone is173

reduced by similar percentages for all months except December. In terms of STE, if we174

run Linoz-2004 tables but turn o! the PSC parameterization, the O3 flux increases by 9%175

from 516 Tg/yr to 563 Tg/yr, with greater increases in the SH. The spatial and temporal176

STE patterns remain roughly the same. Inclusion of the parameterized PSC chemistry177

with Linoz-2004 reduces the STE fluxes globally by 10%, again with greater response in178

the SH. The total shift in STE flux from v1 to v2 (Linoz-2004) is +3% in the NH and179

-7% in the SH. The changes in the photochemical data and inclusion of a parameterized180

PSC loss have corrected the prominent biases in Linoz v1.181



Stratospheric O3 columns calculated with Linoz-2004 with 1997 ECMWF IFS Cycle23r4182

meteorological data are compared with those from observations (McPeters et al., 1997;183

2007) in Fig. 3. We show results for both with and without PSC parameterization. For184

both CTM and observations, the stratosphere is defined as where O3 abundances exceed185

100 ppb (10!7 moles per mole of dry air). The 2007 climatology is an improvement over186

the 1997 climatology, but the changes also reflect the inclusion of more recent years with187

greater ozone depletion, e.g., a deeper Antarctic ozone hole in September and greater188

Arctic loss in March. From 40S to 40N, the Linoz simulation is excellent, with no obvious189

biases and errors less than 25 DU. At high latitudes, the PSC parameterization improves190

the Linoz simulations, and the ozone hole is reasonably well matched.191

To study interannual variability we use continuous ECMWF-IFS T42L40 meteorolog-192

ical fields from years 2000 through 2005. Year 2000 data are extracted from ECMWF193

IFS Cycle 23r4 model whereas the rest are extracted from Cycle 29r2 model. We find194

Cycle 29r2 generates about 20% more STE flux than does version Cycle 23r4 (see below),195

and this di!erence is much greater than the interannual variability. Thus, year 2000 me-196

teorological data are only used to spin up the experiments to approach a steady state197

before continuing with the next five years from January 2001 through December 2005,198

which are analyzed here. This five-year monthly mean climatology of total O3 column,199

plus the interannual variability defined relative to the five-year mean, are compared with200

the recent corrected Earth Probe TOMS observations based on NOAA-16 SBUV/2 ozone201

records as shown in Fig. 4. Note that the missing data for December 2005 are replaced202

with those from December 2004 for convenience. The CTM simulation with Linoz-2004203

captures the general patterns of the observed seasonal cycle and the Antarctic ozone hole204



with its minimum below 190 DU. At mid and high northern latitudes, total O3 column205

is well simulated. In the tropics, the minimum (NDJF at 15N) are likewise matched,206

but the CTM has a spurious high (310 DU contour) in July at 10N and likewise with 270207

contour bulging equatorward to 10S in austral summer. Even worse, the circum-Antarctic208

maximum around 60S is consistently about 60 DU higher than observed. These anomalies209

do not appear in the previous publications with Linoz v1 using the 1997 and 2000-2001210

Cycle 23r4 meteorological data. Using Linoz-2004BrT reduces the total ozone error by211

20 DU confined to poleward of 60S and over the spring artic vortex. The spurious errors212

remain evident and large regardless of the chemistry used.213

Analysis of the monthly latitude-height ozone profiles from the CTM (not shown) reveals214

a deep sinking motion near the edge of the Antarctic polar vortex that persists through the215

seasons and a spurious downward shift of contours in the top model layers at 10N in July216

and 10S in January. We presume these errors stem from a poorly resolved stratosphere217

with a top lid in the middle stratosphere (2 hPa). To test this point, we acquired year218

2005 using IFS Cycle 29r2 but with much finer vertical resolution, T42L60, in which219

the whole stratosphere is resolved with layers at most 1.5 km thick from 15 to 0.5 hPa.220

Linoz-2004 with the T42L60 meteorological data corrects the worst errors seen with the221

T42L40 meteorological data as shown in Fig. 5, viz, the tropical bubble disappears and the222

circum-Antarctic high columns now are lower and closer to observations. Unfortunately,223

the T42L60 data was only available to us for year 2005, and so our analysis of interannual224

variability continues with the T42L40 data.225

Monthly anomalies in zonal-mean total O3 column for years 2001-2005 are shown in226

Fig. 4c and 4d (N.B. Contour intervals for the CTM simulation are 10 DU, but those227



for the TOMS data are only 5 DU.) In October 2002, the CTM matches the extremely228

high column anomaly over Antarctica, which was caused by a sudden warming event and229

the transport of ozone-rich air into the vortex (e.g. Simmons et al., 2005). In general230

the phases of alternating high and low anomalies are well captured by the CTM and a231

two-year QBO-like signal is evident. In spite of the coarsely resolved lower stratosphere,232

the ECMWF IFS 40-layer model produces a QBO with alternating descending easter-233

lies and westerlies in the stratosphere (not shown). The forecast model is re-initialized234

with observations every 24 hours and appears to generate a QBO pattern in the lower235

stratospheric transport. The magnitude of the modeled equatorial and SH interannual236

variability in O3 column, however, is often twice as large as observed.237

To understand the modeled interannual variability and its relation to the O3 STE flux,238

we isolate the QBO signal following the regression procedure of Randel and Wu (1996). A239

QBO time series is defined by determining the linear combination of the equatorial zonal240

wind at 20 and 40 hPa that best correlates the equatorial total O3 column interannual241

variability. This time series is then regressed against the time series of total O3 anomalies242

at all latitudes (Fig 4d). Fig. 6 shows (a) the modeled column O3 anomaly correlated243

with the QBO and (b) the residuals. The QBO signal in O3 column shows large positive244

(negative) equatorial ozone anomalies during equatorial westerlies (easterlies) as observed.245

The subtropical QBO signal is correctly out-of-phase with the equatorial signal. However,246

this signal is more confined to the subtropics than is observed (see Fig. 1 of Randel247

and Wu, 1996), and the observed 6-month phase lag between the maxima at subtropical248

and mid latitudes in the two hemispheres is absent. Also unlike earlier observations249

(e.g. Randel and Cobb 1994), the SH midlatitude QBO signal does not continue into250



to the spring Antartic polar region but directly changes sign before 60S. This points to251

the possibility that for the coarsely resolved stratosphere of the ECMWF L40 model,252

the interaction of the annual cycles and the QBO as well as the high-latitude planetary253

waves modulated by the QBO (See Baldwin et al., 2001) are completely missing or even254

mispresented. The residuals are about the same magnitude as the QBO signal and show a255

large-scale, low-frequency, coherent structure in the SH that is roughly out of phase with256

the equator.257

4. Stratosphere-to-Troposphere Fluxes and Tropospheric Ozone

Following Hsu et al. (2005), the ozone STE flux is calculated based on the mass balance258

of a latitude-by-longitude tropospheric ozone columns. In this study, the diagnostic is259

improved by further including the first moment of the horizontal ozone flux within the260

troposphere (see equation 1 in Hsu et al., 2005) when computing this term. As a result,261

the overall noise level such as the dipole structures within the Pacific jet stream noted in262

the previous study is diminished. The average seasonal cycle of STE O3 fluxes calculated263

from the CTM simulations from years 2001-2005 (contour lines in Fig. 7) has a similar264

pattern to that published in Hsu et al (2005) for years 1997 and 2000. Also shown is265

the average zonal-mean zonal wind at 200 hPa (shaded contours). The STE maximum in266

the NH occurs just poleward of the tropospheric zonal jet, peaks during late spring and267

early summer when the zonal jet weakens, and migrates with the subtropical jet up to268

45N. In the SH, the STE maximum stays around 30S, does not migrate poleward with269

the jet in summer, but does peak during austral spring when the jet weakens. However,270

the global STE flux is on average 20 % larger than the previous estimates despite the271

fact that as discussed in Section 3, the global STE flux should be slightly reduced using272



Linoz-2004 and including the PSC parameterization. The larger magnitude is mostly due273

to the excessive decent in the Antarctic circumpolar region with the Cyc29r2 L40 fields.274

Indeed, recalculating the STE flux using the ECMWF 2005 Cyc29r2 L60 data, we find275

that the L40 SH STE flux is about 25 % too large.276

The QBOs role in modulating the STE O3 flux is derived with the same method as for277

total O3 column. STE variability attributed to the QBO (Fig. 6c) is small: peak contour278

intervals in mid-latitudes are +0.20 g m!2 yr!1 (30S in 2004) as compared with a global279

average of about 1.2 g m!2 yr!1 (i.e., 610 Tg per year). These mid-latitude STE QBO280

signals are out of phase with the mid-latitude total ozone QBO, and, not surprisingly, there281

are no QBO signals over the tropics and high latitudes where STE fluxes are negligible.282

This pattern indicates that the induced sinking part of the overturning circulation in the283

subtropics during the easterly phase of QBO creates a dynamical condition that disfavors284

the mixing of the mid-latitude, lower stratospheric ozone into the troposphere. The STE285

residual field (Fig. 6d) has much larger amplitudes than the QBO signal. In SH both286

STE and column O3 residuals show the same low-frequency variability and are positively287

correlated with STE lagging by a few months. This relationship is evident for years 2001-288

2002 (negative, blue patch) and 2003-2004 (positive, red patch). In the NH, the STE289

residuals lack coherence and have smaller amplitudes.290

Influx of O3 from the stratospheric is a principal component of the tropospheric O3291

budget, the others being in situ photochemical production and loss and surface deposition.292

The amount of tropospheric O3 ozone that can be assigned a stratospheric origin, however,293

is not well defined. There is a wide range of reported STE fluxes and various accounting294

methods for tropospheric production and loss (e.g., Prather et al., 2001; Stevenson et al.,295



2006). In this study, we quantify the variability in tropospheric O3 caused by changes in296

STE by using a simplified, uniform chemistry in the troposphere, viz, O3 in the lowest297

600m is forced to 20 ppb. We diagnose tropospheric O3 column (adopting common usage of298

TCO for the tropospheric column ozone) hourly as the vertically integrated ozone burden299

for all CTM layers with abundances less than 100 ppb. Thus, with these simulations, the300

latitudinal and seasonal variations in TCO (color-filled contours in Fig. 8a) are driven301

by both STE flux (line contours in Fig. 8a) and the changing size of the troposphere.302

In the NH, the monthly zonal-mean TCO varies from 16 to 26 DU with lowest values303

in the tropics (where the troposphere is largest). In northern mid-latitudes, the seasonal304

peak-to-peak range is 8 DU, and although this tends to follow the troposphere mass (i.e.,305

the tropopause peaks in late summer) a large fraction appears to follow the STE flux.306

This large seasonality is driven without tropospheric chemistry. We expect the TCO to307

lag the STE by a month (i.e., the tropospheric lifetime of an STE perturbation).308

Comparing to the recently observed TCO derived from OMI and MLS measurements309

(Fig. 6 of Ziemke et al, 2006), our modelled TCO without tropospheric chemistry surpris-310

ingly matches the observed latitude-by-month pattern, much better than the comparison311

with the full chemistry CTM in Ziemke et al. Our TCO is, however, consistently biased312

low by about 8 DU over the tropics and wintertime mid-latitudes. Furthermore, we un-313

derestimate the magnitude of the buildup to maximum TCO seen at 40N in July and314

at 30S in November by 18 DU, indicating the importance of tropospheric photochemical315

net production that is not simulated here. Renormalizing the TCO pattern to the tro-316

pospheric mass shows the variation in mean tropospheric ozone abundance (ppb in Fig317

8b). Here, we see that peak abundances tend to follow the STE flux. Given that the318



tropospheric parameterization pushes to a uniform mixing ratio, this simulation provides319

a measure of the seasonality and amplitude of tropospheric O3 variability driven by STE.320

5. Ozone Depletion and STE fluxes

The reduction in STE O3 flux from the halogen-catalyzed depletion of stratospheric O3321

is evaluated with di!erent Linoz v2 models. We simulate pre-depletion O3 with Linoz-322

1979 using the meteorological data for years 2001-2005. For post-depletion, we use the323

same meteorological data and Linoz-2004. Additional experiments, Linoz-2004Br and324

Linoz-2004BrT are used to investigate the e!ect of enhanced bromine and higher PSC325

temperature threshold for post-ozone depletion. Because the additional bromine repre-326

sents a natural tropospheric source, we simulate the pre-depletion ozone, Linoz-1979Br in327

pair with the latter experiments. Note that because we have only meteorological data for328

recent years, this study cannot elucidate the role of changing transport in the observed329

depletion.330

All three Linoz-2004 variants calculate qualitatively similar patterns of column O3 de-331

pletion as observed in the merged satellite data based on TOMS/SBUV measurements332

from 1979-2000 (see Fig. 11 of Fioletov et al., 2002). The depth of overall ozone depletion333

and its seasonal di!erences, particularly in the NH, are largely underestimated unless the334

activation temperature of the PSC parameterization is raised from 195K to 199K. En-335

hanced bromine has only a small e!ect on further ozone depletion in these calculations336

compared to Salawitch et al. (2005). As noted before, we speculate that this could be the337

result of our PSC parameterized loss being independent of bromine level.338

Fig. 9 shows the latitude-by-month changes in total O3 column (DU) and STE O3339

flux (g m!2 yr!1) between Linoz-1979Br and Linoz-2004BrT for the five year average of340



2001-2005 meteorology. We can compare the ozone depletion in Fig. 9a directly to the341

vertically integrated SAGE/sonde stratospheric ozone data over 1979-2005 (Fig. 10 of342

Randel and Wu, 2007). Overall, the simulations agree with observations, both in shape343

and magnitude. In the tropics, the di!erence of about 8 DU in total O3 column relative344

to the pre-ozone depletion is as large as seen in SAGE/sonde data (Fig. 10a of Randel345

and Wu, 2007), but disagrees with the equatorial total ozone change of about zero seen in346

the merged TOMS/SBUV data (Fig. 10b of Randel and Wu, 2007). In our simulation at347

T42L40, the top layer from 2 to 20 hPa does not accurately cover the diversity in upper348

stratospheric O3 chemistry whereby chlorine-driven depletion above 3 hPa results in more349

penetration of solar ultraviolet and hence more O3 production below. Other errors in our350

simulation include: missing the second maximum in NH ozone depletion during the fall;351

and simulating Antarctic ozone depletion to be about twice as large as the observed. The352

latter discrepancy could be reduced if there were some PSC-induced ozone loss already in353

1979, which is not modeled here.354

The latitude-month change in the STE ozone flux is quite di!erent than that in total355

ozone. For the NH STE change, the di!erence pattern follows roughly the seasonal pattern356

in Fig. 7 but with the maximum depletion in summer lagging the seasonal STE maximum357

by 2 months. The maximum change in the NH STE fluxes is about -0.3 g m!2 yr!1 (less358

than 10 % change) and is about twice that obtained when the PSC threshold is lowered359

to 195 K. The SH STE change pattern does not resemble its seasonal pattern. The360

maximum decrease, -0.7 g m!2 yr!1, occurs in the austral summer around 40S with a six361

month time-lag from the Antarctic ozone hole. Antarctic ozone, primarily in the isolated362



vortex, requires more time to propagate to the midlatitude STE exchange than its NH363

counterpart.364

Fig. 10 shows the seasonal changes in the ozone vertical profiles at NH mid latitudes365

for all three pairs of experiments. Decreases are most evident near 11-14 km in the lower366

stratosphere as the peak moves down in height from spring to summer. The annual367

average decrease in the lower stratosphere from Linoz-2004BrT pairs is about 11 %. This368

trend corresponds to about 8 % per decade for the 1979-1990 period (see Randel and Wu,369

2007) and is comparable with some estimates for a similar period ( e.g. Randel et al.,370

1999) but is much smaller than reported in Fusco and Logan (2003). With Linoz-2004Br371

pairs, the net ozone depletion is at most 5 %. The relatively uniform and weak decrease in372

the winter season is distinctively di!erent from the profiles of the other seasons and from373

observations. It might point to the importance of a trend in winter circulation (Hood and374

Soukharev, 2005) lacking in this study.375

6. Discussion and Conclusions

To separate changes in STE flux due to the ozone depletion from those due to natural376

variability, we use a regression model to fit the hemispheric monthly STE flux from the five-377

year sequence with the hemispheric mean, the seasonal harmonics and the monthly time378

series of the QBO and NAO (North Atlantic Oscillation) indices. The QBO accounts for 20379

% and the NAO for only 4 % of the NH interannual variability. The total NH interannual380

variability is about 25 Tg/yr (r.m.s), the same as the long-term change driven by ozone381

depletion (Linoz2004BrT-Linoz1979Br). Thus, as detection of a NH trend in column382

ozone is obscured by transport variability (e.g. Stolarski et al, 2006), so is any long-term383

change in STE flux. The SH STE flux shows quite di!erent modes of variability. For384



the SH regression model, we replace NAO index with the AAO (Antarctic Oscillation)385

index. The QBO nearly captures about 45 % of the interannual variability, while the386

AAO captures negligible variance and thus does not contribute to coherent structure of387

the residuals seen in Fig. 6d. The maximum SH STE change due to ozone depletion (70388

Tg/yr) is twice as large as the interannual variability (30 Tg/year r.m.s) over the years389

2001-2005.390

The impact of changes in the STE flux on tropospheric ozone is shown with the scatter391

plot of annual, hemispheric tropospheric ozone abundances versus the mean STE fluxes392

for all 25 years of Linoz v2 simulations (Fig. 11). In both hemispheres there is a distinct393

linear correlation with slightly di!erent slopes: 0.033 ppb/Tg for the NH, and 0.028394

ppb/Tg for the SH. For given STE, the tropospheric ozone burden for NH is slightly395

lower, an indication of stronger ozone sink in the NH due to more vigorous vertical mixing396

probably driven by greater convection over continents and planetary wave activities. The397

SH STE has a much wider range due to the Antarctic ozone hole. The interannual398

variability is similar across the di!erent Linoz chemical models and thus meteorological399

variations appear to overshadow the chemical evolution of the lower stratosphere. The400

interannual variability between the NH and SH is uncorrelated (see the labelled years401

for Linoz-2004BrT). The mean di!erence due to ozone depletion between Linoz-1979Br402

and Linoz-2004BrT in the NH is 25 Tg corresponding to 1 ppb decrease for the mean403

tropospheric ozone abundance and that in the SH is 72Tg, or a 2 ppb decrease in the SH.404

This study has sought to understand and quantify how the stratosphere drives tropo-405

spheric ozone. The scientific results can be summarized as follows:406



• Linoz v2 corrects the low bias in equatorial total ozone found in Linoz v1, and with407

ECMWF IFS data our chemistry-transport model better matches the observed strato-408

spheric ozone columns.409

• Linoz is useful in diagnosing errors in stratospheric circulation for both general cir-410

culation models and assimilated winds, e.g., the change in IFS cycle to 29r2 degraded the411

EC L40 meteorology, but not the L60 version.412

• Observed interannual variability in column ozone is reasonably well modeled with413

Linoz and the EC meteorology; however, the modeled magnitude in the SH, including414

QBO, is twice as large. Other EC products such as ERA-40 with L60 resolution have also415

shown unrealistic, large SH interannual variability (Fleming et al, 2007).416

• Our best estimate for the current STE ozone flux is 290 Tg/yr in the NH and 225417

Tg/yr in the SH with interannual variability over years 2001-2005 of ± 25 Tg and ± 30418

Tg/year respectively. Enhanced STE flux can be correlated with more rapid tropospheric419

mixing and removal of ozone.420

• The STE shows negative anomalies over the mid-latitudes during the easterly phases421

of the QBO and vice versa. The QBO-induced overturning circulations over mid-latitudes422

during the easterly phase creates conditions that reduce STE.423

• The STE flux alone drives a large seasonal change in the tropospheric ozone burden424

with NH mid-latitude peak-to-peak changes of about 8 DU, mimicking summertime pho-425

tochemical production but with half the amplitude. Part of this seasonal change is due426

to increasing tropospheric air mass (i.e., rising tropopause height).427



• When the observed column ozone depletion from 1979 to 2004 is modeled with Linoz428

v2, we predict STE reductions of about 10 %, corresponding to about 1 ppb in tropospheric429

O3 of the northern hemisphere, much less than anticipated by Fusco and Logan (2003).430

References

Baldwin, M. P., et al. (2001), The quasi-biennial oscillation, Rev. Geophys., 39(2), 179230.431

Bojkov, R. D., V. E. Fioletov (1997), Changes of the lower stratospheric ozone over Europe432

and Canada, J. Geophys. Res., 102(D1), 1337-1348, 10.1029/96JD00095.433

Cariolle, D., Lasserre-Bigory, A. and Royer, J. F. (1990), A general circulation model sim-434

ulation of the springtime antarctic ozone and its impact on mid-latitudes. J. Geophys.435

Res., 95, 1883-1898.436

Cariolle, D. and H. Teyssedre (2007), A revised linear ozone photochemistry parameter-437

ization for use in transport and general circulation models: multi-annual simulations.438

Atmos. Phys. Disscuss, 7, 1655-1697, 2007.439

Eyring, V., D. E. Kinnison, and T. G. Shepherd (2005), Overview of planned coupled440

chemistry-climate simulations to support upcoming ozone and climate assessments,441

SPARC Newsletter No. 25, 11-17.442

Fleming, E. L., C. H. Jackman, D. K. Weisenstein, and M. K. W. Ko (2007), The impact443

of interannual variability on multidecadal total ozone simulations, J. Geophys. Res.,444

112, D10310, doi:10.1029/2006JD007953.445

Fusco A. C., J. A. Logan (2003), Analysis of 19701995 trends in tropospheric ozone at446

Northern Hemisphere midlatitudes with the GEOS-CHEM model, J. Geophys. Res.,447

108 (D15), 4449, doi:10.1029/2002JD002742.448



Hsu J., M. J. Prather, O. Wild (2005), Diagnosing the stratosphere-to-troposphere449

flux of ozone in a chemistry transport model, J. Geophys. Res., 110, D19305,450

doi:10.1029/2005JD006045.451

IUPAC (2004), International Union of Pure and Applied Chemistry, Subcommittee on Gas452

Kinetic Data Evaluation, http://www.iupac-kinetic.ch.cam.ac.uk/, Datasheets down-453

loaded 18 May 2004.454

McPeters, R. D., G. J. Labow, and B. J. Johnson (1997), A satellite-derived ozone clima-455

tology for balloonsonde estimation of total column ozone, J. Geophys. Res., 102(D7),456

88758886.457

McPeters R. D., G. J. Labow, J. A. Logan (2007), Ozone climatological profiles for satellite458

retrieval algorithms, J. Geophys. Res., 112, D05308, doi:10.1029/2005JD006823.459

McLinden, C.A., S. C. Olsen and B.J. Hannegan, O. Wild and M.J. Prather (2000),460

Stratosphere ozone in 3-D models: A simple chemistry and the cross-tropopause flux,461

J. Geophys. Res., 105(D11), 14653-14666, 10.1029/2000JD900124.462

Michelsen, H. A., G. L. Manney, M. R. Gunson, R. Zander (1998), Correlations of strato-463

spheric abundances of NOy, O 3, N 2 O, and CH 4 derived from ATMOS measurements,464

J. Geophys. Res., 103(D21), 28347-28360, 10.1029/98JD02850.465

Michelsen, H. A., G. L. Manney, M. R. Gunson, C. P. Rinsland, R. Zander (1998), Correla-466

tions of stratospheric abundances of CH4 and N2O derived from ATMOS measurements,467

Geophys. Res. Lett., 25(15), 2777-2780, 10.1029/98GL01977.468

Nassar R., P. F. Bernath, C. D. Boone, G. L. Manney, S. D. McLeod, C. P. Rinsland, R.469

Skelton, K. A. Walker (2005), Stratospheric abundances of water and methane based on470

ACE-FTS measurements, Geophys. Res. Lett., 32, L15S04, doi:10.1029/2005GL022383.471



Nagatani, R. M., and J. E. Rosenfield, Temperature (1993), net heating and circulation,472

in The Atmospheric E!ects of Stratospheric Aircraft: Report of the 1992 Models and473

Measurements Workshop, NASA Reft Publ. 1292, edited by M. J. Prather and E. E.474

Remsberg, pp. A1-A47.475

Prather, M. (1992), Catastrophic loss of stratospheric ozone in dense volcanic clouds, J.476

Geophys. Res., 97(D9), 10,18710,191.477

Prather, Michael and E Remsburg (1993), The atmospheric e!ects of stratospheric air-478

craft. Report of the 1992 models and measurements Workshop. NASA Publ 1292.479

Prather, M., D. Ehhalt, F. Dentener, R. G. Derwent, E. Dlugokencky, E. Holland, I. S. A.480

Isaksen, J. Katima, V. Kirchho!, P. Matson, P. M. Midgley, and M. Wang, Chapter 4481

(2001), Atmospheric Chemistry and Greenhouse Gases, in Climate Change 2001: The482

Scientific Basis, J.T. Houghton et al., eds., Cambridge U. Press, pp. 239-287.483

Randel, W. J., and J. B. Cobb (1994), Coherent variations of monthly mean total ozone484

and lower stratospheric temperature, J. Geophys. Res., 99(D3), 54335447.485

Randel, W.J., B.A. Boville, J.C. Gille, P.L. Bailey, S.T. Massie, J.B. Kumer, J.L. Mer-486

genthaler, and A.E. Rochem (1994), Simulation of Stratospheric N2O in the NCAR487

CCM2: Comparison with CLAES Data and Global Budget Analyses. J. Atmos. Sci.,488

51, 2834-2845.489

Randel, W.J., and F. Wu, 1996: Isolation of the Ozone QBO in SAGE II Data by Singular-490

Value Decomposition. J. Atmos. Sci., 53, 2546-2559.491

Randel W. J., R.S. Stolarski, D.M. Cunnold, J.A. Logan, M. J. Newchurch, and J. M.492

Zawodny, 1999: Trends in the Vertical Distribution of Ozone, Science 285 (5434), 689,493

DOI:10.1126/science.285.5434.1689.494



Randel W. J., F. Wu (2007), A stratospheric ozone profile data set for 19792005: Variabil-495

ity, trends, and comparisons with column ozone data, J. Geophys. Res., 112, D06313,496

doi:10.1029/2006JD007339.497

Salawitch R. J., D. K. Weisenstein, L. J. Kovalenko, C. E. Sioris, P. O. Wennberg, K.498

Chance, M. K. W. Ko, C. A. McLinden (2005), Sensitivity of ozone to bromine in the499

lower stratosphere, Geophys. Res. Lett., 32, L05811, doi:10.1029/2004GL021504.500

Sander et al. (2006), Chemical kinetics and photochemical data for use in atmospheric501

studies, Evaluation Number 15. JPL Publication 06-2., Jet Propul. Lab., Pasadena,502

Calif.503

Schoeberl, M.A. et al. (1993), The evolution of ClO and NO along air parcel trajectories,504

Geophys. Res. Lett., 20(22), 2511-2514505

Simmons A., M. Hortal, G. Kelly, A. McNally, A. Untch and S. Uppala (2005), ECMWF506

analyses and forecasts of stratopsheric winter polar vortex breakup: September 2002 in507

the southern hemisphere and related events. J. Atmos. Sci., 62, 668-689.508

Stevenson, D. S., et al. (2006), Multimodel ensemble simulations of present-day and near-509

future tropospheric ozone, J. Geophys. Res., 111, D08301, doi:10.1029/2005JD006338.510

Stolarski, R.S., A.R. Douglass, S. Steenrod, and S. Pawson (2006), Trends in Stratospheric511

Ozone: Lessons Learned from a 3D Chemical Transport Model. J. Atmos. Sci., 63,512

10281041.513

Strahan, S. E. (1999), Climatologies of lower stratospheric NOy and O3 and correlations514

with N2O based on in situ observations, J. Geophys. Res., 104(D23), 30463-30480,515

10.1029/1999JD900775.516



Sudo, K., M. Takahashi, and H. Akimoto (2003), Future changes in stratosphere-517

troposphere exchange and their impacts on future tropospheric ozone simulations, Geo-518

phys. Res. Lett., 30(24), 2256, doi:10.1029/2003GL018526.519

Thornton, B. F. et al. (2005), Variability of active chlorine in the lowermost Arctic strato-520

sphere, J. Geophys. Res., 110(D22), D22304.521

Thuillier, G., L. Floyd, T. N. Woods, R. Cebula, E. Hilsenrath, M. Herse, D. Labs (2004),522

Solar Irradiance Spectra for Two Solar Activity Levels, Adv. Space Res., 34, 256-261.523

Wamsley, P. R., et al. (1998), Distribution of halon-1211 in the upper troposphere and524

lower stratosphere and the 1994 total bromine budget, J. Geophys. Res., 103(D1), 1513-525

1526.526

Wild, O., J. K. Sundet, M. J. Prather, I. S. A. Isaksen, H. Akimoto, E. V. Browell,527

and S. J. Oltmans (2003), Chemical transport model ozone simulations for spring528

2001 over the western Pacific: Comparisons with TRACE-P lidar, ozonesondes, and529

Total Ozone Mapping Spectrometer columns, J. Geophys. Res., 108(D21), 8826,530

doi:10.1029/2002JD003283.531

WMO (World Meteorological Organization), Scientific Assessment of Ozone Depletion:532

2006, Global Ozone Research and Monitoring ProjectReport No. 50, 572 pp., Geneva,533

Switzerland, 2007.534

Woodbridge, E. L., et al. (1995), Estimates of total organic and inorganic chlorine in the535

lower stratosphere from in situ and flask measurements during AASE II, J. Geophys.536

Res., 100(D2), 3057-3064.537

Ziemke, J. R., S. Chandra, B. N. Duncan, L. Froidevaux, P. K. Bhartia, P. F. Levelt, and538

J. W. Waters (2006), Tropospheric ozone determined from Aura OMI and MLS: Evalu-539



ation of measurements and comparison with the Global Modeling Initiative’s Chemical540

Transport Model, J. Geophys. Res., 111, D19303, doi:10.1029/2006JD007089.541



0.1

0.20.30.4

0.50.6

0.70.80.9

0.1

0.2

0.20.20.3 0.3

0.3

0.4

0.4

0.5

0.5

0.6

0.60.7

0.70.8

0.8 0.9

0.10.2

0.30.4

0.50.6

0.70.8 0.9

0.10.2 0.30.4

0.50.6

0.7

0.80.9

1 1

0.2

0.30.40.50.60.7

0.80.9

Latitude

1.1

1.21.

31.41.5

1.6

1.7

N2O (318 ppb) CH4 (1.78 ppm) H2O (3.65 ppm)

NOy (19.4 ppb)

Cly (3.44 ppb) Bry (15.6 ppt)

!50 0 50!50 0 50 !50 0 50

Hei

ght (

km)

0

10

20

30

40

50

0

10

20

30

40

50

Figure 1. The Linoz climatologies of trace gases for January. For N2O and CH4 the

patterns are normalized relative to their mean tropospheric abundances; and for H2O, to

the tropopause value. The trace gas families (NOy, Cly, Bry) are normalized relative

to their maximum values (in the upper stratosphere). For year 2004, the normalization

values are noted.



!1.0

!0.5

0.0

0.5

1.0

x106

P!L

(cm

!3s!1

)

10 20 30 40 50 60

4

5

6

7

Height (km)

!10

!8

!6

!4

!2

0

x104

d(P!

L)/d

T (c

m!3

s!1

K!1

)

0

2

4

6

8

10

12

x105

d(P

!L)/

dO3co

l (cm

!3 s

!1 D

U!1

)

JPL97!S3

JPL97!S2

JPL97!S1

JPL00!S1

JPL02!S1

JPL06!S1

P!L

d(P!L)/dO3

d(P!L)/dT

d(P!L)/d(O col)3

!2

8!

log

d

(P!L

)/dO

(

s )!1

10

3

Figure 2. The sensitivity of Linoz terms (Equation 1) to di!erent radiation and chemical

updates using the standard ATMOS-profile on May 31 at 30N as the basic state. See the

text for details.



200

250

300

350

400

MAR

DU

150

200

250

300

350

400 JUN

DU

150

200

250

300

350

400 SEP

DU

!50 0 50

150

200

250

300

350

400DEC

latitude

DU

450Linoz-2004 w/o PSCObs - 1997Obs - 2007

Figure 3. Stratospheric O3 columns (DU) as a function of latitude for March, June,

September and December. The stratospheric column is integrated over the atmosphere

where O3 > 100 ppb. The four di!erent profiles are: McPeters et al. (1997) climatology

(black line with o); McPeters et al. (2007) climatology (black line with +); Linoz-2004

in the UCI CTM with the 1997 ECMWF IFS met data (red solid line); the same Linoz

simulation without PSC parameterization (red dashed).



170190210230 250

250

250270270

270

270

290290

290

290290

310

310

330

330

350350

310

31037

039

0

330

310410

350

330

(a) EP TOMS Clim

J F M A M J J A S O N D J

−60

−30

0

30

60

−40−35−20−15

−15

−15−10 −10

−10

−10

−10

−10

−5−5

−5

−5 −5−5

−5 −5

−5−5 −5

−5−5

5

5

5

5

5 55

55

5

5

5 55

5

5

5

5

5

5

55

10

10

10

10

10

10

15

15

15

2025303540

(c) EP TOMS Ano

J J J J J

−60

−30

0

30

60

230

250250 270

270

270

270

290

290

290

290

310

310

310

310

330

330

350

350

330350370390

410

370

370

330

430

350

450

250

370

310

390

390

(b) UCI CTM/ECMWF−IFS Clim


−60

−30

0

30

60

−50

−40

−40

−40−30 −30

−30

−30

−30−30

−20 −20

−20

−20

−20

−20

−20

−20

−10

−10

−10

−10

−10

−10

−10

−10

−10

−10

−10

−10

−10

−10

−10

−10

−10

−10

10

10

10

10 10

10

1010

10

10

1010

10

10

1010

10

10

20

20 20

20

20

20

20 20

30

30 30

30 30

30

40

40

40

40

50

50 607080

(d) UCI CTM/ECMWF−IFS Ano

J J J J J

−60

−30

0

30

60

Figure 4. Five-year (2001-2005) monthly zonal-mean climatology of total O3 column

(DU) as a function of latitude and for (a) corrected Earth Probe TOMS and (b) Linoz-

2004 CTM simulation using Oslo/EC meteorological data at T42L40 resolution from IFS

Cycle 29r2. The anomalies reltive to this climatology from Jan 2001 to Dec 2005 are

shown for (c) TOMS and (d) CTM.



175200

225

250

250250

250

250

250

250

250

275

275275

275

275

275

275

275275

300

300300300

300

300 300

300

300

325 325

325

325

325

325

325

325325

350

350

350

350

350350

375

375 325

325

400400

375

375

375

350

300

300

425

450

400

400

375

22535

0

225

toatal cloumn ozone (DU) 2005T42L40


−80

−60

−40

−20

0

20

40

60

80

175

200

225

250

250250

250

250 250

250

250

250

275

275275

300300

300

275275

275

275

275

325

325

325

325325

300 300

300

300

300

300

350

350

325

325

325

325

325

375

375

400

400

350

350

350350

425350

375375

375

450

325

375

475

400

toatal cloumn ozone (DU) 2005T42L60


−80

−60

−40

−20

0

20

40

60

80

Figure 5. Year 2005 monthly zonal mean total column O3 (DU) simulated with Linoz-

2004 for (a) T42L40 CTM and (b) T42L60 CTM.



time

Latit

ude

(a) total ozone QBO fit (DU)

J J J J J

−60

−30

0

30

60

(b) total ozone residual (DU)

time

Latit

ude

J J J J J

−60

−30

0

30

60

time

latit

ude

(c) STE QBO fit (g/m2/year)

J J J J J

−60

−30

0

30

60

(d) ozone STE residual (g/m2/year)

time

latit

ude

J J J J J

−60

−30

0

30

60

Figure 6. Monthly zonal mean anomalies in the modeled total O3 column split into

(a) QBO signal and (b) residuals, taken from the 2001-2005 simulation shown in Fig. 4d.

Contour intervals are +5, +10, +15, DU (solid red) and -5, -10, -15, (dashed blue).

Corresponding monthly zonal mean anomalies in the STE O3 fluxes are shown for (c)

QBO signal with contour intervals of ±0.05 g m!2 yr!1 and (d) the residuals with contour

intervals of ±0.10 g m!2 yr!1. Zero contour lines are omitted.



Latit

ude

O3 STE flux (g/m2/year) and zonal mean zonal wind (m/sec)

0 0

1 1

1

1

1 1

1

11

1

11

1

1

1 1

2

2

22

2

2

2

2

33

3

0.5

0.5

0.5 0.5

0.5

0.50.5

0.5 0.5

0.5

0.5 0.5

0.50.5

1.5

1.5

1.5

1.5

1.5

1.5

1.5

1.51.5

1.51.5

2.5

2.52.5

2.5

3.5


−70

−50

−30

−10

10

30

50

70

−10

−5

0

5

10

15

20

25

30

35

40

Figure 7. Latitude by month average STE O3 fluxes (white-line contours at 0 to +3.5

g m!2 yr!1) from UCI CTM with Linoz-2004 driven by ECMWF IFS T42L40 2001-2005

met data. Zonal-mean zonal wind at 200 hPa (grey-scale contours at -5, +5, +15, +25,

+35 ms!1) from the same met data.



Latit

ude

STE flux (g/m2/year) and TOC (in DU)

0 0

1 1

1

1

1 11

11 1

11 1

1

11

2

2

22

2

2

2

23 33

0.5 0.5

0 5 0.5

0.50.5 0.5

0.5 0.50.5

0.5 0.5

0.50.5

1.5

1.5

1.5

1.5

1.5

1.5

1.5

1.51.5

1.5 1.5

2.5

2.52.5

2.5

3.5


−70

−50

−30

−10

10

30

50

70

10

12

14

16

18

20

22

24

26La

titud

e

STE flux (g/m2/year) and Trop. O3 Concentration (ppb)

0 0

1 1

1

1

1 1

1

1

11

11 1

1

11

2

2

2

2

2

2

2

2 3 33

0.5 0.50.5 0.5

0.5 0.50.5

0.50.5 0.5

0.5 0.5

0.50.5

1.5

1.5

1.5

1.5

1.5

1.5

1.5

1.51.5

1.5 1.5

2.52.5

2.5

2.5

3.5


−70

−50

−30

−10

10

30

50

70

20

25

30

35

40

45

Figure 8. (a) Latitude by month average STE O3 fluxes (see Fig 7) on top of simulated

tropospheric column ozone (TCO, color-filled contours at 12, 14, 16, 18, 20, 22, 24 and

26 DU). (b) Monthly zonal-mean tropospheric O3 abundance (ppb).



−160−140−120

−120

−100 −100−80

−80−60

−60

−40

−40

−40

−40 −40−3

0

−30

−30

−30−30

−30

−20−20

−20

−20−20

−20

−16−16

−16

−16−16

−16

−12 −12

−12

−12

−12 −12 −12

−8−8−8

−8

−8 −8 −8


−60

−40

−20

0

20

40

60

−0.7

−0.7−0.5

−0.5−0.5

−0.5

−0.3 −0.3−0.3

−0.3−0.3

−0.3

−0.3

−0.3

−0.3

−0.1 −0.1−0.1

−0.1

−0.1

−0.1

−0.1

−0.1

−0.1

−0.1 −0.1−0.1

−0.1


−60

−40

−20

0

20

40

60

Figure 9. (a) Monthly zonal-mean di!erences in total O3 column (DU) for Linoz-

2004BrT minus Linoz-1979Br calculated with the same 5-year met data. Contour intervals

are 8, 12, 16, 20, 30, 40, DU. (b) Di!erences in STE O3 flux for the same simulation.

Contour intervals are 0.1, 0.2, 0.3, 0.4, and 0.5 g m-2 yr-1.



−15 −10 −5 05

10

15

20

heig

ht (k

m)

%

DJF: ozone change

−15 −10 −5 05

10

15

20

heig

ht (k

m)

%

MAM: ozone change

−15 −10 −5 05

10

15

20

heig

ht (k

m)

%

JJA: ozone change

−15 −10 −5 05

10

15

20

heig

ht (k

m)

%

SON: ozone change

Figure 10. Seasonal profile changes in O3 abundance (%) over the lower strato-

sphere and upper troposphere at 40N-50N for 2004 relative to 1979. Base calculations

use Linoz-1979 and Linoz-1979Br and perturbation calculations use Linoz-2004 (crosses),

Linoz-2004Br (circles) and Linoz-2004BrT (triangles). All calculations use the same me-

teorological data.



200 250 300 350 40030

31

32

33

34

35

36

O3 STE flux (Tg)

Trop

O3 (p

pb)

1

2

3

4

5

12 3

4

5

1

2

3

4

5

12 3

4

5

Linoz−1979Linoz−2004Linoz−1979BrLinoz−2004BrLinoz−2004BrT

Figure 11. Mean tropospheric O3 (ppb) vs. STE O3 flux (Tg/yr) by hemisphere for

five di!erent Linoz models: -1979 (red crosses), -1979Br (black squares), -2004 (green

left-triangles), -2004Br (blue circles), and -2004BrT (blue asterisks). Values for the five

meteorological years, 2001-2005, are shown as 1-5 and only labeled for Linoz-2004BrT.

The upper dashed-dot line is a fit to SH data with the slope of 0.028 (ppb/Tg), and the

lower dashed blue line, to NH data with the slope of 0.033 (ppb/Tg).



Species abundances\Linoz v2 Linoz-1979 Linoz-2004 Linoz-1979Br Linoz-2004Br Linoz-2004BrT

N2O (ppbv) 300.4 318.4 300.4 318.4 318.4

NOy (ppbv) 18.2 19.4 18.2 19.4 19.4

Cly (pptv) 2242 3437 2242. 3437 3437

Bry (pptv) 8.7 15.6 14.7 21.6 21.6

CH4 (ppbv) 1555 1777 1555 1777 1777

H2O (ppmv) 3.65 3.65 3.65 3.65 3.65

PSC activa. Temp (K) 195 195 195 195 199

Table 1. Prescribed abundances of long-lived species and activation temperatures for

the PSC parameterization used for deriving the 5 Linoz tables indicated in the column

headings. See text for details.


Date post:	24-Jun-2020
Category:	Documents
Upload:	others
View:	3 times
Download:	0 times

1 Ozonejuno/HsuPrather_JGR2008.pdfHSU AND PRATHER: STRATOSPHERIC VARIABILITY AND TROPOSPHERIC OZONE...

Documents