EPA 430/K94/021
MONTREAL PROTOCOL
i
i
ON SUBSTANCES THAT DEPLETE
i
THE OZONE LAYER
UNEP
Scientific Assessment of Ozone Depletion.
1994
-------
-------
World Meteorological Organization
Global Ozone Research and Monitoring Project — Report No. 37
SCIENTIFIC ASSESSMENT OF
OZONE DEPLETION: 1994
National Oceanic and Atmospheric Administration
National'Aeronautics and Space Administration
United Nations Environment Programme
World Meteorological Organization
-------
World Meteorological Organization
Global Ozone Observing System (GO3OS)
41 Avenue Giuseppe Motta
P.O. Box 2300
Geneva 2, CH 1211
Switzerland
United Nations Environment Programme
United Nations Headquarters
Ozone Secretariat
P.O. Box 30552
Nairobi
Kenya
U.S. Department of Commerce
National Oceanic and Atmospheric Administration
14th Street and Constitution Avenue NW
Herbert C. Hoover Building, Room 5128
Washington, DC 20230
USA
National Aeronautics and Space Administration
Office of the Mission to Planet Earth
Two Independence Square
300 E Street SW
Washington DC 20546
USA
ISBN 92-807-1449-X
Requests for extra copies by scientific users should be directed to:
WORLD METEOROLOGICAL ORGANIZATION
attn. Dr. Rumen Bojkov
P.O. Box 2300
1211-Geneva, Switzerland
-------
LIST OF INTERNATIONAL AUTHORS,
CONTRIBUTORS, AND REVIEWERS
I
Assessment Co-chairs
Daniel L. [Albritton, Robert T. Watson, and Piet J. A.ucamp
Chapter Lead Authors
1: Neil R.P. Harris 8: Keith P. Shine
2: Eugenio Sanhueza : g: Richard L. McKenzie
3: David W.Fahey ; IQ: Stuart A. Penkett
4: Roderic L. Jones j J1: Andreas Winner and Marvin A. Geller
5: Andreas Volz-Thomas and Brian A!. Ridley 12: R.A. Cox
6: Malcolm K.W. Ko : 13: Susan Solomon and Donald J. Wuebbles
7: FrodeStordal ! •
Coordinating Editor
'•! Christine A. Ennis
Authors, Contributors, and Reviewers
.!_
Daniel L. Albritton ; US Byron Boville US
MarcAllaart The Netherlands Kenneth P. Bowman US
FredN.Alyea US GeirBraathen Norway
Gerard Ancellet ! France Guy P. Brasseur US
MeinratO.Andreae Germany Carl Brenninkmeijer New Zealand
James K.Angell •! US Christoph Briihl Germany
Frank Arnold , Germany William H. Brune US
Roger Atkinson ,; US James H. Butler US
ElliotAtlas i US Sergio Cabrera Chile
PietJ.Aucamp South Africa Bruce A. Callander UK
L.Avallone : US Daniel Cariolle France
Helmuth Bauer . Germany R. Cebula US
SlimaneBekki : UK William L. Chameides US
TiborBerces ; Hungary ' S.Chandra US
T. Bemtsen , Norway Marie-Lise Chanin France
Lane Bishop US J.Christy US
Donald R. Blake US Ralph J. Cicerone US
NJ'BIake : US G.J.R. Coetzee South Africa
Mario Blumthaler •'• Austria Peter S. Connell US
Greg E. Bodeker South Africa D. Considine US
Rumen D. Bojkov Switzerland R.A. Cox , UK
Charles R. Booth | US Paul J. Crutzen . Germany
-------
AUTHORS, CONTRIBUTORS, AND REVIEWERS
Derek N. Cunnold
John Daniel
Malgorzata Deg6rska
John J. DeLuisi
Dirk De Muer
Frank Dentener
Richard G. Derwent
Terry Deshler
Susana B. Diaz
Russell Dickerson
J. Dignon
Ed Dlugokencky
Anne R. Douglass
Tom Duafala
James E. Dye
Dieter H. Ehhalt
James W. Elkins
Christine Ennis
D. Etheridge
David W.Fahey
T. Duncan Fairlie
Donald A. Fisher
Jack Fishman
E. Fleming
Frank Flocke
L. Flynn
P.M. de F. Forster
James Franklin
Paul J. Fraser .
John E. Frederick
Lucien Froidevaux
J.S. Fuglestvedt
Reinhard Furrer
. Ian E. Galbally
Brian G. Gardiner
Marvin A. Geller
Hartwig Gemandt
James F. Gleason
S. Godin
Amram Golombek
Ulrich Gorsdorf
Thomas E. Graedel
Claire Granier
William B. Grant
L. Gray
William L. Grose
J. Gross,
US
US
Poland
US
Belgium
The Netherlands
UK
US
Argentina
US
US
US
US
US
US
Germany
US
US
Australia
US
US
US
US
US
Germany
US
UK
Belgium
Australia
US
US
Norway
Germany
Australia
UK
US
Germany
US
France
Israel
Germany
US
US
US
UK
US
Germany
A. Grossman
Alexander Gruzdev
James E. Hansen
Neil R.P. Harris
ShirO Hatakeyama
D.A. Hauglustaine
Sachiko Hayashida
G.D. Hayman
Kjell Henriksen
Ernest Hilsenrath
David J. Hofmann
Stacey M. Hollandsworth
James R. Holton
Lon L. Hood
0ystein Hov
Carleton J. Howard
Robert D. Hudson
D. Hufford
Linda Hunt
Abdel M. Ibrahim
Mohammad Ilyas
Ivar Isaksen
Tomoyuki Ito
Charles H. Jackman
Daniel J. Jacob
Colin E. Johnson
Harold S. Johnston
Paul V. Johnston
Roderic L. Jones
Torben S. J0rgensen
M. Kanakidou
Igor L. Karol
Prasad Kasibhatla
Jack A. Kaye
Hennie Kelder
James B. Kerr
M.A.K. Khalil
Vyacheslav Khattatov
J.T. Kiehl
S. Kinne
D. Kinnison
Volker Kirchhoff
Malcolm K.W. Ko
UlfKohler
Walter D. Komhyr
Yutaka Kondo
Janusz W. Krzyscin
US
Russia
US
UK
Japan
France
Japan
UK
Norway
US
US
US
US
US
Norway
US
US
US
'us
Egypt
Malaysia
Norway
Japan
US
US
UK
US
New Zealand
UK
Denmark
France
Russia
US
US
The Netherlands
Canada
US
Russia
US
Germany
US
Brazil
US
Germany
US
Japan
Poland
IV
-------
AUTHORS, CONTRIBUTORS, AND REVIEWERS
Antti Kulmala
Michael J. Kurylo
K. Labitzke
Murari Lai
K.S. Law
G. LeBras
Yuan-Pern Lee
Frank Lefevre
Jos Lelieveld
Robert Lesclaux
J.S. Levine
Joel Levy
J. Ben Liley
Peter Liss
David H. Lister
Zenobia Litynska
Shaw C. Liu
Jennifer A. Logan
Nicole Louisnard
Pak Sum Low
Daniel Lubin
Sasha Madronich
Jerry Mahlman
Gloria L. Manney
Huiting Mao
W. Andrew Matthews
Konrad Mauersberger
Archie McCulloch
Mack McFarland
M.E. Mclntyre
Richard L. McKenzie
Richard D. McPeters
Gerard Megie
Paulette Middleton
A.J. Miller
Igor Mokhov
Mario Molina
G.K. Moortgat
Hideaki Nakane
Paul A. Newman
Paul C. Novelli
Samuel J. Oltmans
Alan O'Neill
Michael Oppenheimer
S. Palermi
K. Patten
Juan Carlos Pelaez
, Switzerland
; us
i Germany
; India
'; UK
•,'< France
! Taiwan
1 France
Title Netherlands
i France
•; us
, US
; New Zealand
UK
; UK
i Poland
\ US
US
!
I France
I Kenya
'-! -US
US
i us
: us
; us
New Zealand
, < Germany
.: UK
•i US
i, UK
New Zealand
US
i France
; us
M US
i Russia
!i us
i Germany
: Japan
US
: US
i US
UK
:! us
'. Italy
; us
; Cuba
Stuart A. Penkett
J. Penner
Thomas Peter
Leon F. Phillips
Ken Pickering
R.B. Pierce
S. Pinnock
Michel Pirre
Giovanni Pitari
Walter G. Planet
R.A. Plumb
Jean-Pierre Pommereau
Lament R. Poole
Michael J. Prather
Margarita Prendez
Ronald G. Prinn
Joseph M. Prospero
John A. Pyle
Lian Xiong Qiti
Richard Ramaroson
V. Ramaswamy
William Randel
Phillip Rasch
A.R. Ravishankara
William S. Reeburgh
. C.E. Reeves
J. Richardson
Brian A. Ridley
David Rind
Curtis P. Rinsland
Aiden E. Roche
Michael O. Rodgers
Henning Rodhe
Jose M. Rodriguez
M. Roemer
Franz Rohrer
Richard B. Rood
F. Sherwood Rowland
C.E. Roy
Jochen Rudolph
James M. Russell III
Nelson Sabogal
Karen Sage
Ross Salawitch
Eugenio Sanhueza
K.M. Sarma
T. Sasaki
UK
US
Germany
New Zealand
US
US
UK
France
Italy
US
US
France
US
US
Chile
US
US
UK
China
France
US
US
us
us
us
UK
US
US
US
us
us
us
Sweden
US
The Netherlands
Germany
US
US
Australia
Germany
US
Kenya
US
US
Venezuela
Kenya
Japan
-------
AUTHORS, CONTRIBUTORS, AND REVIEWERS
S.M. Schauffler
Hans Eckhart Scheel
Ulrich Schmidt
Rainer Schmitt
Ulrich Schumann
M.D. Schwarzkopf
Gunther Seckmeyer
Jonathan D. Shanklin
Keith P. Shine
H. Sidebottom
P. Simmonds
Paul C. Simon
H. Singh
Paula Skrivankova
Herman Smit
Susan Solomon
Johannes Staehelin
Knut Stamnes
L. Paul Steele
Leopoldo Stefanutti
Richard S. Stolarski
Frode Stordal
A. Strand
B.H. Subbaraya
N.-D. Sze
Anne M. Thompson
Xue X. Tie
Margaret A. Tolbert
Darin W. Toohey
RalfToumi
Michael Trainer
Charles R. Trepte
US
Germany
Germany
Germany
Germany
US
Germany
UK
UK
Ireland
UK
Belgium
US
Czech Republic
Germany
US
Switzerland
US
Australia
Italy
US
Norway
Norway
India
US
US
US
US
US
UK
US
US
Adrian Tuck
R. Van Dorland
Karel Vanicek
Geraint Vaughan
G. Visconti
Andreas Volz-Thomas
Andreas Wahner
W.-C. Wang
D.I. Wardle
David A. Warrilow
Joe W. Waters
Robert T.Watson
E.C. Weatherhead
Christopher R. Webster
D. Weisenstein
Ray F. Weiss
Paul Wennberg
Howard Wesoky
Thomas M.L. Wigley
Oliver Wild
Paul H. Wine
Peter Winkler
Steven C. Wofsy
Donald J. Wuebbles
Vladimir Yushkov
Ahmed Zand
Rudi J. Zander
Joseph M. Zawodny
Reinhard Zellner
Christos Zerefos
Xiu Ji Zhou
US
The Netherlands
Czech Republic
UK
Italy
Germany
Germany
US
Canada
UK
US
US
US
US
US
US
US
US
US
UK
US
Germany
US
US
Russia
Iran
Belgium
US
Germany
Greece
China
-------
SCIENTIFIC ASSESSMENT OF OZONE DEPLETION: 1994
PREFACE
EXECUTIVE SUMMARY ......,'
COMMON QUESTIONS ABOUT
PART1. OBSERVED CHANGES
CHAPTER 1: OZONE MEASUREMENTS
Lead Author: Neil R. P. Harris
Scientific Summary
1-1 Introduction
Total Ozone
Ozone Profiles
Ozone and Aerosol since 1991
Antarctic Ozone Depletion.....
1.2
1.3
1.4
1.5
References,
TABLE OF CONTENTS
OZONE
xi
xiii
.... xxv
UN OZONE AND SOURCE GASES
CHAPTER 2: SOURCE GASES: TRENDS AND BUDGETS
Lead Author: Eugenia Sqnhueza
Scientific Summary i
2.1 Introduction i
... 1.1
... 1.5
.. 1.5
1.23
1.37
1.43
1.48
Halocarbons i
Str '
2.2
2.3 Stratospheric Inputs of Chlorine and Particulates from Rockets
2.4 Methane J
2.5 Nitrous Oxide
2.6
2-16
Short-Lived Ozone Precursor Gasesi 2-2°
2.7 Carbon Dioxide • 2-22
References.
2.26
2.27
PART 2. ATMOSPHERIC PROCESSES RESPONSIBLE FOR THE OBSERVED CHANGES IN OZONE
CHAPTERS: POLAR OZONE
Lead Author: David W. Fahey
Scientific Summary
3.1
3.2
3.3
3.4
3.5
References
Introduction
Vortex Formation and Tracer Relations.
Processing on Aerosol Surfaces ...
Destruction of Ozone
Vortex Isolation and Export to Midlatitudes
... 3.1
... 3.3
... 3.5
3.10
3.27
3.34
3.41
vu
-------
TABLE OF CONTENTS
CHAPTER 4: TROPICAL AND MIDLATITUDE OZONE
Lead Author: Roderic L. Jones
4 1
Scientific Summary *
4.1 General Introduction
I. Chemical Processes Influencing Middle Latitude and Tropospheric Ozone 4.3
4.2 Introduction.... T " " "
4.3 Eruption of Mount Pinatubo
4.4 Photochemical Ozone Loss Processes at Midlatitudes 4-15
4.5 The Solar Cycle and Quasi-Biennial Oscillation (QBO) Effects on Total Ozone 4.16
n. Transport Processes Linking the Tropics, Middle, and High Latitudes 4.18
4.6 Introduction • ; —•—••• • •"'..'-,
4.7 Transport of Air from Polar Regions to Middle (Latitudes .,..„,..*.....:....*../..'..; ,'.-...:....:. - '-..: 4.23
References « '•'"". ' "" '"
CHAPTERS: TROPOSPHERIC OZONE
Lead Authors: Andreas Volz-Thomas and Brian A. Ridley
Scientific Summary • '
5.1 Introduction - '• "
5.2 Review of Factors that Influence Tropospheric Ozone Concentrations - • 5-3
5.3 Insights from Field Observations: Photochemistry and Transport - 5.8
5.4 Feedback between Tropospheric Ozone and Long-Lived Greenhouse Gases • 5.20
References
PART 3. MODEL SIMULATIONS OF GLOBAL OZONE
CHAPTER 6: MODEL SIMULATIONS OF STRATOSPHERIC OZONE
Lead Author: Malcolm K.W. Ko
Scientific Summary '
6.1 Introduction • • ' '
6.2 Components in a Model Simulation :
6.3 Comparison of Model Results with Observation • 6-12
6.4 Results from Scenario Calculations - : •
6.5 Conclusions "
_ „ ' 6.33
References
CHAPTER 7: MODEL SIMULATIONS OF GLOBAL TROPOSPHERIC OZONE
• "'•'" Lead Author: FrodeStordal
Scientific Summary ' '
7.1 Introduction '*"
7.2 3-D Simulations of the Present-Day Atmosphere:
7 4
Evaluation with Observations .- '
7.3 Current Tropospheric Ozone Modeling • _ '
7:4 Applications """ "
7.5 Intercomparison of Tropospheric Chemistry/Transport Models ••• '•">
References -
vm
-------
TABLE OF CONTENTS
PART 4. CONSEQUENCES OF OZONE CHANGE
CHAPTERS: RADIATIVE FORCING AND TEMPERATURE TRENDS
Lead Author: Keith P. Shine
Scientific Summary ,1 o .
8.1 Introduction o -
8.2 Radiative Forcing Due to Ozone [Change o 3
8.3 Observed Temperature Changes"'.. o yy
8.4 Halocarbon Radiative Forcing.... g ,o
References .; „ „.,
CHAPTER 9: SURFACE ULTRAVIOLET RADIATION
Lead Author: Richard L. McKenzie
Scientific Summary n ,
9.1 Introduction ,; j „ ,
9.2 Update on Trend Observations g 3
9.3 Spectro-Radiometer Results |. _ g 4
9.4 Implications of Recent Changes.;. , g ^
9.5 Update on Predictions i. 9 14
9.6 Gaps in Knowledge j. o ,«
References
PART 5. SCIENTIFIC INFORMATIpN FOR FUTURE DECISIONS
CHAPTER 10: METHYL BROMIDE!
Lead Author: Stuart A. Penkett
Scientific Summary ............................. ,,i [[[ 10 1
10.1 Introduction ...... , ........... . ............. ..; ........ . ............................ ,QT
10.2 Measurements, Including Interhemispheric Ratios ....................................... ..
10.3 Sources of Methyl Bromide ......... 1 [[[
10.4 Sink Mechanisms ........................ ;; [[[ 1011
10.5 The Role of the Oceans .............. . [[[ 10 13
10.6 Modeled Estimates of Global Budget [[[ JQ 15
10.7 Stratospheric Chemistry: Measurements and Models.......... [[[ 10 ig
10.8 The Ozone Depletion Potential of Methyl Bromide [[[ 10 20
10.9 Conclusions ................................. j ............................... . 1023
References ..................................... , ....... r .......................................... ,Q 23
CHAPTER 1 1: SUBSONIC AND SUPERSONIC AIRCRAFT EMISSIONS
Lead Authors: Andreas Wahner and Marvin A. Getter
Scientific Summary ............................... |. [[[ I j ,
11.1 Introduction ................................. .'. ................................ , 1 1 3
1 1.2 Aircraft Emissions ...................... .[ [[[ I j 4
11.3 Plume Processes ......................... .1 [[[ jj JQ
11.4 NOx/H2O/Sulfur Impacts on Atmospheric Chemistry [[[ 11 13
-------
TABLE OF CONTENTS
CHAPTER 12: ATMOSPHERIC DEGRADATION OF HALOCARBON SUBSTITUTES
Lead Author: R.A. Cox
Scientific Summary ' .^ 3
12.1 Background 123
12.2 Atmospheric Lifetimes of HFCs and HCFCs •
12.3 Atmospheric Lifetimes of Other CFC and Halon Substitutes |Ł«
12.4 Atmospheric Degradation of Substitutes '
12.5 Gas Phase Degradation Chemistry of Substitutes •
12.6 Heterogeneous Removal of Halogenated Carbonyl Compounds • ^ ^
12.7 Release of Fluorine Atoms in the Stratosphere •
12.8 CF3OX and FC(O)OX Radical Chemistry in the Stratosphere — .
Do These Radicals Destroy Ozone? - '
12.9 Model Calculations of the Atmospheric Behavior of HCFCs and HFCs J •
References
CHAPTER 13: OZONE DEPLETION POTENTIALS, GLOBAL WARMING POTENTIALS, AND FUTURE
CHLORINE/BROMINE LOADING
Lead Authors: Susan Solomon and Donald J. Wuebbles
Scientific Summary ^ 3
13.1 Introduction 13 4
13.2 Atmospheric Lifetimes and Response Times '
13.3 Cl/Br Loading and Scenarios for CFC Substitutes ^
13.4 Ozone Depletion Potentials - "" '2Q
13.5 Global Warming Potentials • : ; 13"32
References
APPENDICES A 1
A List of International Authors, Contributors, and Reviewers • ^ ^
B Major Acronyms and Abbreviations • c"j
C Chemical Formulae and Nomenclature -.-
-------
PREFACE
The present document is a scientific assessment that will be part of the information upon which the Parties to the
Montreal Protocol will base their future; decisions regarding protection of the stratospheric ozone layer.
Specifically, the Montreal Protocol on Substances That Deplete the Ozone Layer states (Article 6): "... the Parties
shall assess the control measures ... on the basis of available scientific, environmental, technical, and economic infor-
mation." To provide the mechanisms whereby these assessments are conducted, the Protocol further states: "... the
Parties shall convene appropriate panel;>:of experts" and "the panels will report their conclusions ... to the Parties."
Three assessment reports have been prepared during 1994 to be available to the Parties in advance of their meeting
in 1995, at which they will consider the meed to amend or adjust the Protocol. The two companion reports to the present
scientific assessment focus on the environmental and health effects of ozone layer depletion and on the technology and
economic implications of mitigation approaches.
The present report is the latest in a series of seven scientific assessments prepared by the world's leading experts
in the atmospheric sciences and under the international auspices of the World Meteorological Organization (WMO) and
the United Nations Environment Programme (UNEP). The chronology of those scientific assessments and the relation
to the international policy process are summarized as follows:
Year
1981
1985
1987
1988
1989
1990
1991
1992
1992
1994
(1995)
Policy Process
Vienna Convention
Montreal Protocol ]
London Amendment
Copenhagen Amendment
Vienna Amendment (?)
Scientific Assessment
The Stratosphere 1981 Theory and Measurements.
WMO No. 11.
Atmospheric Ozone 1985. 3 vol. WMO No. 16.
International Ozone Trends Panel Report 1988.
2vol. WMO No. 18.
Scientific Assessment of Stratospheric Ozone:
1989. 2vol. WMO No. 20..
Scientific Assessment of Ozone Depletion: 1991.
WMO No. 25.
Methyl Bromide: Its Atmospheric Science, Technology, and
Economics (Assessment Supplement). UNEP (1992).
Scientific Assessment of Ozone Depletion: 1994.
WMO No. 37 (This report.)
The genesis of Scientific Assessment of Ozone Depletion: 1994 occurred at the4 th meeting of the Conference of the
Parties to the Montreal Protocol in Copenhagen, Denmark, in November 1992, at which the scope of the scientific needs
of the Parties was defined. The formal planning of the present report was a workshop that was held on 11 June 1993 in
XL
-------
Virginia Beach, Virginia, at which an international steering group crafted the outline and suggested scientists from the
world community to serve as authors. The first drafts of the chapters were examined at a meeting that occurred on 2 - 4
March 1994 in Washington, D.C., at which the authors and a small number of international experts improved the coor-
dination of the text of the chapters.
The second draft was sent out to 123 scientists worldwide for a mail peer review. These anonymous comments
were considered by the authors. At a Panel Review Meeting in Les Diablerets, Switzerland, held on 18 - 21 July 1994,
the responses to these mail review comments were proposed by the authors and discussed by the 80 participants. Final
changes to the chapters were decided upon, and the Executive Summary was prepared by the participants.
The final result is this document. It is the product of 295 scientists from the developed and developing world1 who
contributed to its preparation and review (230 scientists prepared the report and 147 scientists participated in the peer
review process).
What follows is a summary of their current understanding of the stratospheric ozone layer and its relation to hu-
mankind.
' Participating were Argentina, Australia, Austria, Belgium, Brazil, Canada, Chile, Cuba, Czech Republic, Denmark, Egypt, France, Germany,
Greece, Hungary, India, Iran, Ireland, Israel, Italy, Japan, Kenya, Malaysia, New Zealand, Norway, Poland, Russia. South Africa. Sweden, Switzer-
land, Taiwan, The Netherlands. The People's Republic of China, United Kingdom, United States of America, and Venezuela.
Xll
-------
EXECUTIVE SUMMARY
Recent Major Scientific Findings and Observations
The laboratory investigations, atmospheric observations, and theoretical and modeling studies of the past few years
have provided a deeper understanding of the human-influenced and natural chemical changes in the atmosphere and
their relation to the Earth's stratospheric ozone layer and radiative balance of the climate system. Since the last interna-
tional scientific assessment of the state of understanding, there have been several key ozone-related findings,
observations, and conclusions: i
I
• The atmospheric growth rates of several major ozone-depleting substances have slowed, demonstrating the
expected impact of the Montreal Protocol and its Amendments and Adjustments. The abundances of the
chlorofluorocarbons (CFCs), fcarbon tetrachloride, methyl chloroform, and halons in the atmosphere have been
monitored at global ground-based sites since about 1978. Over much of that period, the annual growth rates of
these gases have been positive. However, the data of recent years clearly show that the growth rates of CFC-11,
CFC-12, halon-1301, and halpn-1211 are slowing down. In particular, total tropospheric organic chlorine in-
creased by only about 60 ppt/year (1.6%) in 1992, compared to 110. ppt/year (2.9%) in 1989. Furthermore,
tropospheric bromine in halons increased by only about 0.25 ppt/year in 1992, compared to about 0.85 ppt/year in
1989. The abundance of carton tetrachloride is actually decreasing. The observed trends in total tropospheric
organic chlorine are consistent with reported production data, suggesting less emission than the maximum al-
lowed under the Montreal Protocol and its Amendments and Adjustments. Peak total chlorine/bromine loading in
the troposphere is expected to occur in 1994, but the stratospheric peak will lag by about 3-5 years. Since the
stratospheric abundances of chlorine and bromine are expected to continue to grow for a few more years, increas-
ing global ozone losses are predicted (other things being equal) for the remainder of the decade, with gradual
recovery in the 21st century.
• The atmospheric abundances of several of the CFC substitutes are increasing, as anticipated. With phase-
, out dates for the CFCs and other ozone-depleting substances now fixed by international agreements, several
hydrochlorofluorocarbons (HdFCs) and hydrofluorocarbons (HFCs) are being manufactured and used as substi-
tutes. The atmospheric growth of some of these compounds (e.g., HCFC-22) has been observed for several years,
and the growth rates of others (e.g., HCFC-142b and HCFC-141b) are now being monitored. Tropospheric
chlorine in HCFCs increased by 5 ppt/year in 1989 and about 10 ppt/year in 1992.
• Record low global ozone levels were measured over the past two years. Anomalous ozone decreases were
observed in the midlatitudes of both hemispheres in 1992 and 1993. The Northern Hemispheric decreases were
larger than those in the Southern Hemisphere. Globally, ozone values were 1 - 2% lower than would be expected
from an extrapolation of the trend prior to 1991, allowing for solar-cycle and quasi-biennial-oscillation (QBO)
effects. The 1994 global ozone levels are returning to values closer to those expected from the longer-term
downward trend.
Xlll
-------
EXECUTIVE SUMMARY
• The stratosphere was perturbed'by a major volcanic eruption. The eruption of Mt. Pinatubo in 1991 led to a
large increase in sulfate aerosol in the lower stratosphere throughout the globe. Reactions on sulfate aerosols
resulted in significant, but temporary, changes in the chemical partitioning that accelerated the photochemical
ozone loss associated with reactive hydrogen (HOX), chlorine, and bromine compounds in the lower stratosphere
in midlatitudes and polar regions. Absorption of terrestrial and solar radiation by the Mt. Pinatubo aerosol result-
ed in a transitory rise of 1°C (globally averaged) in the lower-stratospheric temperature and also affected the
distribution of ozone through circulation changes. The observed 1994 recovery of global ozone is qualitatively
consistent with observed gradual reductions of the abundances of these volcanic particles in the stratosphere.
Downward trends in total-column ozone continue to be observed over much of the globe, but their magni-
tudes are underestimated by numerical models. Decreases in ozone abundances of about 4-5% per decade at
midlatitudes in the Northern and Southern Hemispheres continue to be observed by both ground-based and satel-
lite-borne monitoring instruments. At midlatitudes, the losses continue to be much larger during winter/spring
than during summer/fall in both hemispheres, and the depletion increases with latitude, particularly in the South-
ern Hemisphere. Little or no downward trends are observed in the tropics (20°N - 20°S). While the current two-
dimensional stratospheric models simulate the observed trends quite well during some seasons and latitudes, they
underestimate the trends by factors of up to three in winter/spring at mid- and high latitudes. Several known
atmospheric processes that involve chlorine and bromine and that affect ozone in the lower stratosphere aret
difficult to model and have not been adequately incorporated into these models.
Observations have demonstrated that halogen chemistry plays a larger role in the chemical destruction of
ozone in the midlatitude lower stratosphere than expected from gas phase chemistry. Direct in situ measure-
ments of radical species in the lower stratosphere, coupled with model calculations, have quantitatively shown
that the in situ photochemical loss of ozone due to (largely natural) reactive nitrogen (NOX) compounds is smaller
than that predicted from gas phase chemistry, while that due to (largely natural) HOX compounds and (largely
anthropogenic) chlorine and bromine compounds is larger than that predicted from gas phase chemistry. This
confirms the key role of chemical reactions on sulfate aerosols in controlling the chemical balance of the lower
stratosphere. These and other recent scientific findings strengthen the conclusion of the previous assessment that
the weight of scientific evidence suggests that the observed middle- and high-latitude ozone losses are. largely due
to anthropogenic chlorine and bromine compounds.
• The conclusion that anthropogenic chlorine and bromine compounds, coupled with surface chemistry on
natural polar stratospheric particles, are the cause of polar ozone depletion has been further strengthened.
Laboratory studies have provided a greatly improved understanding of how the chemistry on the surfaces of ice,
nitrate, and sulfate particles can increase the abundance of ozone-depleting forms of chlorine in the polar strato-
spheres. Furthermore, satellite and in situ observations of the abundances of reactive nitrogen and chlorine
compounds have improved the explanation of the different ozone-altering properties of the Antarctic and Arctic.
• The Antarctic ozone "holes" of 1992 and 1993 were the most severe on record. The Antarctic ozone "hole"
has continued to occur seasonally every year since its advent in the late- 1970s, with the occurrences over the last
several years being particularly pronounced. Satellite, balloon-borne, and ground-based monitoring instruments
revealed that the Antarctic ozone "holes" of 1992 and 1993 were the biggest (areal extent) and deepest (minimum
amounts of ozone overhead), with ozone being locally depleted by more than 99% between about 14 - 19 km in
' October, 1992 and 1993. It is likely that these larger-than-usual ozone depletions could be attributed, at least in
part, to sulfate aerosols from Mt. Pinatubo increasing the effectiveness of chlorine- and bromine-catalyzed ozone
destruction. A substantial Antarctic ozone "hole" is expected to occur each austral spring for'many more decades
because stratospheric chlorine and bromine abundances will approach the pre-Antarctic-ozone-"hole" levels
(late-1970s) very slowly during the next century.
xiv
-------
I EXECUTIVE SUMMARY
i
Ozone losses have been detected in the Arctic winter stratosphere, and itheir links to halogen chemistry
have been established. Studies in the Arctic lower stratosphere have been expanded to include more widespread
observations of ozone and key reactlye species. In the late-winter/early-spring period, additional chemical losses
of ozone up to 15 - 20% at some altitudes are deduced from these observations, particularly in the winters of 1991/
2 and 1992/3. Model calculations constrained by the observations are also consistent with these losses, increasing
the confidence in the role of chlorine and bromine in ozone destruction. The interannual variability in the photo-
chemical and dynamical conditions: of the Arctic polar vortex continues to limit the ability to predict ozone
changes in future years. ;
, I
The link between a decrease in stratospheric ozone and an increase in surface ultraviolet (UV) radiation
has been further strengthened. Measurements of UV radiation at the surface under clear-sky conditions show
that low overhead ozone yields high UV radiation and in the amount predicted by radiative-transfer theory Large
increases of surface UV are observeci in Antarctica and the southern part of South America during the period of
the seasonal ozone "hole." Furthermore, elevated surface UV levels at mid-to-high latitudes were observed in the
Northern Hemisphere in 1992 and 1993, corresponding to the low ozone levels of those years. However, the lack
of a decadal (or longer) record of accurate monitoring of surface U V levels and the variation introduced by clouds
and other factors have precluded the Unequivocal identification of a long-term trend in surface UV radiation.
Methyl bromide continues to be viewed as a significant ozone-depleting compound. Increased attention has
been focused upon the ozone-depleting role of methyl bromide. Three potentially major anthropogenic sources of
atmospheric methyl bromide have been identified (soil fumigation, biomass burning, and the exhaust of automo-
biles using leaded gasoline), in additibn to the natural oceanic source. Recent laboratory studies have confirmed
the fast rate for the BrO + HO2 reaction and established a negligible reaction pathway producing HBr, both of
which imply greater ozone losses due to emissions of compounds containing bromine. While the magnitude of
the atmospheric photochemical removal is well understood, there are significant uncertainties in quantifying the
oceanic sink for atmospheric methyl Bromide. The best estimate for the overall lifetime of atmospheric methyl
bromide is 1.3 years, with a range of 0.8 -1.7 years. The Ozone Depletion Potential (ODP) for methyl bromide is
calculated to be about 0.6 (relative to;an ODP of 1 for CFC-11).
<
Stratospheric ozone losses cause a global-mean negative radiative forcing. In the 1991 scientific assessment,
it was pointed out that the global ozorje losses that were occurring in the lower stratosphere caused this region to
cool and result in less radiation reaching the surface-troposphere system. Recent model studies have strengthened
this picture. A long-term global-mean cooling of the lower stratosphere of between 0.25 and 0.4°C/decade has
been observed over the last three decades. Calculations indicate that, on a global mean, the ozone losses between
1980 and 1990 offset about 20% of die radiative forcing due to the well-mixed greenhouse-gas increases during
that period (i.e., carbon dioxide, methane, nitrous oxide, and halocarbons).
Tropospheric ozone, which is a greenhouse gas, appears to have increased in many regions of the Northern
Hemisphere. Observations show that tropospheric ozone, which is formed by chemical reactions involving
pollutants, has increased above many, locations in the Northern Hemisphere over the last 30 years. However, in
the 1980s, the trends were variable, being small or nonexistent. In the Southern Hemisphere, there are insufficient
data to draw strong inferences. At the South Pole, a decrease has been observed since the mid-1980s. Model
simulations and limited observations suggest that tropospheric ozone has increased in the Northern Hemisphere
since pre-industrial times. Such changes would augment the radiative forcing from all other greenhouse gases by
about 20% over the same time period:
-------
EXECUTIVE SUMMARY
The atmospheric residence times of the important ozone-depleting gases, CFC-11 and methyl chloroform
and "house ~ — * « nOW bŁtter ^^ A ^tr °f ;5boC± r "andl 4
known emissions using an atmospheric model has led to a best-estimate lifetime of 50 years for CFC-1 1 and 54
SSTcbloLiin. witn uncertainties of about 10%. These lifetimes provide an accurate standard for
eaTs desLe^ only in the stratosphere (such as CFCs and nitrous oxide) and for those also reacting with tropo-
Lheric hylxyl radical, OH (such as HCFCs and HFCs), respectively. Recent mode. simulations of methane
J±LlT^
^dOHhavedemonstratedthatmemanepenurbationsdecayw
12 17 years, as compared with the 10-year lifetime derived from the total abundance and losse . This longer
resonse ^e and other indirect effects increase the estimate of the effectiveness of emissions of methane as a
se7as by a factor of about two compared to the direct-effect-only values given in the 1991 assessment.
Supporting Scientific Evidence and Related Issues
OZONE CHANGES IN THE TROPICS AND MroLATrrcDES AND THEIR INTERPRETATION
. Analysis of global total-column ozone data through early 1994 shows substantial decreases of ozone in all sea-
udlaLdes (30° - 60°) of both hemispheres. For example, in the middle latitudes of the Northern
re downward trends of about 6% per decade over 1979 - 1994 were observed in winter and spnng and
sonewhaUess but the midlatitude trends averaged a similar 4% to 5% per decade. There are no statistical^
sSfTcatt trends in the tropics (20°S - 20°N). Trends through 1 994 are about 1 % per decade more negative in the
Norfem Hemisphere (2% per decade in the midlatitude winter/spring in the Northern Hemisphere) compare Uo
mordulated'without using data after May 1991. At Northern midlatitudes, ft*. downwarc 1 txend in ozone
between 1981 - 1991 was about 2% per decade greater compared to that of the period 1970 - 1980.
^^
SAGE I/II and SBUV yield downward trends of 10 and 5% per decade, respectively.
Simultaneous in situ measurements of a suite of reactive chemical species have directly ^confirmed modeling
suZs implying that the chemical destruction of ozone in the midlatitude lower stratosphere is more strongly
flue c" by HOX and halogen chemistry than NOX chemistry. The seasonal cycle of CO in the lower s^>
phere at midlatitudes in both hemispheres supports a role for in situ heterogeneous perturbations (,, on sulfate
ae^ols Tut does not appear consistent with the timing of vortex processing or dilution. These studies provide
S^^« view It sulfate aerosol chemistry plays an important role in determining nudlaUtude chem-
ical ozone destruction rates.
XVI
-------
EXECUTIVE SUMMARY
The model-calculated ozone depletions in the upper stratosphere for 1980 - 1990 are in broad agreement with the
measurements. Although these model-calculated ozone depletions did not consider radiative feedbacks and tern
perature trends, including these effects is not likely to reduce the predicted ozone changes by more than 20%.
Models including the chemistry involving sulfate aerosols and polar stratospheric clouds (PSCs) better simulate
the observed total ozone depletions of the past decade than models that include only gas phase reactions How
ever, they still underestimate the ozone loss by factors ranging from 1 .3 to 3.0.
Some unresolved discrepancies between observations and models exist for the partitioning of inorganic chlorine
species, which could impact model predictions of ozone trends. These occur for the C1O/HC1 ratio in the uooer
stratosphere and the fraction of Hp to total inorganic chlorine in the lower stratosphere.
°f °Z°ne-dePleted ^ from P°lar "Sions has the potential to influence ozone concentrations at
middle latitudes. While there are uncertainties about the importance of this process relative to in situ chemistry
for midlatitude ozone loss, both directly involve ozone destruction by chlorine- and bromine-catalyzed reactions
Radiosonde and satellite data continue to show a long-term cooling trend in globally annual-average lower-strato-
spheric temperatures of about 0.3 - 0.4°C per decade over the last three decades. Models suggest that ozone
depletion is the major contributor to this trend.
i
Anomalously large downward ozone trends have been observed in midlatitudes of both hemispheres in 1992 and
1 993 0 e the first two years after the eruption of Mt. Pinatubo), with Northern-Hemispheric decreases larger than
those of the Southern Hemisphere.; Global-average total-ozone levels in early 1993 were about 1% to 2% below
that expected from the long-term trend and the particular phase of the solar and QBO cycles, while peak decreases
of about 6 - 8% from expected ozone levels were seen over 45 - 60°N. In the first half of 1994 ozone levels
returned to values closer to those expected from the long-term trend.
The sulfur gases injected by Mt. Pihatubo led to large enhancements in stratospheric sulfate aerosol surface areas
(by a maximum factor of about 30 - 40 at northern midlatitudes within a year after the eruption), which have
subsequently declined. •
i i
Anomalously low ozone was measured at altitudes below 25 km at a Northern-Hemispheric midlatitude station in
1992 and I 1993 and was correlated with observed enhancements in sulfate-aerosol surface areas, pointing towards
3 cdusnl link.
i
Observations indicate that the eruption of Mt. Pinatubo did not significantly increase the HC1 content of the
stratosphere. i .
The recent large ozone changes at midlatitudes are highly likely to have been due, at least in part to the greatly
increased sulfate aerosol in the lower stratosphere following Mt. Pinatubo. Observations and laboratory studies
have demonstrated the importance of heterogeneous hydrolysis of -N2O5 on sulfate aerosols in the atmosphere
Evidence suggests that C1ONO2 hydrolysis also occurs on sulfate aerosols under cold conditions. Both processes
perturb the chemistry m such a way as to increase ozone loss through coupling with the anthropogenic chlorine
and bromine loading of the stratosphere.
xvn
-------
EXECUTIVE SUMMARY
/
Global mean lower stratospheric temperatures showed a marked transitory rise of about 1°C following the erup-
tion of Mt. Pinatubo in 1991, consistent with model calculations. The warming is likely due to absorption of
radiation by the aerosols.
POLAR OZONE DEPLETION
In 1992 and 1993, the biggest-ever (areal extent) and deepest-ever (minimum ozone below 100 Dobson units)
ozone "holes" were observed in the Antarctic. These extreme ozone depletions may have been due to the chem-
ical perturbations caused by sulfate aerosols from Mt. Pinatubo, acting in addition to the well-recognized chlorine
and bromine reactions on polar stratospheric clouds.
Recent results of observational and modeling studies reaffirm the role of anthropogenic halocarbon species in
Antarctic ozone depletion. Satellite observations show a strong spatial and temporal correlation of CIO abun-
dances with ozone depletion in the Antarctic vortex. In the Arctic winter, a much smaller ozone loss has been
observed. These losses are both consistent with photochemical model calculations constrained with observations
from in situ and satellite instruments.
Extensive new measurements of HC1, CIO, and C1ONO2 from satellites and in situ techniques have confirmed the
picture of the chemical processes responsible for chlorine activation in polar regions and the recovery from those
•processes, strengthening current understanding of the seasonal cycle of ozone depletion in both polar regions.
New laboratory and field studies strengthen the confidence that reactions on sulfate aerosols can activate chlorine
under cold conditions, particularly those in the polar regions. Under volcanically perturbed conditions when
aerosols are enhanced, these processes also likely contribute to ozone losses at the edges of PSC formation
regions (both vertical and horizontal) just outside of the southern vortex and in the Arctic.
Satellite measurements have confirmed that the Arctic vortex is much less denitrified than the Antarctic, which is
likely to be an important factor in determining the interhemispheric differences in polar ozone loss.
Interannual variability in the photochemical and dynamical conditions of the vortices limits reliable predictions of
future ozone changes in the polar regions, particularly in the Arctic.
COUPLING BETWEEN POLAR REGIONS AND MroLATiruDES
Recent satellite observations of long-lived tracers and modeling studies confirm that, above 16 km, air near the
center of the polar vortex is substantially isolated from lower latitudes, especially in the Antarctic.
Erosion of the vortex by planetary-wave activity transports air from the vortex-edge region to lower latitudes.
Nearly all observational and modeling studies are consistent with a time scale of 3 - 4 months to replace a substan-
tial fraction of Antarctic vortex air. The importance of this transport to in situ chemical effects for midlatitude
ozone loss remains poorly known.
Air is readily transported between polar regions and midlatitudes below 16km. The influence of this transport on
midlatitude ozone loss has not been quantified.
XVlll
-------
TROPOSPHERIC OZONE
EXECUTIVE SUMMARY
There is observational evidence that tropospheric ozone (about 10% of the total-column ozone) has increased in
the Northern Hemisphere (north of 20°N) over the past three decades. The upward trends are highly regional
They are smaller in the 1980s than in the 1970s and may be slightly negative at some locations. European
measurements at surface sites also indicate a doubling in the lower-tropospheric ozone concentrations since ear-
lier this century. At the South Pole, a decrease has been observed since the mid-1980s. Elsewhere in the Southern
Hemisphere, there are insufficient data to draw strong inferences.
There is strong evidence that o^one levels in the boundary layer over the populated regions of the Northern
Hemisphere are enhanced by mcjre than 50% due to photochemical production from anthropogenic precursors,
and that export of ozone from North America is a significant source for the North Atlantic region during summer
It has also been shown that biomass burning is a significant source of ozone (and carbon monoxide) in the tropics
during the dry season. ;
An increase in UV-B radiation (e.g., from stratospheric ozone loss) is expected to decrease tropospheric ozone in
the background atmosphere, but, in some cases, it will increase production of ozone in the more polluted regions.
Model calculations predict that a 20% increase in methane concentrations would result in tropospheric ozone
increases ranging from 0.5 to 2.5'ppb in the tropics and the northern midlatitude summer, and an increase in the
methane residence time to about il4 years (a range of 12 - 17 years). Although there is a high degree of consis-
tency in the global transport of short-lived tracers within three-dimensional chemical-transport models, and a
general agreement in the computation of photochemical rates affecting tropospheric ozone, many processes con-
trolling tropospheric ozone are not adequately represented or tested in the models, hence limiting the accuracy of
these results.
i
TRENDS IN SOURCE GASES RELATING TO OZONE CHANGES
CFCs, carbon tetrachloride, methyl chloroform, and the halons are major anthropogenic source gases for strato-
spheric chlorine and bromine, and hence stratospheric ozone destruction. Observations from several monitoring
networks worldwide have demonstrated slowdowns in growth rates of these species that are consistent (except for
carbon tetrachloride) with expectations based upon recent decreases in emissions. In addition, observations from
several sites have revealed accelerating growth rates of the CFC substitutes, HCFC-22, HCFC-141b, and HCFC-
142b, as expected from their increasing use.
Methane levels in the atmosphere affect tropospheric and stratospheric ozone levels. Global methane increased
by 7% over about the past decade.: However, the 1980s were characterized by slower growth rates, dropping from
approximately 20 ppb per year in |l980 to about 10 ppb per year by the end of the decade. Methane growth rates
slowed dramatically in 1991 and 1992, but the very recent data suggest that they have started to increase in late
1993. The cause(s) of this behavior are not known, but it is probably due to changes in methane sources rather
than sinks.
Despite the increased methane levels
was a decade ago. Recent analyses of global
early 1980s to about 1987 and have declined
not been identified.
;, the total amount of carbon monoxide in today's atmosphere is less than it
carbon monoxide data show that tropospheric levels grew from the
from the late 1980s to the present. The cause(s) of this behavior have
xix
-------
EXECUTIVE SUMMARY
CONSEQUENCES OF OZONE CHANGES
. The only general circulation model (GCM) simulation to investigate the climatic impacts of observed ozone
depletions between 1970 and 1990 supports earlier suggestions that these depletions reduced the model-predicted
wanning due to well-mixed greenhouse gases by about 20%. This is consistent with radiative forcing calcula-
tions.
Model simulations suggest that increases in tropospheric ozone since pre-industrial times may have made signif-
icant contributions to the greenhouse forcing of the Earth's climate system, enhancing the current total forcing by
about 20% compared to that arising from the changes in the well-mixed greenhouses gases over that period.
. Large increases in ultraviolet (UV) radiation have been observed in association with the ozone hole at high south-
ern latitudes. The measured UV enhancements agree well withmodel .calculations. ........
Clear-sky UV measurements at midlatitude locations in the Southern Hemisphere are significantly larger than at
a corresponding site in the Northern Hemisphere, in agreement with expected differences due to ozone column
and Sun-Earth separation.
Local increases in UV B were measured in 1992/93 at mid- and high latitudes in the Northern Hemisphere. The
spectral signatures of the enhancements clearly implicate the anomalously low ozone observed in those years,
rather than variability of cloud cover or tropospheric pollution. Such correlations add confidence to the ability to
link ozone changes to UV-B changes over relatively long time scales.
Increases in clear-sky UV over the period 1979 to 1993 due to observed ozone changes are calculated to be
greatest at short wavelengths and at high latitudes. Poleward of 45°, the increases are greatest m the Southern
Hemisphere.
. Uncertainties in calibration, influence of tropospheric pollution, and difficulties of interpreting data from broad-
band instruments continue to preclude the unequivocal identification of long-term UV trends. However, data
from two relatively unpolluted sites do appear to show UV increases consistent with observed ozone trends.
Given the uncertainties of these studies, it now appears that quantification of the natural (i.e., pre-ozone-reduc-
tion) UV baseline has been irrevocably lost at mid- and high latitudes.
Scattering of UV radiation by stratospheric aerosols from the Mt. Pinatubo eruption did not alter total surface-UV
levels appreciably.
RELATED PHENOMENA AND ISSUES
Methyl Bromide
. Three potentially major anthropogenic sources of methyl bromide have been identified: (i) soil fumigation: 20 to
60 ktons per year, where new measurements reaffirm that about 50% (ranging from 20 - 90%) of the methyl
bromide used as a soil fumigant is released into the atmosphere; (ii) biomass burning: 10 to 50 ktons per year; and
(iii) the exhaust of automobiles using leaded gasoline: 0.5 to 1.5 ktons per year or 9 to 22 ktons per year (the two
studies report emission factors that differ by a factor of more than 10). In addition, the one known major natural
source of methyl bromide is oceanic, with emissions of 60 to 160 ktons per year.
-------
I EXECUTIVE SUMMARY
Recent measurements have confiimed that there is more methyl bromide in the Northern Hemisphere than in the
Southern Hemisphere, with an imerhemispheric ratio of 1.3.
There are two known sinks for atmospheric methyl bromide: (i) atmospheric, with a lifetime of 2.0 years (1.5 to
2.5 years); and (ii) oceanic, with an estimated lifetime of 3.7 years (1.5 to 10 years). The overall best estimate for
the lifetime of atmospheric methyl bromide is 1.3 years, with a range of 0.8 to 1.7 years. An overall lifetime of
less than 0.6 years is thought to bp highly unlikely because of constraints imposed by the observed interhemi-
spheric ratio and total known emissions.
The chemistry of bromine-induced stratospheric ozone destruction is now better understood. Laboratory mea-
surements have confirmed the fast rate for the BrO + HO2 reaction and have established a negligible reaction
pathway producing HBr, both of which imply greater ozone losses due to emissions of compounds containing
bromine. Stratospheric measurements show that the abundance of HBr is less than 1 ppt.
Bromine is estimated to be about 50 times more efficient than chlorine in destroying stratospheric ozone on a per-
atom basis. The OOP for methyl bromide is calculated to be about 0.6, based on an overall lifetime of 1.3 years.
An uncertainty analysis suggests that the ODP is unlikely to be less than 0.3.
Aircraft
Subsonics: Estimates indicate that present subsonic aircraft operations may be significantly increasing trace
species (primarily NOX, sulfur dioxide, and soot) at upper-tropospheric altitudes in the North-Atlantic flight cor-
ridor. Models indicate that the NOX emissions from the current subsonic fleet produce upper-tropospheric ozone
increases as much as several percent, maximizing at northern midlatitudes. Since the results of these rather
complex models depend critically Jan NOX chemistry and since the tropospheric NOX budget is uncertain, little
confidence should be put in these (quantitative model results at the present time.
Supersonics: Atmospheric effects of supersonic aircraft depend on the number of aircraft, the altitude of opera-
tion, the exhaust emissions, and die background chlorine and aerosol loadings. Projected fleets of supersonic
transports would lead to significant changes in trace-species concentrations, especially in the North-Atlantic
flight corridor. Two-dimensional model calculations of the impact of a projected fleet (500 aircraft, each emitting
15 grams of NOX per kilogram of fuel burned at Mach 2.4) in a stratosphere with a chlorine loading of 3.7 ppb,
imply additional (i.e., beyond 'those from halocarbon losses) annual-average ozone column decreases of
0.3 - 1.8% for the Northern Hemisphere. There are, however, important uncertainties in these model results,
especially in the stratosphere below 25 km. The same models fail to reproduce the observed ozone trends in the
stratosphere below 25 km between; 1980 and 1990. Thus, these models may not be properly including mecha-
nisms that are important in this crucial altitude range.
Climate Effects: Reliable quantitative estimates of the effects of aviation emissions on climate are not yet avail-
able. Some initial estimates indicate that the climate effects of ozone changes resulting from subsonic aircraft
emissions may be comparable to those resulting from their Cp2 emissions.
xxi
-------
EXECUTIVE SUMMARY
Ozone Depletion Potentials (ODPs)
If a substance containing chlorine or bromine decomposes in the stratosphere, it will destroy some ozone.
HCFCs have short tropospheric lifetimes, which tends to reduce their impact on stratospheric ozone as compared
to CFCs and halons. However, there are substantial differences in ODPs among various substitutes. The steady-
state ODPs of substitute compounds considered in the present assessment range from about 0.01 - 0.1.
Tropospheric degradation products of CFC substitutes will not lead to significant ozone loss in the stratosphere.
Those products will not accumulate in the atmosphere and will not significantly influence the ODPs and Global
Warming Potentials (GWPs) of the substitutes.
• Trifluoroacetic acid, formed in the atmospheric degradation of HFC-134a, HCFC-123, and HCFC-124, will enter
into the aqueous environment, where biological, rather than physico-chemical, removal processes may be'effec-
tive.
• It is known that atomic fluorine (F) itself is not an efficient catalyst for ozone loss, and it is concluded that the
F-containing fragments from the substitutes (such as CF3OX) also have negligible impact on ozone. Therefore,
ODPs of HFCs containing the CF3 group (such as HFC-134a, HFC-23, and HFC-125) are likely to be much less
than 0.001.
• New laboratory measurements and associated modeling studies have confirmed that perfluorocarbons and sulfur
hexafluoride are long-lived in the atmosphere and act as greenhouse gases.
The ODPs for several new compounds, such as HCFC-225ca, HCFC-225cb, and CF3I, have been evaluated using
both semi-empirical and modeling approaches, and are found to be 0.03 or less.
Global Warming Potentials (GWPs)
• Both the direct and indirect components of the GWP of methane have been estimated using model calculations.
Methane's influence on the hydroxyl radical and the resulting effect on the methane response time lead to substan-
tially longer response times for decay of emissions than OH removal alone, thereby increasing the GWP. In
addition, indirect effects including production of tropospheric ozone and stratospheric water vapor were consid-
ered and are estimated to range from about 15 to 45% of the total GWP (direct plus indirect) for methane.
• GWPs, including indirect effects of ozone depletion, have been estimated for a variety of halocarbons, clarifying
the relative radiative roles of ozone-depleting compounds (i.e., CFCs and halons). The net GWPs of halocarbons
depend strongly upon the effectiveness of each compound for ozone destruction; the halons are highly likely to
have negative net GWPs, while those of the CFCs are likely to be positive over both 20- and 100-year time
horizons.
Implications for Policy Formulation
The research findings of the past few years that are summarized above have several major implications as scientific
input to governmental, industrial, and other policy decisions regarding human-influenced substances that lead to deple-
tion of the stratospheric ozone layer and to changes of the radiative forcing of the climate system:
xxu
-------
EXECUTIVE SUMMARY
The Montreal Protocol and its Amendments and Adjustments are reducing the impact of anthropogenic
halocarbons on the ozone layer and should eventually eliminate this ozone depletion. Based on assumed
compliance with the amended Montreal Protocol (Copenhagen, 1992) by all nations, the stratospheric chlorine
abundances will continue to grow from their current levels (3.6 ppb) to a peak of about 3.8 ppb around the turn of
the century. The future total bromine loading will depend upon choices made regarding future human production
and emissions of methyl bromide. After around the turn of the century, the levels of stratospheric chlorine and
bromine will begin a decrease that will continue into the 21st and 22nd centuries. The rate of decline is dictated
by the long residence times of the CFCs, carbon tetrachloride, and halons. Global ozone losses and the Antarctic
ozone hole were first discernible in the late 1970s and are predicted to recover in about the year 2045 other
things being equal. The recovery of the ozone layer would have been impossible without the Amendments and
Adjustments to the original Protocol (Montreal, 1987). wumenis ana
, °CCUr dUrf "g thC ^ SeVeral years* The ozone laver ™» be most
affected by human-influenced perturbations and susceptible to natural variations in the period around the year
1998, since the peak stratosphericichlorine and bromine abundances are expscted to occur then. Based on extrap-
olation of current trends, observations suggest that the maximum ozone loss, relative to the late 1960s, will likely
(i) about 12 - 13% at Northern, Imidlatitudes in winter/spring (i.e., about 2.5% above current levels)-
n) about 6 - 7% at Northern midlatitudes in summer/fall (i.e., about 1.5% above current levels)- and
(in) about 11% (with less cerfciinty) at Southern midlatitudes on a year-round basis (Le., about 2.5% above
current levels/. ;
' and'13% inCreaSCS' resF-tiv^ "> surface erythemal radia-
f p r' if *« Were to * a ™J°r «*«uc eruption
like that
f p ' c erupon
f M, Pmatubo, or if an extremely cold and persistent Arctic winter were to occur, then the ozone losses
and UV increases could be larger in individual years.
, I
Approaches to lowering stratospheric chlorine and bromine abundances are limited. Further controls on
ozone-depleting .substances would not be expected to significantly change the timing or the magnitude of the peak
stratospheric halocarbon abundances and hence peak ozone loss. However, there are four approaches that would
steepen the initial fall from the peak halocarbon levels in the early decades of the next century:
(i) If emissions of methyl bronze from agricultural; structural, and industrial activities were to be eliminated
in the year 2001, then the integrated effective future chlorine loading above the 1980 level (which is related
to the cumulative future loss of ozone) is predicted to be 13% less over the next 50 years relative to full
compliance to the Amendments and Adjustments to the Protocol
(ii) If emissions of HCFCs weieito be totally eliminated by the year 2004, then the integrated effective future
chlorine loading above the 1980 level is predicted to be 5% less over the next 50 years relative to full
compliance with the Amendments and Adjustments to the Protocol.
(iii) If halons presently contained in existing equipment were never released to the atmosphere then the inte-
grated effective future chlorine loading above the 1980 level is predicted to be 10% less over the next 50
years relative to full compliance with the Amendments and Adjustments to the Protocol
(iv) If CFCs presently contained in existing equipment were never released to the atmosphere, then the integrat-
ed effective future chlorine loading above the 1980 level is predicted to be 3% less over the next 50 years
relative to full compliance wi;th the Amendments and Adjustments to the Protocol
xxm
-------
EXECUTIVE SUMMARY
. Failure to adhere to the international agreements will delay recovery of the ozone layer. If there were to be
additional production of CFCs at 20% of 1992 levels for each year through 2002 and ramped to zero by 2005
(beyond that allowed for countries operating under Article 5 of the Montreal Protocol), then the integrated effective
future chlorine loading above the 1980 level is predicted to be 9% more over the next 50 years relative to full
compliance to the Amendments and Adjustments to the Protocol.
. Many of the substitutes for the CFCs and halons are also notable greenhouse gases. Several CFC and halon
substitutes are not addressed under the Montreal Protocol (because they do not deplete ozone), but, because they
are greenhouse gases, fall under the purview of the Framework Convention on Climate Change. There is a wide
range of values for the Global Warming Potentials (GWPs) of the HFCs (150 - 10000), with about half of them
having values comparable to the ozone-depleting compounds they replace. The perfluorinated compounds, some
of which are being considered as substitutes, have very large GWPs (e.g., 5000 - 10000). These are examples of
compounds whose current atmospheric abundances are relatively small, but are increasing or could increase in the
future.
- Consideration of the ozone change will be one necessary ingredient in understanding climate change. The
extent of our ability to attribute any climate change to specific causes will likely prove to be important scientific
input to decisions regarding predicted human-induced influences on the climate system. Changes in ozone since
pre-industrial times as a result of human activity are believed to have been a significant influence on radiative
forcing; this human influence is expected to continue into the foreseeable future.
XXIV
-------
COMMON QUESTIONS ABOUT OZONE
Ozone is exceedingly rare in our atmosphere,
averaging about 3 molecules of ozone for
every ten million air molecules. Nonethe-
less, atmospheric ozone plays vital roles that belie its
small numbers. This Appendix to the World Meteoro-
logical Organization/United Nations Environment
Programme (WMO/UNEP) Scientific 'Assessment of
Ozone Depletion: 1994 answers some of the questions
that are most commonly asked about ozone and the
changes that have been occurring in recent years. These
common questions and their answers were discussed by
the 80 scientists from 26 countries who participated in
the Panel Review Meeting of the Scientific Assessment of
Ozone Depletion: 1994. Therefore, thrs information is
presented by a large group of experts from the interna-
tional scientific community. '.-.
Ozone is mainly found in two regions of the Earth's atmo--
sphere. Most ozone (about 90%) resides in a layer
between approximately 10 and 50 kilometers (about 6 to
30 miles) above the Earth's surface, in the region of the
atmosphere called the stratosphere. This stratospheric
ozone is commonly known as the "ozone layer." The re-
maining ozone is in the lower region of the atmosphere,
the troposphere, which extends from the! Earth's surface
up to about 10 kilometers. The figure bellow shows this
distribution of ozone in the atmosphere.>!
While the ozone in these two regions is ctii emically iden-
tical (both consist of three oxygen atoms and have the
chemical formula "03"), the ozone molecules have very
different effects on humans and other living things de-
pending upon their location. !;
!i
Stratospheric ozone plays a beneficial rolfe by absorbing
most of the biologically damaging ultraviolet sunlight
called UV-B, allowing only a small amount to reach the
Earth's surface. The absorption^ UV radiation by ozone
creates a source of heat, which actually forms the strato-
sphere itself (a region in which the temperature rises as
one goes to higher altitudes). Ozone thus plays a key
role in the temperature structure of the! Earth's atmo-
sphere. Furthermore, without the.filtering action of the
ozone layer, more of the Sun's UV-B radiation would
penetrate the atmosphere and would reach the Earth's
surface in greater amounts. Many experimental studies
of plants and animals, and clinical studies of humans,
have shown the harmful effects of excessive exposure to
UV-B radiation '(these are discussed in the WMO/UNEP
reports on impacts of ozone depletion, which are com-
panion documents to the WMO/UNEP scientific assess-
ments of ozone depletion).
At the planet's surface, ozone comes into direct contact
with life-forms and displays its destructive side. Be-
cause ozone reacts strongly with other molecules, high
levels are toxic to living systems and can severely'dam-
age the tissues of plants and animals. Many studies
have documented the harmful effects of ozone on crop
production, forest growth, and human health. The sub-
stantial negative effects of surface-level tropospheric
ozone from this direct toxicity contrast with the benefits
of the additional filtering of UV-B radiation that it pro-
vides.
With these dual aspects of ozone come two separate en-
vironmental issues, controlled by different forces in the
atmosphere. In the troposphere, there is concern about
increases in ozone. Low-lying ozone is a key component
of smog, a familiar problem in the atmosphere of many
cities around the world. Higher than usual amounts of
surface-level ozone are now increasingly being observed
in rural areas as well. However, the ground-level ozone
concentrations in the smoggiest cities are very much
smaller than the concentrations routinely found in the
stratosphere.
There is widespread scientific and public interest and
concern about losses of stratospheric ozone. Ground-
based and satellite instruments 'have'measured
decreases in the amount of stratospheric ozone in our
atmosphere. Over some parts of Antarctica, up to 60% of
.the total overhead amount of ozone (known as the "col-
umn ozone") is depleted during September and October.
This phenomenon has come to be known as the Antarctic
"ozone hole." Smaller, but still significant, stratospheric
decreases have been seen at other, more-populated re-
gions of the Earth. Increases in surface UV-B radiation
have been observed in association with decreases in
stratospheric ozone.
The scientific evidence, accumulated over more than two
decades of study by the international research communi- -
ty, has shown that human-made chemicals are
responsible for the observed depletions of the ozone lay-
er over Antarctica and likely play a major role in global
ozone losses. The ozone-depleting compounds contain
various combinations of the chemical elements chlorine,
fluorine, bromine, carbon, and hydrogen, and are often
described byjhe general term halocarbons. The com-
XXV
-------
COMMON QUESTIONS
pounds that contain only carbon, chlorine, and fluorine
are called chlorofluorocarbons, usually abbreviated as
CFCs. CFCs, carbon tetrachloride, and methyl chloro-
form are important human-made ozone-depleting gases
that have been used in many applications including re-
frigeration, air conditioning, foam blowing, cleaning of
electronics components, and as solvents. Another im-
portant group of human-made halocarbons is the
halons. which contain carbon, bromine, fluorine, and (in
some cases) chlorine, and have been mainly used as fire
extinguishants. Governments have decided to discon-
tinue production of CFCs, halons, carbon tetrachloride,
and methyl chloroform, and industry has developed
more "ozone-friendly" substitutes. .
Two responses are natural when a new problem has been
identified: cure and prevention. When the problem is the
destruction of the stratospheric ozone layer, the corre-
sponding questions are: Can we repair the damage
already done? How can we prevent further destruction?
Remedies have been investigated that could (i) remove
CFCs selectively from our atmosphere, (ii) intercept
ozone-depleting chlorine before much depletion has tak-
en place, or (iii) replace the ozone lost in the stratosphere
(perhaps by shipping the ozone from cities that have top
much smog or by making new ozone). Because ozone
reacts strongly with other molecules, as noted above, it
is too unstable to be made elsewhere (e.g., in the smog
of cities) and transported to the stratosphere. When the
huge volume of the Earth's atmosphere and the magni-
tude of global stratospheric ozone depletion are carefully
considered, approaches to cures quickly become much
too expensive, impractical, and potentially damaging to
the global environment. Prevention involves the interna-
tionally agreed-upon Montreal Protocol and its
Amendments and Adjustments, which call for elimina-
tion of the production and use of the CFCs and other
ozone-damaging compounds within the next few years.
As a result, the ozone layer is expected to recover over
the next fifty years or so as the atmospheric concentra-
tions of CFCs and other ozone-depleting compounds
slowly decay.
The current understanding of ozone depletion and its re-
lation to humankind is discussed in detail by the leading
scientists in the world's ozone research community in the
Scientific Assessment of Ozone Depletion: 1994. The
answers to the common questions posed below are
based upon that understanding and on the information
given in earlier WMO/UNEP reports.
Atmospheric Ozone
Stratospheric Ozone
(The Ozone Layer)
Tropospheric Ozone
Contains 90% of Atmospheric
Ozone
Beneficial Role:
Acts as Primary UV Radiation
Shield
• Current Issues:
- Long-term Global
Downward Trends
- Springtime Antarctic Ozone
Hole Each Year
• Contains 10% of Atmospheric
Ozone
• Harmful Impact: Toxic Effects
on Humans and Vegetation
• Current Issues:
- Episodes of High Surface
Ozone in Urban and
Rural Areas
0 5 10 15 20 25
Ozone Amount
(pressure, milli-Pascals)
XXVI
-------
COMMON QUESTIONS
How Can Chlorofluordcarbons (CFCs) Get to the Stratosphere
If They're Heavier than Air?
Although the CFC molecules are indeed several times
heavier than air, thousands of measurements have been
made from balloons, aircraft, and satellites demonstrat-
ing that the CFCs are actually present in the stratosphere.
The atmosphere is not stagnant. 'Winds1 mix the atmo-
sphere to altitudes far above the top of the stratosphere
much faster than molecules can settle according to their
weight. Gases such as CFCs that are insoluble in water
and relatively unreactive in the lower atmosphere (below
about 10 km) are quickly mixed and therefore reach the
stratosphere regardless of their weight. ; |
Much can be learned about the atmospheric fate of com-
pounds from the measured changes in' concentration
versus altitude. For example, the two gases carbon tet-
rafluoride (CF4, produced mainly as a by-product of the
manufacture of aluminum) and CFC-11 (c'ci3F, used in a
variety of human activities) are both much heavier than
air. Carbon tetrafluoride is completely unreactive in the
lower 99.9% of the atmosphere, and measurements
show it to be nearly uniformly distributed throughout the
atmosphere as shown in the figure. There have also been
measurements over the past two decades of several other
completely unreactive gases, one lighter than air (neon)
and some heavier than air (argon, krypton), which show
that they also mix upward uniformly through the strato-
sphere regardless of their weight, just as observed with
carbon tetrafluoride. CFC-11 is unreactive in the lower
atmosphere (below about 15 km) and is similarly uni-
formly mixed there, as shown. The abundance of
CFC-11 decreases as the gas reaches higher altitudes,
where it is broken down by high energy solar ultraviolet
radiation. Chlorine released from this breakdown of
CFC-11 and other CFCs remains in the stratosphere for
several years, where it destroys many thousands of mol-
ecules of ozone.
Measurements of CFC-11 and CF*
40
"S 30
_g
15
^ 20
TJ
< 10
_L
O.OI ! O.I i.O IO.O IOO
: Atmospheric Abundance
: (in parts per trillion )
Stratosphere
IOOO
/WW
I
Troposphere
JL
xxvn
-------
COMMON QUESTIONS
What is the Evidence that Stratospheric Ozone
is Destroyed by Chlorine and Bromine?
Laboratory studies show that chlorine (Cl) reacts very
rapidly with ozone. They also show that the reactive
chemical chlorine oxide (CIO) formed in that reaction
can undergo further processes which regenerate the
original chlorine, allowing the sequence to be repeated
very many times (a "chain reaction"). Similar reactions
also take place between bromine and ozone.
But do these ozone-destroying reactions occur in the real
world? All of our accumulated scientific experience dem-
onstrates that if the conditions of temperature and
pressure are like those in the laboratory studies, the
same chemical reactions will take place in nature. How-
ever, many other reactions including those of other
chemical species are often also taking place simulta-
neously in the stratosphere, making the connections
among the changes difficult to untangle. Nevertheless,
whenever chlorine (or bromine) and ozone are found to-
gether in the stratosphere, the ozone-destroying
reactions must be taking place.
Sometimes a small number of chemical reactions is so
" important in the natural, circumstance that the connec-
tions are almost as clear as in laboratory experiments.
Such a situation occurs in the Antarctic stratosphere dur-
ing the springtime formation of the ozone hole. During
August and September 1987 - the end of winter and be-
ginning of spring in the Southern Hemisphere - aircraft
equipped with many different instruments for measuring
a large number of chemical species were flown repeated-
ly over Antarctica. Among the chemicals measured were
ozone and chlorine oxide, the reactive chemical identi-
fied in the laboratory as one of the participants in the
ozone-destroying chain reactions. On the first flights
southward from the southern tip of South America, rela-
tively high concentrations of ozone were measured
everywhere over Antarctica. By mid-September, howev-
er, the instruments recorded low concentrations of ozone
in regions where there were high concentrations of chlo-
rine oxide and vice versa, as shown in the figure. Flights
later in September showed even less ozone over Antarc-
tica, as the chlorine continued to react with the
stratospheric ozone.
Independent measurements made by these and other in-
struments on this and other airplanes, from the ground,
from balloons, and from satellites have provided a de-
tailed understanding of the chemical reactions going on
in the Antarctic stratosphere. Regions with high concen-
trations of reactive chlorine reach temperatures so cold
(less than approximately -80°C, or -112°F) that strato-
spheric clouds form, a rare occurrence except during the
polar winters. These clouds facilitate other chemical re-
actions that allow the release of chlorine in sunlight. The
chemical reactions related to the clouds are now well
understood through study under laboratory conditions
mimicking those found naturally. Scientists are working
to understand the role of such reactions of chlorine and
bromine at other latitudes, and the involvement of parti-
cles of sulfuric acid from volcanoes or other sources.
Measurements of Ozone and Reactive Chlorine
from a Flight into the Antarctic Ozone Hole
3000
Reactive Chlorine
(Scale at Right)
< =
03 CO
63
64
65
66 67 68 69
Latitude (Degrees South)
'xxviii
70
-------
COMMON QUESTIONS
Does Most of the Chlorine in the Stratosphere
Come from Human or Natural Sources?
Most of the chlorine in the stratosphere is there as a re-
sult of human activities.
;i
Many compounds containing chlorine are released at the
ground, but those that dissolve in water cannot reach
stratospheric altitudes. Large quantities of chlorine are
released from evaporated ocean spray asisea salt (sodi-
um chloride) aerosol. However, because sea salt
dissolves in water, this chlorine quickly iis taken up in
clouds or in ice, snow, or rain droplet:;, and does not
reach the stratosphere. Another ground-level source of
chlorine is its use in swimming pools and as household
bleach. When released, this chlorine is rapidly convert-
ed to forms that dissolve in water and therefore are
removed from the lower atmosphere, never reaching the
stratosphere in significant amounts. Volcanoes can-emit
large quantities of hydrogen chloride, but this gas is rap-
idly converted to hydrochloric acid in rainwater, ice, and
snow and does not reach the stratosphere. Even in ex-
plosive volcanic plumes that rise high in trie atmosphere,
nearly all of the hydrogen chloride is scrubbed out in
precipitation before reaching stratospheric altitudes.
In contrast, human-made halocarbons - such as CFCs,
carbon tetrachloride (CCI4) and methyl chloroform
(CHgCCIa) - are not soluble in water, dp: not react with
snow or other natural surfaces, and are not broken down
chemically in the lower atmosphere. While the exhaust
from the Space Shuttle and from some rockets does in-
ject some chlorine directly into the stratosphere, this
input is very small (less than one percent of the annual
input from halocarbons in the present stratosphere, as-
suming nine Space Shuttle and six Titan IV rocket
launches per year).
Several pieces of evidence combine to establish human-
made halocarbons as the primary source of stratospheric
chlorine. First, measurements (see the figure below)
have shown that the chlorinated species that rise to the
stratosphere are primarily manufactured compounds
(mainly CFCs, carbon tetrachloride, methyl chloroform,
and the HCFC substitutes for CFCs), together with small
amounts of hydrochloric acid (HCI) and methyl chloride
(CHaCI) which are partly natural in origin. The natural
contribution now is much smaller than that from human
activities, as shown in the figure below. Second, in 1985
and 1992 researchers measured nearly all known gases
containing chlorine in the stratosphere. They found that
human emissions of halocarbons plus the much smaller
contribution from natural sources could account for all of
the stratospheric chlorine compounds. Third, the in-
crease in total stratospheric chlorine measured between
1985 and 1992 corresponds with the known increases in
concentrations of human-made halocarbons during that
time.
Primary Sources of Chlorine Entering the Stratosphere
Entirely
Human-
Made
Natural
Sources
Contribute
XXIX
-------
COMMON QUESTIONS
Can Changes in the Sun's Output Be Responsible
for the Observed Changes in Ozone?
Stratospheric ozone is primarily created by ultraviolet
(UV) light coming from the Sun, so the Sun's output af-
fects the rate at which ozone is produced. The Sun's
energy release (both as UV light and as charged particles
such as electrons and protons) does vary, especially
over the well-known 11-year sunspot cycle. Observa-
tions over several solar cycles (since the 1960s) show
that total global ozone levels decrease by 1-2% from the
maximum to the minimum of a typical cycle. Changes in
the Sun's output cannot be responsible for the observed
long-term changes in ozone, because these downward
trends are much larger than 1-2%. Further, during the
period since 1979, the Sun's energy output has gone
from a maximum to a minimum in 1985 and back
through another maximum in 1991, but the trend in
ozone was downward throughout that time. The ozone
trends presented in this and previous international sci-
entific assessments have been obtained by evaluating
the long-term changes in ozone concentrations after ac-
counting for the solar influence (as has been done in the
figure below).
Global Ozone Trend (60°S-60°N)
I960 I982 I984 1986 I988
Year
I990
1992
I994
XXX
-------
COMMON QUESTIONS
When Did the Antarctic Ozone Hole First Appear?
The Antarctic ozone hole is a new phenomenon. The fig-
ure shows that observed ozone over the British Antarctic
Survey station at Halley Bay, Antarctica first revealed ob-
vious decreases in the early 1980s compared to data
obtained since 1957. The ozone hole;is formed each
year when there is a sharp decline (currently up to 60%)
in the total ozone over most of Antarctica for a period of
about two months during Southern Hemisphere spring
(September and October). Observations from three other
stations in Antarctica, also covering several decades, re-
veal similar progressive, recent decreases in springtime
ozone. The ozone hole has been shown to result from
destruction of stratospheric ozone by gases containing
chlorine and bromine, whose sources are mainly hu-
man-made halocarbon gases. ;j
Before the stratosphere was affected by human-made
chlorine and bromine, the naturally occurring springtime
ozone levels over Antarctica were about 30-40% lower
than springtime ozone levels over the Arctic. This natu-
ral difference between Antarctic and Arctic conditions
was first observed in the late 1950s by Dobson. It stems
from the exceptionally cold temperatures and different
winter wind patterns within the Antarctic stratosphere as
compared to the Arctic. This is not at all the same phe-
nomenon as the marked downward trend in total ozone in
recent years referred to as the ozone hole and shown in
the figure below.
Changes in stratospheric meteorology cannot explain
the ozone hole. Measurements show that-wintertime
Antarctic stratospheric temperatures of past decades'
have not changed prior to the development of the hole
each September. Ground, aircraft, and satellite measure-
ments have provided, in contrast, clear evidence of the
importance of the chemistry of chlorine and bromine
originating from human-made compounds in depleting
Antarctic ozone in recent years.
A single report of extremely low Antarctic winter ozone in
one location in 1958 by an unproven technique has been
shown to be completely inconsistent with the measure-
ments depicted here and with all credible measurements
of total ozone.
Historical Springtime Total Ozone Record
for Halley Bay, Antarctica (76°S)
400
300
o
Q
o>
200
o
October Monthly Averages
I955
I965
I975
Year
I985
I995
XXXI
-------
COMMON QUESTIONS
Why is the Ozone Hole Observed
When CFCs Are Released Mainly
Human emissions of CFCs do occur mainly in the North-
ern Hemisphere, with about 90% released in the
latitudes corresponding to Europe, Russia, Japan, and
North America. Gases such as CFCs that are insoluble in
water and relatively unreactive are mixed within a year or
two throughout the lower atmosphere (below about 10
km). The CFCs in this well-mixed air rise from the lower
atmosphere into the stratosphere mainly in tropical lati-
tudes. Winds then move this air poleward - both north
and south - from the tropics, so that air throughout the
stratosphere contains nearly the same amount of chlo-
rine. However, the meteorologies of the two polar
regions are very different from each other because of
major differences at the Earth's surface. The South Pole
is part of a very large land mass (Antarctica) that is com-
over Antarctica
in the Northern Hemisphere?
pletely surrounded by ocean. These conditions produce
very low stratospheric temperatures which in turn lead to
formation of clouds (polar stratospheric clouds). The
clouds that form at low temperatures lead to chemical
changes that promote rapid ozone loss during Septem-
ber and October of each year, resulting in the ozone hole.
In contrast, the Earth's surface in the northern polar re-
gion lacks the land/ocean symmetry characteristic of the
southern polar area. As a consequence, Arctic strato-
spheric air is generally much warmer than in the
Antarctic, and fewer clouds form there. Therefore, the
ozone depletion in the Arctic is much less than in the
Antarctic.
Schematic of Antarctic Ozone Hole
I979
I986'
I99I
-------
CHAPTER 1
Ozone Measurements
Lead Author:
N.R.P. Harris
Co-authors:
G. Ancellet
L. Bishop
D.J. Hofmann
J.B. Kerr
R.D. McPeters
M. Prendez
W. Randel
J. Staehelin
B.H. Subbaraya
A. Volz-Thomas
J.M. Zawodny
C.S. 2«refos
Contributors:
M. Allaart
J.K. Angell
R.D. Bojkov
K.P. Bowman
G.J.R, Coetzee
M. Degorska
J.J. DeLuisi
D. De Muer
T. Deshler
L. Froidevaux
R. Furcer
E.G. Gardiner
H. Gernandt
J.F. Gleason
U. Gorsdorf
K. Henriksen
E. Hilsenrath
S.M. Hollandsworth
0. Hov
H. Kelder
V. Kirchhoff
U. Kohler
W.D. Komhyr
J.W. Krzyscin
Z. Litynska
J.A. Logan
PS. Low
A.J. Miller
SJ. Oltmans
W.G. Planet
J.-P. Pommereau
H.-E. Scheel
J.D. Shanklin
P. Skfivankova
H. Smit
J. Waters
P. Winkler
-------
-------
SUMMARY.
1.1
1.2
1.3
1.4
INTRODUCTION
CHAPTER 1
OZONE MEASUREMENTS
Contents
.1.1
.1.5
1.5
1.5
TOTAL OZONE ;
1.2.1 Total Ozone Data Quality.,;
1.2.1.1 Ground-Based Observations 1Ł
: l.O
1.2.1.2 Satellite-Based Observations 1Q
• • l.o
1.2.1.3 Data Quality Evaluation I g
1.2.2 Trends in Total Ozone , j2
1.2.2.1 Statistical Models for Trends ! 13
1.2.2.2 Total Ozone Trendis: Updated through 1994 1 13
1.2.2.3 The Effect of the 1992-1994 Data Z"IZZZZZZ 1 18
1.2.2.4 Acceleration of Ozone Trends 1 20
1.2.3 Discussion j. , 29
OZONE PROFILES ',. j 23
1.3.1 Ozone Profile Data Quality . 1 23
1.3.1.1 Umkehr
1.3.1.2 Ozonesondes
1.23
1.23
1.3.1.2a Background Current 124
1.3.1.2b S02 :i IIIIIIIIII"II"IIlL24
1.3.1.2c Operational Changes 1 25
1.3.1.2d Intercomparisons 1 25
1.3.1.2e Correction Factors '. 1 25
1.3.1.3 Satellite Measurements of the Ozone Profile 1 26
1.3.2 Trends in the Ozone Profiled 1 27
1.3.2.1 Trends in the Upper Stratosphere 1 2g
1.3.2.2 Trends in the Lower Stratosphere 1 29
1.3.2.3 Trends in the Free Troposphere 1 31
1.3.2.4 Trends Inferred from Surface Observations 1 34
1.3.3 Discussion
.1.36
OZONE AND AEROSOL SINCE11991 } 37
1.4.1 Total Ozone Anomalies 1 37
1.4.2 Vertical Profile Information; 1 40
1.4.3 Stratospheric Aerosol after lie Eruption of Mt. Pinatubo 1.40
1.4.4 Dynamical Influences •. . 142
1.5 ANTARCTIC OZONE DEPLETION L 43
1.5.1 Introduction and Historical Data 1 43
1.5.2 Recent Observations .'. 1 44
REFERENCES 1; ! 48
-------
-------
OZONE MEASUREMENTS
SCIENTIFIC SUMMARY
The quality of the total ozone measurements made by ground-based and satellite systems has been assessed and
trends calculated where appropriate.
i
• Trends in total ozone since 1979 have been updated through early 1994:
- Northern Hemisphere middle latitude trends are significantly negative in all seasons, but are much larger in
winter/spring (about 6%/decade); than in summer/fall (about 3%/decade).
- Tropical (approx. 20°S - 20°N) trends are slightly negative, but not statistically significant when suspected drift
in the satellite data is incorporatfjid into the uncertainty.
- Southern midlatitude trends are significantly negative in all seasons, and increase in magnitude for high latitudes.
• Representative trends (annual averages, in % per decade) for north and south midlatitudes and the tropics are as
follows. '
Latitude
; Mid South Equatorial . Mid North
Recent: '
1/79 to 5/94 SBUV+SBUV/21 -4.9 ± 1.5 -1.8 ± 1.4 -4.6 ± 1.8
1/79 to 2/94 Dobson network; -3.2 ±1.3 -1.1 ±0.6 -4.8 ±0.8
I !
1/79 to 2/94 Ozonometer (former USSR) na na -4.9 ± 0.8
Pre-Pinatubo: j •
1/79 to 5/91 ' SBUV+SBUV/2! -4.9 ± 2.3 -0.8 + 2.1 -3.3 ± 2.4
1/79 to 5/91 TOMS ;. -4.5 ± 2.1 +0.4 + 2.1 -4.0 ± 2.1
1/79 to 5/91 Dobson network:1 -3.8 + 1.3 +0.2+1.2 -3.9 + 0.7
1/79 to 5/91 Ozonometer (former USSR) na na -3.8 ± 1.0
Note: Uncertainties (+) are expressed at the 95% confidence limits (2 standard errors).
ii
• The corresponding ozone loss (in %) accumulated over 15.3 years for trends calculated through 1994
are: !
Latitude
! Mid South Equatorial Mid North
i| ' .
SBUV+SBUV/27 -7.4 ±2.3 -2.7 ±2.2 -7.0 + 2.7
Dobson network/ -4.8 ±2.1 -1.7 ±0.9 -7.3 ± 1.3
Ozonometer (former USSR) na na -7.5 ± 1.3
-------
OZONE MEASUREMENTS
• There was a statistically significant increase (about 2%/decade) in the average rate of ozone depletion at the
Dobson stations north of 25°N in the period 1981-1991 compared to the period 1970-1980.
• We have confidence in the trends deduced from the ground-based network, particularly in the Northern Hemi-
sphere. The record is longer than for the satellite instruments, although the geographic coverage is patchy, with
most stations situated in the'Northern Hemisphere midlatitudes. The absolute calibration of the International
Standard Dobson spectrophotometer has been maintained at ± 1%/decade. The quality of the data from the
ground-based network has improved since the last assessment, partly as a result of improvements to the existing
records and partly as a result of the improving quality control in the ground-based network.
• An extensive revision and reanalysis of the measurements made using the filter ozonometer data from the vast
area of the former USSR has recently been performed. Trend estimates from these revised data substantiate those
made at similar latitudes by Dobson and satellite instruments.
• During the 1980s,-the Total Ozone Mapping Spectrometer (TOMS) total ozone calibration drifted by 1-2% rela-
tive to the Dobson instruments, depending on latitude. In addition, a systematic bias of 1-2%/decade may be
present in measurements made at high solar zenith angles (and so is most important at high latitudes in winter).
Our confidence in the trends presented in the 1991 Ozone Assessment, which covered the period through March
•1991, is unchanged. . ,
iHowever after this time, a problem developed in the TOMS instrument that lasted until the instrument became
inoperative in May 1993. This problem resulted in systematic errors dependent on both season and latitude, and
caused, on average, a drift of 1-2% between 1991 and 1993. TOMS satellite measurements made after May 1991
were, therefore, not used for trend analyses. A TOMS instrument was launched on the Meteor-3 satellite in
August 1991. The satellite orbit is not ideal and the measurements from this instrument have not yet been suffi-
ciently assessed to allow use in trend analyses.
The drift in the calibration of total ozone by the Solar Backscatter Ultraviolet (SBUV) instrument from January
1979 to June 1990 was 1% or less relative to Dobson instruments, and any seasonal differences in the Northern
Hemisphere were less than 1%. The SBUV/2 instrument on board the NOAA-11 satellite has measurements
available from January 1989. The drift relative to Dobson instruments in the Northern Hemisphere has been less
than 1 %. However, there is an apparent seasonal cycle in the differences of about 1-2% (minimum to maximum).
• Nearly all ground-based instruments are now on the calibration scale of the World Standard Dobson Instrument
#83. The quality of the measurements made at individual stations is tested using satellite data; any revision of the
data is based on available instrumental records. Satellite measurements are independently calibrated by checking
the internal consistency. However, the satellite record is tested for possible drift by comparison with the collec-
tion of station data. Thus, the ground-based and satellite records are~not completely independent from one
another.
Trends in the Vertical Distribution of Ozone
The state of knowledge about the trends in the vertical distribution of ozone is not as good as that about the total
ozone trends. The quality of the available data varies considerably with altitude.
1.2
-------
; OZONE MEASUREMENTS
U s AGEwnlRnv111'^? T? S°nable agrcemCnt betWCen ^ StratosPheric Ae-°l and Gas Experiment I/
I (SAGE MI), SBUV, and Umkehr, that during 1979-1991, ozone declined 5-10% per decade at 30-50°N and
- an
S°Uf™dIa<;tudes- In *e "OP-- SAGE mi gives larger trends (ca. -10% per decade) than
SBUV (ca. -5% per decade) at these altitudes.
•ji .
At altitudes between 25 and 30 km, there is reasonable agreement between SAGE I/II, SBUV Umkehr and
ozonesondes that, during the 1979,1991 period, there was no significant ozone depletion at any' latitude ' The
agreement continues down to about 20 km, where statistically significant reductions of 7 ± 4% per decade were
observed between 30 and 50°N by bod, ozonesondes and SAGE I/II. Over the longer period from 1968- 199 l*e
ozonesonde record indicates a trend of -4 ± 2% per decade at 20 km at northern midlatitudes.
1979'1991 P^ in me 15-20 km region in
f-n ™ , - m regon n
Tol Tis; d?ed istd;^7erntr the magnitude of ** reduction' with SAGE indica^ <->* « S «
-20 ±8% per decade at 16-17 km and the ozonesondes indicating an average trend of -7 ± 3% per decade in the
TOMHndT^
TOMS, and the ground-based network, but the uncertainties are too large to evaluate the consistency
1968-1991 *e
1L s°?nCSdHT rr1*1011 aItitUdCS betWee" 15 3nd 2° ""» is made diffi«"< by the small ozone
amounts. In addmon, the large vert.cal ozone gradients make the trends very sensitive to small vertical displace-
ments of the profile. The SAGE I/II record indicates large (-20 to -30% (± 18%) per decade) trends in me iS
km reglon (-10% (± 8%) at 20 km). Limited tropical ozonesonde data sets at Natal, 6°S and HUo 20^ * not
indicate s.gmficant trends between 16 and 17 km or at any other altitude for this time period '
e ' • rea U-
are large. The effect on the trend ,n rhe total column from any changes at these altitudes would be small.
* NoS freH ^^f Cre' ^ limiteC':data (a" from °*>nes0ndes) are available for trend determination. In the
Northern Hemisphere, trends are higKly variable between regions. Upward trend, in the 1970s over Europe have
deemed s.gnificantly ,n the 1980s, have been smaU or non-existent over North America, and continue upward
over Japan. The determinate of the size of the change over North America requires a proper treatment of the
relative tropospheric sensitivities for the type of sondes used during different time periods.
• Surface measurements indicate that ozone levels at the surface in Europe have doubled since the 1 950s. Over the
last two decades there has been a downward trend at the South Pole, and positive trends are observed at high
al itude sues in the .Northern Hemisphere. When considering the latter conclusion, the regional nature of trend
the Northern Hemisphere must be borne in mind.
Observations of Ozone and Aerosols in 1 991 -1 994
• Global total ozone values in 1992/93 were 3-4% lower than the 1980s average. If the trend, solar cycle and quasi-
biennial osc, lation (QBO) effects inferred from the 1980s record are extrapolated, an additional global anomaly
ot oetween - 1 and -2% remains.
• The most negative anomalies were observed in the Northern Hemisphere springs in 1992 and 1993 with peak
deviations of 6- 10% in February-April 1993. '
1.3
-------
OZONE MEASUREMENTS
A reduction of 3-4% occurred in the tropics in the six months following the eruption of Mt. Pinatubo.
Overall the smallest effects were observed in the extra-tropical Southern Hemisphere, where total ozone amounts
were at the low end of the range observed in the 1980s, as would be expected from the long-term downward trend
observed in that region.
In 1994, global ozone levels are also at the low end of the 1980s range, again in line with expectations of a
continuation of the observed long-term trend.
Following the June 1991 eruption of Mt. Pinatubo, stratospheric aerosol levels increased globally, with northern
midlatitude peak particle surface areas increasing by factors of 30-40 above pre-emption values about one year
after the eruption. Since that time, they have been decreasing.
Several mechanisms have been suggested as causes of the total ozone anomalies, though the relative importance
is not yet clear. The possible influences include: radiative, dynamical, and chemical perturbations resulting from
the Mt. Pinatubo volcanic aerosol; and global and regional dynamical perturbations, including the El Nino-South-
ern Oscillation.
Antarctic Ozone Depletion
Record low mean values for October were observed at three Antarctic ground-based stations with continuous
records since the late 1950s and early 1960s. There is no evidence of major springtime ozone depletion in
Antarctica at any of the four Dobson stations prior to 1980.
In early October 1993, a record low daily value of total ozone of 91 ± 5 Dobson units was observed with an
ozonesonde at the South Pole. During this flight (and in several others), no detectable ozone (less than 1%) was
found over a 5 km range from 14 to 19 km, implying that complete chemical destruction of ozone had occurred.
The geographical extents of the ozone holes in 1992 and 1993 were the two largest on record.
• A comparison of ozonesonde measurements made at the South Pole from 1967-1971 with those made between
1986 and 1991 reaffirms that the Antarctic depletion that has developed since the early period occurs at altitudes
between 14 and 20 km, and that the largest changes occur in-September, October, and November.
1.4
-------
OZONE MEASUREMENTS
1.1 INTRODUCTION |
Ozone in the atmosphere is easy t6 detect. Several
techniques have been successfully used: most are opti-
cal, using absorption or emission of; light in many
regions of the spectrum; others are chemical; and some
are a mixture of the two. However, while it is relatively
easy to detect ozone in the atmosphere, i[ has proved dif-
ficult to make sufficiently precise and numerous
measurements to determine credible changes of a few
percent on a decadal time scale. Difficulties include:
knowing what the absolute calibrations of the instru-
ments are and how they change with time; assessing how
much variability in any set of measurements is caused by
the instrument and how much by the natural variability
in the atmosphere; and making meaningful comparisons
of measurements made by different instruments, espe-
cially when different techniques are used. Detailed
descriptions of the major techniques and instruments
were given in the report of the International Ozone
Trends Panel (IOTP) (WMO, 1990a) and are not repeat-
ed here. \
We first consider the quality of total ozone mea-
surements, particularly those made by the ground-based
observing network, the Total Ozone Mapping Spectrom-
eter (TOMS), and the Solar Backsca.tter Ultraviolet
spectrometers (SBUV). The ground-based and satellite
instruments have proven invaluable in assessing each
others' data quality. Nearly all ground-based instru-
ments are now on the calibration scale of the World
Standard Dobson Instrument #83. The quality of the
measurements made at individual stations is tested using
satellite data; any revision of the data is based on avail-
able' instrumental records. Satellite measurements are
independently calibrated by checking this internal con-
sistency. However, the satellite record is tested for
possible drift by comparison with the collection of sta-
tion data. Thus, the ground-based and satellite records
are not completely independent from one another.
Given this perspective, we next present the trends
in total ozone calculated to May 1994. Special attention
is paid to how the trends are affected by the record low
ozone values that were observed in 1992 and 1993. This
theme is taken up again later, in Section 1.4, where we
describe the ozone changes seen in this period. The evo-
lution of stratospheric aerosol following the eruptions of
Mt. Pinatubo in June 1991 and Volcan Hudson in August
J991, and possible links with the low ozone values, are
briefly discussed, along with other potentially important
influences on ozone at this time.
In Section 1.3 we discuss the quality of the various
techniques (remote and in situ) that measure the vertical
distribution of ozone in the atmosphere. Although
progress has been made, a good deal of work remains
before a clear picture can emerge, especially in the re-
gion near the tropopause, which is so important in
determining the impact of ozone changes on climate.
Last, the development of the Antarctic ozone hole
in 1992 and 1993 is described in Section 1.5, together
with some new analyses of some old measurements.
1.2 TOTAL OZONE
1.2.1 Total Ozone Data Quality
Total column ozone has been measured using
Dobson instruments since the 1920s. The number of
monitoring stations has increased through the years, and
since the 1960s a large enough network has existed to
monitor ozone over most of the world with particularly
good coverage in the northern midlatitudes and in Ant-
arctica. Truly global monitoring has been possible only
since the introduction of satellite-based instruments.
The 1988 IOTP (WMO, 1990a) examined the quality of
ozone measurements from both ground-based systems
(Dobson, M83, MI24) and satellite systems. They re-
ported great variability in the quality of the data from
ground-based instruments and found large calibration
drifts in the SBUV and TOMS instruments caused by
imperfectly corrected degradation of the on-board dif-
fuser plates. When the 1991 assessment of ozone trends
was made (WMO, 1992a), improvements in the quality
of ozone data were noted. The re-evaluation of historical
Dobson data records initiated by Bojkov et al. (1990)
had been carried out at a small number of stations. Sim-
ilarly, the quality of the satellite data had improved,
though unresolved problems were still apparent. The
entire TOMS ozone data record had been reprocessed
using the version 6 algorithm, which improved the in-
strument calibration through the requirement that ozone
amounts measured by different wavelength pairs main-
tain relative stability (Herman et al., 1991). Comparison
with the World Standard Dobson instrument number 83
7.5
-------
OZONE MEASUREMENTS
(183) at Mauna Loa indicated that good, long-term preci-
sion had been achieved (McPeters and Komhyr, 1991).
1.2.1.1 GROUND-BASED OBSERVATIONS
Since January 1992, all ground-based measure-
ments have been reported using the Bass and Paur (1985)
ozone cross sections. This change should increase the
accuracy of the ozone record for direct comparison with
other measurement systems, but should have no effect on
the core time series of observations made with the AD
wavelength pairs since the conversion from the old Vi-
groux (1953) scale is defined (Komhyr et ai, 1993).
Since the last assessment the number of Dobson stations
at which the historical records have been reanalyzed by
the responsible personnel has increased to over 25, with
many more in the process of reanalysis.
In a full re-evaluation, the station log books and
lamp calibration records are carefully examined and cor-
rections are made where appropriate (WMO, 1992b).
Measurements are treated on an individual basis, in con-
trast to the "provisionally revised" data described and
used in IOTP and subsequent assessments where month-
ly averages are treated. Comparisons with external data
sets (total ozone records in the same synoptic region,
meteorological data and, since 1978, satellite overpass
data) are made to identify periods where special atten-
tion should be paid. The data are only corrected if a
cause is found based on the station records. The goal of
re-evaluation is to produce a high quality, long-term total
ozone record. Increasingly frequent international inter-
comparisons of ground-based instruments bring more
consistency to the global network. Recent intercompari-
sons were made at Arosa (Switzerland) in 1990, at
Hradec Kralove (Czech Republic) in 1993, and at Izana
(Canary Islands) in 1994. In addition, the practice of
using traveling standard lamps to check the calibration
of individual instruments has become more frequent in
recent years, with a consequent reduction in the ob-
served scatter (Grass and Komhyr, 1989; WMO, 1994a).
Several important concerns about the quality of
the ground-based data remain - in particular, how reli-
able are trends determined from Dobson data in the
1960-1980 period? While the program to reanalyze
Dobson records is important, there are limits to what can
be achieved. Not only is sufficient information not avail-
able in many cases, particularly in the early years, but,
even in recent TOMS overpass comparisons, apparent
calibration shifts are identified for which no cause has
been found. Another issue of concern is whether uni-
form data quality can be maintained when a Dobson
instrument is replaced by a Brewer. Brewer instruments
replaced Dobsons at 4 Canadian sites (Churchill, Ed-
monton, Goose Bay, and Resolute) in the mid-1980s.
During the changeover at each site, both instruments
were operated for a period of at least 3 years in order to
quantify possible biases and differences in seasonal re-
sponse. In order to ensure continuity, a simulated
Dobson AD direct sun measurement is reported for these
sites (Kerr et al., 1988). The data records for these sites
must be monitored for possible biases and differences in
seasonal response that might affect trend analyses.
In Section 1.2.2, trend analyses of the measure-
ments from 43 stations are reported. The records from
many more were examined for possible inclusion, but
were not used for a variety of reasons. First, only
records starting before 1980 were considered sufficient-
ly long for meaningful analyses to be made. Second, a
minimum of 12 days of observation were required for a
monthly mean to be included. In the case of three high
latitude stations, all midwinter monthly means were
missing, a situation that cannot be handled by the cur-
rent, well-documented statistical technique, at least as
far as computation of seasonal average trends is con-
cerned. For this assessment, no analysis of data from
such stations is made. Third, some station records show
large variations against nearby stations or satellite over-
passes that cannot be explained in terms of any natural
phenomena. These records are few and were not used. A
number of points requiring corrections have been identi-
fied in the measurements submitted to the World Ozone
Data Center. The re-evaluations used in the records used
here for trend analysis will be documented in WMO Re-
port No. 35 (appendix by Bojkov). About half of these
corrections result from the WMO intercomparison pro-
gram and individual instrument's calibration procedures,
and most of the remainder are made from information
made available from the instrument log books by the op-
erating agency (Bojkov, private communication). A few
obvious calibration shifts- for which no instrumentally
derived correction can be found are treated in the statisti-
cal analysis (see Section 1.2.2). An empirical technique
has been used to correct the air mass dependencies at a
few stations, as insufficient instrumental information
1.6
-------
OZONE MEASUREMENTS
exists in these cases (see appendix by Bojkov in WMO,
1994b). .<:
The chief instrument used in the former USSR was
a filter ozonometer. Various improvements have been
made over the years, and the record of l:he M-83 and
M-124 versions since 1973 has been assessed by Bojkov
et at. (1994) using recently available information on the
individual instruments' performance and calibration his-
tories. The errors associated with these instruments are
larger than those of Dobson or Brewer instruments, and
Bojkov et al. combine the individual station data into re-
gional averages. ;;
Sulfur dioxide (SO2) absorbs ultraviolet at the
wavelengths used by Dobson and Brewer instruments to
measure total ozone. The presence of SC>2 causes a false
increase of total ozone measured by Dobson instruments
for both the AD and CD wavelengths. As part of a de-
tailed revision of the total ozone record at Uccle,
Belgium, De Muer and De Backer (1992) considered the
effect of the locally measured reduction in1 surface SC>2
on the total ozone record from 1972 to 1991. Over this
period, surface SOj levels dropped by a factor of about
5. The size of the downward correction to the observed
total ozone was found to be 3-4% in 1972 and just under
1% in 1990, a change of similar size to the trend calcu-
lated in IOTP (WMO, 1990a). The trends calculated
using the revised data for 1978-1991 are in reasonable
agreement with TOMS version 6 overpass measure-
ments (WMO, 1992). ;
This analysis clearly raises the questibn as to how
many records might be similarly affected (De Muer and
De Backer, 1993). Most North American stations are in
unpolluted areas and measurements made there will not
have been influenced by tropospheric SO2- In Canada,
surface SO2 measurements made since 197|4 are report-
ed by Environment Canada in the National Air Pollution
Surveillance Series for sites in Toronto, the worst affect-
ed station in Canada. In 1974 the average:'surface SO2
concentration in Toronto was 42 u.g m"3, about 40% that
measured near Uccle in the same year. A review of these
data indicates that about 1% (3-4 Dobsoij units, DU)
false total ozone may have occurred at Toronto in the
early part of the record. This dropped to 0:3% (1.2 DU
false ozone) in the early to mid-1980s and has remained
level since then, in good agreement with Kerr et al.
(1985, 1988). There is greater uncertainty in the earlier
data made by wet chemical instruments (which may be
sensitive to other pollutants besides SO2) than there is in
the later data made by pulse fluorescence techniques.
Similar measurements made at Edmonton indicate that
interference due to SO2 is less than 0.2% throughout the
record (in good agreement with Kerr et al., 1989). The
effects of SO2 on the three other non-urban sites in Can-
ada are thought to be negligible. In the United States,
anthropogenic emissions of SO2 decreased by 27-29%
from 1970 to 1988 (Placet, 1991). None of the U.S. sta-
tions is in as heavily populated a region as Uccle and so
should not have been as affected.
A model study of SO2 concentrations in Europe,
based on emission estimates, indicated that the largest
changes in SO2 concentrations since 1970 have occurred
over Belgium, Holland, and Northern France, and that in
1960 the SO2 concentrations calculated for Belgium
were among .the highest in Europe (Mylona, 1993). De-
creases by a factor of 50-75% were calculated for this
region, while elsewhere in Europe and Scandinavia the
reductions since 1970 seem to have been 50% at most
and are often less. Thus while some Dobson measure-
ments in Europe were affected by the decreasing SO2
concentrations, it is likely that Uccle is one of the most
heavily influenced.
Elsewhere, stations are in polluted regions where
the SO2 trends are different from those in Europe and
parts of North America. Work still needs to be done to
assess the impact of SO2 on ©3 measurements at many
individual stations.
In June of 1991 the eruption of Mt. Pinatubo re-
sulted in the injection of large amounts of material into
the stratosphere. The plume included large amounts of
SO2, but this had decreased to low levels by the end of
July (Bluth et al., 1992). Of greater concern is the high
level of stratospheric aerosol that spread over the globe
and produced large aerosol optical depths for more than
a year. But Komhyr (private communication) notes that
the data record from the World Standard Dobson instru-
ment 1-83 shows little apparent disturbance when the
initial, dense aerosol cloud passed over Mauna Loa Ob-
servatory in early July. The initial error appeared to be
only a tenth of a percent or so. A small change (< 1 %) in
the calibration of 1-83 was seen in June 1992. In June
1993, when the stratosphere over Mauna Loa was much
cleaner, the calibration of 1-83 was the same as in 1991.
Thus ozone measurement errors due to Mt. Pinatubo
aerosols most likely did not exceed ±1%, for direct sun
1.7
-------
OZONE MEASUREMENTS
observations made by a well-maintained Dobson instru-
ment using the fundamental AD wavelength pairs. This
result should be expected, as the wavelength pairs were
originally chosen to minimize the effect of aerosol on the
measurement (Dobson, 1957).
1.2.1.2 SATELLITE-BASED OBSERVATIONS
Total ozone data are now available from a number
of satellite systems. The Nimbus 7 TOMS produced glo-
bal ozone maps (except in polar night) on nearly every
day from November 1978 until May 6, 1993, when the
instrument failed. Another TOMS instrument was
launched on the Russian Meteor 3 spacecraft in August
of 1991 and continues to operate, so a continuous TOMS
data record has been maintained, although, because of its
drifting .orbit, the geographic coverage of the Meteor 3
TOMS is not as extensive as that of Nimbus 7 TOMS.
TOMS has been used as the "most reliable" satel-
lite-based monitor of total ozone because it gives daily
global coverage and has a 14.5-year record of observa-
tions. The version 6 TOMS data were produced using a
calibration based on data up through May 1990, and
there is concern that its calibration may have drifted
since then. This issue will be addressed through compar-
isons with other instruments. There is a known error at
large solar zenith angles (>70°) demonstrated by com-
parison with Systeme d'Analyse par Observation
Zenithale (SAOZ) spectrometers (Pommereau and Gou-
tail, 1988), which make zenith sky measurements of
ozone at sunset and sunrise and thus avoid the concerns
about airmass or temperature dependencies that arise
with the shorter wavelengths used in the Dobson, Brew-
er, TOMS, and SBUV instruments. This error in TOMS
is caused by a dependence on the shape of the ozone pro-
' file when the ultraviolet light, used to measure the
ozone, no longer penetrates well to the ground. For the
Nimbus 7 TOMS, this problem is only important at high
latitudes in the winter hemisphere. Wetlemeyer et al.
(1993) estimate that the 60° latitude winter trend will be
in error by less than 1-2% per decade; errors at lower
latitudes should be insignificant.
The sensitivity of TOMS to volcanic aerosol has
been analyzed in detail (Bhartia et al., 1993). There are
systematic errors depending on scan angle, but on a zon-
al mean basis the errors largely cancel. Aerosol-related
effects on the TOMS observation were only observed in
the tropics for a few months, so there should not be a
significant effect on trends.
Data from Meteor 3 TOMS have been available
since its launch in August 1991, but the consistency of
the data from the two TOMS instruments (Nimbus 7
TOMS and Meteor 3 TOMS) has not been properly as-
sessed yet The comparison is complicated by the orbit
of Meteor 3, which drifts from near-noon observations to
near-terminator observations every 53 days. Periodical-
ly, all data from Meteor 3 TOMS are collected at very
large solar zenith angles, so that the problems connected
with high latitude measurements occur at all "latitudes.
In the light of these problems and the lack of a more de-
tailed assessment of the data quality, no use is made of
the Meteor 3 TOMS measurements for the trends pre-
sented in this assessment.
The SBUV instruments also measure total ozone,
viewing directly below the orbital track of the space-
craft. The SBUV instrument on Nimbus? operated from
November 1978 to June 1990. While SBUV and TOMS
were separate instruments, they shared the diffuser plate
measuring the extraterrestrial solar flux and so did not
have completely independent calibrations. The same
basic algorithm is used to calculate both the TOMS and
the SBUV total ozone measurements. Data are also
available from the NOAA-11 SBUV/2 beginning in Jan-
uary 1989 through May 1994. The SBUV/2, which has
suffered much less degradation than SBUV, maintains
calibration using on-board calibration lamps and com-
parison with periodic flights of the Shuttle SBUV
instrument (Hilsenrath et al., 1994), and so its data
record is truly independent of the other systems. There
is a concern that, as the NOAA-11 orbit has drifted from
an initial 1:30 PM equator crossing time to a 4:30 PM
equator crossing time in 1994, zenith angle dependent
errors could be aliased into the ozone trend from SBUV/2.
The TOVS (TIROS Operational Vertical Sounder)
instruments (flown on a number of platforms) monitor
total ozone using the 9.6 u.m channel, which makes them
most sensitive to ozone near the ozone maximum. This
fact and the unresolved problem of possible calibration
differences between the series of TOVS instruments lim-
it the current usefulness of TOVS for trend analysis.
1.8
-------
OZONE MEASUREMENTS
TOMS vs Hohenpeissenberg
93
1.2.1.3 DATA QUALITY EVALUATION •!
The large natural variations in ozipne complicate
the evaluation of the quality of total ozone measure-
ments. The comparison of simultaneous measurements
of the same quantity by independent instruments is an
effective means of checking the quality of the individual
instruments. For the early Dobson record there are no
independent, simultaneous1 measurements of total ozone
except during rare intercomparisons. (An exception oc-
curred at Arosa, where two instruments have been
operated simultaneously since 1968). The quality of the
early record thus depends on how well the individual in-
struments and their calibrations were maintained.
Evaluations of the early Dobson records are based on
comparisons with data from other stations in the same
synoptic region, with meteorological dzita such as the
100 hPa temperature series, and critical examination of
available log books. Such methods were discussed at
length in the IOTP (WMO, 1990a) and have been de-
scribed further in WMO Report No. 29 (1992b). Details
of the calibration histories at individual stations will be
published in WMO Report No. 35 (1994b).
Since the launch of TOMS in 1978, a total ozone
•measurement has been made almost daily from space
within 1° of every Dobson station. Figure 1-1 shows an
example of a TOMS-Dobson comparison for Hohen-
peissenberg, Germany. Similar comparisons have been
made for each of 142 ground-based stations (Dobson,
Brewer, and M-124) with relatively complete records
over the life of TOMS (Ozone Data for the World, 1993).
A single such comparison shows the relative differences
between the two measurement systems; examination of
many such plots can reveal the cause for differences be-
tween the systems. Changes relative to TOMS that
occur at one station but not at other nearby stations can
be presumed to be caused by that one station, but a
change that is seen at most stations can be presumed to
be caused by TOMS. Two simple indicators of data
1.9
-------
OZONE MEASUREMENTS
quality that can be derived from these plots are the aver-
age bias and drift relative to TOMS. Figure 1-2 shows
the first-year bias and trend relative to TOMS of 18 Dob-
son stations, including 183 in its measurements at Mauna
Loa each summer. The average offset of TOMS relative
to Dobson of 3-4% is almost certainly due to small pre-
launch calibration errors in TOMS. The scatter in this
diagram is noticeably less now that revised total ozone
records are used, indicating an improvement in the qual-
ity of these measurements. The average drift of TOMS
relative to the Dobson network of about -2% per decade
is discussed below.
TOMS - ground stations
x>
s
0)
5'2
o
9-,
U)
O
Edmonton 133
Poona
^Toronto Per'h
Melbourne ° Wallops
Belsk
Delhi
Hohenp'brg Tateno
Oslo
Reykjavik
0 2 *
TOMS - Dobson Bias (%)
Figure 1-2. The average bias relative to TOMS in
the first year (usually 1979) and the drift relative to
TOMS over 14 years for a sample of 18 Dobson
stations. The Dobson station data were taken from
the World Ozone Data Center in December 1993.
183 at Mauna Loa and the regular Mauna Loa
record are shown separately.
Despite the variability of individual Dobson sta-
tions, random errors should largely cancel in a network
of Dobson stations, so that conclusions can be made
about the performance of TOMS. Figure 1-3 shows
comparisons of TOMS with ground-based measure-
ments, including 183 both at Mauna Loa and at Boulder,
a network of 30 Northern Hemisphere (25-60°N) Dob-
son'stations that have complete data records through
May 6, 1993, and summer-only averages for the same
stations. TOMS is stable relative to 183 over its life. The
error bars shown for the 183 comparisons are statistical
uncertainties (95% confidence limits) for each summer's
Figure 1 -3. Percent difference between TOMS and
World Standard Dobson #83, at both Mauna Loa
(solid circles) and at E3oulder (empty circles);
monthly average differences for an average of 30
Northern Hemisphere Dobson stations; and sum-
mer only (JJA) differences for the same stations
(squares). The uncertainties shown are 95% confi-
dence limits for the mean value.
set of match-ups; the ±0.5% or so year-to-year variation
represents the limit of accuracy for a single site compar-
ison, since many errors are systematic and not random as
the statistical error calculation assumes. A preliminary
comparison of 183 observations made in Boulder, where
fewer measurements were made with 183, shows a drift
relative to TOMS that is very similar to that seen in the
30-stations average, which implies a TOMS latitude (or
zenith angle) dependent drift. The comparison with the
ensemble of 30 Northern Hemisphere Dobson stations
was made using monthly averages. There is a seasonal
cycle in the TOMS-Dobson difference of about 1% am-
plitude in 1985 and increasing thereafter.
- An initial decline of TOMS ozone relative to Dob-
son (or increase in Dobson ozone relative to TOMS)
between 1979 and 1984 is followed by a period of appar-
ent lesser drift between 1984 and 1990 and, after 1990, a
significant decline of about 2'/2%. Evidence of this de-
cline beginning in about 1989 can also be seen in Figure
1-1, the comparison with Hohenpeissenberg. The initial
decline of TOMS relative to Dobson could be caused by
an error in TOMS not resolved by the internal calibration
method or, possibly, it could be partly due to a change in
the average calibration of the Dobson network in the
1.10
-------
OZONE MEASUREMENTS
TOMS - satellite
93
rM 1"1" We^'y avera9e differences between TOMS and SBUV, NOAA-11 SBUV/2, and Meteor 3
TOMS, and monthly average differences between TOMS and TOVS.
early 1980s before the strong program of intercompari-
son was extended. Figure 1-3 shows that; such a decline
is common to the Dobson records for most stations. Pos-
sible solar zenith angle dependent errors (in either
Dobson or TOMS) can be minimized by comparing
summer average values .(where summer is defined as
June-July-August). There is a similar time dependence,
though of lesser magnitude" (1-1.5%). Most of the sea-
sonal cycle must then be due to TOMS. The decline of
TOMS relative to 183 at Boulder, coupled with the stabil-
ity relative to 183 at Mauna Loa, indicates a TOMS error
that depends on the signal level, because UV signal lev-
els are generally lower at Boulder (more pzone) than at
Hilo. It is most likely that the TOMS photomultiplier
has developed a small nonlinearity in it;;: response that
,has increased with time. If true, the equatorial and sum-
mer midlatitude trends from TOMS should be accurate,
but the high and winter midlatitude trends could be too
large by 1-2% per decade. ;;
Comparisons with TOMS have been done with an
average of 9 Brewer stations (not shown). The data
record is simply not long enough for definitive compari-
sons, but the seasonal dependence is larger, probably
because the Brewers tend to be at high latitude sites.
There is a decline between 1990 and 1992 that is consis-
tent with the Dobson results.
Figure 1-4 is a comparison of Nimbus 7 TOMS
with other satellite instruments: Nimbus-7 SBUV,
NOAA-11 SBUV/2, TOVS, and Meteor 3 TOMS. Com-
parisons have been done of weekly zonal mean ozone,
except for TOVS, where monthly means are used. The
comparisons for the 30°-50°N, 30°-50°S, and 20°S-20°N
zones are shown. Although 3% higher than SBUV,
TOMS is quite stable relative to SBUV, not surprising
since both were recalibrated using similar techniques
and the two instruments use the same diffuser plate, al-
beit with different viewing geometries. There is a
seasonal variation of about 1% magnitude that again is
-------
OZONE MEASUREMENTS
likely caused by nonlinearity in the TOMS photomulti-
plier. There is no evidence for nonlinearity in the SBUV
photomultiplier. TheNOAA-11 SBUV/2 calibration is
completely independent and is maintained through use
of on-board calibration lamps. There is a decline of
TOMS relative to SBUV/2 of 1% or so between 1989
and 1993. A comparison of SBUV/2 with an ensemble
of ground-based stations between 20° and.60°N indicates
that there has been little drift and that there is an apparent
seasonal cycle of about 1-2% (minimum to maximum).
Finally, comparisons with monthly average TOVS
zonal means for 30°-50°N are shown (Figure 1-4). The
TOVS data show significant variance, presumably re-
sulting from the sensitivity to stratospheric tempera-
tures, and cannot currently be used for trend analysis.
1.2.2 Trends in Total Ozone
Trends in total ozone were reported in the last as-
sessment (WMO, 1992a; see also Stolarski et al, 1992),
using TOMS satellite data from November 1978 through
March 1991, and ground-based data through March
1991 where available. A number of recent studies have
examined the available records, either on large scales
(KrzyScin, 1992,1994a; Reinsel et al, 1994a) or at indi-
vidual stations (Deg6rska et al, 1992; Henriksen et al,
1992,1993; Kundu and Jain, 1993; Lehmann, 1994). In
addition, a number of studies investigated the effects of
interannual variability, and its various causes, on total
ozone trends (Hood and McCormack, 1992; Shiotani,
1992; Marko and Fissel, 1993; Krzyscin, 1994b, c; Ran-
del and Cobb, 1994; Zerefos et al., 1992, 1994). In
general, the conclusions of these studies agree well with
those presented in WMO (1992a) and here. One excep-
tion is the analysis by Henriksen et al. (1992, 1993) of
the total ozone record from Troms0 (70°N). Measure-
ments have been made there using a Dobson
spectrophotometer that show no long-term change from
1939 to 1989. Two difficulties arise in the interpretation
of this record. First, there is a gap between 1969 and
1984 during which the instrument was overhauled. Un-
fortunately the amount of adjustment caused by this
Overhaul cannot be given (Henriksen et al., 1992). Sec-
ond, the natural variability of ozone is such that there are
geographic differences in the trends (WMO, 1992a), so
that one would expect the trends measured at some indi-
vidual stations to be zero.
For this assessment, trends have been updated
through the most recent available data. The trend update
is complicated by the failure of the Nimbus 7 TOMS in-
strument on May 6, 1993, and concerns about the
correction of its calibration after 1990 (see Section
1.2.1.3). However, SBUV data have been re-evaluated
since the 1991 assessment, and are now suitable for trend
analysis when combined with the SBUV/2 data from the
NOAA-11 satellite. In the following section, trend anal-
yses of SBUV data extended with SBUV/2 after 1988,
abbreviated SBUV(/2), are updated through May 1994.
Trends from the Dobson network are updated
through February 1994 at the majority of stations, and
several new stations have been added. In addition, since
the 1991 assessment, a number of Dobson stations have
revised data for part or all of their historical records
based on detailed re-evaluations. These data have been
used if submitted to the World Ozone Data Center or di-
rectly to the chapter authors, in addition, at some
stations, revisions were made by R. Bojkov (private
communication) from the WMO intercomparison pro-
gram results or from information in the station log books
(see Section 1.2.1.1). Furthermore, data from 45 filter
ozonometer stations in the former USSR have been
thoroughly assessed and revised by Bojkov et al. (1994).
Regional average data for the four regions discussed in
that paper have been obtained from the authors and
trends calculated using the same statistical fit as for the
Dobson stations; the trends calculated for this report are
close to those tabulated by the authors.
As discussed in detail in Section 1.4, ozone levels
declined a few months after the eruption of Mt. Pinatubo
in June 1991, and at northern midlatitudes they remained
abnormally low through the fall of 1993 (Gleason et al.,
1993; Herman and Larko, 1994; Bojkov et al, 1993;
Kerr et al, 1993; Komhyr et al, 1994a). Whatever the
cause of these low values, the calculation of trends with
abnormally low data at the end of the time period may
lead to substantially more negative values for the calcu-
lated trend. This presents difficulties in interpretation of
the results, since the use of the word "trend" implies a
generally consistent, continuing change over a given
period. By the inclusion of very recent data in late 1993
and the first half of 1994, this effect is lessened, except in
the Jun-Jul-Aug season where the very low 1993 data are
at the end of the series. Section 1.2.2.3 compares
SBUV(/2) trends through May 1991 versus trends
1.12
-------
OZONE MEASUREMENTS
through May 1994 as an analysis of the effect of includ-
ing this period of anomalous ozone. ;
1.2.2.1 STATISTICAL MODELS FOR TRENDS
i
As discussed in previous reports (WMO 1990a;
WMO 1990b; WMO 1992a), proper trend analysis of
ozone series uses a statistical regression [model that fits
terms for seasonal variation in mean ozone, seasonal
variation in ozone trends, and the effects of other identi-
fiable variables such as the 11-year solar cycle,
quasi-biennial oscillation (QBO), and atmospheric nu-
clear tests (if data from the early 1960s are used). The
residuals from the model are autocorrelated, and this
autocorrelation should be fitted as part of the statistical
estimation procedure to ensure reliable standard errors
for the calculated trends (see, for example;, Reinsel el al.,
1987; 1994a; Bojkov et a/., 1990). Also, proper error
analysis requires a weighted regression t
-------
OZONE MEASUREMENTS
Table 1-1 Set of 43 Dobson stations used for the trend analyses, with dates of usable data
station authorities, Rev = revised as discussed in Section 1 .2.1 .1 .
St. Petersburg
Churchill
Edmonton
Goose
Belsk
Uccle
Hradec Kralove
Hohenpeissenberg
, Caribou
Arosa
Bismarck
Sestola
Toronto
Sapporo
Vigna Di Valle
Boulder
Shiangher
Lisbon
Wallops Island
Nashville
Tateno
Kagoshima
Quetta
C^iro
New Delhi
Naha
Varanasi
Kunming
Ahmedabad
Mauna Loa •
.Kodaikanal
Mahe
Natal
Huancayo
Samoa
Brisbane
Perth
Buenos Aires
Aspendale
Hobart
Invercargill
MacQuarie Island
60.0 N
58.8 N
53.6 N
53.3 N
51.8 N
50.8 N
50.2 N
47.8 N
46.9 N
46.8 N
46.8 N
44.2 N
43.8 N
43.1 N '
42.1 N
40.0 N
39.8 N
38.8 N
37.9 N
36.3 N
36.1 N
31.6 N
30.2 N
30.1 N
28.7 N
26.2 N
25.3 N
25.0 N
23.0 N
19.5 N
10.2 N
1.3 N
4.7 S
5.8 S
12.1 S
14.3 S
27.4 S
31.9S
34.6 S
38.0 S
42.8 S
46.4 S
54.5 S
68-08
65-01
58-03
62-01
63-04
71-07
62-03
68-05
62-09
57-07
62-12
76-11
60-01
58-02
57-07
76-09
79-01
67-08
57-07
62-08
57-07
63-02
69-08
74-11
75-01
74-04
75-01
80-01
59-01
64-01
76-08
79-02
75-11
78-12
64-02
75-12
57-07
69-03
65-10
57-07
67-07
70-07
63-03
Last
94-02
93-10
94-02
94-02
93-12
94-02
94-02
94-02
94-02
94-02
94-02
94-02
94-02
94-02
94-02
94-02
93-08
94-02
94-02
94-02
94-02
94-02
93-02
94-02
94-02
94-02
94-02
94-02
92-12
94-02
94-02
93-10
93-10
94-02
92-12
94-02
93-07
94-02
94-02
93-07
92-04
94-02
93-06
Dec-Feb
est 2se
-5.7 4.2
-5.6 4.7
-2.8 5.4
-9.1 5.4
-5.9 5.4
-7.3 5.3
-8.4 4.7
-5.3 4.4
-5.9 4.7
-1.9 3.5
-5.4 4.7
-4.5 3.7
-6.8 3.7
-8.0 4.3
-2.5 3.2
-5.1 3.2
-1.3 3.4
-6.5 3.5
-5.0 3.3
-3.6 3.7
-2.6 3.1
-5.3 4.3
-1.7 4.0
-2.2 3.3
-2.3 3.0
-2.2 2.4
-0.5 2.6
-1.1 2.7
-0.6 3.4
1.1 2.6
1.03.1
-0.7 1.8
-0.3 2.5
-0.7 1.7
-1.6 1.9
-2.2 1.8
-0.4 1.4
-2.1 1.5
-2.9 1.6
-4.4 2.1
-5.2 1.6
-6.8 2.6
Mar-May
est 2se
-6.9 3.4
-7.6 3.3
-5.7 4.4
-6.7 4.0
-7.4 3.8
-6.4 3.8
-6.1 4.6
-6.5 2.7
-4.5 3.8
-6.8 2.9
-6.8 4.0
-5.9 2.8
-5.6 3.1
-5.5 5.1
-7.5 3.2
-3.8 3.6
-6.7 2.8
-5.4 3.5
-4.4 3.9
-1.23.3
-1.83.1
-1.64.2
-3.1 3.0
-2.0 3.2
-2.0 2.9
-1.42.5
-1.8 3.5
-1.6 3.4
0.2 3.1
0.2 2.6
-0.4 4.0
-1.0 2.4
1.6 2.0
-1.4 2.0
-2.5 1.8
-2.1 1.7
-1.7 2.0
-1.4 2.4
-3.5 1.6
-5.2 2.7
-2.02.1
-3.4 3.0
Jun-Aug
est 2se
-4.5 2.3
-5.5 2.2
-7.6 3.0
-4.0 2.4
-1.3 2.4
-4.4 2.4
-3.6 2.6
-2.6 2.2
-2.2 2.0
-2.1 2.3
-4.3 2.2
-2.7 1.8
-4.0 2.6
-3.8 2.4
-1.7 1.6
-0.4 2.7
-4.1 1.7
-4.4 2.2
-2.9 2.6
-0.8 2.2
-0.6 1.9
0.7 2.7
-0.2 1.6
0.3 2.9
-0.3 1.7
-0.2 2.5
0.2 1.8
-4.3 1.7
-0.1 2.3
-0.8 2.8
-1.1 3.0
-2.0 2.5
-1.6 2.4
-3.4 2.8
-1.33.1
-1.8 3.6
-1.4 3.4
-4.2 3.3
-3.2 2.8
-5.2 3.4
-1.2 2.6
-6.5 4.8
Sep-Nov
est 2se
-2.4 3.2
-3.4 3.2
-3.4 2.8
-1.4 3.2
-0.3 3.4
-0.8 2.9
-2.03.1
-3.1 3.3
-1.1 2.6
-1.82.1
-0.9 3.0
-0.5 3.1
-2.2 2.6
-4.8 2.7
-1.7 2.6
-1.0 2.8
-1.5 2.7
-3.0 3.3
-1.33.1
0.5 2.3
0.3 2.0
-0.2 2.5
-0.9 1.6
-0.4 1.5
-1.02.0
-1.2 1.9
-1.2 1.7
1.2 2.5
-0.4 1.9
-1.02.9
-l.l 3.3
-1.7 2.3
-l.l 2.4
-0.5 2.1
-1.9 2.5
-1.9 2.4
-0.9 2.0
-2.0 3.4
-2.1 2.4
-2.7 2.7
-3.2 2.6
-6.0 3.2
Year
est 2se
-6.0 2.3
-5.0 1.8
-5.6 1.9
-4.92.4
-5.52.3
-4.0 2.2
-4.92.2.
-5.2 2.4
-4.5 1.8!
-3.6 2.1
-3.3 1.5
-4.6 2.0
-3.6 1.6
-4.8 1.8
-5.6 2.4
-3.6 1.6
-2.7 1.8
-3.6 1.4
-4.9 1.9
-3.5 1.9
-1.3 1.7
-1.2 1.6
-1.6 2.5
-1.5 1.7
-1.1 1.9
-1.4 1.5
-1.2 1.5
-0.8 1.6
-1.5 1.7
-0.2 1.8
-0.22.1
-0.4 2.9
-1.4 1.6
-0.4 1.6
-1.5 1.5
-1.8 1.7
-2.0 1.5
-1.1 1.3
-2.5 1.6
-2.9 1.2
-4.3 1.6
-2.9 1.2
-5.7 1.9
Src
Sta
Sta
Sta
Sta
Rev
WODC
WODC
Sta
Sta
Sta
Sta
Rev
Sta
WODC
Rev
Sta
,« — ™^— ^^^
WODC
Sta
Sta
Sta
Sta
Sta
Rev
Sta
WODC
WODC
Rev
Rev
Rev
Sta
WODC
Rev
Rev
Rev
Rev
Sta
Rev
Rev
Sta
Rev
Rev
Rev
Rev
1.14
-------
OZONE MEASUREMENTS
2
0
-2
-4
-6
•8
-10
-12
-14
Individual Dobson Station Trends 1/79 to 2/94
-90
-60
(a) - Deii-Jan-Feb
—I—
-30
-i—3 r
\?
Latitude
i—i—i
30
60
90
2
0
-2
-4
•6
•e
-10
-12
-14
(b) - Mar-Apr-May
-90
-60
—I—
-30
~i—I r
0
Latitude
~i — I
30
r-rn
60
90
(o) - Jun-Jul-Aug
o
-2
-4
-6 •
-8 •
•10 •
-12
-14 •
-9
! .
" ^^^"^^i
"**^ • ; . *yj|L
• •
• i
. i ; •
'!:
•60-30 0 30 60
0 x
(d) - Sep-Oct-Nov
Latitude
Ł.
o
-2 -
-4
-6 •
-e •
-10 •
-12 •
-14 •
-9
•
^~^"~-«"*~&^
*.'*'
*
n-f-T r— i i i I i " "i — i r '"f rri
-€0-30 0 30 60
0 9C
Latitude
(e) - Year Round
Ł.
Q
•2
-4
-6
•a
-10
-12
-14 •
-9
;
•^i^*^^5*^.
** ' jl*
" |l *«'»t
('.
!|
i
i
n~i ' ' i ' ' r. T — i — i — i — i — r~n
-60 -30 0 30 60
0 ii «
Latitude
^ h o/S^T a^'0n seasonal trends in tota! ozone in %/decade against latitude, over the
t <» ^^.^ V* available). The gray curves are the averages of the individual
n M In the Mto*™? latitudinal zones: 55-30°S, 30°S-0, 0-20°N, 20-30°N, 30-40°N 40-50°N and
50-60 N. These averages (plus standard errors) are tabulated in Table 1-3. '
within the following latitudinal zones: 55°S-3.0°S,
•30°S-0, 0-20°N, 20-30°N, 30°N-40°N, 4Q°N-508N,'and
50°N-65°N. Figure 1-5 shows the individual station
trends together with the zonal averages for the period
1/79 through 2/94. Although there is substantial scatter
among individual stations, the latitudinal pattern is clear-
ly represented by the zonal averages, which will be used
in the following analyses for comparison to satellite
trends. Seasonal trends from the reassessed filter ozo-
nometer in four large regions of the former USSR are
plotted as separate points in Figure 1-6; they are consis-
tent with and support the Dobson data analysis.
SBUV(/2) trends through May 1994 are given by
season and latitudinal zone in Table 1-2. Ground-based
1.15
-------
OZONE MEASUREMENTS
Updated SBUV(/2) and Dobson Trends
(a) - Deo-Jan-Feb
(b) - Mar-Apr-May
—I—I
-30
30
60
UMuda
(d) - Sep-Ocl-Nov
~i—I—r-
0
Latitude
30
90
(e) - Year Round
2
0
•J
3 *
I *
f -8
•10
-12
-14
•90
i—I—
•30
Dobson (sh
-------
OZONE MEASUREMENTS
Table 1-2. SBUV(/2) trends in %/decade by season and latitudinal zone over the period
1/79 to 5/94, with 95% uncertainty limits (two standard errors, labeled 2se).
Zone
65N
55N
45N
35N
25N
15N
5N
5S
15S
25S
35S
45S
55S
65S
Dec-Feb
-5.6
-6.0
-6.4
•4.9
-3.2
-2.0
-1.3
-1-5
-0.7
-3.1
-4.4
-4.4
-4.6
-5.8
2se
4.2
3.4
2.9
2.4
2.0
1.6
1.8
1.3
1.1
0.9
1.0
1.4
1.6
1.4
Mar- May
-6.3 !
-6-1 ,i
-5.7 :
-4-5 "
-2.7 ;
1
-1.8 ;!
-1.7
-1.8 ;
-0.3 •;
-2.7
' -5.3 ',
-5.0 !,
-6.3 ii
-7.6 :
2se
2.9
2.6
2.3
2.5
2.5
1.8
2.2
1.7
1.1
1.3
1.8
1.7
1.9
2.1
Jun-Aug
-3.5
-3.0
-2.8
-3.1 '
-2.7
-2.0
-1.8
-2.5
-1.5
-3.6
-6.5
-6.6
•-10.T-.
-14.3
2se
1.4
1.6
1.5
1.5
1.6
1.9
1.5
1.3
1.8
2.6
2.6
2.5
3.0
3.8
Sep-Nov
4.3
-3.7
-3.1
-2.9
-3.0
-2.6
-1.6
-2.1
-1.0
-2.7
-3.9
-3.5
-6.3
-13.6
2se
1.6
1.5
'1.5
1.5
1.3
1.4
1.8
1.5
1.5
1.8
2.0
2.2
3.3
5.2
Year
-5.0
-4.8
-4.6
-3.9
-2.9
-2.1
-1.6
-2.0
-0.9
-3.0
-5.0
-4.9
-7.0
-10.4
2se
2.0
1.9
1.8
1.8
1.6
1.4
1.6
1.2
1.1
1.4
1.5
1,5
2.1
' 2.4
Table 1-3. Short-term Dobson trends in %/decade using data from 1/79 to 2/94 Tabled
numbers are averages of individual trends within latitude zones, with 95% uncertainty limits (two
standard errors, labeled 2se). !,
Zone
50-65 N
40-50 N
30-40 N
20-30 N
0-20 N
30- OS
55-30 S
N
7
9
8
5
3
5
6
Dec-Feb
-6.2
-5.4
-3.9
-1.7
0.5
-l.l
-3.6
2se
1.5
1.5
1.3
0.7
1.1
0.7
1.9
i Mar-May
; -6.9
-6.1
-3.5
,: -1.8
-0.0
i -1:1
-2.9
2se
0.5
0.6
1.4
0.2
0.4
1.4
1.2
Jun-Aug
-4.6
-3.0
-1.6
-0.9
-0.7
-2.0
-3.6
2se
1.4
0.6
1.4
1.7
0.6
0.7
1.7
Sep-Nov
-2.2
-2.0
-0.9
-0.5
-0.8
-1.4
-2.8
2se
1.1
0.9
0.8
0.9
0.4
0.5
1.4
Year
-5.2
-4.3
-2.5
-1.2
-0.3
-1.4
-3.2
2se
0.5
0.5
1.0
0.2
0.2
0.6
1.3
Table 1-4. Long-term Dobson trends in %/decade using data from 1/64 to 2/94 (trends
.from 1/70). Tabled numbers are averages of individual trends within latitude zones, with 95%
uncertainty limits (two standard errors;' labeled 2se).
Zone
50-65 N
40-50 N
30-40 N
20-30 N
0-20 N
30- OS
55-30 S
N
7
9
8
5
3
5
6
Dec-Feb
-4.0
-3.7
-2.4
-1.5
0.4
-1.2
-1.8
2se
1.0
0.8
1.0
0.8
0.8
0.7
1.0
Mar-May
i-3.4
-3.6
it-1.8
irl.l
i^O.O
-1.2 •
fl.9
2se
0.5
0.9
0.7
0.5
0.4
1.5
0.8
Jun-Aug
-1.4
-1.8
-0.6
0.0
-0.8
-1.7
-2.5
2se
0.4
0.6
0.5
0.3
0.7
0.3
0.6
Sep-Nov
-1.2
-1:3
-0.7
-0.4
-0.7
-1.4
-1.6
2se
0.5
0.4
0.5
0.7
0.9
0.7
0.8
Year
-2.6
-2.7
-1.4
-0.7
-0.3
-1.4
-2.0
2se
0.4
0.5
0.6
0.4
0.3
0.7
0.7
1.17
-------
OZONE MEASUREMENTS
-7%/decade. In the Southern Hemisphere, extremely
large ozone depletion is seen in the southern winter (Jun-
Aug) and spring (Sep-Nov).
The agreement between SBUV(/2) satellite and
Dobson ground-based trends is not as good as seen in the
1991 assessment between TOMS satellite and ground-
based trends. In the 1991 assessment, TOMS trends
averaged slightly more negative than the Dobson trends,
but only by 1 %/decade or less. As seen in Figure 1-6, the
SBUV(/2) trends average 1 to 2%/decade more negative
than the short-term Dobson trends in all seasons and at
all latitudes except mid- to high northern latitudes. In
the case of the mid- to high northern latitudes, the agree-
ment is much better. In the equatorial regions, while the
Dobson network shows essentially no trend in total
ozone in concurrence with previous assessments, the
SBUV(/2) analysis indicates a seasonally independent
trend of about -2%/decade; these are just statistically
significant in many cases, since two standard errors of
the trend estimates are about 2%/decade in low latitudes.
This is particularly so in the Jun-Jul-Aug period; how-
ever, due to the timing of this assessment, we cannot
update trends in that period beyond the extremely low
1993 values discussed in Section 1.2.2.4.
In order to check the consistency of the SBUV(/2)
trends versus both Dobson and TOMS, Figure 1-7 shows
seasonal trends in total ozone using data through May
1991 for all three. The TOMS trends through May 1991
are similar to those reported in the 1991 assessment
(only an additional two months of data are used), al-
though the seasonal definitions were different in the
1991 assessment (Dec-Jan-Feb-Mar, May-Jun-Jul-Aug,
Sep-Oct-Nov, with April not reported). The Dobson and
•TOMS curves in Figure 1-7 are close to those given in
Reinsel et al (1994a) for the period 11/78 through
12/91; slight differences in the recent Dobson results are
primarily due to use of Dobson station revisions.
Over the same time period, SBUV(/2) trends tend
to be consistently more negative than both TOMS and
Dobson at low latitudes, say 30°S to 30°N. TOMS trends
are also slightly more negative than Dobson trends on
the average, as noted above and in the previous assess-
ment. SBUV(/2) trends average close to -2%/decade in
the tropics, even when data from the low 1992-1993
period are excluded.
Reinsel et al. (1994a) used a set of 56 Dobson sta-
tion records, publicly available from the World Ozone
Data Center, to analyze trends through 1991. Figure 1-8
shows the year-round trends calculated for this report as
discussed above compared to the year-round trends from
the 56 Reinsel et al. stations records, updated with pub-
licly available data. The data used for the comparison
analysis were obtained from the World Ozone Data Cen-
ter, except that newly revised data for the U.S. stations
and Arosa were used as obtained directly from the sta-
tions. The same statistical interventions as used in
Reinsel et al. were also used in the comparison analysis,
with an additional one at Mauna Loa as discussed above.
The results from the larger Reinsel et al. set of stations,
using in many cases data that have not yet been pro-
cessed using current quality control procedures (WMO,
1992b), show much more variation in the trends; howev-
er the average across stations v/ithin each latitude zone is
close to the analogous average for the 43-station analysis
discussed here.
1.2.2.3 THE EFFECT OF THE 1992-1994 DATA
As discussed in the preamble to Section 1.2.2, it is
desirable to update trend estimates through the most re-
cently available data. However, interpretation of these
trends that include the recent period must be made with
caution, since global total ozone was low over the period
late 1991 through late 1993.
Figure 1-9 shows the effect of using data over the
period 1992-1994, compared to stopping the trend anal-
yses at December 1991, for SBUV(/2) and Dobson data.
The comparison is not made for TOMS, because of the
concerns about the TOMS calibration in the last couple
of years of the instrument's life, and because of difficul-
ties in extending the TOMS data beyond May 1993.
The effect of excluding the 1992-1994 data from
the trend calculations is less than one might expect, giv-
en the size of the 1992-1993 anomaly, although certainly
on the average the updated trends are slightly more neg-
ative. The largest consistent effects are in the Jun-Jul-
Aug period in the tropics (note the latest Jun-Jul-Aug
data in this analysis are from 1993) seen in both SBUV
and Dobson analyses; the effect is to make the trends
about 1 %/decade more negative. The Dobson data show
about a 2%/decade effect in winter and spring in the mid-
to high north latitudes, which is not so clear from SBUV
except in the high northern latitudes. In other seasons/
latitudes, the effects are typically less than about !%/•
decade.
1.18
-------
2
0
-2
-4
-6
-8
-10
-12
-14
-90
| OZONE MEASUREMENTS
SBUV(/2), Dobson, and TOMS Trends 1/79 to 5/91
(a) - Dec-Jani-Feb
-60 -30
0
Latitude
30 60
90
(b) - Mar-Apr-May
Latitude
(c) - Jun-Jul-Aug
(d) - Sep-Oct-Nov
Latitude
Latitude
2
0
-2
-4 -
-6 -
-8 -
-10 •
-12 -
-14
(e) - Year Round
-60
-90
—i—
-30
0
Latitude
I~TT
60
90
SBUV-SBUV/2
1/79 .. 5/91
Dobson
1/79 .. 5/91
-a TOMS
1/79 .. 5/91
1.19
-------
OZONE MEASUREMENTS
Year Round Trend from Current Station Set vs. Set from Reinsel et al.
(a) Current static-• set
(b) Reinsel et al. set
•8
0 •
-5
-10
•flf
ii i '
-60
i — '
-30
-90
T r—r
0
Latitude
i
30
i
60
-------
0)
T3
a
I
-90
-60
-90
3
2
1
0
-1
-2
-3
-4
-90
-60
OZONE MEASUREMENTS
v E ffect of Using 1992-1994 Data
(a) - Dec-Jan-Feb
(c) - Jun-Jul-Aug
(e) - Year Round
-30
Latitude
30 60
90
(b) •• Mar-Apr-May
Latitude
(d) - Sep-Oct-Nov
Latitude
-«• SBUV(/2) diffs
1/79.. 5/94 vs. 1/79.. 5/91
* 4 Dobson difts
1/79 .. 2/94 vs. 1/79 .. 5/91
Figure 1-9. Effect on trends of using 1992-1994 data. Triangles denote the difference in the trends calculat-
ed from Dobson data (1/79 to 2/94 minus 1/79 to 5/91). Circles denote the difference in the trends (1/79 to 5/
94 minus 1/79 to 5/91) calculated from SBUV(/2) data.
1.21
-------
OZONE MEASUREMENTS
Table 1-5 Difference in trends 1981-1991 vs. 1970-1980 from the double trends model,
averaged over 24 Dobson stations north of 25°N. The column labeled 2se represents 95%
uncertainty limits (two standard errors) for the difference in trend.
Season
Dec-Jan-Feb
Mar-Apr-May
Jun-Jul-Aug
Sep-Oct-Nov
Year round
Average Trend
Difference
-2.0
-2.8
-1.9
-0.4
-1.8
2se
1.5
1.1
.5
1.2
0.7
Differences between Year Round Trends 81-91 and 70-80
From Double Trends Analysis of 34 Dobson Stations
1
•90
•60
•30
Latitude
—T~
30
Figure 1-10. Differences between trends 1981-
1991 and 1970-1980 at 34 Dobson stations from
double trends analysis.
to-station variability as determined when TOMS is used
as a transfer standard to look for short-term shifts. It is
important that stations' records continue to be main-
tained and improved.
When the TOMS trends through May 1990 were
evaluated (Stolarski et al., 1991) the trend error was esti-
mated to be 1.3% per decade (two sigma error). As a
result of a recent evaluation it appears that the Nimbus 7
TOMS calibration has drifted by 1-2% since 1990. The
changing seasonal cycle in the TOMS-Dobson and
TOMS-SBUV differences appears to be caused by
changing nonlinearity in the TOMS photomultiplier re-
sponse. While the previous error estimate may be
appropriate for equatorial and midlatitude summer data,
the photomultiplier nonlinearity may be introducing as
much as 2% per decade error into midlatitude winter
trends.
The SBUV record has benefited greatly from the
work done on the TOMS measurements (the same basic
algorithm is used; the diffuser plate correction is the
same). The drift in the calibration of total ozone by the
SBUV instrument from January 1979 to June 1990 was
1% or less, and any seasonal differences relative to Dob-
son instruments in the Northern Hemisphere were less
than 1%. The SBUV2 instrument has the on-board cali-
bration lamps and has been compared with the Shuttle
Solar Backscatter Ultraviolet (SSBUV) flights since
1989. There was good agreement during the 18 months
that both SBUV and SBUV2 made measurements. The
main problems with the combined SBUV/SBUV2
record are the possible aliasing of trends resulting from
the changing orbit of the NOAA-11 satellite and the pos-
sibly linked seasonal difference of 1-2% (minimum to
maximum) relative to the ground-based network in the
Northern Hemisphere. The TOMS non-winter measure-
ments agree well with those from SBUV and SBUV/2.
Given these factors, and the extra year of data in the
combined SBUV(/2) record, it is best at this time to fo-
cus on trends derived from the SBUV(/2) measurements.
The most obvious features of the total ozone
trends have been commented upon in previous assess-
ments. Statistically significant negative trends are seen
at mid- and high latitudes in all seasons. The largest neg-
ative ozone trends at mid- and high latitudes in the
northern Hemisphere are seen in winter (Dec-Feb) and
spring (Mar-May); these trends are about -4 to -7%/de-
cade. In the Southern Hemisphere, the annual variation
in the trends at midlatitudes is smaller, though the aver-
age trend is similar to the Northern Hemisphere average.
7.22
-------
OZONE MEASUREMENTS
The effect of including the 1992-1994 data in the
trend calculations is less than one might expect, given
the size of the 1992-1993 anomaly, although certainly on
the average the updated trends are slightly more nega-
tive. The largest consistent effects are in the Jun-Jul-
Aug period in the tropics (note the latest Jun-Jul-Aug
data in this analysis are from 1993) seen in both SBUV
and Dobson analyses; the effect is to make the trends
about 1%/decade more negative. The Dobson data show
about a 2%/decade effect in winter and spring in the mid-
to high north latitudes, which is not so clear from SBUV
except in the high north. In other seasons/latitudes, the
effects are typically less than about 1 %/decade.
Analysis of TOMS, SBUV(/2), Dobson, and ozo-
nometer data through 5/91 reconfirms the results in the
1991 assessment (WMO, 1992), which were based on
TOMS and Dobson analyses through 3/91. The SBUV(/2)
trends tend to be slightly more negative than either
TOMS or Dobson trends, particularly in the tropics,
while as pointed out in the 1991 report, TOMS trends are
also slightly more negative than Dobson. However, the
differences between the instrument systems are within
the 95% confidence limits.
SBUV(/2) trends in the tropics over the period
1/79 through 5/94 are estimated to be about -2%/decade
in all seasons, with formal 95% confidence limits in the
tropics of 1.5 to 2%/decade. This appears to be due to a
combination of two effects: (1) SBUV/2 trends are about
1 %/decade more negative than TOMS and Dobson in the
tropics, raising suspicions of an instrumental drift; and
(2) the inclusion of the low 1992-1994 data makes the
trends an additional 1%/decade more: negative in the
tropics. !
There was a statistically significant increase
(about 2%/decade) at the Dobson stations north of 25°N
in the average rate of ozone depletion in the period 1981 -
1991 compared to the period 1970-1980.
|i
1.3 OZONE PROFILES
I |
1.3.1 Ozone Profile Data Quality
Various techniques have been vised to measure
ozone profiles. However only a few of these have pro-
duced data sets that are long enough, and of sufficient
quality, to be considered for trends. In this section we
consider two ground-based methods that have been in
use since the 1960s (Umkehr and ozonesondes) and two
satellite instruments (SBUV and the Stratospheric Aero-
sol and Gas Experiment, SAGE).
1-3.1.1 UMiiEHR
The long-term records of Umkehr observations are
made using Dobson spectrophotometers at high solar
zenith angles using zenith sky observations (e.g., Gotz,
1931; Dobson, 19(58). A new inversion algorithm has
been developed (Mateer and DeLuisi, 1992), and all the
Umkehr records submitted to the World Data Center
have been recalculated. The new algorithm uses new
temperature-dependent ozone absorption coefficients
(Bass and Paur, 1985) and revised initial estimates of the
ozone profiles. The correction for the presence of aero-
sols is still calculated after the ozone retrieval (DeLuisi
et al., 1989), and the aerosol corrections needed for the
new and old retrievals are similar (Reinsel et al., 1994b).
Mateer and DeLuisi concluded that reliable ozone trends
can only be found for Umkehr layers 4-8 (19-43 km).
Lacoste et al. (1992) compared the lidar and Umkehr
measurements made at Observatoire de Haute-Provence
from 1985-1988. Ifhey found good agreement between
these two measurements systems from layers 4-7 (the li-
dar was not sensitive below layer 4) and attributed the
poor agreement in layer 8 to the low return signals in the
lidar system from this high altitude at that time.
Some information is available below layer 4 be-
cause total ozone must be balanced within the complete
profile. DeLuisi et al. (1994) have compared Umkehr
observations (calculated using the old algorithms) with
SBUV ozone profile data in the 30-50°N latitude band
for 1979-1990 and showed that there is good agreement
in layer 4 and above. The agreement in the SBUV and
Umkehr profiles is not so good lower down, but useful
trend information may be present, a situation that could
improve when the new algorithm is used. For now, as in
recent assessments, only the trends in layers 4 and above
will be considered.
1.3.1.2 OZONESONDES
Ozonesondes are electrochemical cells sensitive to
the presence of ozone that are carried on small balloons
to altitudes above 30 km. Several versions have been
used, and the impoitant ones for ozone trend determina-
tion are the Brewer-Mast (BM), the electrochemical
1.23
-------
OZONE MEASUREMENTS
concentration cell (ECC), and the OSE (used principally
in Eastern Europe). The principle on which they work is
that the current produced in the cell from the reaction of
ozone with potassium iodide solution is proportional to
the amount of ozone passing through the cell. This is not
true if other sources of current exist. Two such cases are
discussed here: the zero-ozone current output possibly
caused by chemicals other than O3; and the interfering
gas, SO2. Changes in operational procedures can also
strongly influence the ozonesonde data quality. Two
ways by which the quality of the ozonesondes can be
assessed are also discussed: intercomparisons and cor-
rection factors.
The ozonesonde network is geographically un-
even, with the large majority of stations in Europe and
North America. The highest density of stations is in Eu-
rope, where they are all located between 44 and 52°N
' and between 5 and 21°E. The long-term records in Eu-
rope all involve BM sondes or OSE sondes. With the
exception of Wallops Island, the North American sta-
tions do not have continuous records longer than 15
years, as Brewer Mast sondes were used at Canadian sta-
tions until about 1980, when there was a switch to ECC
sondes. The frequency of launches at the Japanese sta-
tions has been quite low at times, which has the effect of
increasing the uncertainties associated with the long-
term trends. However, the most obvious shortage of data
is in the Southern Hemisphere, where the only long-
term, non-Antarctic records are at Natal (6°S) and
Melbourne (38°S: Aspendale/Laverton). Unfortunately,
the launch frequency at these sites has been irregular as
well. Last, it should be noted that many stations have
ongoing programs to assess and improve the quality of
the measurements.
. 1.3.1.2a Background Current
The presence of a background (zero ozone) cur-
rent has long been recognized in the ECC sonde and the
standard operating procedures include a method for cor-
rection (Komhyr, 1969). For most ECC sondes that have
been flown, a correction has been applied that assumes
that the background current decreases with altitude
(Komhyr and Harris, 1971). Measurements are sensitive
to errors in the correction for the background current in
regions where the ozone concentration is low, i.e., at or
near the tropopause. Such errors have the potential to be
large as the background current can become similar in
magnitude to the ozone-generated current, for, example,
in the tropical upper troposphere. In the stratosphere,
where ozone concentrations are much higher, the errors
associated with background current corrections are
small.
The response time of the ECC sonde to ozone is
about 20 seconds. Laboratory studies indicate that there
is an additional component of the background current
with a response time of 20-30 minutes (Hofmann, Smit,
private communications). For this component there is a
memory effect as the balloon rises and the background
current would vary through the flight. Earlier studies
(Thornton and Niazy, 1983; Barnes et ai, 1985) con-
cluded that the background current remained constant in
the troposphere. No correction is made for the zero cur-
rent in BM sondes, although some stations measure it
before launch. For BM sondes, the procedure is to re-
duce this zero current to a very low value by adjusting
the sonde output, possibly at the expense of the sonde
sensitivity. Any changes in the magnitude of the back-
ground current over the last 20-30 years will most
strongly affect the trends calculated for the free tropo-
sphere. More work is needed to assess the size and
impact of any changes in the background current in the
different ozonesondes.
1.3.1.2b SO2
The presence of SO2 lowers the ozonesonde read-
ings (one SO2 molecule roughly offsets one O3
molecule), an effect that can linger in the BM sonde be-
cause the SO2 can accumulate in the bubbler (Schenkel
and Broder, 1982). The SO2 contamination is a problem
at Uccle, where the measured SO2 concentrations were
high in the 1970s and have dropped by a factor of about 5
over the last 20 years. A procedure has been developed
to correct for the SO2 effect at Uccle, and the influence is
found to be greatest in the lower troposphere (De Muer
and De Backer, 1994). Logan (1994) argues that the
Hohenpeissenberg, Tateno, and Sapporo ozonesonde
measurements in the lower troposphere may have been
affected by SO2. This interference is worst in winter
when the highest concentrations of SO2 occur. Staehelin
et al. (1994; personal communication) have found that
SO2 levels in Switzerland were too low to have a notice-
able effect at Payerne. Other stations are also likely to
have been less affected.
1.24
-------
OZONE MEASUREMENTS
1.3.1.2c Operational Changes
Changes in operational procedures at an ozone-
sonde station can have dramatic effects on the ozone
measurements, particularly in the troposphere. Two
clear examples are: (a) the change from BM to ECC
sondes at the Canadian stations that took place in the ear-
ly 1980s, when there was an apparent jump in the
amount of tropospheric ozone measured at most of these
stations; and (b) the change in launch tinib at Payerne in
1982, which affected the measurements' in the lowest
layer of the troposphere (Staehelin and Schmid, 1991).
Logan (1994) argues that there is a jump in lower and
mid-tropospheric ozone values in the combined Berlin/
Lindenberg record, corresponding to the change in
ozonesonde launch site from Berlin to Lindenberg and to
the simultaneous change in sonde type from BM to OSE.
Alterations in the manufacture of the seniors and in the
pre-launch procedures can also have an effect.
Another possible cause of error is a change in the
efficiency of the pump. The air flow through the ozone
sensor is not measured, but is calculated from laboratory
tests performed at a number of pressures (Gorsdorf et al.,
1994; Komhyr et al., 1994b, and references therein). It
is possible that there have been some changes in the de-
sign of the pump that may have changed its efficiency
over time and that primarily affect measurements made
at altitudes above 25-28 km.
l.3.1.2d Intercomparisons :
A series of campaigns have been mounted in
which different ozonesondes have been co;mpared to see
whether the quality of any type of ozxmesonde has
changed overtime and to find out what systematic differ-
ences exist between different types of sonde and
between the sondes and other instruments (lidar, UV
photometer). In each campaign good agreement was
found between the various ECC sondes flown simulta- .
neously. However in the most recent WMO campaign
held at Vanscoy, Canada, in May 1991, the BM gave re-
sults 15% higher than the ECC in the:! troposphere,
whereas in the previous campaigns (1970; 1978, 1984)
the BM was reported as measuring 12% less tropospher-
ic ozone than the ECC (Kerr et al., 1994, and references
therein). This result may indicate a change in the sensi-
tivity of the BM to ozone. This conclusion is supported
to some extent by the findings of a study atObservatoire
de Haute Provence, where comparisons involving BM
and ECC sondes, lidars, and UV photometers made in
1989 and 1991 showed a change in the BM sensitivity
relative to ECC in the troposphere. However, operation-
al practices maintained during campaigns can be
different from those used at home, and it is hard to assess
how representative die measurements made under cam-
paign conditions are. The implications of such findings
on trends in tropospheric ozone are discussed in Section
1.3.2.3. The measurements in the stratosphere show
good agreement in all the comparisons.
Although one must be careful in the comparison of
the regular Brewer-Mast sondes with the GDR sondes
manufactured in the former East Germany, results of two
intercomparison campaigns in Germany (Attmann-
spacher and Outsell, 1970, 1981) showed similar
differences between BM and OSE of 3 and 5 nbar, re-
spectively, in the free troposphere (a difference of about
5% of the measured ozone concentration) and no differ-
ences in the stratosphere. This may be a good indication
that OSE sonde quality remained the same, at least over
the time period 1970-1978; and therefore differences be-
tween trend estimates obtained at various stations need
not be strongly dependent on the type of sondes used,
unless changes in sonde type occurred.
1.3.I.2e Correction Factors
Ozonesonde readings are normalized so that the
integrated ozone of the sonde (corrected for the residual
ozone at altitudes above the balloon burst level) agrees
with the total ozone amount given by a nearby Dobson
(or other ground-based) instrument. This is a good way
to assess the overall sounding quality - an unusually
high or low correction factor indicates that something
might be wrong with a particular sounding. A correction
factor of 1 is not a guarantee that the profile is correct.
However, care is needed in using correction fac-
tors, as new errors can be generated. First, the process
relies on the quality of the local total ozone measure-
ment. For instance, errors can be introduced either by a
single, erroneous reading or through changes in the cali-
bration of the ground-based instrument. It is important
to ensure that the ozonesonde records are updated in line
with the ground-based revisions. Second, errors in the
pressure and ozone reading at the burst level will affect
the value of the residual ozone, which in turn influences
the rest of the profile through an inaccurate correction
1.25
-------
OZONE MEASUREMENTS
factor. Third, any variation of the sonde sensitivity to
ozone changes with altitude leads to an incorrectly
shaped profile, which the use of a correction factor
(based only on total column amounts) cannot adjust.
ECC sondes are thought to have a more constant re-
sponse with altitude than the BM sondes which tend to
underestimate tropospheric ozone amounts.
No significant long-term trend in the correction
factor has been seen at Hohenpeissenberg, Payeme, and
Uccle, a fact which suggests that there has not been a
change in sensitivity of the BM sonde, possibly indicat-
ing that the result of the intercomparisons arose from the
different operational conditions used in the intercompar-
isons. Changes in correction factor over shorter times
have occurred, e.g.. at Payerne in the early 1970s and
since 1990 (Logan, 1994). Logan (1994) has compared
the trends estimated using measurements calculated with
and without correction factors and found only small
changes in the ozone trends in all but 3 of the 15 ozone-
sonde records.
13.13 SATELLFTE MEASUREMENTS OF THE OZONE
PROFILE
The SAGE I and SAGE H instruments were de-
scribed in detail in the IOTP (WMO, 1990a). SAGE I
operated from February 1979 to November 1981 and
SAGE II from October 1984 to the present day. They are
solar occultation instruments measuring ozone absorp-
tion at 600 nm. Correction is made for attenuation by
molecular and aerosol scattering and NO2 absorption
along the line of sight by using the observations made at
other wavelengths. Comparisons of SAGE II number
density profiles with near-coincident balloon and rocket
measurements have shown agreement on average to
within±5-10% (Attmannspacheref a/., 1989; Chuefa/.,
1989; Cunnold et al., 1989; De Muer et al. 1990; Barnes
era/., 1991).
The SAGE I and SAGE II instruments are different
in some respects, but, in principle, there are few reasons
for calibration differences between the two instruments.
One reason is the altitude measurements of the two in-
struments, which are now thought to be offset by 300 m.
The effects of such an offset would be felt most at alti-
tudes between 15 and 20 km, where the ozone concen-
trations vary rapidly with altitude. Two independent
investigations have found a potential error in the altitude
registration of the SAGE I data set. From a detailed in-
tercomparison with sondes and lidars, Cunnold (private
communication) has found that agreement between
SAGE I and these other measurements can be signifi-
cantly improved if the SAGE I profiles are shifted up in
altitude by approximately 300 meters. The prelaunch
calibration archives for SAGE I have been reexamined,
and together these data show that the spectral location of
the shortest wavelength channel may be in error by 3 nm
(382 nm instead of 385 nm). Since this channel is used
to correct the altitude registration via the slant path Ray-
leigh optical depth, a shift of 3 nm to shorter wave-
lengths would result in an upward altitude shift of about
300 ± 100 m. The full impact of this wavelength error is
being studied and a preliminary version of the shifted
SAGE I ozone data set is used in this assessment.
The presence of aerosols increases the errors asso-
ciated with the measurement, as the aerosols are
effective scatterers of light at 600 nm. Comparisons of
SAGE II ozone measurements with those made by Mi-
crowave Limb Sounder (MLS) (which should be
unaffected by aerosol) indicate that errors become ap-
preciable when the aerosol extinction at 600 nm is larger
than 0.003 km'1, which corresponds to about 8 times the
background aerosol at 18 km. Only measurements made
with an aerosol extinction less than 0.001 km'1 are used
in the trend analyses presented, in the next section. Using
the 0.001 per km extinction value as a screening criteri-
on, the following general observations follow. The
SAGE II ozone measurements were interrupted for a pe-
riod of one year following the eruption of Mt. Pinatubo
at 22 km near 40°N and 40°S. At the equator the gap in
the series was two years at the same altitude. Extratrop-
ical measurements were uninterrupted at altitudes of 26
km and above (30 km and above at the equator). By ear-
ly 1994, SAGE II was making measurements at all
latitudes down to the tropopause.
SBUV operated from November 1978 to June
1990. The total ozone measurements are described in
Section 1.2.1. Ozone profiles are found by measuring
the backscatter from the atmosphere at wavelengths be-
tween 252 and 306 nm. The wavelengths most strongly
absorbed by ozone give information about the higher al-
titudes. There is little sensitivity to the shape of the
profile at or below the ozone maximum. As with the
Umkehr measurements, some information is available
below the ozone maximum because the complete profile
must be balanced with the total ozone. Hood et al.
1.26
-------
OZONE MEASUREMENTS
(1993) considered the partial column from the ground to
32 mbar (26 km) as the most useful quantity in this re-
gion. ;
Corrections have been made for the diffuser plate
degradation using the pair justification method (Herman
et al., 1991; Taylor et ai, 1994), so the quality of the
SBUV profile measurements has improved since the
IOTP(WMO, 1990). The shorter wavelengths were cor-
rected using a form of the Langley technique: near the
summer pole, ozone measurements are made at each lat-
itude with two solar zenith angles. If the zonal mean
ozone values are constrained to be the same, the wave-
length dependence of the correction to the diffuser plate
degradation can be determined. The accuracy of any de-
rived trend in the ozone profile is no betteirthan 2-3% per
decade. Above 25 km, the vertical resolution of SBUV
is about 8 km, and this increases below 25 km to about
15 km. A limit on the independence of the SBUV ozone
profile data in trend determination is that the retrieval
algorithm requires further information on the shape of
the ozone profile within these layers. It is thus possible
that a trend in the shape of the profile within a given lay-
er could induce a trend in the retrieved Jayer amount,
even though the actual layer amount remains unchanged.
A problem with the synchronization of the chop-
per in the SBUV instrument occurred after February
1987'. After corrections are made, there is,no evidence of
bias at the 1-2% level between the data collected before
and after this date, although the latter dataware somewhat
noisier (Gleason et al., 1993; Hood et a/.,;'1993).
McPeters et al. (1994) have compared the SAGE II
and SBUV measurements from 1984 to 1990, the period
when both instruments were in operation. Co-located
data were sorted into 3 latitude bands (2b°S-20°N, 30-
50°N, and 30-50°S). Agreement is usually better than
5% (Figure 1-11, 20°S-20°N not shown).'iThe main ex-
ceptions are near and below 20 km, where SBUV has
reduced vertical resolution, and above 50 km, where the
sampling of the diurnal variation of ozone is not ac-
counted for in the comparison. A discrepancy between
the SAGE sunrise and sunset data was found in the upper
stratosphere near the equator. This may be related to the
SAGE measurements made at sun angles, which causes
the measurements to be of long duration so that the
spacecraft motion during the event can be ,bn the order of
10 great circle degrees. i
The drift from 1984-1990 between the two mea-
surements above 32 mb is less than 5% and is statistically
insignificant (Figure 1-12). Below this, the drift is 10%
per decade in the tropics and becomes smaller (4-6%) at
midlatitudes. These are roughly consistent with the dif-
ference in the ozone trends from the two instruments.
Some, or all, of this apparent drift may be caused by the
requirement of information about the shape of the ozone
profile in the SBUV retrievals (McPeters et al., 1994).
In contrast, the relative drift between SBUV and the
Umkehr measurements (all between 30 and 50°N) is less
than 2%. However, below the ozone maximum the aver-
age ozone amounts from SBUV and Umkehr differ by as
much as 20% (DeLuisi et al., 1994).
cr
o.
c
o
20
100 -
-15-10-5 0 5
BIAS (%)
10 15 20
Figure 1-11. Ozone profile bias of SBUV relative to
SAGE sunset data in northern midlatitudes (o) and
southern midlatitudes (0) for 1984-1990. The solid
symbols are for layers 3+4 combined to represent
the low SBUV resolution in the lower stratosphere.
Standard deviations; of the appropriate daily values
used in calculating the average biases are also
plotted. (McPeters et a/., 1994.)
1.3.2 Trends in the Ozone Profile
Ozone trends in three altitude ranges received
special attention in the 1991 report. In the upper strato-
sphere (35 km and above) the ozone losses reported from
two observational systems (Umkehr and SAGE) were
1.27
-------
OZONE MEASUREMENTS
-12
-8
-40 4
Relative Drill (percent)
Figure 1-12. Linear drift of SBUV relative to SAGE
II over the 1984-1990 time interval for layers 5-10
and for layer 3+4 combined, derived from a linear fit
applied to percent difference data. (SBUV-SAGE)/
SAGE in percent is plotted. Symbols on X axis give
drift of layer 3-10 integrated ozone amounts. For
comparison, drift relative to an average of five
Umkehr stations (1984-1990) is also shown. The
1 a errors from the standard regression analysis are
given. (McPeters et al., 1994).
qualitatively similar but quantitatively different. These
high altitude decreases have long been calculated in at-
mospheric models and are caused by gas phase
chlorine-catalyzed ozone loss. Ozone losses were also
reported below 25 km, though there were discrepancies
between the values inferred from ozonesonde and SAGE
measurements, especially below 20 km. In the free tro-
posphere, long-term ozone increases were reported at
three European ozonesondes sites. Ozone is an impor-
tant radiative component of the free troposphere and a
better understanding of ozone changes on a global scale
is important. No significant ozone losses were reported
around 30 km altitude or near the tropopause, where the
lower stratospheric decrease switched to an upper tropo-
spheric increase.
In this assessment the same altitude ranges are ex-
amined (starting in the upper stratosphere and working
down) in the light of some new analyses of both the data
quality (see Section 1.3.1) and of the data themselves. In
addition, there is discussion of some ground-level mea-
surements from which inferences can be drawn regard-
ing changes in free tropospheric ozone.
McCormick et al. (1992) calculated trends using
the combined SAGE I/II data set. The SAGE data used
here are slightly different, as the altitude correction has
been applied to the SAGE I data. Also, the base year
used to calculate percentage changes is ,1979 here (not
1988, used by McCormick et ai), so that the percentage
changes in the lower stratosphere, where SAGE reports
the largest decreases, are smaller. Hood et al. (1993)
used the Nimbus 7 SBUV data from 1978 to 1990 to es-
timate trends. In this assessment we use the combined
SBUV/SBUV2 data to extend the record, but the calcu-
lated trends are similar.
1.3.2.1 TRENDS IN THE UPPER STRATOSPHERE
In Section 1.3.1, we described an intercomparison
of the various ozone data sets over a limited time inter-
val. Upper stratospheric trends in ozone have been
estimated from Umkehr, SAGE, and SBUV measure-
ments using the full data sets. While the periods of time
represented by each differ, they all represent, to first or-
der, the changes observed from 1980 through 1990.
Figure 1-13 shows the observed decadal trends as a func-
tion of altitude and latitude from the SBUV and SAGE I/
II data sets. The two are now in reasonable agreement in
the upper stratosphere. The altitude of the maximum
percentage ozone loss is around 45 km and relatively in-
dependent of latitude. The magnitude of this peak
decrease is smallest at the equator (about 5%/decade)
and increases towards the poles in both hemispheres,
reaching values in excess of 10% per decade poleward of
55 degrees latitude.
Figure 1-14 shows the ozone trend profile as a
function of altitude in the latitude band from 30 to 50°N
from SBUV and SAGE, along with the average Umkehr
and ozonesonde trend profiles. SBUV, SAGE, and
Umkehr all show a statistically significant loss of 5-10%
per decade at 40-45 km, although there is some uncer-
tainty as to its exact magnitude. Below 40 km the trends
become smaller and are indistinguishable from zero near
25km.
The seasonal dependence of the trends in the upper
stratosphere has been investigated using the SBUV and
Umkehr data (Hood et a/., 1993; DeLuisi etai, 1994;
Miller et al., 1994). The Umkehr records between 19°N
1.28
-------
OZONE MEASUREMENTS
and 54°N have been examined and their combined data
do not show a significant seasonal variation in the trend.
This is slightly at odds with the analysis,of the SBUV
measurements, which shows that the largest ozone de-
creases have occurred in winter at high latitudes in both
hemispheres, though this difference may not be signifi-
cant given the problems associated with measurements
made at high solar zenith angles. i
13.2.2 TRENDS IN THE LOWER STRATOSPHERE
; i
As discussed in Section 1.3.1, we rely on SAGE
and ozonesondes for information on ozone trends in the
lower stratosphere, as the SBUV and Umikehr capabili-
ties are limited at these altitudes. SAGE measures ozone
from high altitudes to below 20 km. Ozoriesondes oper-
ate from the ground up to about 30 km. '!
In the 1991 assessment, the SAGE 111 midlatitude
trends below 20 km were reported as greiter than 20%
per decade, twice as large as were measured at two
ozonesonde stations or than found from an average of
five Umkehr records in the Northern Hemisphere. The
size of the SAGE trends at these altitudes has provoked a
great amount of discussion, partly because of the sensi-
tivity of climate to changes in ozone in thils region. As
mentioned earlier, the SAGE I/II trends shown here dif-
fer in two respects from those reported preyiously. First,
an altitude correction of 300 m has been applied to the
SAGE I measurements. Second, the year used to calcu-
late the percentage change is now 197$, not 1988.
Below 20 km the effect of both these changes is to re-
duce the SAGE I/II trends because oz&ne changes
rapidly with altitude and because the largest losses are
observed at these altitudes so that the change in the base
value is greatest. Two other factors complicate the
SAGE measurement below 20 km: (i) ozone concentra-
tions are smaller than at the maximum, so lhat the signal
is lower; and (ii) the amount of aerosol is greater, which
attenuates the signal and acts as an additional interfer-
ence. These are well-recognized difficulties for which
allowance is made in the calculation of the ozone
amount and which contribute to the size of die uncertain-
ties in SAGE ozone trends in the lower stratosphere.
Figures 1-13 and 1-14 show the lower stratospher-
ic ozone trends in the 1980s from SBUV, SAGE, and the
non-satellite systems. At altitudes between 25 and 30
km, there is reasonable agreement between SAGE I/II,
SBUV, Umkehr, and ozonesondes that there was no sig-
nificant ozone depletion at any latitude. The agreement
continues down to about 20 km, where statistically sig-
nificant reductions of 7 ± 4% per decade were observed
between 30 and 50°N by both ozonesondes and SAGE II
II. In the equatorial region, the combined SAGE I/II
record (1979-1991) shows decreases of more than 20%
per decade in a region just above the tropopause between
about 30°N and 30°S, although in absolute terms this loss
in the tropics is quite small as there is not much ozone at
these altitudes. The height of the peak decrease in ozone
is about 16 km, and the region of decrease becomes
broader away from the equator. At northern midlatitudes
(Figure 1-14) the SAGE I/II trend at 16-17 km is -20 ±
8% per decade, compaired with an average trend from the
ozonesondes of -7 ± 3% per decade.
The SAGE I/II trends in the column above 15 km
have been compared with the total ozone trends found
from TOMS, SBUV(/2), and the ground-based network
for 1979-1991. This comparison implicitly assumes little
or no change in the ozone amount below 15 km. The
SAGE I/II trends are larger than those found with the
other data sets, peaking at -6% per decade in the northern
midlatitudes, but the associated uncertainties are too
large for firmer conclusions to be drawn. Hood et al.
(1993) compared the tropical trend from SBUV for the
partial column from the ground to 26 km with the SAGE
I/II trends reported by McCormick et al. (1992). They
decided that no conclusive comparison could be made,
although they found trends of about +3 ± 4% per decade
for SBUV for 1979-1990 (see Figure l-13(a) for an up-
dated version), while McCormick et al. found trends
similar to the ones shown in Figure l-13(b) for 1979-
1991. While not shown here, comparisons of SBUV
with SAGE II have recently been completed (McPeters
et al., 1994). Comparisons of the sum of ozone in
Umkehr layers 3-10 (15 km-55 km) show that SBUV in-
creased relative to SAGE (or SAGE decreased relative to
SBUV) by 1.1% between 1984 and 1990.
Logan (1994), London and Liu (1992), and Miller
et al. (1994) have reviewed the global long-term ozone-
sonde data records. Furrer et al. (1992, 1993) and
Litynska (private communication) have analyzed the
records at Lindenberg, Germany, and Legionowo, Po-
land, respectively. These studies handled data quality
issues differently and used different statistical models,
but they gave broadly similar results in the lower strato-
sphere. The large natural variability of ozone concen-
1.29
-------
OZONE MEASUREMENTS
SBUV/SBUV2 Ozone Annual Trend (%/Decade) Thru 6/91
-15
-60
Figure 1-13. (a) Trends calculated for the combined SBUV/SBUV2 data set for 1/79 to 6/91. Hatched areas
in the upper panel indicate that the trends are not significant (95% confidence limits). The lower panel shows
the trends in the partial column between the ground and 32 mbar. Error bars in the lower panel represent
95% confidence levels.
trations is compounded at some stations by a low sam-
pling frequency. It is hard to draw firm conclusions
about seasonal effects. The following results are thus
general and not true for all stations.
Figure 1-15 shows the ozone trends calculated
from the ozonesonde records for the period 1970-1991.
In the northern midlatitudes, a maximum trend of -8 to
-12%/decade was found near 90 mbar from the early
1970s to 1991. Decreases extend from about 30 mbar
down to near the troposphere. Significant ozone loss
certainly appears to have occurred between 90 and 250
mbar. Few conclusions about the seasonal nature of the
trends are statistically significant. A possible difference
exists between the Canadian ozonesonde records where
the summer trends are similar to, and possibly even
greater than, the winter trends. At Wallops Island, U.S.,
and at the European stations, the winter loss is greater
than the summer loss. These features are also seen in the
total ozone record from 1978-1991 observed; at these sta-
tions.
1.30
-------
SAGE!
OZONE MEASUREMENTS
I&II O3 Trend (1979-1991)
"•i , r'xL •- " ->• /#" •:- -y-'.'^Xc-->-,-, ' >v--\-, >-i
-4 .-•- J.V''•Sfe-H'^
•••••••••••--•••':-;;v;;;;^
'•'. '.'.'.: "?P%vHn-ii:JB^&iaMflffiSi-i"^TvS\v;---~-^38.-..r -.-•.•?:-•.«".'«.•.-.•.-• -2.Q'. ...... '\ •
, • r ••!^uft\'^^piroft™i-r^^^ •• .• ..;;;.
" ——' ' ' 1 1 1—i—i 1 1 i i • • r '
15
-60
0
Latitude
Figure 1 -13. (b) Trends calculated for SAGE l/ll for 1979-91. Hatched areas indicate trends calculated to
.ns,gn,f,cant at the 95% confidence level. The dashed line indicates the
SAGE I measurements have been adjusted by 300 meters.
In the tropics, only Natal (6°S) has an ozonesonde
record longer than. 10 years) The trend found by Logan
at 70-90 mbar is -10 ( ± 15)%/decade. AtHilo, Hawaii
(20°N), ozonesondes from 1982 to 1994 indicate insig-
nificant trends of -12 ± 15% per decade near the
tropopause (17-18 km) and -0.7 ± 6% per decade in the
•lower stratosphere at 20 km (Oltmans aid Hofmann,
1994). Trends from both ozonesonde records are small-
er than the calculated SAGE tropical trends; but the large
uncertainties mean that the two trends are not inconsis-
tent. In the Southern Hemisphere, the only iong-running
station outside Antarctica is at Melbourne, Australia,
where a trend of about -10% per decade is observed in
the lower stratosphere.
13.23 TRENDS IN THE, FREE TROPOSPHERE
Only ozonesonde data are available for ozone
trend analyses in the free troposphere. As discussed in
Section 1.3.1, the quality of the ozonesonde data in the
troposphere is worse than in the stratosphere. The strong
likelihood of regional differences in trends further con-
fuses attempts to assess the consistency of a limited
number of ozonesonde records. In the last report, ozone
in the free troposphere at Payerne was shown to have
1.31
-------
OZONE MEASUREMENTS
55
03 Trends During the 1980s for 30N-50N
15F-r-i =
-20
-15
-10 -5 0
Trend (% / Decade)
Figure 1-14. Comparison of trends in the vertical distribution of ozone during the 198Os. Ozonesonde and
Umkehr trends are those from Miller et al. (1994). 95% conf.dence limits are shown.
increased by 30-50% since 1969. An assessment of data
from several stations through 1986 was made (WMO,
1990b) that showed regional effects with increases at
the European and Japanese stations. A tropospheric in-
crease was also reported at Resolute (75°N), but
decreases were found at the three midlatitude Canadian
sites.
Since then, Logan (1994) and Miller et al. (1994)
have analyzed the global ozonesonde record, paying
particular attention to inhomogeneities in the data. A
similar study by London and Liu (1992) did not account
for instrumental changes at some sites. There is now
evidence that the upward trend over Europe is smaller
since about 1980 than before. The Hohenpeissenberg
ozone measurements show no increase since the early to
mid-1980s. The Payeme record shows a somewhat sim-
ilar behavior until 1990. This conclusion is supported by
the recent analyses of the Berlin/Lindenberg record
(Furrer et al., 1992, 1993) and of the Legionpwo record
(Litynska and Kois, private communication). Furrer et
al. found a large tropospheric trend from 1967-1988 of
about +15%/decade, but this seems to have been at least
partly caused by a jump in the measured ozone levels at
the change of station in the early 1970s. Logan (1994)
finds no significant trend at 500 mbar for 1980-1991 and
points out that this trend is sensitive to changes in the
correction factor over this period and could be negative.
At Legionowo, an upper tropospheric trend of -10
(± 4.4)%/decade is reported for 1979-1993, a trend that
is dominated by changes in spring.
Some of the trends, particularly those in Europe,
'might be influenced by changes in SO2 levels. De Muer
and De Backer (1994) have corrected the Uccle data set
allowing for all known instrumental effects, including
SO2. The ozone trend in the upper troposphere was only
slightly reduced (10-15%/decade, 1969-1991) and re-
1.32
-------
- . r . • ""
OZONE MEASUREMENTS
10
20
50
100
200
500
100C
10
20
50
100
200
_§ 500
.§, 1000
« 10
w
Ł 20
Q.
50
100
200
500
1000
10
20
50
100
200
500
1000
Resolute
75 N
: i
i
: 1
}':...,
r 70-91
•
:
•
j
Churchill
59 N
: j
/74-91
•— m
^_
Edmonton
h53N
i
73-91'
'•
-
•
Goose_Bay
53 N
I
±
_Ł
T 70-91
;
•
1 Wallops Island
'38 N
70-91
]
•
':
Hohenpeissenberq
48 N \
;
:
70-91
Payerne
"47 N t< 70-91
lij
i : ^4
< n
~
i .
:
,
1 :
Bsrlin/Lindenberq
52 N
j
4
—
70-91
.
:
_
\
Sai
43 N
; 3
1
sporo
70-91
•
I ~
[ •
[ :
t -
Tateno
36 N
i
: i
i J
; ^
r 70-91
•M.
i i
.
j
Melbourne
38 S
: ,
! 4
i
i
' 70-90
.
: ;
,
Svowa
69 S ""
»
- 1
^
%
70-91
\-
•
Kaaoshima
"32 N j
I
4
4
-^
70-91
_^_
\_ \
Lindenbera
:
':
52 N
1
|
I
1
•
75-91
•
:
.
t_ ;
-4-20 2 4
Figure 1-15. Trends for the periods shown in the pzonesonde measurements at different altitudes 95%
confidence limits are shown. (Adapted from Logan, 1994.)
1.33
-------
OZONE MEASUREMENTS
mained statistically significant. However, below 5 km,
the trend was reduced and became statistically insignifi-
cant, going from around +20%/decade to +10%/decade.
Logan (1994) argues, using SC>2 emission figures and
nearby surface measurements of ozone and SC>2, that
measurements made at Hohenpeissenberg, Lindenberg,
and possibly other European stations might be influ-
enced by SO2 and points out that any such effect would
be largest in winter. In polluted areas, local titration of
ozone by NOX can also influence measurements of
ozone at low altitude. However neither of these effects
should have much influence except in the lower tropo-
sphere (<4 km).
Tropospheric ozone over Canada decreased be-
tween 1980 and 1993 at about -1 ± 0.5%/year (Tarasick
et al., 1994). The positive trend observed at Wallops Is-
land has diminished and from 1980-1991 was close to
zero (Logan, 1994). Prior to 1980 the situation is more
confused. Wallops Island is the only station in North
America with a homogeneous record from 1970 to 1991,
and a trend of just under +10% per decade was observed
(Figure 1-15). In two cases, the critical factor needed to
deduce the long-term tropospheric ozone changes over
North America is the ratio of the tropospheric ozone
measured by BM and ECC sondes. First, the Canadian
stations changed from BM to ECC sondes around 1980,
and a conversion factor is needed if the two parts of the
record are to be combined into one. Second, BM ozone-
sonde measurements were made at Boulder in
1963-1966 (Diitsch etal., 1970), while ECC sondes have
been used in the soundings made since 1985 (Oltmans et
al., 1989). Logan (1994) has compared the Boulder data
by multiplying the BM data at 500 and 700 mbar by 1.15
and concluded that (a) no increase has occurred in the
middle or upper troposphere, and (b) a 10-15% increase
occurred in the lower troposphere caused by local pollu-
tion. The factor of 1.15 was based on a reanalysis of the
intercomparisons in 1970, 1978, and 1984 (see Section
1.3.1.2d). Bojkov (1988; private communication) com-
pared the concurrent measurements made by several
hundred ECC sondes at Garmisch-Partenkirchen and
BM sondes at Hohenpeissenberg, and concluded that the
ratio should be between 1.04 and 1.12 depending on alti-
tude. This approach would produce a larger change at
Boulder in the lower troposphere and would indicate a
small increase at 500 mbar. However, it is possible that
the differences depend on the pre-launch procedures in
use at the different sites, in which case no single factor
can be used: this possibility is supported by the apparent-
ly different jumps seen at the changeovers at the four
Canadian stations. Anyway, there is no sign that ozone
concentrations over Boulder rose by the 50% observed at
Hohenpeissenberg or Payeme since 1967; at most a
10-15% increase has occurred, similar to the increase
observed at Wallops Island.
A reanalysis of the ozonesonde records from the
three Japanese stations from 1969-1990 (Akimoto et al.,
1994) found annual trends of 25 ± 5%/decade and 12 ±
3%/decade for the 0-2 km and 2-5 km layers, respective-
ly. Between 5-10 km the trend is 5 ± 6%/year. There is
no evidence for a slowing of trends in the 1980s.
In the tropics, Logan (1994) reports that Natal
shows an increase between 400 and 700 mbar, but which
is only significant at 600 mbar. The Melbourne record
shows a decrease in tropospheric ozone that is just sig-
nificant between 600 and 800 mbar and is largest in
summer.
13.2.4 TRENDS INFERRED FROM SURFACE OBSERVATIONS
Some information about free tropospheric ozone is
contained in measurements of ozone at the Earth's sur-
face, although care has to be taken in the interpretation
of these data as they are not directly representative of
free tropospheric levels.
Ground-based measurements were made during
the last century, mostly with the Schoenbein method
(e.g. Anfossiefa/., 1991;Sandronief a/., 1992;Marenco
et al., 1994), which is subject to interferences from wind
speed (Fox, 1873) and humidity (Linvill et al., 1980).
Kley et al. (1988) concluded that these data are only
semi-quantitative in nature and should not be used for
trend estimates. Recent improvements in the analysis
are still insufficient to allow simple interpretation of
such data. A quantitative method was used continuously
from 1876 until 1911 at the Observatoire de Montsouris,
Paris (Albert-Levy, 1878; Bojkov, 1986; Volz and Kley,
1988). The average ozone concentration was around 10
ppbv, about a factor of 3-4 smaller than is found today in
many areas of Europe and North America. However, the
measurements at Montsouris were made close to the
ground and, hence, are not representative of free tropo-
spheric ozone concentrations during the last century.
Staehelin et al. (1994) reviewed occasional mea-
surements by optical and chemical techniques at a num-
1.34
-------
OZONE MEASUREMENTS
her of European locations in the, 1930s and measure-
ments made at Arosa in the 1950s (Gotz and Volz, 1951;
Perl, 1965). Figure 1-16 shows a comparison of the
ozone concentrations found in the 1930s land 1950s with
measurements made at Arosa and other European loca-
tions in the late 1980s. On average, orone concentra-
tions in the troposphere over Europe (0-4 km) today
seem to be a factor of two larger than in tjhe earlier peri-
od. The Arosa data also suggest that the' relative trends
are largest in winter. The measurements in the 1950s
were made by iodometry and are potentially biased low
from SO2 interferences caused by local sources, al-
though Staehelin et al. (1994) argue thait SO2 at Arosa
was probably less than a few ppbv.
Figure 1-16 also shows that, because of the vari-
ance between the different sites, little cm be inferred
about a possible increase in tropospheric ozone before
1950. In this context, it is interesting to mate that the data
from Montsouris (1876-1911; 40 m ASL) and those
from Arosa (1950-1956; 1860 m ASL) do not show a
single day with ozone concentrations above 40 ppb
(Volz-Thomas, 1993; Staehelin etal., 1994).
"Modern" ozone measurements, e.g., using UV-
absorption, were started in the 1970s at several remote
coastal and high altitude sites (Scheel et al., 1990, 1993;
Kley et al., 1994; Oltmans and Levy, 19S>4; Wege et al.,
1989). The records for Mauna Loa, Hawaii, and Zug-
spitze, Southern Germany, are shown in Figure .1-17. A
summary of the trends observed at the rpmote sites is
presented in Figure 1-18. All stations north of about
20°N exhibit a positive trend in ozone that is statistically
significant. On the other hand, a statistically significant
negative trend of about -7%/decade is observed at the
South Pole. For the most part, the trends increase from
-7%/decade at 90°S to +7%/decade at 70CIN. Somewhat
anomalous are the large positive trends observed at the
high elevation sites in Southern Germany (10-20%/de-
cade); these large trends presumably reflbct a regional
influence (Volz-Thomas, 1993). It must be noted, how-
ever, that the average positive trends observed at the high
altitude sites of the Northern Hemisphere lire largely due
to the relatively rapid ozone increase that<5Łcurred in the
seventies. If the measurements had started in the 1980s
when the ozone concentrations tended to be at their peak
(Figure 1-17), no significant ozone increase would have
been found. i
4000
3000
ro
o> 2000
T3
1000
O
Jungfraujoch
Gronds-Mulets
Zugspitze
o
, Wank
,Fi
O
Fichtelber
gx
Pfoender .
Q
Chamonix
O
Louterbrunnen
O
Friedrichshafen
. Schouinslond
A A ^ohenPe'ssen~
\ ^
Brotjacklriegel
« Oeuselbach
Montsouris Arkona
— _ o n^,
Westerland
.. ,
20 40
Ozone [ppb]
60
Figure 1-16. Measurements of surface ozone con-
centrations from different locations in Europe
performed before the end of the 1950s (circles) and
in recent years (1990-1991; triangles) during Au-
gust and September, as function of altitude.
(Reprinted from Atmospheric Environment, 28,
Staehelin et al., Trends in surface ozone concen-
trations at Arosa [Switzerland], 75-87, 1994, with
kind permission from Elsevier Science Ltd., The
Boulevard, Langford Lane, Kidlington OX5 1GB,
UK.)
Unlike ozonesondes, and sites such as Mauna Loa
and Zugspitze, where data are specifically identified as
free tropospheric or otherwise (Oltmans and Levy, 1994;
Sladkovic et al., 1994), the ground-based instruments do
not often sample free: tropospheric air. However, the
marine boundary layer sites like Samoa, Cape Point, and
Barrow are representative of large geographical regions,
and although the absolute concentrations may be differ-
ent from those in the free troposphere, this fact should
exhibit only a second-order influence on the trends.
1.35
-------
OZONE MEASUREMENTS
b
M
3 20
o
10
Mauna Loa
3 74 75 76 77 78 79 80 81 82 63 84 85 86 87 88 89 90 91 92 93 94
YEAR
S 20
o
Zugspitze
73 74 75 76 77 78 79 80 81 82 83 84 85 86 67 88 89 90 9V 92 93 94
YEAR •
Fiqure 1-17. Surface ozone concentrations observed during the past two decades at Mauna Lpa (Hawaii,
20°N, 3400m) (adapted from Oltmans and Levy, 1994) and Zugspitze (Germany, 47°N, 3000 m) (SladkovIC
era/.,'1994).
Whether this is the case for Barrow is open to some
question, as Jaffe (1991) has suggested it may be influ-
enced by local ozone production associated with the
nearby oil fields.
1.3.3 Discussion
The state of knowledge about the trends in the ver-
tical distribution of ozone is not as good as that about the
total ozone trends. The quality of the available data var-
ies considerably with altitude.
The global decreases in total ozone are mainly due
to decreases in the lower stratosphere, where the uncer-
tainties in the available data sets are largest. SBUV and
Umkehr measurements are most reliable around and
above the ozone maximum.. Information at lower alti-
tudes is available from these techniques, but it is not
clear at the present time whether much can be learned
about trends in these regions. Ozonesondes make reli-
able measurements in the lower stratosphere, but the
natural variability is such that the uncertainties associat-
ed with trends calculated for individual stations are
large. Only in the northern midlatitudes do enough
ozonesonde records exist for trends to be calculated with
uncertainties smaller than 5%/decade. SAGE can mea-
sure ozone down to 15 km altitude. Two factors
complicate the SAGE measurement below 20 km: (i)
ozone concentrations are smaller than at the maximum.
2.S
0.5
0
-O.S
SP
•
US
MLOT
1
TS
i
HPB
WFM
. B
0
LHHud«
Figure 1-18. Trends in tropospheric ozone ob-
served at different latitudes, including only coastal
and high-altitude sites (after Volz-Thornas, 1993).
CP: Cape Point, 34°S (Scheel et al., 1990); SP:
South Pole, 90°S, 2800m ASL; AS: American Sa-
moa, 14°S; MLO: Mauna Loa, 20°N, 3400m; B:
Barrow, 70°N (Oltmans and Levy, 1994); WFM:
Whiteface Mountain, 43°N, 1600m (Kley ef al.,
1994); ZS: Zugspitze, 47°N, 3000m; HPB: Hohen-
peissenberg, 48°N, 1000m (Wege et al., 1989).
i
so that the signal is lower; and (ii) the amount of aerosol
is greater, so that there is an additional .interference.
These are well recognized difficulties for which allow-
ance is made in the calculation of the ozone amount.
At altitudes of 35-45 km, there is reasonable
agreement between SAGE I/II, SBUV(/2), and Umkehr
1.36
-------
OZONE MEASUREMENTS
that, during 1979-1991, ozone declined 5-10% per
decade at 30-50°N and slightly more at southern midlat-
itudes. In the tropics, SAGE I/II gives larger trends (ca.
-10% per decade) than SBUV (ca. -5%; per decade) at
these altitudes. '
At altitudes between 25 and 30 iim, there is rea-
sonable agreement between SAGE MI, SBUV(/2),
Umkehr, and ozonesondes that, during'the 1979-1991
period, there was no significant ozone depletion at any
latitude. The agreement continues down to about 20 km,
where statistically significant reductions of 7 ± 4% per
decade were observed between 30 and, 50°N by both
ozonesondes and SAGE I/II. Over the longer period
from 1968 to 1991, the ozonesonde record indicates a
trend of -4 ± 2% per decade at 20 km at iaorthern midlat-
itudes.
There appear to have been sizeable ozone reduc-
tions during the 1979-1991 period in the 15-20 km
region in midlatitudes. There is disagreement on the
magnitude of the reduction, with- SAGE indicating
trends as large as -20 ±8% per decade ai 16-17 km and
the ozonesondes indicating an average trend of -7 ± 3%
per decade in the Northern Hemisphere. The trend in the
integrated ozone column for SAGE is liirger than those
found from SBUV, TOMS, and the grobnd-based net-
work, but the uncertainties are too large'to evaluate the
consistency between the data sets properly. Over the
longer period from 1968 to 1991, the ozonesonde record
indicates a trend of -7 ± 3% per decade al .16 km at north-
em midlatitudes. :-i
In the tropics, trend determination at altitudes be-
tween 15 and 20 km is made difficult by jthe small ozone
amounts. In addition, the large vertical ozone gradients
make the trends very sensitive to small vertical displace-
ments of the profile. The SAGE I/II record indicates
large (-20 to -30% (±18%) per decade) trends in the 16-
17 km region (-10% ±8% at 20 km). Limited tropical
ozonesonde data sets at Natal, 6°S and Hilo, 20°N do not
indicate significant trends between 16 amd 17 km or at
any other altitude for this time period. :With currently
available information it is difficult to evaluate the trends
below 20 km in the tropics, as the relatil uncertainties
are large. The effect on the trend in the total column
from any changes at these altitudes would be small.
In the free troposphere, only limited data (all from.
ozonesondes) are available for trend determination. In
the Northern Hemisphere, trends are highly variable be-
tween regions. Upward trends in the 1970s over Europe
have declined significantly in the 1980s, have been small
or non-existent over North America, and continue up-
ward over Japan. The determination of the size of the
change over North America requires a proper treatment
of the relative tropospheric sensitivities for the type of
sondes used during different time periods.
Surface measurements indicate that ozone levels
at the surface in Europe have doubled since the 1950s.
Over the last two decades there has been a downward
trend at the South Pole and positive trends are observed
at high altitude sites in the Northern Hemisphere. When
considering the latter conclusion, the regional nature of
trends in the Northern Hemisphere must be borne in
mind.
1.4 OZONE AND AEROSOL SINCE 199T
Since the last report, record low ozone values have
been observed. This section describes the ozone mea-
surements in this period to allow the updated trends
given in Section 1.2.2 to be put into perspective. There is
also a brief discussion of a variety of possible causes,
including the aftermath of the eruption of Mt. Pinatubo,
aspects of which will be discussed at greater length in
later chapters.
1.4.1 Total Ozone Anomalies
Figure l-19(a) shows the daily global average
(50°N-50°S) ozone amount during 1992-1994, together
with the envelope of 1979-1990 observations. Persistent
low ozone levels are observed beginning in late 1991
'(not shown), with values completely below the 1979-
1990 envelope from March 1992-January 1994. During
1993 total ozone was about 10-20 DU (3-6%) below the
1980s average. Total ozone in early 1994 recovered
somewhat and was at the bottom end of the range ob-
served in the 1980s.
Figure 1- 19(b)-(d) shows similar plots for the lati-
tude bands 30-50°S, 20°S-20°N, and 30-50°N. The
largest and longest-lived anomalies are seen at the north-
ern midlatitudes (15-50 DU lower in 1993), with 1980s
values reached again in January 1994. Ground-based
measurements made at sites with long records show that
the anomalies in the northern midlatitudes were the larg-
est since measurements began, and that values in early
1.37
-------
OZONE MEASUREMENTS
300
250
SBUV/SBUV2 GLOBAL OZONE 50N-50S
79-90 Ozone Range
92-94 SBUV/SBUV2 Ozone
Jin 92 )u!92 Jin 93 Jul93 Jan 94
Jul94
Jan 95
g 270
260
230
Jan 92
SBUV/SBUV2 20N-20S OZONE
79-90 Ozone Range
92-94 SBUWSBUV2 Ozone
Jul92
Jan 93 Jul 93 Jan 94 Jul 94 Jan 95
SBUV/SBUV2 30S-50S OZONE
SBUV/SBUV2 30N-50N OZONE
3801
360
340
320
300
280
260
240
Jan 92
79-90 Ozone Range
92-94 SBUV/SBUV2 Ozone
380
Jul92 Jan93 Jul 93 Jan 94 Jul 94 Jan 95
Jan 92
|ul 92 Jan 93 Jul 93
)an 94 Jul 94
Jan 95
Figure 1-19. Total ozone measured by SBUV and SBUV(/2) since January 1992 compared with the 1980s
range and average: (a) 50°N-50°S, Global ozone; (b) 30°-50°S; (c) 20°N-20°S; (d) 30°-50°N.
1993 were about 15% lower than the average values
observed before 1970 (Bojkov et ai, 1993; Kerr el aL,
1993; Komhyr et aL, 1994a). The largest ozone losses
occurred at higher latitudes in early 1993; deviations
were in excess of 60 DU (15% lower than the 1980s
mean). Total ozone values over North America in 1994
were in line with the long-term decline, but no longer
below it (Hofmann, 1994).
In southern midlatitudes, total ozone values during
1993 were about 15-20 DU below the 1980s mean and
were close to the low end of the 1980s range. In the
tropics, the maximum negative anomaly was about 10
DU, and from late 1992 to early 1993 total ozone was
slightly higher than the 1980s average. Locally, larger
anomalies were seen, with negative ozone anomalies of
about 15 DU (6%) occurring near the equator in Septem-
ber-November 1991 and in the southern tropics in
mid-1992.
The solar cycle and the quasi-biennial oscillation
(QBO) affect total ozone .levels by a few percent and it is
thus useful to remove these influences. Figure l-20(a)
shows the 60°S-60°N average total ozone from SBUV(/2)
after these effects (and the annual cycle) have been re-
moved by the statistical analysis described in Section
1.2.2. The most obvious remaining feature is the long-
term decrease in total ozone, which has been fitted with a
1.38
-------
-2 -
Total Ozone Deviations over 60S - 60N from 1/79 to 5/94
Deseasonalized and Adjusted tor Solar and QBO Components
Solid line is least squares fit to
deviations 1/79 to 5/91. then extended
(cloned line) to 5/94. '
1960
1982
1984
1986
1988
1990
1992
1994
Year
OZONE MEASUREMENTS
Figure 1-20. (a) Total ozone (60°N-
60°S) from 1/79 to 5/94 measured by
SBUV(/2). The annual cycle and the
effects of the solar cycle and QBO
have been removed. The solid line
shown is a simple least squares fit to
the data-through 5/91. The dashed line
is an extrapolation through 5/94.
Deviations (in %) from SBUV(/2) Mpdel 1/79 through 5/91, Extended to 5/94
Shaded regions represent negative departures more than 2%
1987 1988 1989 1990
nQRMx/o! PJ°tS °f tota' ozone ^siduals as a function of latitude and time from the statistical
fit to the SBUV(/2) satellite data over the period 1/79 to 5/91 . The fitted model was extrapolated through 5/94
to calculate the residuals over the extended period 1/79-5/94. The total ozone data have the seasonal trend
solar, and QBO components removed, and the resulting deviations are expressed as percentages of the
mean ozone level at the beginning of tye series. Shown are contours of constant deviations at intervals of
3 /o, and the shaded areas indicate negative departures of at least 2%. The 1992-1993 low ozone values are
prominent, as well as other periods of very low values in 1 982-1 983 and 1 985.
1.39
-------
OZONE MEASUREMENTS
linear trend (-2.9% per decade) from January 1979 to
May 1991 (pre-Pinatubo). The recent (1992-1993) glo-
bal anomaly is about 2% below the trend line and about
1% less than previous negative anomalies. The 1992-
1993 anomaly also stands out as the most persistent,
spanning nearly 2 years. The only other negative anom-
aly lasting more than one year followed the El Chich6n
eruption in 1982. Figure l-20(b) shows the time evolu-
tion at all latitudes (60°S-60°N) of the total ozone
deviations found after the removal of the trend found for
1/79 to 5/91 (extrapolated to 5/94), the annual cycle, and
the effects of the solar cycle and the QBO. The strong
regional nature of the deviations is again obvious, with
the largest (6-10%) occurring in northern midlatitudes in
January to April 1993. The Southern Hemisphere, by
contrast, was hardly affected.
1.4.2 Vertical Profile Information
Figure 1-21 (a) shows the ozonesonde measure-
ments at Edmonton made in January-April in 1980/
1982, 1988/1991, and 1993 (Km et al, 1993). Similar
results were found at Resolute, Goose Bay, and
Churchill. These indicate that the decrease in early 1993
occurred in the same altitude region as the decline during
the 1980s. The standard deviations are ±8 nbar (1980-
1982 and 1988-1991 profiles) and ±9 nbar (1993
profiles) where the maximum ozone difference is found
(100 mbar). The differences between the 1993 and
1980-1982 profiles are statistically significant (2 stan-
dard deviations) between 200 and 40 mbar. Ozone levels
were depleted by about 25% over approximately 14-23
km (at and below the profile maximum), spatially coin-
cident with the observed aerosol maximum, as shown in
Figure l-21(b) (Hofmann et ai, 1994a). Notably, there
is substantial ozone increase above the profile maximum
(above 25 km) at Boulder, of about 15% of background
levels, which is also seen at Hilo, Hawaii (Hofmann et
al., 1993).
1.4.3 Stratospheric Aerosol after the Eruption
of Mt. Pinatubo
The eruption in the Philippines of Mt. Pinatubo
(15°N, 120°E) in June 1991 injected approximately 20
Tg of sulfur dioxide (SO2) directly into the lower strato-
sphere at altitudes as high as 30 km. Within a month or
so, this SO2 was oxidized to sulfuric acid, which rapidly
condensed as aerosol. In August 1991, Volcan Hudson
(46°S, 73°W) erupted and deposited about 2 Tg of SO2
into the lower stratosphere, mostly below 14 km. Sever-
al studies of the SO2 and aerosol observations have been
published (e.g., Bluth et al, 1992; Lambert et al., 1993;
Read et al., 1993; Trepte et al, 1993; Deshler et al.,
1993; Hofmann et al, 1994b), which are now briefly dis-
cussed. The latitudinal variation of optical depth from
1991 to 1994 is shown in Figure 1-22 as measured by
SAGE II and the Stratospheric Aerosol Measurement
(SAM II) instrument.
The initial aerosol cloud from Mt. Pinatubo dis-
persed zonally but was confined mostly within the
tropics below 30 km for the first several months. By
September 1991 the Mt. Pinatubo aerosol had moved
into the midlatitude Southern Hemisphere at altitudes
between 15 and 30 km. It did not enter into the Antarctic
vortex in 1991, unlike the aerosol from Volcan Hudson,
which was observed at altitudes of 10-13 km over Mc-
Murdo station, 78°S (Deshler et al., 1992). In the tropics
the Mt. Pinatubo plume rose to altitudes of 30 km during
December 1991-March 1992. Strong dispersal from the
tropics into northern middle-high latitudes was observed
during the 1991-1992 winter, and enhanced aerosol lev-
els have been detected over 15-25 km in the Northern
Hemisphere since that time.
The total mass of the stratospheric aerosol maxi-
mized several months after the eruption at about 30 Tg
and thereafter remained fairly constant until mid-1992,
since when it has been declining with an approximate e-
folding time of one year. The total aerosol loading in
January 1994 was about 5 Tg, still 5-10 times higher
than .the background levels observed before the Mt. Pi-
natubo eruption.
The size distribution of the aerosol particles
evolved significantly over time, increasing in effective
radius from approximately 0.2 |im just after the eruption
to a peak of some 0.6-0.8 |im a year or so later, since
when it has slowly decreased (Deshler et al, 1993). At
northern midlatitudes, the aerosol surface area peaked at
about 40 urn2 cm'3 (Figure 1-23). The altitude of the
maximum surface area has episodically decreased since
early 1992.
Negative total ozone anomalies of about 15 DU,
6%, occurred near the equator in September-November
1991 (Schoeberl et al, 1993; Chandra, 1993), at the
same time that the maximum temperature increase,
1.40
-------
OZONE MEASUREMENTS
4O
30
20
.C
o>
"2 may also,have played a part (Bekki et
al., 1993). \\
In addition to radiative and dynamical 'influences,
the Mt. Pinatubo aerosol provides a surface on which
chemical reactions can occur, possibly leading to chemi-
cal ozone loss, as discussed in Chapters 3 and 4. These
reactions tend to proceed faster at lower temperatures
and the ozone depletion process is more effective at low
light levels. In this context it is worth noting that both
the 1991/1992 and 1992/1993 northern winters were
cold with later-than-average final warmings (e.g., Nau-
jokat et al., 1993), and that the cold temperatures
occurred both within and on the edge of the Arctic vor-
tex, so that there was the opportunity for large areas to be
affected.
For comparison, the maximum aerosol surface
area and its peak altitude following the eruption of El
Chichon in early 1982 are shown in Figure 1-23. The
Mt. Pinatubo eruption provided twice the aerosol surface
area as that from El Chichon. The total ozone anomalies
in 1982/1983 (as compared with 1980,1981, 1985, 1986
TOMS values) are now thought to have been smaller
than the earlier initial estimates, about 3-4% in the 1982/
1983 winter rather than 10% (Stolarski and Krueger,
1988).
1.41
-------
OZONE MEASUREMENTS
SAM II and SAGE II Stratospheric Aerosol
SON
60N-
40S-
60S-
80S
1991
<10'3
1992 1993
1-Micron Optical Depth
io-2
1994
1995
Pinatubo
Hudson
>2xlO"
Figure 1-22. Aerosol optical depths from 1991-1994 measured by SAM II and SAGE II showing the effects
of the Mt. Pinatubo (*) and Volcan Hudson (+) volcanic eruptions. (Updated by L. Thomason from data
shown in Trepte et al., 1993.)
1.4.4 Dynamical influences
Natural variations in ozone are induced by meteo-
rological phenomena such as the El Nino-Southern
Oscillation (ENSO), in addition to the QBO (e.g., Zere-
'fos, 1983; Bojkov, 1987; Komhyrera/., 1991; Zerefos et
al., 1992). Thus the observed ozone anomalies since
1991 will have been affected to some degree by the pro-
longed El Nino event that lasted throughout 1992/1993.
The amplitude of the El Nino effect in total ozone is
2-6%, but such anomalies are highly localized. While
ENSO effects for zonal or large-area means were about
1 %, ozone in specific areas may have been reduced by an
additional 2-3% in 1992-1993 (Zerefos et al., 1992;
Shiotani, 1992; Zerefos et al., 1994; Randel and Cobb,
1994). Other dynamical influences can strongly affect
total ozone on a regional basis; one clear example was
the persistent blocking anti-cyclone in the northeast At-
lantic from December 199! to February 1992 (Farman et
al., 1994).
1.42
-------
i i i i i i i i i i i
• EL CHICHON
o PINATUBO
OZONE MEASUREMENTS
Q.
1985 EL CHICHON
1994 PINATUBO
• EL CHICHON
o PINATUBO
1982
1991
1983
1992
1984
1993
1985 EL CHICHON
1994 PINATUBO
Figure 1-23. The maximum surface area and its altitude observed at Laramie, Wyoming, in the years follow-
ing the El Chichon and Mt. Pinatubci|eruptions (Deshler et a/., 1993).
1.5 ANTARCTIC OZONE DEPLETION
1.5.1 Introduction and Historical Data
Total ozone records obtained with Dobson spec-
trophotometers with a traceable calibration are available
for Antarctica from 1957 at the British stations Halley
(76°S, formerly Halley Bay) and Faraday (65°S, former-
ly Argentine Islands). They are available from the
American station at the South Pole (Amundsen-Scott,
90°S) since 1962 and at the Japanese: station Syowa
(69°S) since 1966, although measurements had been ob-
tained at Syowa in 1961. Figure 1-24 shows October
monthly means for these four stations. Ilj the case of the
South Pole station, the average is for October 15-31
since inadequate sunlight precludes accurate total ozone
measurements from the surface before about October 12.
Halley and Amundsen-Scott show similar long-
term total ozone declines in October, presumably
reflecting the fact that the region of most severe ozone
depletion is generally shifted off the pole towards east
Antarctica. The decline in ozone above these stations
began in the late 1960s, accelerated around 1980, and
after 1985 remained! relatively constant at a total ozone
value of about 160 DU. In 1993, record low values
(about 116 DU) were recorded at Halley and Amundsen-
Scott.
The decline in total ozone at Faraday and Syowa in
October was more subtle, if existent at all, prior to 1980.
The major decline occurred between 1980 and 1985, lev-
1.43
-------
OZONE MEASUREMENTS
400
0
u3°°
o
M
O
O 200
10
if f T
*y A*;: ?. i*
m
OCTOBER MONTHLY MEANS
. SOUTH POLE (90°S)
O HALLEY BAY (76'Sj
. FARADAY (65°S)
ft SYOWA (69°S)
1965
1975
YEAR
1985 1995
Figure 1-24. The historical springtime total ozone
record for Antarctica as measured by Dobson spec-
trophotometers during October at Halley Bay,
Syowa, and Faraday and from 15-31 October at
South Pole. (Data courtesy J. Shanklin, T. Ito, and
D. Hofrnann.)
elling off with a value of total ozone of about 260 DU
thereafter. An unusually low value of about 160 DU was
observed at Syowa in 1992, a feature not seen at Faraday.
Although the earliest ozbne vertical profiles show-
ing the 1980 rapid ozone hole onset were obtained at
Syowa in 1983 (Chubachi, 1984), the most extensive set
of ozone profile data for trend studies has been obtained
at the South Pole using ECC ozonesondes throughout
(Oltmans et al., 1994). This data set includes the ap-
proximately 500 year-round profiles measured between
1986 and 1993, and a series of about 85 profiles made
between 1967 and 1971. Winter data for the earlier peri-
od do not extend to as high an altitude because rubber
balloons were used. Figure 1-25 shows a comparison of
smoothed monthly average ozone mixing ratio values at
pressure levels 400 hPa (-6.5 km), 200 hPa (-10.5 km),
100 hPa (-14.5 km), 70 hPa (-16.5 km), 40 hPa (-19.5
km) and 25 hPa (-22.5 km) for these two periods. The
major springtime ozone depletion has occurred in the
14-22 km region at the South Pole between the 1967-
1971 and 1986-1991 periods, and it has worsened since
1992. The 1967-1971 data indicate a weak minimum in
the spring in the 40-100 hPa (14-19 km) region. This
feature might result from heterogeneous ozone loss re-
lated to considerably lower stratospheric chlorine levels,
consistent with the weak downward trend in total ozone
at South Pole for this period shown in Figure 1-24. In
1992 and 1993, ozone was almost completely destroyed
in the 70-100 hPa range (14-171cm).
Summer (December to February) ozone levels in
1986-1991 are tower in the 70-200 hPa (10-17; km) re-
gion than they were in 1967-1971. The ozone that is
transported to the South Pole following vortex break-
down at these altitudes now replenishes the ozone lost
during the previous spring, rather than causing the
marked late spring maximum which existed in 1967-
1971. At all altitudes, ozone values from March to
August are similar (to within about 10%) in the two
periods.
Rigaud and Leroy (1990) reanalyzed measure-
ments taken at Dumont d'Urville (67°S) in 195& using a
double monochromator with spectrographic plates
(Fabry and Buisson, 1930; Chalonge and Vassey, 1934).
They calculated some very low total ozone values (as
low as 110 DU) that are only observed nowadays in the
ozone hole. De Muer (1990) and Newman (1994) have
examined the available 1958 meteorological and total
ozone data. They find that the early Dumont d'Urville
data are inconsistent with any other source of data from
1958: (a) the variability was greater throughout the year
than that measured with any Dobson spectrophotometer
in Antarctica that year (Figure 1-24); (b) Dumont
d'Urville was not under the vortex that year (see also Alt
et al., 1959), but under the warm belt where ozone values
are high; and (c) while the climatologies of measure-
ments taken by Dobson instruments that year are fully
consistent with those derived from TOMS measure-
ments in the last decade, there is little or no consistency
between the TOMS climatologies and that from. Dumont
d'Urville in 1958. Some doubts concerning a number of
experimental aspects of the spectrographic plate instru-
ment are also raised. These reported values thus appear
to be a good example of being able to detect ozone with-
out necessarily being able to measure it well.
1.5.2 Recent Observations
Figure 1-26 shows monthly average total column
ozone measured at the South Pole by balloon-borne
1.44
-------
OZONE MEASUREMENTS
5
4
3
2
1
I 0
O
I— 2
1 '
X
o
o°-10
0.08
0.06
0.04
0.02
0.00
40 hPa
« 1967-1971
o 1986-1991
1992
-:::;— 1993
—i 1 r
25 hPa
—i i i 1 1 1 r
• 1967-1971
o 1986-1991
1992
1993
JFMAMJJiiASOND JFMAMJJASO'ND
5
100 hPa
• 1967-1971
o 1986-1991
1992
1993
70 hPr,
70 hPa
~\ 1 1 1 1 1 1 1 1 r
• 1967-1971
o 1986-1991
—— 1992
1993
JF.MAMJJASOND JFMAMJJASOND
1.0
400 hPa
• 1967-1971
o 1986-1991
1992
1993
0.8
0.6
6.4
0.2
0.0
—i 1 1 1 1 1 1 1 1 1 r
•7nn KP^ • 1967-1971
200 hPo Q 1986_1991
1992
1993
J'FMAMJJ;,ASOND JFMAMJJ'ASOND
I! MONTH !
Figure 1-25. Comparison of smoothed monthly average ozone mixing ratios at 6 pressure levels for the
1967-1971 period (filled points and full lines), the 1986-1991 perJod (open points and dashed lines), and for
1992 and 1993 (straight and dashed lines, respectively). The error bars represent ±1 standard deviation.
(Adapted from Oltmans etal., 1994.),
1.45
-------
OZONE MEASUREMENTS
ozonesondes since 1986 (Hofmann etal., 1994b). (Total
ozone is obtained by assuming that the ozone mixing ra-
tio is constant above the highest altitude attained, a
procedure that has'an estimated uncertainty of about 2-3
DU.) These data are independent of the Dobson spectro-
photometer data shown in Figure 1-24 and corroborate
the fact that the major springtime depletion started be-
tween the 1967-1971 and 1986-1991 periods.
On 12 October 1993, total ozone at the South Pole
fell to a new low of 91 DU, well below the previous low
of 105 DU measured there in October 1992. Sub-100
DU readings were observed on 4 occasions and readings
in the 90-105 DU range were measured on 8 consecutive
soundings from 25 September to 18 October 1993.
Ozone levels in austral winter prior to the deple-
tion period show no systematic variation, with values of
250 ± 30 DU. Similarly, coming out of the depletion
period, January values show no systematic variation
since 1986, but are lower than the 1967-1971 values.
At the South Pole, both Dobson spectrophotome-
ter and Meteor TOMS measurements showed record low
total ozone levels after the return of adequate sunlight in
mid-October. Similarly, NOAA-11 SBUV2 measure-
ments indicate new record lows for the 70°S-80°S region
in 1993 (NOAA, 1993). Thus, since 1991, the Septem-
ber total ozone decline has continued/worsened.
SOUTH POLE STATION
17-26 AUG 93 272*26 DU
12 OCT 93 91 DU
11 OCT 92 105 DU
0 5 10 15 20
03 PARTIAL PRESSURE (mPo)
Figure 1-27. Comparison of the South Pole pre-
depletion ozone profile in 1993 (average of 4
soundings) with the profile observed when total
ozone reached a minimum in 1992 and 1993.
(Adapted from Hofmann etal., 1994b.)
Q
UI
•z.
o
N
O
400
300
200
O
0 100
• 1967-1971
o 1986-1991
1992
JFM
MJJA
MONTH
SOND
Figure 1-26. Monthly averaged total column ozone
by month measured in balloon flights at South Pole
for the 1967-1971 and 1986-1993 periods, and for
1992 and 1993 (straight and dashed lines, respec-
tively). (Adapted from Oltmans etal., 1994.)
In Figure 1-27 the average of four ozone profiles
before depletion began in August 1993 is compared with
the profiles at the time of minimum ozone in 1992 and
1993 (Hofmann etal., 1994b). Total destruction (>99%)
of ozone was observed from 14 to 19 km in 1993, a 1 km
upward extension of the zero-ozone region from the
previously most severe year, 1992. Unusually cold tem-
peratures in the 20 km region are believed to be the main
cause of lower-than-normal ozone in the 18-23 km
range. These lower temperatures prolong the presence
of polar stratospheric clouds (PSCs), in particular nitric
acid trihydrate (NAT), thought to be the dominant com-
ponent of PSCs. This tends to enhance the production
and lifetime of reactive chlorine and concomitant ozone
depletion at the upper boundary of the ozone hole, be-
cause chlorine in this region is not totally activated in
years with normal temperatures. Temperatures at 20 km
in September 1993 were similar to those of 1987 and
1.46
-------
OZONE MEASUREMENTS
Jun
Nov
Dec
Figure 1-28. Area of the region enclosed by the 220 DU total ozone contour in the Southern Hemisphere.
The white line represents the 1978-1991 average with the shaded area representing the extremes for this
period. The 1992 and 1993 areas are represented by the continuous line and points, respectively. 12 million
square kilometers is about 5% of the surface area of the Southern Hemisphere, so that the maximum extent
of .the region in 1992 or 1993 with total ozone less than 220 DU, if circular, was about 65°S. Data for 1978-
1992 are from Nimbus 7 TOMS; data for 1993 are from Meteor TOMS. Only measurements made south of
40°S were considered, to avoid including any low tropical values recorded. (Courtesy of the Ozone Process-
ing Team, NASA Goddard.)
: i
1989, other very cold years at this altitude.; Cold sulfate
aerosol from Mt. Pinatubo, present at altitudes between
10 and 16 km, probably contributed to the low ozone
through heterogeneous conversion of chlorine species
(see Chapters 3 and 4). ,
Figure 1-28 shows the horizontal extent of the
Antarctic ozone hole in terms of the area contained with-
in the 220 DU total ozone contour from Nimbus TOMS
(1978-1991 shaded region pid 1992 curve) and from
Meteor TOMS (1993 points). These data: indicate that
the 1992 and 1993 ozone hole areas were the largest on
record and that the development of the depleted region
began about 1-2 weeks earlier, a fact also apparent in the
total ozone data in Figure 1-26.
Since 1991, springtime ozone depletion at the
South Pole has worsened in the 12-16 kni
total ozone destruction at 15-16 km in 1992 and 1993
region, with
Similar observations were made in 1992 at McMurdo,
78°S (Johnson et aL, 1994), Syowa, 69°S (T. Ito, private
communication), and Georg Forster stations (71°S)
(H. Gernandt, private communication), indicating that
this depletion at lower altitudes was widespread. In addi-
tion, the 1993 springtime ozone loss was very severe in
the 18-22 km region, effectively extending the ozone de-
pletion region upward by about 1-2 km (Figure 1-27).
This occurred in spite of ozone being considerably high-
er than normal during the preceding winter (Figure
1-26). Complete ozone destruction from 14 to 19 km
was peculiar to 1993 and, combined with lower-than-
normal ozone at 20-22 km, resulted in the record low
total ozone recorded in early October 1993.
The decrease in summer ozone levels at 10-17 km
since the late 1960s is not apparent in the 1986-1993
data, possibly because the record is too short.
1.47
-------
OZONE MEASUREMENTS
REFERENCES
Akimoto, H., N. Nakane, and Y. Matsumoto, The chem-
istry of oxidant generation: Tropospheric ozone
increase in Japan, in The Chemistry of the Atmo-
sphere: Its Impact on Global Change, edited by
J.G. Calvert, Blackwell Sci. Publ., Oxford, 261-
273, 1994.
Albert-Levy, Analyse de 1'air, Annulaire de I'Obser. de
Montsouris, Gauthier-VUlars, Paris, 495-505,
1878.
Alt, J., P. Astapenko, and N.J. Roper, Some aspects of
the Antarctic atmospheric circulation in 1958, IGY
General Report Series, 4, 1-28, 1959.
Anfossi, D., S. Sandroni, and S. Viarengo, Tropospheric
ozone in the nineteenth century: The Montalieri
series,/ Geophys. Res., 96, 17349-17352, 1991.
Attmannspacher, W., and H. Diitsch, International ozone
sonde intercomparison at the Observatory of
Hohenpeissenberg, Berichte des Deutschen Wet-
terdienstes, 120, 85 pp., 1970.
Attmannspacher, W., and H. Dutsch, 2nd International
ozone sonde intercomparison at the Observatory
of Hohenpeissenberg, Berichte des Deutschen
Wetterdienstes, 157, 65 pp., 1981.
Attmannspacher, W., J. de la Noe, D. De Muer, J. Leno-
ble, G. Megie, J. Pelon, P. Pruvost, and R. Reiter,
European validation of SAGE II ozone profiles, J.
Geophys. Res., 94, 8461-8466, 1989.
Barnes, R.A., .A.R- Bandy, and A.L. Torres, Electro-
chemical concentration cell ozonesonde accuracy
and precision, J. Geophys. Res., 90, 7881-7887,
1985.
Barnes, R.A., L.R. McMaster, W.P. Chu, M.P. McCor-
mick, and M.E. Gelman, Stratospheric Aerosol
and Gas Experiment II and ROCOZ-A ozone pro-
files at Natal, Brazil: A basis for comparison with
other satellite instruments, J. Geophys. Res., 96,
7515-7530, 1991.
Bass, A.M., and R J. Paur, The ultraviolet cross-sections
of ozone: I. The measurements, in Atmospheric
Ozone, edited by C.S. Zerefos and A.M. Ghazi,
Reidel, Dordrecht, The Netherlands, 606-610,
1985.
Bekki, S., R. Toumi, and J.A. Pyle, Role of sulphur pho-
tochemistry in tropical ozone changes after the
eruption of Mount Pinatubo, Nature, 362, 331-
333, 1993.
Bhartia, P.K., J. Herman, R.D. McPeters, and 0- Torres,
Effect of Mt. Pinatubo aerosols on total ozone
measurements from backscatter ultraviolet (BUY)
experiments, J. Geophys. Res., 98, 18547-18554,
1993.
Bluth, G.J.S., S.D. Doiron, C.C. Schnetzler, A.J. Krue-
ger, and L.S. Walter, Global tracking of the SO2
clouds from the June 1991 Mount Pinatubo erup-
tions, Geophys. Res. Lett., 19, 151-154, 1992.
Bojkov, R.D., Surface ozone during the second half of
the nineteenth century, J. dim. Appl. Meteorol.,
25, 343-352, 1986.
Bojkov, R.D., The 1983 and 1985 anomalies in ozone
distribution in perspective, Man. Wea. Rev., 115,
2187-2201, 1987.
Bojkov, R.D., Ozone changes at the surface and in the
free troposphere, in Tropospheric Ozone, Regional
and Global Scale Interactions, edited by I.S.A.
Isaksen, NATO ASI Series C, 227, 83-96, Reidel,
Dordrecht, The Netherlands, 1988.
Bojkov, R.D., L. Bishop, WJ. Hill, G.C. Reinsel, and
G.C. Tiao, A statistical trend analysis of revised
Dobson total ozone data over the Northern Hemi-
sphere, J. Geophys. Res., 95, 9785-9807, 1990.
Bojkov, R.D., C.S. Zerefos, D.S. Balis, I.C. Ziomas, and
A.F. Bais, Record low total ozone during northern
winters of 1992 and 1993, Geophys. Res. Lett., 20,
1351-1354, 1993.
Bojkov, R.D., V.E. Fioletov, and A.M. Shalamjansky,
Total ozone changes over Eurasia since 1973
based on reevaluated filter ozonometer data, J.
Geophys. Res., 99, 22985-22999, 1994.
Brasseur, G., and C. Granier, Mount Pinatubo aerosols,
chlorofluorocarbons and ozone depletion, Science,
257, 1239-1242, 1992.
CMlonge, D., and E. Vassy, Spectrographic astigma-
tique a prisme objectif, Rev. Opt. Theor. Instrum.,
73,113-126,1934.
Chandra, S., Changes in stratospheric ozone and temper-
ature due to the eruption of Mt. Pinatubo,
Geophys. Res. Lett., 20, 33-36, 1993.
1.48
-------
OZONE MEASUREMENTS
Chu, W.P., M.P. McCormick, J. Lenoblb, C. Brogniez,
and P. Pruvost, SAGE II inversion algorithm, J.
Geophys. Res., 94, 8339-8351, 1989.
Chubachi, S., Preliminary result of ozone observations
at Syowa Stations from February, 1982 to January
1983, Mem. Natl. Inst. Polar. Res^ Spec. Iss., 34,
13-18, 1984. !;
Cunnold, D.M., W.P. Chu, R.A. Barnes! M.P. McCor-
mick, and R.E. Veiga, Validation of SAGE II
ozone measurements, /. Geophys. Res., 94, 8447-
8460, 1989. .:;
De Muer, D., Comment on "Presumptive; Evidence for a
Low Value of the Total Ozone Content above Ant-
arctica in September, 1958" by P.'iRigaud and B.
Leroy, Annales Geophysicae, 11, 795-796, 1990.
De Muer, D., H. De Backer, R.E. Veiga, and J.M.
Zawodny, Comparison of SAGE ll ozone mea-
surements and ozone soundiiigs at Uccle
(Belgium) during the period February 1985 to Jan-
uary 1986, J. Geophys. Res., 95,.' 11903-11911,
1990. |;
De Muer, D., and H. De Backer, Revision of 20 years of
Dobson total ozone data at Uccle (Belgium): Ficti-
tious Dobson total ozone trends induced by sulfur
dioxide trends, / Geophys. Res., 97, 5921-5937,
1992. !
De Muer, D., and H. De Backer, Influence of sulphur
dioxide trends on Dobson measurements and on
electrochemical ozone soundings, I in Proceedings
of the Society of Photo-Optical Instrumentation
Engineers, 2047, 33, 18-26, 1993.'!
De Muer, D., and H. De Backer, Trend; analysis of 25
years of regular ozone soundings iat Uccle (Bel-
gium), Abstract, EUROTRAC meeting, Garmisch-
Partenkirchen, April 1994. -.'.
Degtfrska, M., B. Rajewska-Wiech, and R.B. Rybka, To-
tal ozone over Belsk: Updating of .trends through
1991 and comparison of the Dobson data with
those from TOMS in 1978-1991, Publs. Inst. Geo-
phys. Pol. Acad. Soc., 252, 53-58, 1992.
DeLuisi, J.J., D.A. Londenecker, C.L. Mateer, and D.J.
Wuebbles, An analysis of Northern middle latitude
Umkehr measurements corrected for stratospheric
aerosols for 1979-1986, /. Geophys. Res., 94,
9837-9846, 1989. ;
DeLuisi, J.J., C.L. Mateer, D. Theisen, P.K. Bhartia, D.
Longenecker, and B. Chu, Northern middle-lati-
tude ozone profile features and trends observed by
SBUV and Umkehr, 1979-1990, J. Geophys. Res.,
99, 18901-18908, 1994.
Deshler, T., A. Adriani, G.P. Gobbi, D.J. Hofmann, G. Di
Donfrancesco, and B.J. Johnson, Volcanic aerosol
and ozone depletion within the Antarctic polar
vortex during the austral spring of 1991, Geophys.
Res. Lett., 19, 1819-1822, 1992.
Deshler, T., B.J. Johnson, and W.R. Rozier, Balloon-
borne measurements of Pinatubo aerosol during
1991 and 1992 at 41°N: Vertical profiles, size dis-
tribution and volatility, Geophys. Res. Lett., 20,
1435-1438, 1993.
Dobson, G.M.B., Observers handbook for the ozone
spectrophotometer, Ann. IGY, 5, 46-89, 1957.
Dobson, G.M.B., Forty years' research on atmospheric
ozone at Oxford: A history, Applied Optics, 7,
401, 1968.
Dutsch, H.U., W. Zuliz, and C.C. Ling, Regular ozone
observations at Thalwill, Switzerland and at Boul-
der, Colorado, Rep. LAPETH 1, Lab. Atmo-
spharenphys. Eidigenoss., Tech. Hochsch., Zurich,
1970.
Fabry, C., and H. Buisson, L'absorption des radiations
dans la haute atmosphere, Memorial Sc. Phys., 11,
1-64, 1930.
Farman, J.C., A. O'Neill, and R. Swinbank, The dynam-
ics of the arctic polar vortex during the EASOE
campaign, Geophys. Res. Lett., 21, 1195-1198,
1994.
Feister, U., and W. Warmbt, Long-term measurements of
surface ozone in the German Democratic Repub-
lic,/ Atmos. Chem., 5, 1-21, 1987.
Fox, C.B., Ozone and Antozone, Churchill Publ., Lon-
don, 1873.
Furrer, R., W. Dohler, H. Kirsch, P. Plessing, and U.
Gorsdorf, Evidence for vertical ozone redistribu-
tion since 1967, J. Atmos. Terr. Phys., 11,
1423-1445, 1992.
Furrer, R., W. Dohler, H. Kirsch, P. Plessing, and U.
Gorsdorf, Evidence for vertical ozone redistribu-
tion since 1967, Surveys in Geophysics, 14,
197-222, 1993.
1.49
-------
OZONE MEASUREMENTS
Gleason, J.F., P.K. Bhartia, J.R. Herman, R. McPeters,
P. Newman, R.S. Stolarski, L. Flynn, G. Labow,
D. Larko, C. Seftor, C. Wellemeyer, W.D. Komhyr,
A.J. Miller, and W. Planet, Record low global
ozone in 1992, Science, 260, 523-526, 1993.
Go'rsdorf, U., H. Steinhagen, and H. Gernandt, Bestim-
mung der Forderleistung von Ozonsondenpumpen
bei veranderlichen Messbedingungen, Z Meteo-
ro., in press, 1994.
G6tz, F.W.P., Zum Strahlungsklima des Spitzbergen-
sommers. Strahlungs- und Ozonmessungen in der
KQnigsbucht 1929, Gerlands Beitr. Geophys., 31,
119,1931. '
G5tz, F.W.P., and F. Volz, Aroser Messungen des
Ozonegehaltes in der unteren Troposphare und
sein Jahresgang, Z Naturforsch., 6a, 634-639,
1951.
Grass, R.D., and W.D. Komhyr, Traveling standard lamp
calibration checks of Dobson ozone spectropho-
tometers during 1985-1987, in Ozone in the
Atmosphere, edited by R.D. Bojkov and P. Fabian,
A. Deepak Publ., Hampton, Virginia, 144-146,
1989.
Henriksen, K.,T. Sven0e, and S.H.H. Larsen, On the sta-
bility of the ozone layer at Troms0, /. Atmos. Terr.
Phys.,54, 1113-1117, 1992.
Henriksen, K., EJ. Terez, G.A. Terez, V. Roldugin, T.
Sven0e, and S.H.H. Larsen, On the stationarity of
the ozone layer in Norway and U.S.S.R., J. Atmos.
Terr. Phys., 55, 145-154, 1993.
Herman, J.R., R.D. Hudson, R.D. McPeters, R.S. Stolar-
ski, Z. Ahmad, X-Y. Gu, S.L. Taylor, and C.G.
Wellemeyer, A new self-calibration method ap-
plied to TOMS and SBUV backscattered data to
determine long-term global ozone change, /. Geo-
phys. Res., 96. 7531-7545, 1991.
Herman, J.R., and D. Larko, Low ozone amounts during
1992-1993 from Nimbus 7 and Meteor 3 Total
Ozone Mapping Spectrometers, J. Geophys. Res.,
99, 3483-3496, 1994.
Hilsenrath, E., D.E. Williams, R.T. Caffrey, R.P. Cebula,
and S.J. Hynes, Calibration and radiometric stabil-
ity of the Shuttle Solar Backscattered Ultra-Violet
, (SSBUV) experiment, Metrologia, in press, 1994.
Hofmann, D.J., S.J. Oltmans, I.M. Harris, W.D. Komh-
yr, J.A. Lathrop, T. DeFoor, and D. Kuniyuki,
Ozonesonde measurements at Hilo, Hawaii fol-
lowing the eruption of Pinatubo, Geophys. Res.
Lett., 20, 1555-1558, 1993.
Hofmann, D.J., S.J. Oltmans, W.D. Komhyr, J.M. Har-
ris, J.A. Lathrop, A.O. Langford, T. Deshler, B.J.
Johnson, A. Torres, and W.A. Matthews, Ozone
loss in the lower stratosphere over the United
States in 1992-1993: Evidence for heterogeneous
chemistry on the Pinatubo aerosol, Geophys. Res.
Lett., 21, 65-68, 1994a.
Hofmann, D.J., S.J. Oltmans, J.A. Lathrop, J.M. Harris,
and H. Voemel, Record low ozone at the South
Pole in the spring of 1993, Geophys. Res.,Lett., 21,
421-424, 1994b.
Hofmann, D.J., Recovery of stratospheric ozone over the
U.S. in the winter of 1993-1994, Geophys. Res.
Lett., 21, 1779-.1782, 1994. ,
Hood, L.L., and J.P. McCormack, Components of inter-
annual ozone change based on Nimbus 7 TOMS
data, Geophys. Res. Lett., 19, 2309-2312, 1992.
Hood, L.L., R.D. McPeters, J.P. McCormack, L.E.
Flynn, S.M. Hollandsworth, and J.F. Gleason, Al-
titude dependence of stratospheric ozone trends
based on Nimbus 7 SBUV data, Geophys. Res.
Lett., 20, 2667-2670, 1993.
Jaffe, D.A., Local sources of pollution in the Arctic:
From Prudhoe Bay to the Taz Peninsula, Pollution
of the Arctic Atmosphere, edited by W.T. Sturges,
Elsevier, New York, 255-287, 1991. ]
Johnson, B.J., T. Deshler, and W.R. Rozier, Ozone pro-
files at McMurdo station, Antarctica during the
austral spring of 1992, Geophys. Res. Lett., 21,
269-272, 1994. :
Kerr, J.B., C.T. McEIroy, D.I. Wardle, R.A. Olafson, and
W.F.J. Evans, The automated Brewer spectropho-
tometer, in Proceedings 1984 Quadrennial Ozone
Symposium, edited by C.S. Zerefos and A. Ghazi,
Reidel Publ., Dodrecht, The Netherlands, 396,
1985. :
Kerr, J.B., LA. Asbridge, and W.F.J. Evans, Intercompar-
ison of total ozone measured by the Brewer and
Dobson spectrophotometers at Toronto, J. Geo-
phys. Res., 93, 11129-11140, 1988.
1.50
-------
Kerr, J.B., LA. Asbridge, and W.FJ. Evans, Long-term
intercomparison between the Brewer and Dobson
spectrophotometers at Edmonton, in Ozone in the
Atmosphere, edited by R.D. Bojkov and P. Fabian,
A. Deepak Publ., Hampton, Virginia, 105, 1989.
Kerr, J.B., D.I. Wardle, and D.W. Tarasick, Record low
ozone values over Canada in early 1993, Geophys.
Res. Lett., 20, 1979-1982, 1993. '•
Kerr, J.B., H. Fast, C.T. McElroy, S.J. pltmans, J.A.
Lathrop, E. Kyro, A. Paukkunen, H. Claude, U.
Kohler, C.R. Sreedharan, T. Takao, 'and Y. Tsuka-
goshi, The 1991 WMO international ozonesonde
intercomparison at Vanscoy, Canada Atmosphere-
Ocean, in press, 1994.
Kley, D., A. Volz, and F. Miilhiems, Ozone measure-
ments in historic perspective, in'Tropospheric
Ozone, Regional and Global Scale Interactions,
edited by I.S.A. Isaksen, NATO ASI Series C, 227,
63-72, Reidel Publ., Dordrecht, The Netherlands
1988.
Kley, D., H. Geiss, and V.A. Mohnen, Concentrations
and trends of tropospheric ozone and precursor
emissions in the USA and Europe, in The Chemis-
try of the Atmosphere: Its Impact on Global
Change, edited by J.G. Calvert, Blackwell Sci.
Publ., Oxford, 245-259, 1994. ;
Komhyr, W., andT.B. Harris, Note on flow rate measure-
ments made on Brewer-Mast ozone sensor pumps,
Mon..Wea.. Rev., 93, 267-268, 1965.
Komhyr, W.D., Electrochemical concentration cells for
gas analysis, Ann. Geophys., 25, 203-210, 1969.
Komhyr, W.D., and T.B. Harris, Development of the
ECC ozonesonde, NOAA Technical Report ERL
200-APCL 18, U.S. Department of, Commerce,
NOAA Environmental Research Laboratories,
Boulder, Colorado; 54 pp., 1971.
Komhyr, W.D., S.J. Oltmans, R.D. Grass, and R.K.
Leonard, Possible influence of long-term sea sur-
face temperature anomalies in the tropical Pacific
on global ozone, Canadian. J. Phys., 65, 1093-
1102,1991. ;
Komhyr, W.D., C.L. Mateer, and R.D. Hudson, Effective
Bass-Paur 1985 ozone absorption coefficients for
use with Dobson ozone spectrophotometers, /.
Geophys. Res., 98, 451-465, 1993.
OZONE MEASUREMENTS
Komhyr, W.D., R.D. Grass, R.D. Evans, R.K. Leonard,
D.M. Quincy, D.J. Hofmann, and G.L. Koenig,
Unprecedented! 1993 ozone decrease over the
United States from Dobson spectrophotometer
observations, Geophys. Res. Lett., 21 210-214
1994a.
Komhyr, W.D., J.A., Lathrop, D.P. Opperman, R.A.
Barnes, and G.B. Brothers, ECC ozonesonde per-
formance evaluation during STOIC 1989 J
Geophys. Res., in press, 1994b.
Krzyscin, J.W., Interannual changes in the atmospheric
ozone derived from ground-based measurements,
Papers in Met. and Geophys., 43, 133-164, 1992.
Krzyscin, J.W., Total ozone changes in the Northern
Hemisphere mid-latitudinal belt (30-60°N) de-
rived from Dobson spectrophotometer measure-
ments, 1964-1988, J. Afmos. Terr. Phys 56
1051-1056, 1994a.
Krzyscin, J.W., Long-term changes in the statistical dis-
tribution of Dobson total ozone in selected
Northern Hemisphere geographical regions, in
Proceedings of the Quadrennial Ozone Sympo-
sium 1992, Charlottesville, Virginia, NASA
CP-3266, 207-210, 1994b.
Krzyscin, J.W., Statistic analysis of annual total ozone
extremes for the period 1964-1988, in Proceed-
ings of the Quadrennial Ozone Symposium 1992,
Charlottesville, Virginia, NASA CP-3266, 203-
206, 1994c.
Kundu, N., and M. Jain, Total ozone trends over low lat-
itude Indian stations, Geophys. Res. Lett., 20
• 2881-2883, 1993,
Labitske, K., and M.P. McCormick, Stratospheric tem-
perature increases due to Pinatubo aerosols,
Geophys. Res. Lett,, 19, 207-210, 1992.
Lacoste, A.M., S. Godin, and G. Megie, Lidar measure-
ments and Umkehr observations of the ozone
vertical distribution at the Observatoire de Haute-
Provence, J. Atmos. Terr. Phys., 54, 571-582
1992. •
Lambert, A., R.G. Grainger, JJ. Remedies, D.C. Rod-
gers, M. Corney, and F.W. Taylor, Measurements
of the evolution of the Mt. Pinatubo aerosol clouds
by ISAMS, Geophys. Res. Lett., 20, 1287-1290
1993.
1.51
-------
OZONE MEASUREMENTS
Lehmann, P., A statistical seasonal analysis of winter
decreases in ozone at Macquarie Island, Geophys.
Res. Lett., 21, 381-384, 1994.
Linvill, D.E., W.J. Hooken, and B. Olson, Ozone in
Michigan's environment (1876-1880), Man. Wea.
Rev., 108, 1883-1891, 1980.
Logan, J.A., Trends in the vertical distribution of ozone:
An analysis of ozonesonde data, J. Geophys. Res.,
in press, 1994.
London, J., and S. Liu, Long-term tropospheric and
lower stratospheric ozone variations from ozone-
sondes observations, J. Atmos. Terr. Phys., 5,
599-625, 1992.
Marenco, A., H. Gouget, P. N6delec, J.-P. Pages, and
R Karcher, Evidence of a long-term increase in
tropospheric ozone from Pic du Midi data series
—Consequences: Positive radiative forcing, /.
Geophys. Res., 99, 16617-16632, 1994.
Marko, J.R., and D.B. Fissel, Empirical linkages be-
tween Arctic sea ice extents and Northern
Hemisphere mid-latitude column ozone levels,
Geophys. Res. Lett., 20, 37-40, 1993.
Mateer, C.L., and J.J. DeLuisi, A new Umkehr inversion
algorithm, J. Atmos. Terr. Phys., 54, 537-556,
1992.
McCormick, M.P., R.E. Veiga, and W.P. Chu, Strato-
spheric ozone profile and total ozone trends
derived from the SAGE I and SAGE II data, Geo-
phys. Res, Lett., 19, 269-272, 1992.
McPeters, R.D., and W.D. Komhyr, Long-term changes
in TOMS relative to World Primary Standard Dob-
son Spectrometer 83, J. Geophys. Res., 96,
2987-2993, 1991.
McPeters, R.D., T. Miles, L.E, Flynn, C.G. Wellemeyer,
and J.M. Zawodny, Comparison of SBUV and
SAGE II ozone profiles: Implications for ozone
trends,/. Geophys. Res., 99, 20513-20524, 1994.
Miller, A.J., G.C. Tiao, G.C. Reinsel, D. Wuebbles, L.
Bishop, J. Kerr, R.M. Nagatani, and J.J. DeLuisi,
Comparisons of observed ozone trends in the
stratosphere through examination of Umkehr and
balloon ozonesonde data, in preparation, 1994.
Mylona, S., Trends of sulphur dioxide emissions, air
concentrations and depositions of sulphur in Eu-
rope since 1880, in European Monitoring and
Evaluation Program/Meteorological Synthesis
Center-West Report 2/93, August 1993.
Naujokat, B., K. Petzoldt, K. Labitzke, R. Lenschow, B.
Rajewski, M. Wiesner, and R-C. Wohlfart, The
stratospheric winter 1992/1993: A cold winter
with a minor wanning and a late final warming,
Beilage zur Berliner Wetterkarte, SO 21/93, 1993.
Newman, P.A., Antarctic ozone hole in 1958, Science,
264, 543-546, 1994.
NO AA (National Oceanic and Atmospheric Administra-
tion), Southern Hemisphere Winter Summary,
Selected indicators of stratospheric climate, Cli-
mate Analysis Center, 1993.
Oltmans, S.J., W.D. Komhyr, P.R. Franchois, and W.A.
Matthews, Tropospheric ozone: Variations from
surface and ECC ozonesonde observations, in
Ozone in the Atmosphere, edited by R.D. Bojkov
and P. Fabian, A. Deepak Publ., Hampton, Virgin-
ia, 539-543, 1989.
Oltmans, S.J., D.J. Hofmann, W.D. Komhyr, and J.A.
Lathrop, Ozone vertical profile changes over
South Pole, in Proceedings of the Quadrennial
Ozone Symposium 1992, Charlottesville, Virginia,
13 June 1992, NASA CP-3266,578-581, 1994.
Oltmans, S.J., and H. Levy II, Surface ozone measure-
ments from a global network, Atmos. Environ., 28,
9-24, 1994.
Oltmans, S.J., and D.J. Hofmann, Trends in the vertical
profile of ozone at Hilo, Hawaii, frorh 1982 to
1994, in preparation, 1994.
Ozone Data for the World, Atmospheric Environment
Service, Department of the Environment in coop-
eration with the World Meteorological Organiza-
tion, Downsview, Ontario, Canada, 34, 3, 1993.
Paur, R.J., and A.M. Bass, The ultraviolet cross-sections
of ozone: II. Results and temperature dependence,
in Atmospheric Ozone, edited by C.S. Zerefos and
A.M. Ghazi, Reidel Publ., Dordrecht, The Nether-
lands, 611-616, 1985.
1.52
-------
OZONE MEASUREMENTS
Perl, G., Das bodennahe Ozon in Arosa,, iseine regelmas-
sigen und unregelmassigen Schwankungen, Arch.
Met. Geophys. Bioklimat., 14, 449r458, 1965.
Placet, M., Emissions involved in acidic; deposition pro-
cesses, in Acidic Deposition: State of Science and
Technology, Summary report of die U.S. National
Acid Precipitation Assessment Program, edited by
P.M. Irving, 1991. ;|
Pommereau, J.-P., and F. Goutail, O3 aiiid NO2 ground-
based measurements by visible spectrometry
during Arctic winter and spring 1988, Geophys.
Res. Lett., 15, 891-895, 1988. }\
Randel, W.J., and J.B. Cobb, Coherent variations of
monthly mean total ozone and lovifer stratospheric
temperature, J. Geophys. Res., 99, 5433-5447,
1994. !
Read, W.G., L. Froidevaux, and J.W. Waters, Microwave
Limb Sounder measurement of stratospheric 862
from the Mt. Pinatubo volcano, | Geophys. Res.
Lett., 20, 1299-1302, 1993. !|
Reinsel, G.C., G.C. Tiao, A.J. Miller, D.J. Wuebbles,
P.S. Connell, C.L. Mateer, and J. D'eLuisi, Statisti-
cal analysis of total ozone arid stratospheric
Umkehr data for trends and solar cycle relation-
ship, J. Geophys. Res., 92, 2201-2209, 1987.
Reinsel, G.C., G.C. Tiao, D.J. Wuebblesj J.B. Kerr, A.J.
Miller, R.M. Nagatani, L. Bishop,, land L.H. Ying,
Seasonal trend analysis of published ground-based
and TOMS total ozone data through 1991, /. Geo-
phys. Res., 99, 5449-5464, 1994a. |
Reinsel, G.C., W.-K. Tarn, and L.H. Ying, Comparison
of trend analyses for Umkehr data using new and
previous inversion algorithms, Geophys. Res.
Lett., 21, 1007-1010, 1994b. :;
Rigaud, P., and B. Leroy, Presumptive evidence for a low
value of the total ozone content above Antarctica
in September 1958, ^Annales Gepphysicae, II,
791-794, 1990. !!
Sandroni, D., D. Anfossi, and S. Viarengo, Surface
ozone levels at the end of the nineteenth century in
South America, /. Geophys. Res., 97, 2535-2540,
1992. ||
Scheel, H.E., E.G. Brunke, and W. Seiler, Trace gas mea-
surements at the monitoring station Cape Point,
South Africa, between 1978 and' 1988, J. Atm.
Chem., 11, 197-210, 1990.
Scheel, H.E., R. Sladovic, and W. Seiler, Ozone related
species at the stations Wank and Zugspitze:
Trends, short-term variations and correlations with
other parameters, in Photo-Oxidants: Precursors
and Products, Proc. EUROTRAC Symp. 1992,
edited by P.M. Borell, P. Borell, T. Cvitas, and W.
Seiler, Acad. IPubl., The Hague, The Netherlands,
104-108, 1993.
Schenkel, A., and B. Broder, Interference of some trace
gases with ozone measurements by the Kl-meth-
od, Atmos. Environ., 16, 2187-2190, 1982.
Schoeberl, M.R., P.K. Bhartia, and E. Hilsenrath, Tropi-
cal ozone loss following the eruption of Mt.
Pinatubo, Geophys. Res. Lett., 20, 29-32, 1993.
Shiotani, M., Annual, quasi-biennial and El Nino-South-
ern Oscillation (ENSO) time scale variations in
equatorial total ozone, / Geophys. Res., 97, 7625-
7633, 1992.
Sladkovic, R., H.E. Scheel, and W. Seiler, Ozone clima-
tology of the mountain sites, Wank and Zugspitze,
in Transport and Transformation of Pollutants in
the Troposphere, Proc. EUROTRAC Symp. April
1994, edited by P.M. Borrcll, P. Borrell, P. Cvitas,
and W. Seiler, in press, 1994.
Staehelin, J., and W. Schmid, Trend analysis of tropo-
spheric ozone concentrations utilizing the 20-year
data set of ozone balloon soundings over Payerne,
Atmos. Environ., 9, 1739-1749, 1991.
Staehelin, J., J. Thudium, R. Buhler, A. Volz-Thomas,
and W. Graber,. Trends in surface ozone concentra-
tions at Arosa (Switzerland), Atmos. Environ., 28,
75-87, 1994.
Stolarski, R.S., and A.J. Krueger, Variations of total
ozone in the North Polar region as seen by TOMS,
paper presented at the Polar Ozone Workshop,
NASA/NOAAfNSF/CMA, Snowmass, Colorado,
May 9-13, 1988.
Stolarski R.S., P. Bloomfield, R. McPeters, and J. Her-
man, Total ozone trends deduced from Nimbus 7
TOMS Data, Geophys. Res. Lett., 18, 1015-1018,
1991.
Stolarski, R., R. Bojkov, L. Bishop, C. Zerefos, J. Stae-
helin, and J. Zawodny, Measured trends in
stratospheric ozone, Science, 256, 342-349, 1992.
1.53
-------
OZONE MEASUREMENTS
Tarasick, D.W., D.I. Wardle, J.B. Kerr, J.J. Bellefleur,
and J. Davies, Tropospheric ozone trends over
Canada: 1980-1993, submitted to Geophys. Res.
Lett., 1994.
Taylor, S.L., R.D. McPeters, and P.K. Bhartia, Proce-
dures to validate/correct calibration error in solar
backscattered ultraviolet instruments, in Proceed-
ings of the Quadrennial Ozone Symposium 1992,
Charlottesville, Virginia, NASA CP-3266, 923-
926, 1994.
Thornton, D., and N. Niazy, Effects of solution mass
transport on the ECC ozonesonde background cur-
rent, Geophys. Res. Lett., 10, 148-151, 1983.
Trepte, C.R., R.E. Veiga, and M.P. McCormick, The
poleward dispersal of Mount Pinatubo volcanic
aerosol, J. Geophys. Res., 98, 18563-18573,1993.
Vigroux, E., Determination des coefficients moyen
d'absorption de 1'ozone, Ann. Phys., 8, 709-762,
1953.
Volz, A., and D. Kley, Evaluation of the Montsouris se-
ries of ozone measurements made in the nine-
teenth century. Nature, 332, 240-242, 1988.
Volz-Thomas, A., Trends in photo-oxidant concentra-
tions, in Photo-Oxidants: Precursors and
Products, Proc. EUROTRAC Symp. 1992, edited
by P.M. Borrell, P..Borrell, T. Cvitas, and W.
Seiler, Acad. Publ., The Hague, The Netherlands,
59-64, 1993.
Wege, K., H. Claude, and R. Hartmannsgruber, Several
results from the 20 years of ozone observations at
Hohenpeissenberg, in Ozone in the Atmosphere,
edited by R.D. Bojkov and P. Fabian, A. Deepak
Publ., Hampton, Virginia, 109-112, 1989.
Wellemeyer, C.G., S.L. Taylor, C.J. Seftor, and R.D.
McPeters, TOMS profile shape errors estimates at
high latitude, in Proceedings of the Symposium on
• High Latitude Optics, Troms0, Norway, 1993.
WMO, Report of the International Ozone Trends Panel:
1988, World Meteorological Organization Global
Ozone Research and Monitoring Project—Report
No. 18, Geneva, 1990a.
WMO, Scientific Assessment of Stratospheric Ozone:
1989, World Meteorological Organization Global
Ozone Research and Monitoring Project—Report
No. 20, Geneva, 1990b.
WMO, Scientific Assessment of Ozone Depletion: 1991,
World Meteorological Organization Global Ozone
Research and Monitoring Project—Report No. 25,
Geneva, 1992a.
WMO, Handbook for Dobson Ozone Data Re-evalua-
tion, World Meteorological Organization Global
Ozone Research and Monitoring Project—Report
No. 29, Geneva, 1992b.
WMO, Proceedings of the Fourth WMO/NOAA Work-
shop on Ozone Data Re-evaluation, World
Meteorological Organization Global Ozone Re-
search and Monitoring Project—Report No. 36,
Geneva, in preparation, 1994a.
WMO, Proceedings of a Workshop on Homogenizing
Total Ozone Records for Use in Ozone Assess-
ments and Early Warning of Ozone Changes,
World Meteorological Organization Global Ozone
Research and Monitoring Project—Report No. 35,
Saloniki 1993, in preparation, 1994b.
Young, R.E., H. Houben, and O.B. Toon, Radiatively
forced dispersion of the Mt. Pinatubo volcanic
cloud and induced temperature perturbations in
the stratosphere during the first few months fol-
lowing the eruption, Geophys. Res. Lett., 21.
369-372, 1994.
Zerefos, C.S., On the quasi-biennial oscillation in equa-
torial stratospheric temperatures and total ozone,
Adv. Space. Res., 2, 177-181, 1983.
Zerefos, C.S., A.F. Bais, I.C. Ziomass, and R.D. Bojkov,
On the relative importance of quasi-biennial oscil-
lation and El Nino-Southern Oscillation in the
revised Dobson total ozone records, J. Geophys.
Res., 97, 10135-10144, 1992. '•
Zerefos, C., K. Tourpali, and A. Bais, Further studies on
possible volcanic signal in total ozone, J. Geo-
phys. Res., in press, 1994.
1.54
-------
CHAPTER 2
Soufce Gases: Trends and Budgets
Lead Author:
E. Sanhueza
Co-authors:
PJ. Fraser
R.J. Zander
Contributors:
F.N. Alyea
M.O. Andreae
J.H. Butler
D.N. Cunnold
J. Dignon
E. Dlugokencky
D.H. Ehhalt
J.W. Elkins
D. Etheridge
D.W. Fahey
D.A. Fisher
J.A. Kaye
M.A.K. Khalil
P. Middleton
P.C. Novell!
J. Penner
M.J. Prather
R.G. Prinn
W.S. Reeburgh
J. Rudolph
P. Simmonds
L.P. Steele
M. Trainer
R.F. Weiss
D.J. Wuebbles
-------
-------
il CHAPTER 2
SOURCE GASES: TRENDS AND BUDGETS
Contents
!' ' -
h
SCIENTIFIC SUMMARY 2.1
2.1 INTRODUCTION ;! 2.3
2.2 HALOCARBONS ,! 2.3
2.2.1 Tropospheric Distributions Jind Trends 2.3
CFCs and Carbon Tetrachloride
Methyl Chloroform and HClpCs
Brominated Compounds |
Perfluorinated Species
OtherHalogenatedSpecies \.
2.2.2 Stratospheric Observations ,1. 2.8
2.2.3 Sources of Halocarbons :: 2.11
2.2.4 Halocarbon Sinks : I... 2.14
2.2.5 Lifetimes ......J 2.14
2.3 STRATOSPHERIC INPUTS OF CHLORINE AND PARTICULATES FROM ROCKETS 2.15
2.3.1 Stratospheric Chlorine Input: 2.15
2.3.2 Particulates from Solid-Fuel Rockets 2.15
2.4 METHANE 2.16
2.4.1 Atmospheric Distribution and Trends 2.16
2.4.2 Sources .>... 2.18
2.4.3 Sinks .; „ 2.20
2.4.4 Potential Feedbacks from a Changed Climate ; 2.20
2.5 NITROUS OXIDE , : 2.20
2.5.1 Atmospheric Distribution and Trends 2.20
2.5.2 Sources |i ..„ 2.21
2.5.3 Sinks ij 2.22
2.6 SHORT-LIVED OZONE PRECURSOR GASES 2.22
2.6.1 Nitrogen Oxides 2.22
2.6.1.1 Tropospheric Distribution 2.22
2.6.1.2 Sources ;.; ;....2.22
2.6.1.3 Sinks i : 2.23
2.6.2 Non-Methane Hydrocarbons;! 2.24
2.6.2.1 Atmospheric Distribution .'. , 2.24
2.6.2.2 Sources .! 2.24
2.6.2.3 Sinks .; 2.24
2.6.3 Carbon Monoxide J 2.24
2.6.3.1 Atmospheric Distribution and Trends .'. 2.24
2.6.3.2 Sources -. 2.25
2.6.3.3 Sinks ':• 2.26
i!
2.7 CARBON DIOXIDE '•;. Z..26
REFERENCES .1 : 2,27
-------
-------
,! SOURCE GASES
SCIENTIFIC SUMMARY . ,'
Tropospheric growth rates of the major anthropogenic source species for stratospheric chlorine and bromine
(chlorofluorocarbons (CFCs), carbon tetrachloride, methyl chloroform, halons) have slowed significantly, in response
to substantially reduced emissions required by the Montreal Protocol. Total tropospheric chlorine grew by about 60 pptv
(1.6%) in 1992 compared to 110 pptv (2.9%) in 1989. Tropospheric bromine in the form of halons grew by 0.2-0.3 pptv
in 1992, compared to 0.6-1.1 pptv in 1989.
Hydrochlorofluorocarbon (HCFC) growth rates are accelerating, as they are being used increasingly as CFG
substitutes. Tropospheric chlorine as HCFCs increased in 1992 by about 10 pptv, thus accounting for about 15% of total
tropospheric chlorine growth, compared to 5 pptv in 1989 (5% of tropospheric chlorine growth).
i
The atmospheric residence times! 'of CFC-11 and methyl chloroform are now better known. Model studies simu-
lating atmospheric abundances using miore realistic emission amounts have led to best-estimated lifetimes of 50 years
for CFC-11 and 5.4 years for methyl chloroform, with uncertainties of about 10%. These models, calibrated against
CFG-11 and methyl chloroform, are used to calculate the lifetimes, and hence ODPs (Ozone Depletion Potentials), of
other gases destroyed only in the stratosphere (other CFCs and nitrous oxide) and those reacting significantly with
tropospheric hydroxyl radicals (HCFCs and hydrofluorocarbons (HFCs)).
Methyl chloride, released from the oceans (natural) and biomass burning (anthropogenic), is a significant source
of tropospheric chlorine, contributing about 15% of the total tropospheric chlorine abundance in 1992 (3.8 ppbv). Data
collected from the late 1970s to the mid-1980s showed no long-term trend. A paucity of published observational data
since means that the likely existence of a global trend in this important species cannot be assessed further.
The total abundance of organic halocarbons in the lower stratosphere is well characterized by in situ and remote
observations of individual species. Observed totals are consistent with abundances of primary species in the tropo-
sphere, suggesting that other source species are not important in the stratosphere. Loss of halocarbons is found as their
residence time in the stratosphere incniases, consistent with destruction by known photochemical processes. Since the
loss of halocarbons produces inorganic: chlorine and bromine species associated with ozone loss processes, these obser-
vations also constrain the abundance of these organic species in the lower stratosphere.
Volcanoes are an insignificant source of stratospheric chlorine. Satellite and aircraft observations of upper and
lower stratospheric hydrochloric acid (HC1) are consistent with stratospheric chlorine being organic, largely anthropo-
genic, hi origin. No significant increase in HC1 was found in the stratosphere following the intense eruption of Mt.
Pinatubo in 1991. Elevated HC1 levels ;were detected in the eruption cloud of the El Chichon volcano in 1982, but no
related change in global stratospheric HC1 was observed.
> i •
The 1980s were characterized by'declining global methane growth rates, being approximately 20 ppbv per year in
1980 declining approximately monotonically to 10 ppbv per year by the end of the decade. Methane growth rates
slowed dramatically in 1991-1992, but probably started to increase in late 1993. During 1992 global methane levels
grew by only 5 ppbv. The causes of this global anomaly (which manifested predominantly at high latitudes in the
Northern Hemisphere) are not known with certainty, but are probably due to changes in methane sources rather than in
methane sinks. Global growth rate anomalies have been observed in methane records in the 1920s and 1970s from air
trapped in Antarctic ice. 11 ;
Despite the increased methane levels, the total amount of carbon monoxide (CO) in today's atmosphere is less
than it was a decade ago. Recent analyses of global CO data show that tropospheric levels grew from the early 1980s to
about 1987 and have declined from the late 1980s to the present. The causes of this behavior have not been identified.
i
11 2.1
-------
-------
2.1 INTRODUCTION
Recent trends of atmospheric trace gases are im-
portant in understanding stratospheric ozone depletion
and changes to the current radiative forcing of climate.
Estimates of budgets and lifetimes are required to pre-
dict future impacts. Likewise, these data are needed to
accurately predict what levels of emission1 reductions are
needed in order to stabilize and/or reduce present con-
centrations. In this assessment we will deal with gases
emitted by natural and/or anthropogenic sources that in-
fluence the chemical composition of the atmosphere. It
includes long-lived gases that contribute ito stratospheric
ozone depletion (/.&, chlorofluorocarbons (CFCs), ha-
lons, nitrous oxide (N2O)) and/or radiative forcing of the
atmosphere (Le., carbon dioxide (CO2), CFCs, methane
(CH4>, N2O), and short-lived compounds that are in-
volved in the 03 chemistry of the troposphere (i.e.,
carbon monoxide (CO), nitrogen oxides (NOX), non-
methane hydrocarbons). Reduced sulfur gases are
important in the formation of tropospheric aerosols and
therefore in the climate system; however, these com-
pounds are not included in this Chapter and the reader is
referred to Chapter 3 of the EPCC 1994 Interim Report
for an updated discussion of these source gases. The cur-
rent concentrations and recent trends of long-lived gases
are summarized in Table 2-1. Lifetimes (Le., total global
burden of the gas divided by its globally integrated sink
strength) .are also given. ''
'!
|
2.2 HALOCARBONS
Halocarbons play an important role; in stratospher-
ic ozone depletion and are powerful greerihouse gases. A
recent, comprehensive review (Kaye et al., 1994) has
provided extensive details on the global distributions,
trends, emissions, and lifetimes of CFCs '(chlorofluoro-
carbons), halons, and related species,; This section
provides an updated summary review. \
2.2.1 Tropospheric Distributions and Trends
Tropospheric measurements are mostly made in
situ at fixed sites distributed between the two hemi-
spheres, supplemented by data collected on ships and
aircraft (WMO, 1992). Information on the free tropo-
spheric burdens of atmospheric gases and their time
variations has further been obtained through spectro-
SOURCE GASES
scopic remote measurements made from various obser-
vational platforms. Recent concentration trends of
halocarbons and those reported in the 1991 Assessment
(WMO, 1992) are summarized in Table 2-2, indicating
that significant changes in trends have been observed for
most gases during the last few years. The total Cl in-
crease in 1992 was -60 pptv/yr, whereas the 1989
increase was -110 pptv/yr (WMO, 1992).
CFCs AND CARBON TETRACHLORIDE
CC13F (CFC-11), CC12F2 (CFC-12), CC12FCC1F2
(CFC-113), and carbon tetrachloride (CCLt) have been
measured in a number of global programs and their tro-
pospheric mixing ratios have been increasing steadily
over the past fifteen years (Fraser etal., 1994a, and refer-
ences therein).
There is now clear evidence that the growth rates
of the CFCs have slowed significantly in recent years
(Figure 2-1), presumably in response to reduced emis-
sions (see Section 2.2.3). CFC-12 and CFC-11 trends in
the late 1970s to late 1980s were about 16-20 pptv/yr
and 9-11 pptv/yr, respectively. These declined to about
16 and 7 pptv/yr, respiectively, around 1990 and to about
11 and 3 pptv/yr by 1993 (Elkins et al., 1993; Khalil and
Rasmussen, 1993a; Simmonds et al, 1993; Cunnold et
al., 1994; Makide etal., 1994; Rowland et al., 1994).
The global CFC-113 data up to the end of 1990
have been reviewed recently (Fraser et al., 1994a). A
global average trend of about 6 pptv/yr was observed for
CFC-113, with no sign of a slowing down such as ob-
served for CFCs-11 and -12. However, data up to the end
of 1992 now indicate that the growth rate has started to
decrease (Fraser et al., 1994a, b, c). Carbon tetrachloride
appears to have stopped accumulating in the atmosphere
and data collected at Cape Grim, Tasmania, indicate that
the background levels of this trace gas may have actually
started to decline (Fraser and Derek, 1994).
METHYL CHLOROFORM AND THE
HYDROCHLOROFLUOROCARBONS (HCFCs)
Global methyl chloroform (CH3CC13) and HCFC-
22 (CHC1F2) data up to the end of 1990 have been
reviewed by Fraser et al. (1994a), with growth rates in
1990 equal to 4-5 and 6-7 pptv/yr, respectively.
Methyl chloroform data up to the end of 1992 are
shown in Figure 2-2, indicating that the slowing of the
2.3
-------
SOURCE GASES
TABLE 2-1. Current atmospheric levels, changes in abundance (1992 minus 1990) and lifetimes of
long-lived trace gases. (Adapted from IPCC, 1994a.)
Species
CFC-11
CFC-12
CFC-113
CFC-114
CCLj
CH3CC13
CH3C1
HCFC-22
HCFC-141b
HCFC-142b
CH3Br
H-1211
H-1301
CF4
C2F6
SF6
N20(N)
C02(C)
Chem.
Formula
CC13F
CQ2F2
Ca2FCClF2
CC1F2CC1F2
CHC1F2
CH3CC12F
CH3CC1F2
(see Chapter 10)
CBrClF2
CBrF3
mixing ratios
(1992)
ppbv
0.268
0.503
0.082
0.020
0.132
0.160
0.600
0.102
0.0003
0.0035
0.0025
0.0020
0.070
0.004
[0.002-0.003]
310
1714
356000
growth
(1992-1990)
ppbv
0.005
0.026
0.005
0.001
-0.002
0.007
0.014
0.0001
0.0003
1.4
14.
2000
burden
(Tg)
6.2
10.3
2.6
3.4
3.5
5.0
1.5
0.08
0.05
0.9
1480
4850
760000
lifetime3
(years)
50 (±5)
102
85
300,
42
5.4 (±0.4)
1.5
13.3
9.4
19.5
20
65
50000
10000
3200
120
10b
(50-200)c
a Lifetimes of additional halocarbons are given in Chapter 13.
b The adjustment time is 12 to 17 years; this takes into account the indirect effect of methane on its own lifetime
(IPCC, 1994a).
c No single lifetime can be defined because of the different rates of uptake by different sink processes (IPCC, 1994b).
2.4
-------
SOURCE GASES
TABLE 2-2. Recent halocarbon trends compared with the values given in the 1991 assessment.
Compound
Period
This Assessment3
pptv/yr %/yr
1991 Assessment6
pptv/yr %/yr
CFC-11
CFC-12
CFC-113
CCLt
CH3CC13
HCFC-22
HCFC-142b
HCFC-141b
H-1211
H-1301
Total Cl
Total Brc
90-92
90-92
90-92
90-92
90-92
92
92
93
90-92
90-92
!:
is
13'
is
-1;
3.5
7.0
— 1
-0.75
1X075
0.16
-60
0.2-6.3
1 i
0.9
2.6
3.1
-0.8
2.2
6.9
-30
-200
3
8
9.3-10.1
16.9-18.2
5.4- 6.2
1- 1.5
4.8-5.1
5-6
n.d.
n.d.
0.2-0.4
0.4-0.7
-110
0.6- LI
3.7-3.8
3.7-4.0
9.1
1.2
3.7
6-7
n.d.
n.d.
15
20
a see text for references
b 1989 increase (WMO, 1992)
c bromine in the form of halons
growth rate observed in 1990 has continued, presumably
due to reduced emissions in 1991-92 as compared to
1990 and in part to increasing OH levels (1 ± 0.8 %/yr,
Prinn et al, 1992). The methyl chloroform calibration
problems detailed in Fraser et al. (1994a) have yet to be
resolved.
Recent global HCFC-22 data (Mpntzka et al.,
1993) indicate a global mixing ratio in 11992 of 102 ± 1
pptv, an interhemispheric difference of 13 ± 1 pptv, and a
globally averaged growth rate of 7.3 ± 03 %/yr, or 7.4 ±
0.3 pptv/yr, from mid-1987 to 1992. Bas
-------
SOURCE GASES
280
240
200
160
280
240
200
160
IE 26°
Ł 220
O 180
O 140
260
220
180
140
T—I—I—I—I—T
90°N-30°N
— Canada (NOAA)
— Akika(NOAA)
— Ireland (GAGE)
— Dragon (GAGE)
— Colorado (NOAA)
T T
T T
'II
30°N - EQ
— Maun* Lt» (NOAA)
— Barbados (QAGE)
1 1 I
30°S - 90°S
— Tasmania (NOAA)
— Tasmania (GAGE)
— South Pote (NOAA)
. \. -~rv
L.
I—I—I—I—L.
500
450
400
350
300
500
"Z 450
Q.
a. 400
.g 350
OJ 300
O>
* 475
s«
5 375
LL 325
° 275
475
425
375
325
275
T—I—I—I—T
90°N - 30°N
— Canada (NOAA)
— AlaiU(NOAA)
— Inland (SAGE)
— Oragon(GAGE)
— Colorado (NOAA)
I
L.
30°N - EQ
— MaunaLoa(NOAA)
— Barbados (GAGE)
EQ-30°S
— Samoa (NOAA)
— Samoa (QAQE)
'. 30°S-90°S
— Tasmania (NOAA)
— Ta
-------
SOURCE GASES
>
0.
0.
CO
Cd
0)
x
O
O
CO
X
O
180
160
140
120
160
140
120
100
140
120
100
80
i—i—i—i—i—r
90°N-30°N
--- Ireland (GAGE)
— Oregon (GAGE)
T 1 1 1—I T
n—r
-J—I—I—I—I
30°N-EQ
— Bartodo* (GAGE)
V
-I—U—1—1—L
J L.
_l L.
EQ-30°S
— Samoa (GAGE)
J—L.
J—I L.
.-1—L_
30°S - 90°S
— T«smml» (GAGE)
78
Figure 2-2. Monthly mean methyl chloroform mix-
ing ratios from the ALE/GAGE global network
(Prinn era/., 1992; Fraser etat., 1994a; Fraserand
Derek, 1994; ALE/GAGE unpublished data).
show distinct equatorial maxima, indicating a tropical
source related to natural biogenic activity.
PERFLUORINATED SPECIES
Perfluorinated compounds have very long life-
times (see Table 2.1) and strong infrared-red absorption
characteristics (efficient greenhouse gases). The j major
loss process appears to be their photolysis in the! upper
stratosphere and the mesosphere (for details see Chapter
12). !
The global mean concentration of carboii tetra-
fluoride (CF4) was measured in 1979 at 70 ± 7 pptv
(Penkett et al., 1981). This gas has been observed'at the
South Pole in the late 1970s and mid-1980s at about 65
and 75 pptv, respectively, growing at about 2%/yr (Kha-
lil and Rasmussen, 1985). Al: northern midlatitudes in the
mid-1980s, Fabian et al. (1987) reported CF4 and C2F6
concentrations at about 70 pptv and 2 pptv, respectively.
There have been no recent reports on CF4 or C^ in the
background atmosphere.
Sulfur hexafluoride (SFs) is a long-lived atmo-
spheric trace gas that is about three times more effective
as a greenhouse gas than CFC-11 (Ko et al., 1993). Cur-
rent global background levels are 2-3 pptv, which are
apparently increasing with lime at about 8.3%/yr (sur-
face measurements; Maiss and Levin, 1994) and 9 ± 1 %/
yr (lower stratosphere measurements; Rinsland et al.,
1993). IR column measurements in Europe (1986-1990)
and North America (1981-1990) indicate increases of
6.9 ± 1.4%/yr and 6.6 ± 3.6%/yr, respectively (Zander et
al, 1991a).
OTHER HALOGENATED SPECIES
Available data on the abundance of methyl chlo-
ride (CH3CI), chloroform (CHC13), dichloromethane
(CH2C12), and chlorinated ethenes have recently been
reviewed (Fraser et al., 1994a). No long-term trends of
these species have been observed, although they all ex-
hibit distinct annual cycles (summer minimum, winter
maximum). These species are relatively short-lived in
the atmosphere (see Table 2-1) and their contribution to
ozone depletion and climate forcing is minimal. Methyl
chloride is a significant source of tropospheric chlorine.
Data collected from the late 1970s to mid-1980s showed
no long-term trends (Khalil and Rasmussen, 1985). Re-
cent measurements of various of these gases have been
made in the Atlantic (45°N-30°S)(Koppmann et al.,
1993) and in the tropical Pacific Ocean (Atlas et al.,
1993). Methyl chloride showed practically no interhemi-
spheric gradient, indicative of a large oceanic or tropical
source, whereas chloroform, dichloromethane, tetra-
chloroethylene, and trichloroethylene showed higher
concentrations in the Northern Hemisphere likely due to
anthropogenic emissions.
Measurements of methyl iodide and chloro-
iodomethane in the NW Atlantic Ocean indicate that the
latter species may be as important as the former in trans-
ferring iodine from the oceans to the atmosphere (Moore
and Tokarczyk, 1993b).
2.7
-------
SOURCE GASES
CCI3F
Northern Mid- Latitudes
CHCIF2
40-
^
J30-
LU
Q 20-
H
H 10-
o-
1 1 1 1 1 =
!\
'"^N -
;»
€ -
!
i t f i
^^v. "
^V. m
\\^
\
\\
L \
X
i i i i -
1 OQ
:/o
k : -
i°| :
: O
O
|OO
-
i 1 I r * 1 i i < 1 i i i 1 i i i 1 i i i 1
0.001 0.01 O.I I 10 100 1000 20 40 60 80 , 20 40
VOLUME MIXING RATIO (pptv)
60 80 100 120
o KFA Julich 05-Feb-87 68°N
* . !4-Feb-87
o • OI-Feb-87 •
* • IO-Feb-88 •
• « l2-Jon-90 «
o • 09-Feb-90
Fobion MAPpfofile 44°N
• MPAE.26 -Mar -87, I7°N
s " , 23-Jun-87,44'N
— ATMOS/SL3 , 1 - May - 85, 30' N
o MPAE
10- Sep -83 44°N
« • OI-Ocl-84 "
l • l9-Jun'-85 «
ATMOS/SL3 01-May-85 30*N
Fobion MAP profile 44"N
Rgure 2-3. Vertical distributions of CCIaF, CHCIF2, and CF4 volume mixing ratios. Source: Adapted from
Fraserefa/., 1994a.
2.2.2 Stratospheric Observations
When investigating the concentrations of halocar-
bons in the stratosphere, the main objectives are to
determine partitioning among chlorine and bromine
"families," their total loading and their time variations. It
is therefore important to measure simultaneously and
regularly the largest possible number of halocarbons in
order to meet these objectives. For obvious technical rea-
sons, such combined stratospheric measurements have
been much sparser during the last decade than tropo-
spheric investigations. The measurements are generally
performed using in situ air sampling techniques aboard
airplane and balloons, and through infrared remote ob-
servations made from airplane, balloon, and orbiting
platforms.
A recent thorough review dealing with measure-
ments of the stratospheric abundance and distribution of
halocarbons can be found in Chapter 1 of the NASA Re-
port (Fraser et al., 1994a). The review is a compilation of
measured concentrations expressed as volume mixing
ratios versus altitude for CC13F, CC12F2, CCLt, CHC1F2,
CH3C1, CH3CC13, C2C13F3, C2Cl4F2, C2C1F5, C2F6,
CC1F3, CF4, CH3Br, CBrF3, and CBrClF2, gathered be-
tween 1984 and 1990. As an example the concentration
profiles for three halogenated methanes at northern mid-
latitudes are shown in Figure 2-3. The relative changes in
stratospheric concentrations are due to different photo-
chemical destruction rates of these compounds in the
stratosphere: CC13F > CHC1F2 » CF4.
The in situ measurements at sub-tropical, mid- and
high northern latitudes of the long-lived chlorinated ha-
locarbons indicate that (i) the concentrations observed in
the sub-tropics decline less rapidly with altitude than at
midlatitudes, because of increased upward motion at
2.8
-------
such latitudes (i.e., Kaye etal., 1991), thus
allowing for
photodissociation to occur at higher altitudes; (ii) the
concentrations of both the halocarbons amd the long-
lived "reference" gases observed in the Arctic show a
much more rapid decline with altitude than at midlati-
tudes, in particular within the winter vortex where
subsidence is often present (Schmidt et al, 1991; Toon
et al., 1992a, b, c). Thus, surfaces of constant mixing ra-
tio of long-lived chlorinated halocarbons slope poleward
and downward in the lower stratosphere. '
During recent years, a few investigations dealing
with simultaneous measurements of many chlorine- and/
or bromine-bearing gases and related inventories have
been reported. One of these concerns the budget of Cl
(sources, sinks, and reservoirs) between 12.5 and 55 km
altitude, near 30° north latitude, based on the 1985
ATMOS (Atmospheric Trace Molecule Spectroscopy
Experiment)/Spacelab 3 measurements of HC1, CH3C1
C10N02) CC14, CC12F2, CC13F, and CHCIF2, comple-
mented by results for CH3Ca3, C2Cl3F3) CIO, HOC1,
and COC1F obtained by other techniques (Zander et al,
1992 and references therein). The main conclusions of
this work indicate that (i) within the observed uncertain-
ty, partitioning among chlorinated source,! sink, and
reservoir species is consistent with the conservation of
Cl throughout the stratosphere; (ii) the mean 1985 con-
centration of stratospheric Cl was found equal to 2.55 ±
0.28 ppbv; (iii) above 50 km altitude, the inorganic chlo-
rine burden is predominantly contained in ithe form of
HC1, thus making this measurement a unique and simple
way of assessing the effective stratospheric chlorine
loading.
Based on historical emissions for the main chlori-
nated source gases, Weisenstein et al. (1992) used a
time-dependent model to calculate the atmospheric total
chlorine as a function of time, latitude, and altitude.
Their results indicate that the total Cl mixing ratio for
1985 reaches an asymptotic value of 2.35 pipbv in the
upper stratosphere. Considering that the source input
fluxes to the model are probably too low by about 15%
because they do not include emission from China, the
former Soviet Union, and Eastern Europe, it can be con-
cluded that the result found by Weisenstein et al. (1992)
for 1985 is in good agreement with the stratospheric Cl
budget derived from the 1985 ATMOS observations
(Zander ef al, 1992).
SOURCE GASES
The ATMOS instrument was flown again in 1992
(Gunson, 1992) and 1993 as part of the Atmospheric
Laboratory for Applications and Science (ATLAS) 1 and
2 Missions to Planet Earth. HC1 mixing ratios in the
range 3.4 ± 0.3 ppbv were measured above 50 km alti-
tude at different latitudes (30°N to 55°S) during
March-April 1992, as icompared to the measured value
of 2.55 ± 0.28 ppbv in April-May 1985 (Gunson et al,
1994). This corresponds to an increase of 35% over the 7
years between both measurements and is in excellent
agreement with model-predicted increases of about 0.11
to 0.13 ppbv per year (Fiather and Watson, 1990; WMO,
1992; Weisenstein etal, 1992).
During the 1991/92 Airborne Arctic Stratospheric
Expedition H (AASE H), a whole air sampler developed
by NCAR-NASA/Amea (Heidt et al., 1989) was operat-
ed on board the NASA ER-2 aircraft, which attempted to
determine the amounts of organic chlorine and bromine
entering the stratosphere:. Over 600 air samples were col-
lected during AASE II. Twelve of these that were
sampled in the latitud^altitude range of the tropical
tropopause, between 23.8°N and 25.3°N, have been ana-
lyzed by Schauffler et al (1993) for the mixing ratios of
12 chlorinated species (CC13F, CC12F2, C2C13F3,
C2C12F4, C2C1F5, CHC1F2, CH3CCIF2, CH3C1, CH2C12,
CHC13, CH3CC13, and CCU) and 5 brominated com-
pounds (CBrF3, CBrOF2, C2Br2F4, CH3Br, and
CH2Br2). From this extensive suite of measurements,
Schauffler et al. (1993) derived average total mixing ra-
tios of 3.50 ± 0.06 ppbv for Cl and 21.1 ± 0.8 pptv for
Br. The natural source of chlorine is -0.5 ppbv of the
total. Since inorganic chlorine species are negligible at
the tropopause, total chlorine at this level is dominated
by the anthropogenic release of chlorinated halocarbons
at the surface. The stratospheric Cl concentrations de-
rived from the March-April 1992 ATMOS
measurements and the January-March 1992 burdens
found by Schauffler et al. (1993) near the tropopause
provide a further confirmation of the conservation of
chlorine throughout the stratosphere. The individual
contributions to the total organic budget of bromine near
the tropical tropopause were found equal to 54% for
CH3Br, -7% for CH2Br2, and the remaining 39% nearly
evenly distributed among the halons CBrF3, CBrCIF2,
and C2Br2F4.
On the NASA DC-8 aircraft that also participated
in the AASE II campaign, Toon et al. (1993) operated a
2.9
-------
SOURCE GASES
600
500
400
300
200
100
0
ex 300
§ 200
-
CFC-12
ecu
CFC-11
methylchloroform
100
150 200 250
N2O (ppbv)
300
350
Figure 2-4. Concentrations of halocarbons in the iower stratosphere from NCAR/NASA Ames Whole Air
Sampler plotter vs. ATLAS N2O. Source: Woodbridge et a/., 1994.
2.10
-------
SOURCE GASES
high-resolution Fourier transform infrared (FTIR) spec-
trometer to determine the stratospheric colurnns above
about 11 km cruising altitude of a number of trace gases,
including CCl2F2 and CCljF. Based on these and other
long-lived gases (e.g., ^O, CH*), they found consider-
ably more uplifting (~4 km) near the equator than in the
sub-tropics.
Above the tropopause, the AASE n data set can be
used to describe the depletion of chlorinated halocar-
bons in the lower stratosphere. As residence time in the
stratosphere increases, destruction primarily by UV pho-
tolysis liberates Cl and Br from individual halocarbon
species, thereby forming the inorganic halocarbon reser-
voir species HC1 and ClONOa- Nitrous oxide can be
used as an index to examine changes in halocarbon abun-
dances (Kawa et al, 1992). N2O has a near-uniform
abundance in the troposphere of approximately 310
ppbv and is destroyed in the mid-stratosphere with a life-
time near 120 years. Figure 2-4 shows the correlation of
several chlorinated halocarbon species withi! simulta-
neous measurements of N2O within the AASE n data set
(Woodbridge et al., 1994). The seven species sltiown rep-
resent approximately 99 percent of organic halocarbon
species with lifetimes over a year. For each species a dis-
tinct correlation is found, with the halocarbon species
decreasing with decreasing N2O. In each case, the de-
crease begins at upper tropospheric altitudes as reported
by Schauffler et al. (1993). The slope of each correlation
near tropospheric values is related to the ratio of the life-
time of the halocarbons species to that of N^t) (Plumb
and Ko, 1992). The compact nature of ranges, of these
correlations demonstrates the systematic degradation of
the chlorinated halocarbons in the stratosphere. The net
loss of these organic species over a range of ^6 bounds
the available inorganic chlorine reservoir in the lower
stratosphere (see Chapter 3). Inorganic speciek partici-
pate in the principal catalytic loss cycles that destroy
stratospheric ozone.
The emission of HC1 from volcanoes c
-------
SOURCE GASES
I
1_
>*
2
700
600
500
400
300
200
100 -
•I CFG 11
D CFG 12
A CFG 1.13
O HCFC22
+ CHgCCIg
1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992
• CFG 114
D CFQ 115
A H1211
O H1301
+ HCFC142b
1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992
Year
Figure 2-5. Annual emissions of halocarbons in kt/yr. The CFC-11 ,-12 and -113 data are estimates of global
emissions, whereas the remaining estimates are based on data only from reporting companies. Source:
AFEAS, 1993; Fisher et a/., 1994; D. Fisher, Du Pont, personal communication to P.F.; P. Midgley, M&D
Consulting, personal communication to P.F.
2.12
-------
SOURCE GASES
600 C
500
300
200
ANNUAL RELEASES
1975
1980
1985
YEAR
1990
1995
Figure 2-6. Annual releases of CCI3F arid CCI2F2 estimated from 13 years of ALE/GAGE data (points are
joined by a full line), and most recent estimates of world releases of these compounds (Fisher et al 1994)
Source: Cunnold et al., 1994. '"
1990 CFC-11 and CFC-12 releases to be 249; + 28 kton
and 366 ± 30 kton, respectively. These values are compa-
rable to the global emissions assembled by Fisher et al.
(1994) (CFC-11: 255.2 ktbn and CFC-12: 385.6 kton)
(Figure 2-6). '.'.
CH2C12 and CHC1CC12 are used as Industrial
cleaning solvents. Sources of 0.9 and 0.6 tg/yr have
been recently estimated from observed atmospheric
abundances (Koppmann et al., 1993). Industry* estimates
of 1992 emissions for CC12CC12, CHC1CC12, and
CH2C12 were 0.24,0.16, and 0.39 Tg, respectively. Total
emissions for these species have declined by 40% since
1982 (P. Midgley, personal communication to:P.F.). The
aluminum refining industry produces CF4 (0.018 Tg/yr)
and C2Fg (0.001 Tg/yr), however, there are no estimates
of other potential sources (Cicerone, 1979). 80% of SFg
production (0.005 Tg in 1989) is used for insulation of
electrical equipment, 5-10% for degassing molted reac-
tive metals, and a small amount as an atmospheric tracer
(Ko et al, 1993). The rate of increase of SF6 in the atmo-
sphere (Zander et al., 1991a; Rinsland et al., 1993;
Maiss and Levin, 1994) implies that its sources are in-
creasing.
Methyl halides are produced during biomass burn-
ing. Annual emissions of 1.5-1.8 Tg/yr (Lobert et al.,
1991; Andreae, 1993) and 30 Gg/yr (Mano and Andreae,
1994) have been estimated for CH3C1 and CH3Br, re-
spectively.
2.13
-------
SOURCE GASES
A major source of methyl halides appears to be the
marine/aquatic environment, likely associated with algal
growth (Sturges et aL, 1993; Moore and Tokarczyk,
1993a). Methyl chloride, present in the troposphere at
about 600 pptv, is the most prevalent halogenated meth-
ane in the atmosphere. Maintaining this steady-state
mixing ratio with an atmospheric lifetime of the order of
two years requires a production of around 3.5 Tg/yr,
most of which comes from the ocean and biomass bum-
ing. The atmospheric budget of methyl bromide is
discussed in Chapter 10. Other halogenated methanes,
such as CHBr3, CHBr2Cl, and CH2CBr2, are produced
by macrophytic algae (seaweeds) in coastal regions
(Manley et al., 1992) and possibly by open ocean phy-
toplankton (Tokarczyk and Moore, 1994), but they do
not accumulate significantly in the atmosphere.
2.2.4 Halocarbon Sinks
Fully halogenated halocarbons are destroyed pri-
marily by photodissociation in the mid-to-upper
stratosphere. These gases have atmospheric lifetimes of
decades to centuries (Table 2-1).
Halocarbons containing at least one hydrogen
atom, such as HCFC-22, chloroform, methyl chloro-
form, the methyl halides, and other HCFCs and HFCs
are removed from the troposphere mainly by reaction
with OH. The atmospheric lifetimes of these gases range
from years to decades, except for iodinated compounds
such as methyl iodide, which have lifetimes of the order
of days to months. However, some of these gases also
react with seawater. About 5-10% of the methyl chloro-
form in the atmosphere is lost to the oceans, presumably
by hydrolysis (Butler et al., 1991). About 2% of atmo-
spheric HCFC-22 is apparently destroyed in the ocean,
mainly in tropical surface waters (Lobert et al., 1993).
Methyl bromide sinks are discussed in Chapter 10.
Recent studies show that carbon tetrachloride may
be destroyed in the ocean. Widespread, negative satura-
tion anomalies (-6 to -8%) of carbon tetrachloride,
consistent with a subsurface sink (Lobert et al., 1993),
have been reported in both the Pacific and Atlantic
oceans (Butler et al, 1993; Wallace et al, 1994). Pub-
lished hydrolysis rates for carbon tetrachloride are not
sufficient to support these observed saturation anomalies
(Jeffers et al, 1989) which, nevertheless, indicate that
about 20% of the carbon tetrachloride in the atmosphere
is lost in the oceans.
Recent investigation of the atmospheric lifetimes
of perfluorinated species CF4, CF3CF3, and SF6 indi-
cates lifetimes of >50,000, >10,000, and 3200 years
(Ravishankara et al., 1993). Loss processes considered
include photolysis, reaction with OOD), combustion, re-
action with halons, and removal by lightning, i
2.2.5 Lifetimes
Lifetimes are given in Table 2-1. An assessment
and re-evaluation of the empirical models used to derive
the atmospheric residence lifetime of two major industri-
al halocarbons, CH3CC13 and CFC-11, have been made
recently (Bloomfield, 1994). The analysis uses four
components: observed concentrations, history pf emis-
sions, a predictive atmospheric model, and an estimation
procedure for describing an optimal model. An optimal
fit to the observed concentrations at the five Atmospheric
Lifetime Experiment/Global Atmospheric Gases Experi-
ment (ALE/GAGE) surface sites over the period
1978-1990 was done with two statistical/atmospheric
models: the ALE/GAGE 12-box atmospheric model
with optimal inversion (Prinn et al, 1992) and the North
Carolina State University/University of California-Irv-
ine 3D-Goddard Institute for Space Studies (NCSU/UCI
3D-GISS) model with autoregression statistics (Bloom-
field, 1994). There are well-defined differences in these
atmospheric models, which contribute to the uncertainty
of derived lifetimes.
The lifetime deduced for CH3CC13 is 5.4 years
with an uncertainty range of ±0.4 yr (IPCC, 1994a).
From this total atmospheric lifetime, the losses to the
ocean and the stratosphere are used to derive a tropo-
spheric lifetime for reaction with OH radicals of 6.6 yr
(±25%); this value is used to scale the lifetimes of
HCFCs and HFCs (e.g., Prather and Spivakousky, 1990).
On the other hand, the semi-empirical lifetime for CFC-
11 of 50 ± 5 years (DPCC, 1994a) provides an important
transfer standard for species that are mainly removed in
the stratosphere, i.e., the relative modeled lifetimes giv-
en in Table 2-1 for CFCs, H-1301, and N2O are scaled to
a CFC-11 lifetime of 50 yr. ;
A more recent analysis of the ALE/GAGE data
(1978-1991) using the ALE/GAGE model anda revised
CFC-11 calibration scale (SIO 93) gives an equilibrium
lifetime for CFC-11 of 44 (+177-10) years (Cunnold et
al, 1994).
2.14
-------
SOURCE GASES
2.3 STRATOSPHERIC INPUTS OF CHLORINE
AND PARTICULATES FROM ROCKETS
Solid-fuel rocket motors of launch vehicles release
chemicals in the stratosphere, including chlorine (main-
ly HC1), nitrogen, and hydrogen compounds that,
directly or indirectly, can contribute to the catalytic de-
.struction of ozone. Chapter 10 of the WMO-Report No.
25 covers this subject (Harwood et al., 1992). Since that
report, which summarized model studies that evaluated
the chlorine buildup in the stratosphere and its impact on
the ozone layer, based on the projected launches of the
larger rocket types (Space Shuttle and Titan IV by Prath-
er et al., 1990, and by Karol et al., 1991; Ariane 5 by
Pyle and Jones, 1991), no additional studies have been
released. The main conclusions arrived at by Harwood et
al. (1992) were: i) within the expanding exhaust trail of a
large rocket, stratospheric ozone can be reduced sub-
stantially, up to >80% at some heights and up to 3 hours
after launch; ii) because of the slant layout of the trajec-
tory, column ozone is probably reduced by less than 10%
over an area of a few hundred square kilometers; iii) the
local plume ozone reductions decrease to neiir zero with-
in 24 hours and the regional effects are too small to be
detected by satellite observations; iv) steady-state model
calculations for realistic launch scenarios of large rock-
ets by NASA and ES A (European Space Agency) show
that for both scenarios, ozone decreases are less than
0.2% locally in the region of maximum chlorine in-
crease, with corresponding changes in ozone column of
much less than 0.1 %.
2.3.1 Stratospheric Chlorine Input
The specific chlorine (Cl) input to the stratosphere
(above 15 km altitude) from rocket exhausts can be esti-
mated if the Cl amount and its time-dependent release
along the ascent are known. Such evaluations were re-
ported by Prather et al. (1990) regarding the Space
Shuttle (68 tons Cl) and the Titan IV launcher (32 tons
Cl), and by Pyle and Jones (1991) for Arianfe 5 (57 tons
Cl). Assuming a projection of 10 launches per year for
each of these chlorine-releasing rocket types, a total of
1570 tons of Cl is then deposited in the stratosphere each
year. This corresponds to only 0.064% of the present-
day stratospheric burden of chlorine (basibd on a Cl
volume mixing ratio of 3.5 ppbv, or a total of 2.45x106
tons of Cl above 15 km altitude). However, at the rate of
increase of the stratospheric chlorine loading measured
between 1985 and 1992, Le., 0.13 ppbv per year (see
Section 2.2.2) caused by the release of 30X104 tons/yr of
Cl from the photodissociation of CFCs in the strato-
sphere (Prather et al., 1990), the scenario of large rocket
launches envisaged here: corresponds to an additional in-
jection of Cl above 15 km equal to about 0.6% per year.
This percentage will increase as CFCs are phased out.
No similar Cl input to the stratosphere can be evaluated
for a large number of smaller rockets, because their ex-
haust characteristics as well as thei.r number of launches
worldwide (maybe some 100, all types combined; Har-
wood et al, 1992) are poorly documented or
inaccessible.
2.3.2 Particulates .from Solid-Fuel Rockets
Besides gases, solid-fuel rocket motors release
particulates in the form of aluminum oxide (A12O-0,
soot, and ice. Attempts to determine the distribution of
exhausted aluminum oxide particles in the rocket ex-
hausts are limited, with only one Shuttle-related set of
measurements made some 10 years ago (Cofer et al.,
1985) indicating a distribution of particles with signifi-
cantly more particles below 1 u,m than above 1 \im in
size. The lack of satisfactory information on rocket par-
ticulate releases significantly hampers the quantification
of impacts that heterogeneous chemistry (Hofmann and
Solomon, 1989; Granier and Brasseur, 1992) may have
on ozone depletion by rockets.
The only research programs that have provided
some indication about the recent evolution of particu-
lates and aerosols in the stratosphere are by Zolensky et
al (1989) and by Hofmann (1990, 1991). From impac-
tion collections sampled in 1978 and 1984, Zolensky et
al (1989) found an order of magnitude increase in alu-
minum-rich particles of >0.5 ujn diameter at 17-19 km
altitude; they suggested that this rise is likely due to the
influx of solid rocket motor exhaust and ablating rocket
and satellite debris into ithe stratosphere in increasingly
larger amounts, with the latter predominating. Hofmann
(1990) observed an increase by about 80% of the back-
ground (non-volcanic) stratospheric sulfate burden at
northern midlatitudes between 1979 and 1990. He spec-
ulated (Hofmann, 1991) that it may be partially caused
by the increase in air traffic during that same period, bas-
ing his evaluation on a representative fleet and-engine
2.75
-------
SOURCE GASES
emission index of sulfur dioxide (SO2), as well as on a
realistic lifetime for the stratospheric aerosol. However,
Bekki and Pyle (1992) concluded that the increase in
aerosol mass between 1979 and 1990 due to the rise of
air traffic is largely insufficient to account for the ob-
served mass trend and -suggest that a rise in
submicrometer particles due to the influx of solid rocket
exhaust and ablating spacecraft material merits further
investigations. Clearly, particulates from solid-fuel rock-
ets deserve careful attention, especially as their
stratospheric abundance may increase in the near future.
2.4 METHANE (CH4)
Methane is an important greenhouse gas that is
also a reactive gas that participates hi establishing the
oxidizing capacity of the troposphere, and therefore af-
fects the lifetime of many other trace gases. In the
stratosphere it is a source of hydrogen and water vapor,
and a sink of atomic chlorine. It is mainly produced from
a wide variety of anaerobic processes and removed by
the hydroxyl radical. Its abundance in the atmosphere
has been rising since the Industrial Revolution with its
global 1992 tropospheric mixing ratio being equal to
1.714 ppmv. A large fraction of methane is released to
the atmosphere from anthropogenic sources (~2/3) and
is therefore susceptible to possible emission controls. A
reduction of about 10% of anthropogenic emissions
would stabilize the concentration at today's level (IPCC,
1994a).
2.4.1 Atmospheric Distribution and Trends
Due to the distribution of CHU sources, there is an
excess Northern Hemispheric source of about 280 Tg/yir,
and atmospheric concentrations in the Southern Hemi-
sphere are -6% lower. Recent modeling (Law and Pyle,
1993) and isotopic (Lassey etal., 1993) studies confirm
that the seasonal cycle of methane (±1.2% at midlati-
tudes) in the Southern Hemisphere is mainly controlled
by the seasonally of methane oxidation by OH radicals
in the lower troposphere and the transport of air from
tropical regions that are affected by biomass burning.
During the past decade, global methane has in-
creased on average by about 7% (Dlugokencky et al,
1994a). The declining atmospheric methane growth
identified in the previous assessment has continued.
Measurements from two global observing networks
show a steady decline in the globally averaged growth
rate since the early 1980s (Steele et al, 1992; Khalil and
Rasmussen, 1993b; Khalil et al\ 1993a; Dlugokencky et
aL, 1994c), being approximately 20 ppbv/yr hi 1979-
1980,13 ppbv/yr in 1983,10 ppbv/yr in 1990, and about
5 ppbv/yr in 1992 (Dlugokencky et al, 1994c). The de-
cline of the growth rate in the 30°-90°N semi-hemisphere
was 2-3 times more rapid than in the other semi-hemi-
sphere. The 1992 increase in the Northern Hemisphere
was only 1.8 ± 1.6 ppbv (Dlugokencky et al., 1994c).
The cause of this global decline in methane growth is not
entirely clear, but could be related to changes in emis-
sions from fossil fuel (particularly natural gas) in the
former Soviet Union (Dlugokencky et al., 1994c) and
from biomass burning in the tropics (Lowe et a/.;, 1994).
Observed methane levels hi the high Arctic (Alert, 83°N)
in 1993 were actually lower than those observed in 1992
(Worthy et al, 1994). Data reported for Antarctica (Aoki
et al, 1992) show the same trend observed by the
NOAA-CMDL station in the same region. Vertical col-
umn abundance measurements above the Jungfraujoch
station, Switzerland, between February 1985 and May
1994 indicate a rate of increase in the atmospheric bur-
den of CH4 equal to 0.73 ± 0.13 %/yr over the period
1985-1989, which slowed to 0.46 ± 0.11 %/yr between
1990 and May 1994 (Zander et al, 1994c; R.| Zander,
personal communication to E.S.).
A significant decrease hi 13CH4 has been observed
in the Southern Hemisphere since mid-1991, coincident
with significant changes in the CH4 growth rate (15 ppb/
yr in 1991; 5 ppb/yr in 1992) (Lowe et al, 1994). The
isotopic data imply that the change in Cftt growth rate is
due to: i) decreasing sources rather than increasing sinks,
and ii) a combination of decreased tropical biomass
burning and a lower release of fossil CH* in the Northern
Hemisphere.
Global measurements of Cftt between 100 and 0.1
mb pressure levels have been performed by various in-
struments aboard the Upper Atmosphere Research
Satellite (UARS). Since October 1991, the UARS Halo-
gen Occultation Experiment (HALOE) has made routine
measurements of methane concentrations at latitudes
ranging from ~80°N to ~80°S. These measurements have
been used in conjunction with other HALOE observa-
tions to evaluate vertical subsidence in the Antarctic
spring polar vortex (Russell et al, 1993); they have un-
2.16
-------
SOURCE GASES
1650
-0 1550
Q.
^ 1450
c 1350
O
•^ 1250
o
Ł 1150
5 1050
o
c 950
o
0 850
X 750
0
650
1 ' '. i,1-, , ' 1 1 1' 1 1 -
-
-
o DE08 ;
° Site J
ii
• Mizuho i;
X Summit (Berrii)
A Summit (Grenoble)
Cape Grim
t
Q-
8-
o
. 0
; §~
'<8
v —
$
ff. -
~ ^S*"1
I . *o —
i; . , P x d>?'
x x x x IAA *^ A»^ ~
1 1 1 — l_.jr* XA i | • , " , - , ,
900 1000 1100 1200 1300 1400 1500 1600 1700 1800 1900
Year AD
2000
dergone intercomparison with ER-2 airplane oltiserva-
tions (Tuck et ai, 1993), as part of validation exercises.
The Cryogenic Limb Array Etalon Spectrometer
(CLAES) UARS experiment also measured CHLj con-
centrations globally, but results have only been reported
so far as sample cases, as additional validation is re-
quired prior to releasing this data base (Kumeref al.,
1993). Although limited in time and in global coverage,
the high spectral-resolution, multiple-species (over two
dozen gases) observations made by the shuttle-bas;ed At-
mospheric Trace Molecule Spectroscopy Experiment
(ATMOS) instrument during the Spacelab 3 (April-May,
1985), ATLAS 1 (March-April, 1992) and ATLAS 2
(April, 1993) missions (Farmer, 1987; Gunson et al.,
1990; Gunson, 1992) are unique "benchmark!;" for
trends evaluations and for validation exercises.
The paleo record of atmospheric methane concen-
tration has been improved by the analysis of new ice
cores (Etheridge et al., 1992; Nakazawa et al., 1993;
Blunier et al., 1993; Jouzel et al., 1993; Chappellaz et
al., 1993). Antarctic ice core data (Law Dome), which
overlap the direct atmospheric measurements, indicate
that the growth rate was not always monotonic, with ap-
parent stabilization periods around the 1920s and again
during the 1970s (see Figure 2-7; Etheridge et al., 1992;
Dlugokencky et al, 1994a). From Greenland and Ant-
arctic ice cores, Nakazawa et al. (1993) conclude that the
pre-industrial natural sources in the Northern Hemi-
sphere were larger than those in the Southern
Hemisphere. New data from Antarctic Vostok ice core
have extended the methane rscord from 160 thousand
year BP (kaBP) through the penultimate glaciation to the
end of the previous interglacial, about 220 kaBP (Jouzel
et al., 1993). Recent analyses of Greenland ice cores
have provided additional climatic and atmospheric com-
position records (Chappellaz et al., 1993). The methane
concentration through the deglaciation is observed to be
in phase with temperature. Warm periods, each lasting
hundreds of years, are associated with methane peaks of
about 100 ppbv. These variations have not been observed
in the Antarctic ice cores, likely due to the coarse sam-
pling interval and the slower pore close-off of the
Antarctic sites.
2.77
-------
SOURCE GASES
2.4.2 Sources
A detailed discussion of the natural and anthropo-
genic sources of methane has been given in previous
assessments (WMO, 1992; IPCC, 1990, 1992) and only
an update is presented here. Methane sources are listed
in Table 2-3.
Wetlands. Natural wetlands are the major source
of methane and in recent years considerable new data on
methane flux from these ecosystems have been pub-
lished. Recent flux data from the Amazon region suggest
that a large fraction of CH4 is emitted from tropical wet-
lands (20°N-30°S), with a global estimate of -60 Tg/yr
(Bartlett et al, 1990; Bartlett and Harris, 1993). High
northern latitude studies indicate emissions ranging
from 20 to 60 Tg/yr (Whalen and Reeburgh, 1992; Ree-
burgh et aL, 1994). Information from large areas of the
world is lacking, particularly in the tropics and the Sibe-
rian Lowland (Bartlett and Harris, 1993). Recently,
atmospheric data have been used to constrain emission
estimates from wetlands in the former Soviet Union
(Dlugokencky et al, 1994b).
Ocean and Freshwater Ecosystems. A re-evalu-
ation of the ocean source was performed by Lambert and
Schmidt (1993). According to these authors only -3.5
Tg/yr are emitted by the open oceans, but emissions
from methane-rich areas could be considerably more
important, producing a total oceanic source of the order
of 50 Tg/yr. There is no new information about the con-
tribution of freshwater ecosystems.
Termites. A recent estimate made by Martius et
al. (1993) for the contribution of termites to the global
CH4 budget agrees well with the central value of 20 Tg/y
given in the 1992 IPCC Supplement.
Other Natural Sources. New estimates have been
made for volcanoes (3.5 ± 2.7 Tg/yr), hydrothermal
emissions (2.3 ± 2.7 Tg/yr), and hydrates (-3 Tg/yr)
(Judd et al, 1993; Lacroix, 1993). .
Fossil Carbon Related Sources. From studies of
the carbon-14 content of atmospheric CRt it was estab-
lished that about 20% (-100 Tg) of total annual CR,
emission is from fossil carbon sources (IPCC, 1992).
However, there are large uncertainties in the contribution
of the various related sources: coal mines, natural gas
and petroleum industry. New global estimates from coal
mines are: 25 Tg/yr (CIAB, 1992), 17 Tg/yr (Miiller,
1992), 43 Tg/yr (Beck, 1993), 49 Tg/yr (Subak et al,
1993), and 45.6 Tg/yr (Kirchgessner et dl, 1993).
Muller (1992) gives an emission from natural |gas activi-
ties of 65 Tg/yr, which is much higher than the values
given in the DPCC (1992) (25^2 Tg/yr). Khalil et al.
(1993b) proposed mat low-temperature combustion of
coal (not included previously) could be a significant
source of methane, with a global emission of -16 Tg/yr.
However, the emission factor derived by Khalil et al is
higher than the values obtained by Fynes et al (1993)
from coal-fired plants and the one quoted for handfired
coal units by the Air Pollution Engineering Manual (Air
and Waste Management Assn., USA, 1992);; further re-
search is clearly required to refine this estimate.
Waste Management Systems. Landfills, animal
waste, and domestic sewage are significant global
sources of methane, with a total emission estimate of
-80 Tg/yr (IPCC, 1992). New global estimates from
landfills are 40 Tg/yr (Muller, 1992), 36 Tg/yr (Subak et
al, 1993), and 22 Tg/yr (Thomeloe et al, 1993), in good
agreement with the mean value (30Tg/yr) given previ-
ously (IPPC, 1992). No additional information has been
published for animal waste and domestic sewage.
Enteric Fermentation. Anastasi and Simpson
(1993) estimated for 1990 an emission of 84 Tg/yr from
enteric fermentation in cattle, sheep, and buffalo. This
result suggests that the strength of enteric fermentation
be in the upper part of the range given in 1992 (65-100
Tg/yr). Furthermore, Minson (1993) in a re-evaluation
of this source in Australia found values 43% higher than
previous estimates for this country. Johnson et al (1993)
estimated a global emission of 79 Tg/yr.
Biomass Burning. New global estimates of this
source are: 30.5 Tg/yr (Hao and Ward, 1993), 36 Tg/yr
(Subak et al, 1993), and 43 Tg/yr (Andreae and Warnek,
1994). These values are within the range of data reported
previously (IPCC, 1992).
Rice Paddies. There is a very large uncertainty
associated with the emissions estimate from rice paddies
(IPCC, 1992). Three-dimensional (3-D) model calcula-
tions constrain estimates of methane emission from rice
cultivation to -100 Tg/yr (Fung et al, 1991; Dlugo-
kencky etal, 1994b). The results reported earlier (IPCC,
1990, 1992) and recent estimates (i.e., Wassman et al,
1993; Delwiche and Cicerone, 1993; Bachelet and Neue,
1993; Subak et al, 1993; Lai et al, 1993; Shearer and
Khalil, 1993; Neue and Roger, 1993) suggest an emis-
2.18
-------
SOURCE GASES
TABLE 2-3. Estimated sources and sinks of methane (Tg CH4 per year).
Range
Likely
Totals
Sources
Natural
Wetlandsa !
Tropics -- -
Northern Latitudes
Others
Termites
Ocean j
Freshwater
Others3 j
Total Natural
Anthropogenic ;
Fossil Fuel Related
Coal Mines
Natural Gas ;
Petroleum Industry i
Coal Combustion t
Waste Management System j
Landfills
Animal Waste |
Domestic Sewage Treatment
Enteric Fermentation
Biomass Burning
Rice Paddies* [
Total Anthropogenic
Total Source j
Sinks
Reaction with OHa j
Removal in Stratosphere3
Removal by Soils
Atmospheric Increase
Total Sink |
20-60
5-10
10-50
5-50*
1-25
3.13
15-453
25-503
5-30
7-303
20-70
20-30
?
65-100
20-80
20-100
330-560
25-55
15-45
30-40
3 indicates revised estimates since previous assessments
b from carbon-14 studies (IPCC, 1992) '
-60
40
10
20
10
5
10
100b
30
25
25
80
40
60
445
40
30
37
155
360
515
552
2.19
-------
SOURCE GASES
sion range of 20-100 Tg/yr with a most likely value of 60
Tg/yr.
2.4.3 Sinks
CH4 is mainly removed through chemical reac-
tions in the troposphere and stratosphere (485 Tg/yr). A
growing number of studies (reviewed by Reeburgh et al.f
1994) show that methane is consumed by soil rnicrobial
communities in the range between 20 and 60 Tg/yr.
Methane oxidation is expected to be particularly impor-
tant in modulating methane emissions from rice paddies,
wetlands, and landfills. Ojima etal. (1993) estimate that
-20 Tg of methane is consumed annually by temperate
soil, and that this sink has decreased by -30% due to soil
disturbance.
2.4.4 Potential Feedbacks from a Changed
Climate
There are several potential climate feedbacks that
could affect the atmospheric methane budget (IPCC,
1990). At present, however, the attention has focused on
northern wetlands and on permafrost.
High-Latitude Wetlands. Changes in surface
temperature and rainfall are predicted by general circula-
tion models (GCMs) to occur in high-latitude regions.
When changes hi temperature are considered alone, an
increase in the emission of CHLt is predicted (Hameed
and Cess, 1983; Lashoff, 1989). Recent calculations
suggest only a moderate increase in CH^ emissions in a
2xCO2 scenario (Harris and Frolking, 1992). On the oth-
er hand, using a hydro-thermal model, Roulet et.al.
(1992) estimated a significant decrease in moisture stor-
age that resulted in an 80% decrease in CH4 fluxes
(negative feedback); the corresponding increase due to
temperature changes is only 15%. These estimates have
been confirmed by measurements indicating a reduced
CH4 flux from drained northern peatlands (Roulet et ai,
1993). It seems that northern wetlands are more sensitive
to changes in moisture than temperature; however, the
biospheric feedback mechanisms are poorly understood
(Reeburgh etal., 1994).
Permafrost The methane content of permafrost
in Fairbanks (Kvenvolden and Lorenson, 1993) and
northern Alaska (Rasmussen et al., 1993) was recently
evaluated at 2-3 mg/kg. Using these concentrations and
the changes in temperature predicted by various
scenarios, Kvenvolden and Lorensen (1993) predict that
in 100 years the maximum methane release rate will be
-27 Tg/yr and that during the first 30 years no significant .
release will occur. The heat transfer and gas diffusion
model of Moraes and Khalil (1993) indicates that in the
future, permafrost is likely to contribute less than 10 Tg
of methane per year.
2.5 NITROUS OXIDE (N2O)
Nitrous oxide has a long atmospheric lifetime. It is
the major source of stratospheric nitrogen oxides, which
are important in regulating stratospheric ozone. N2O is
also a greenhouse gas. It is emitted by several small
sources, which have large uncertainties, and its atmo-
spheric budget is difficult to reconcile. It is removed by
photolysis and oxidation in the stratosphere and rnicro-
bial oxidation in soils. A reduction of more than 50% of
anthropogenic sources would stabilize its concentration
at today's level of about 310 ppbv (IPCC, 19?4a).
2.5.1 Atmospheric Distribution and Trends
The available global nitrous oxide data indicate
that the trend over the past decade is very variable, rang-
ing from 0.5 to 1.2 ppbv/yr (WMO, 1992; Khalil and
Rasmussen, 1992). A recent analysis of seventeen years
of data collected on oceanic expeditions as well as in
Alaska, Hawaii, and Antarctica (Weiss, 1994) and six-
teen years of data from the NOAA global network
(Swanson et al, 1993) shows that the global average
abundance at the beginning of 1976 was 299 ppbv,
which has risen to 310 ppbv at the beginning of 1993.
During 1976-82 the growth rate was about 0.5-0.6 ppbv/
yr, which increased to a maximum of 0.8-1; ppbv/yr in
1988-89, declining to the-current rate of 0.5-0.6 ppbv/yr.
An analysis of IR solar absorption spectra record-
ed at the Jungfraujoch Station (46.6°N) in 1950-51 and
from 1984 to 1992 has been recently performed by
Zander et al. (1994b). The results indicate that the rate of
increase of the column of N2O was 0.23 ± 0.04 for the
period 1951 to 1984, and 0.36 ± 0.06 %/yr from 1984 to
1992. The corresponding volume mixing ratios at the
levels of the site (3.58 km altitude) increased from 275
ppbv to 305 ppbv between 1951 and 1992. The 1951
concentration is quite similar to the pre-industrial values
obtained from ice cores (285 ppbv; IPCC, 1990), sug-
2.20
-------
SOURCE GASES
TABLE 2-4. Estimated sources and slinks of N2O (Tg N per year).
Sources
A. Natural
Oceans
Tropical Soils
Wet forests
Dry savannas
Temperate Soils
Forests
Grasslands
B. Anthropogenic
Cultivated Soils
Animal Waste*
Biomass Burning
Stationary Combustion
Mobile Sources*
Adipic Acid Production
Nitric Acid Production
Sinks
Removal by Soils
Photolysis in the Stratosphere*
Atmospheric increase*
1.4-5,2*
2.2-3.7
0.5-2.0
0.05-2.0
9
1-3
0.2-0.5
0.2-1.0
0.1-0.3
0.1-0.6
0.4-0.6
0.1-0.3
12.3 (9-17)
3.1-4.7
* indicates revised estimates since previous assessment
i
gesting that the pre-industrial level was lower (see be-
low), or that it persisted until the middle of this century
and that the increase occurred thereafter.
Satellite global measurements of N2O have been
made by CLAES and ISAMS (Improved Stratospheric
and Mesospheric Sounder) aboard UARS (Kumer etal.,
1993; Taylor et al., 1993), but no validated results have
been released so far.
Ice core records of N2O show an increase of about
8% over the industrial period (IPCC, 1990). New records
covering the last 45 ka were obtained from Antarctica
and Greenland (Leuenberger and Siegenthaler, 1992).
The Greenland record suggests a pre-industrial level of
about 260 ppbv, 10 to 25 ppbv lower than previous
records (IPCC, 1990). The Antarctic core shjaws that
N2O was lower during glacial periods, consistent with
the hypothesis that soils are a major natural source of
nitrous oxide.
2.5.2 Sources
A detailed presentation of N2O sources was made
in IPCC (1990) and a revised budget was given in the
1991 ozone assessment CVVMO, 1992). N2O is emitted
by a large number of small sources, most of them diffi-
cult to evaluate and the estimates are very uncertain.
Here we will only present new information not included
in previous assessments. The updated budget is present-
ed in Table 2-4. The overall uncertainty in the N2O
budget suggests that it could be balanced with the cur-
rently identified sources.
2.27
-------
SOURCE GASES
N2O fluxes from an upwelling area of the Indian
Ocean (Law and Owen, 1990) and the Peruvian up-
welling region (Codispoti et al., 1992) indicate that the
oceans may be a larger source of this gas. Weiss (1994)
calculated that the total pre-industrial source of N2O was
~9 TgN/yr, of which -3 TgN/yr was oceanic. An isotopic
study (nitrogen-15 and oxygen-18) of atmospheric N2O
suggests a large gross ocean-atmosphere flux (Kim and
Craig, 1993). Therefore, the upper range for that source
has been extended to 5.2 TgN/yr in this assessment
Recent emission estimates from some anthropo-
genic sources made by Subak et al. (1993) agree well
with previous values. The increasing use of catalytic
converters in cars stimulated the evaluation of the global
contribution of this source: from tailpipe emission mea-
surements, Dasch (1992) derived a global emission of
0.13 Tg N/yr, Khalil and Rasmussen (1992) from mea-
surements in crowded highways in California estimate a
global emission of 0.06-0.6 TgN/yr, Berges et al. (1993)
from measurements in two tunnels (Stockholm and
Hamburg) estimate a global emission of 0.24 ± 0.14
TgN/yr. This new information on N2O emissions from
catalytic converters, together with previous estimates
(WMO, 1992) results in a revised emission range of 0.1-
0.6 TgN/yr.
Important emissions are produced by agricultural
activities. Recent global estimates from fertilized soils
are 0.9 TgN/yr (Kreileman and Bouwman, 1994) and 2
TgN/yr (Pepper et al., 1992). A source that was not in-
cluded in the 1992 Report is cattle and feed lots. Based
on the ratios of excess N2O to excess CH4 in barn stud-
ies, Khalil and Rasmussen (1992) estimate a source of
0.2-0.5 TgN/yr from cattle. Kreileman and Bouwman
(1994) estimate for 1990 a global emission of 0.6 TgN/
yr for the animal waste source. New information in trop-
ical land use change indicates that the flux of N2O
depends on the age of the pasture, with young pastures
(<10 years) emitting 3-10 times more N2O than tropical
forests, whereas older pastures emit less (Keller et al.,
1993). A rather low source of 0.2 TgN/yr was estimated
by Kreileman and Bouwman (1994) due to enhanced
soil N2O emission following deforestation. More re-
search on tropical agricultural systems is required before
conclusions can be reached concerning the relative im-
portance of tropical agricultural systems as a growing
N2O source (Keller and Matson, 1994).
2.5.3 Sinks
The major sink bf N2O is photodissociation in the
stratosphere; a secondary loss of about 10% occurs
through reaction with O('D). The lifetime is 120 ± 30 yr
(Prather and Remsberg,' 1992). Important evidence of
N2O consumption by soils was reported by Donoso et al.
(1993), but there are insufficient data to determine
whether soil provides a significant global N2O sink.
Based on recent data (Swanson et al., 1993; Weiss,
1994) the atmospheric increase is estimated to be 3.1-4.7
TgN/yr. The estimated sinks (including the atmospheric
increase) range from -12 to -21 TgN/yr, therefore, to
balance the N2O atmospheric budget, all sources should
be near their upper limits. This is in agreement with cal-
culations based on ice core records and atmospheric
concentrations that suggest a total anthropogenic emis-
sion of -4.5 TgN/yr and -9.5 TgN/yr for the natural
sources (Khalil and Rasmussen, 1992).
2.6 SHORT-LIVED OZONE PRECURSOR
GASES
Tropospheric ozone is a greenhouse gas, of partic-
ular importance in the upper troposphere. It also plays a
significant role in the oxidizing capacity of the atmo-
sphere. A detailed evaluation of tropospheric ozone is
made in Chapter 5. Since the concentration of O3 de-
pends on the levels of its precursors (i.e., NOX, CO, CH4,
NMHC), we assess their sources, sinks and atmospheric
distributions in the following sub-sections.
2.6.1 Nitrogen Oxides (NOX = NO + NOa)
2.6.1.1 TROPOSPHERIC DISTRIBUTION
Because of its complex geographical source pat-
tern and its short lifetime, the spatial and;temporal
distribution of tropospheric NOX is complex and highly
variable, over 3 orders of magnitude in non urban areas
(Carroll and Thompson, 1994). A detailed discussion of
the tropospheric distribution of NOX is presented in
Chapter 5. :
2.6.1.2 SOURCES ,
Estimated NOX source strengths are summarized
in Table 2-5. '
2.22
-------
SOURCE GASES
TABLE 2-5. Estimated sources of NOX (TgN/yr).
Range
Likely
Natural Soils
Lightning
Biomass Burning
Subsonic Aircraft
Fossil Fuel
Agricultural Soils
5-12
3-20
3-13
0.2-1
21-25
?
I -
: 7
. 7
8
; 0.4
: 24
?
!
Soils. Soil microbial activity is an important natu-
ral source of NOX, but a very large uncertainty aiffects its
estimate (IPCC, 1992). Recently, Williams etal. (1992)
derived an emission of only ~0.1 TgN/yr from natural
soils (grassland, forest, and wetlands) within ;the U.S.
From studies in Venezuela, Sanhueza (1992) estimated
an emission of ~4 TgN/yr for the global savannali region.
Tropical forest soils produce large amounts of NO; how-
ever, due to removal processes inside the forest itself,
most of the NO never reaches the "open" atmosphere
(Bakwin et al., 1990). Recent global estimates of this
source include: Davidson (1991), 13 TgN/yr,: Miiller
(1992), 4.7 TgN/yr; and Dignon et al. (1992), 5 TgN/yr.
Agricultural soils could be an important siource of
NOX, but no reliable global budgets exist. Cultivated
soils from the U.S. emit 0.2 TgN/yr (Willianis et al.,
1992); plowing of tropical savannah soil produces a
large increase of NO emissions (Sanhueza et di:, 1994),
but the impact to the global budget has not been estimat-
ed. ;;
Lightning. The global estimates of NOX 'produc-
tion by lightning show a very large uncertainty (Liaw et
al., 1990). Using a global chemistry, transport, arid depo-
sition model, Atherton et al. (1993) found that a
lightning source of 5-10 TgN/yr (with an upper limit of
20 TgN/yr) is compatible with the NOy levels in remote
locations. This is in agreement with Logan (19:83), who
indicates that the distribution of nitric acid (HINOs) in
the remote troposphere is consistent with a lightning
NOX source of <10 TgN/yr. ;
Biomass burning. Tropical biomass burn ing is an
important source of NOX (Crutzen and Andreae, 1990;
Lobertefa/., 1991; Andreae, 1991; Penner era/., 1991;
Miiller, 1992), ranging from 2 to 8 TgN/yr. Estimates in-
cluding extratropical burning indicate global production
of 9.6 TgN/yr (Andreae, 1993) and 13 TgN/yr (Dignon
and Penner, 1991).
Aircraft. Emissions fErom aircraft are a relatively
small source of NO. However, since a large fraction of
the NOX is released at altitudes between 9-13 km, this
has a large impact on the photochemistry of the free tro-
posphere (Johnsonetal., 1992; Eecketal., 1992), and is
likely responsible for a larges fraction of the NOX found at
those altitudes at northern midlatitudes (Ehhalt et al.,
1992). Estimates of the global source from aircraft range
from 0.23 to 1.0 TgN/yr (Egli, 1990; Johnson et al.,
1992; Beck et al., 1992; Penner et al, 1994). A very de-
tailed evaluation of this source has been recently
completed (Baughcum et al., 1993). This study includes
emissions from scheduled airliner and cargo, scheduled
turboprop, charter, military, and former Soviet Union
aircraft. The results indicate a global emission of 0.44
TgN/yr, with 7% of the emission occurring in the South-
ern Hemisphere. A detailed geographical distribution of
this source is given in Chapter 11.
Fossil Fuel Combustion. This is the largest
source of NOX (24 TgN/yr) and its global distribution is
relatively well known (Dignon and Hameed, 1989;
Hameed and Dignon, 1991; Miiller, 1992; Dignon,
1992). According to Hameed and Dignon (1991), the
emission of NOX increased from 18.1 TgN/yr in 1970 to
24.3 TgN/yr in 1986 (25% increase).
2.6.13 SINKS
The removal processes of NOX (atmospheric oxi-
dation of NOX and dry deposition of NO2) are
reasonably well known. However, it is not possible to
make a direct estimate of the global NOX sink since the
global distribution of NOX is too poorly known.
2.23
-------
SOURCE GASES
TABLE 2-6. Estimated sources of NMHC (TgC/yr).
Vegetation*
Oceans
Biomass Burning
Technological
Range
230-800
20-150
30-90
60-100
Likely
500 i
?
40
70 ;
* mainly isoprene and terpenes
2.6.2 Non-Methane Hydrocarbons (NMHCs)
2.6.2.1 ATMOSPHERIC DISTRIBUTION
Most NMHCs (heavier alkanes, alkenes, alkyl
benzenes, isoprene, terpenes) have atmospheric life-
times of less than a week (sometimes less than a day). In
this case the atmospheric distributions reflect the source
pattern and the regional transport situation, and the mix-
ing ratios generally range from several ppbv in the
boundary layer near the sources to a few pptv or less in
the background atmosphere. NMHCs with predominant-
ly anthropogenic sources exhibit a maximum in winter,
reflecting the seasonality of the removal by OH radicals.
Biogenic NMHCs (i.e., isoprene, terpenes) present very
low mixing ratios in winter and highest abundance in
summer, a consequence of the seasonality of the emis-
sion rate (Fehsenfeld et al., 1992).
For NMHCs with lifetimes of few weeks or more
(Le., ethane, acetylene, propane) there is a better under-
standing of their atmospheric distributions (Ehhalt,
1992; Rudolph et ai, 1992). Seasonal cycles and long-
term trends in the vertical column abundances of ethane
and acetylene above the Jungfraujoch station, Switzer-
land, have been investigated by Ehhalt et al. (1991) and
Zander et al. (1991b).
2.6.2.2 SOURCES
Estimated source strengths of NMHCs are report-
ed in Table 2-6.
Vegetation. Foliar emissions are, by far, the most
important sources of NMHC. Rasmussen (1972) esti-
mated a global emission ranging from 230 to 440 TgC/
yr. Zimmerman et al. (1978) found that vegetation emits
350 TgC/yr of isoprene and 480 TgC/yr of terpenes. Re-
cently, Miiller (1992) reported the following values (in
TgC/yr): isoprene 250, terpenes 147, aromatics 42, and
paraffins 52 (total 491 TgC/yr); Allwine et at. (1994) es-
timate a total NMHC emission of 827 Tg/yr. ,
Oceans. Ehhalt and Rudolph (1984X estimate a
global rate from the ocean of 21 TgC/yr (C^-C^ hydro-
carbons), whereas Bonsang et al. (1988) report a much
larger rate of 52 TgC/yr. Based on the results of Donahue
and Prinn (1990), Muller (1992) indicates that there is a
large uncertainty in marine emissions and gives a range
of 30-300 TgC/yr.
Biomass burning. Emissions of NMHC from bio- •
mass burning range from 36 to 90 TgC/yr (Lobert et al.
1991; Muller, 1992; Andreae, 1993). Ethane, ethene,
propene, acetylene, and benzene are emitted with a rate
>2TgC/yr (Lobert etal., 1991; Bonsang etai, 1991).
Technological sources. These include gasoline
handling, natural gas, refuse disposal, and chemical
manufacturing, and produce a global emission ranging
between 60 and 140 TgC/yr (Wameck, 1988; Muller,
1992; PiccotetaL, 1992; Bouwman, 1993).
2.6.2.3 SINKS
NMHCs react rapidly with the OH radical (unsatu-
rated compounds also react with ©3) and with the
exception of ethane (lifetime 2-3 months), their atmo-
spheric lifetimes are shorter than one month; isoprene
and terpenes have lifetimes of only a few hours.
2.6.3 Carbon Monoxide (CO)
2.63.1 ATMOSPHERIC DISTRIBUTION AND TRENDS
The atmospheric distribution and trends of CO
were reviewed previously (WMO, 1992; IPCC, 1992).
CO mixing ratios in the troposphere present systematic
latitudinal and seasonal variations, ranging from around
2.24
-------
SOURCE GASES
TABLE 2-7. Estimated sources and sinks of CO (Tg/yr)
Range
Sources
Technological
Biomass Burning
Biogenics
Oceans
Methane Oxidation
NMHC Oxidation
Sinks
OH Reaction
Soil Uptake
Stratospheric Remotion
300-900
400-700
60-160
20-190
400-1000
300-1300
1400-26JOO
250-640
~100:
40 to 200 ppbv. Annual mean CO levels in the high lati-
tudes of the Northern Hemisphere are about aifactor of 3
greater than those at similar latitudes in the Southern
Hemisphere.
During the 1980s there was evidence that atmo-
spheric CO was increasing at ~l%/year in the Northern
Hemisphere, whereas no significant trend wsis observed
in the Southern Hemisphere (WMO, 1992). Recent mea-
surements indicate that global CO levels have fallen
sharply from the late 1980s. Novelli et a/. (1994) found
that in the Northern Hemisphere, CO decreasipd at a spa-
tially and temporally average rate of 7.3 ± 0.9 ppbv/yr
(6.1 %/yr) (June 1990 to June 1993), whereas in the
Southern Hemisphere it decreased at 4.2 ± 0.5 ppbv/yr
(7.0 %/yr). Khalil and Rasmussen (1994) for the period
1987 to 1992 reported a decrease of 1.4 ± 0.9 %/yr in the
Northern Hemisphere and 5.2 ± 0.7 %/yr in the Southern
Hemisphere. While the above results concern surface "
levels of CO, total vertical column abundances of CO
above the Jungfraujoch station, Switzerland, also show a
mean rate of decrease equal to 1.15 ± 0.32 %/yr between
1985 and 1993 (Zander et al., 1994c). The causes of this
behavior have not been identified, but decreases in trop-
ical biomass burning and Northern Hemisphere urban
emissions have been suggested: The total amount of CO
in today's atmosphere is less than it was a deca'de ago.
Preliminary global and seasonal variations of CO
between 30 and 90 km altitude have been reported by
Likely
500
600
100
?
600
600
2100
250
100
Lopez-Valverde et al. (1993), based on ISAMS/UARS
infrared limb emission measurements. These are the first
global measurements of CO in the middle atmosphere,
with data validation being still in progress.
2.63.2 SOURCES
Estimated strengths of sources and sinks of CO are
summarized in Table 2-7.
Technological sources. Technological sources in-
clude transportation, combustion, industrial processes,
and refuse incineration. There are several evaluations
(Jaffe, 1973; Logan, 1980; Seiler and Conrad, 1987;
Cullis and Hirshler, 1989; Khalil and Rasmussen, 1990;
Miiller, 1992; Subak, et al., 1993) ranging from 300 to
'900 TgCO/yr.
Biomass burning. Recent estimates for the tropics
range from 400-700 TgCO/yr (Lobert et al., 1991;
Miiller, 1992; Andreae, 1993; Subak et al., 1993). In-
cluding extratropical burning, Andreae (1993) derives a
global source equal to 621 TgCO/yr.
Terrestrial biogenic sources. These include vege-
tation, soils, and animals (i.e., termites). Based on the
emission rates found on higher plants of the temperate
region, Seiler and Conrad (1987) estimated a global
source of 75 ± 15 TgCO/yr. Assuming CO emissions
proportional to net primary productivity (NPP) and us-
ing the flux reported by Kirchhoff and Marinho (1990)
for tropical forests, Miiller (1992) evaluated a global
2.25
-------
SOURCE GASES
biogenic source at 165 TgCO/yr. Photoproduction of CO
from dead plant matter has been reported (Valentine and
Zepp, 1993; Tarr et aL, 1994), however, no global evalu-
ation of this source has been made.
Oceans. Early estimates of CO emissions from the
oceans (IPCC, 1992) range from 20 to 190 TgCO/yr.
Using an atmospheric general circulation model, Erick-
son (1989) calculated a global ocean source equal to 165
±80 TgCO/yr.
Hydrocarbon oxidation. This is the most impor-
tant source of atmospheric CO. The production of CO
from methane oxidation ranges from 400 to 1000 TgCO/
yr, and 300 to 1300 TgCO/yr from NMHC (Zimmerman
et al., 1978; Logan, 1980; Khalil and Rasmussen, 1990;
Crutzen and Zimmerman, 1991).
2.633 SINKS
Reaction with the OH radical is the major sink for
CO. Soil uptake and removal in the stratosphere are mi-
nor sinks. In principle, the atmospheric removal rates for
CO can be calculated from the atmospheric CO distribu-
tion, the distribution of the OH radical concentration,
and the related reaction rate. Model calculations predict
a removal rate of about 2000 Tg(CO)/yr (WMO, 1986;
Seller and Conrad, 1987; Khalil and Rasmussen 1990;
Crutzen and Zimmerman, 1991).
2.7 CARBON DIOXIDE (CO2)
The change in atmospheric concentration of CO2,
from 280 ppmv pre-industrial to -360 ppmv in 1993, is
the major contributor to the calculated increase in radia-
tive forcing since the pre-industrial period (i.e., 1.5 W
m-2). An updated review of the CO2 budget has been
made in the 1994 IPCC report (IPCC, I994b).
Observations of CO2 since the 1950s show sys-
tematic upward trends, in both concentration and rate of
concentration increase, albeit with substantial variation
in the rate of increase from year to year. During the peri-
od 1991 to 1993, the rate of increase of CO2 per year
slowed substantially (to as low as 0.5 ppmv/yr from the
long-term average of 1.5 ppmv/yr). There are numerous
examples in the record of short periods where growth
rates are higher or lower than the long-term mean. The
most recent observations indicate that growth rates are
now increasing again (IPCC, 1994b).
CO2 emission from industrial processes (mainly
fossil fuel combustion and cement production) in 1991 is
estimated at 6.2 GtC/yr (Andres et al., 1994), compared
with 6.0 ± 0.5 GtC in 1990 (IPCC, 1992). The^ cumula-
tive input since the pre-industrial period is estimated at
230 GtC (Andres et aL, 1994). Recent satellite remote
sensing measurements of land clearing rates in the Bra-
zilian Amazon have resulted in substantially reduced
estimates (by -50%) for this area (INPE, 1992; Skole
and Tucker, 1993). However, deforestation rates for the
rest of the tropics remain poorly quantified. Current net
flux estimates (in GtC/yr) that include regrowtri after de-
forestation are: 0.6 for Latin America, 0.7 for South and
Southeast Asia, 0.3 for Africa, and -0.3 to -1.1 for mid/
high latitudes, producing a global mean for the 1980s of
1.1 ± 1.2 GtC/yr (IPCC, 1994b). The oceans represent a
significant sink of atmospheric COa, averaging 2.0 ± 0.8
GtC/yr over the decade 1980-89.
The imbalance between atmospheric concentra-
tion changes, estimated emissions, and estimated ocean
uptake, as well as the discrepancies between the ob-
served and calculated inter-hemispheric gradients of
CO2, indicate the existence of an unaccounted-for terres-
trial sink of 1.2 ± 1.6 GtC/yr, probably attributable to a
' combination of COi-induced plant growth (0.5-2.0 Gt/
yr), nitrogen fertilization (0.2-1.0 Gt/yr), and possible
climatic effects (0-1.0 Gt/yr) (IPCC, 1994b).
Climatic feedback appears to be positive, amplify-
ing the effect of anthropogenic emissions, although this
amplification may be reduced due to feedbacks and
compensating processes within the marine and terrestrial
systems. It is likely that the effect of CC^ fertilization on
plant production will be substantially smaller than the
20-40% observed in most agricultural plants. An impor-
tant body of data supports the view that responses of
plant production to elevated CO2 are restricted in nutri-
ent-limited ecosystems (e.g., Diazetal., 1993); however,
it is possible that N deposition arising from the use of
fertilizers and fossil fuel combustion will reduce the in-
tensity or spatial distribution of nitrogen limitation.
Carbon cycle modeling studies of CC^'concentra-
tions, under a range of emissions scenarios and for a
range of stabilized CO2 concentrations up toi 750 ppmv,
yield the following results (IPCC, 1994b): i) because of
the long residence time for carbon dioxide, stabilization
of anthropogenic emissions at projected 2000 levels
(from IS92a scenario) leads to a nearly constant rate of
2.26
-------
SOURCE GASES
increase in atmospheric concentrations for atj least two
centuries; modeled concentrations reach 4804540 ppbv
by 2100; ii) stable CO2 concentration at values up to 750
ppmv can be maintained only with anthropogenic emis-
sions that eventually drop below 1990 levels; iiii) there is
a close relationship between the eventual stabilized con-
centration and the integrated CO2 emission from now
until the time of stabilization. Integrated emissions for
stabilization at levels lower than 750 ppmv are less than
those calculated for the IS92 a, b, e, and f scenarios.
REFERENCES ;|
i
APEAS, Production, Sales and Atmospheric Release of
Fluorocarbons through 1992, Alternative Fluoro-
carbons Environmental Acceptability Study,
SPA-AFEAS, Inc., Washington, D.C., USA, 1993.
AFEAS, Historic Production, Sales and Atinospheric
Release of HCFC-142b, Alternative Huorocar-
bons Environmental Acceptability! Study,
SPA-AFEAS, Inc., Washington, D.C., USA, 1994.
Albritton, D., and R. Watson, Methyl bromide and the
ozone layer: A summary of current understanding,
in Methyl Bromide: Its Atmospheric Science, Tech-
nology, and Economics, Montreal 'Protocol
Assessment Supplement, edited by R. Watson, D.
Albritton, S. Anderson and S. Lee-Bapty, United
Nations Environment Programme, Nairobi, Ken-
ya, 1992. i!
Allwine, et al.r An initial global inventory of biogenic
VOC emitted from terrestrial sources, submitted to
/. Geophys. Res., 1994. !
Anastasi, C, and VJ. Simpson, Future methaine emis-
sions from animals, / Geophys. Res, 98,
7181-7186, 1993. .i.
Andreae, M.O., Biomass burning: Its history, use, and
distribution and its impact on environmental qual-
ity and global climate, in Global Biomass
Burning: Atmospheric, Climatic and Biospheric
Implications, edited by J.S. Levine, MIT Press, 3-
21, 1991. i
Andreae, M.O., The influence of tropical biomass burn-
ing on climate and the atmospheric chemistry, in
Biogeochemistry of Global Change: Radiatively
Active Trace Gases, edited by R.S. Orelard, Chap-
man and Hall, New York, 113-150, 1993.
Andreae, M.O., and P. Warneck, Global methane emis-
sions from biomass burning and comparison with
other sources, Pure andAppl. Chem.,66, 162-169,
1994.
Andres, T.J., G. Marland, T. Boden, and S. Bischoff,
Carbon dioxide emissions from fossil fuel com-
bustion and cement manufacture 1751-1991 and
an estimate for their isotopic composition and lati-
tudinal distribution, Global Change Institute,
Snowmass, Colorado, USA, 1994.
Aoki, S., T. Nakazawa, S. Murayzuna, and S. Kawagu-
chi, Measurements of atmospheric methane at the
Japanese Antarctic Station, Syowa, Tellus, 44B,
273-281, 1992.
Atherton, C.S., I.E. Penner, and J.J. Walton, The role of
lightning in the tropospheric nitrogen budget:
model investigations, in preparation, 1993.
Atlas, E., W. Pollock, J. Greenberg, L. Heidt, and A.M.
Thompson, Alkyl nitrates, nonmethane hydrocar-
bons, and halocarbon gases over the equatorial
Pacific Ocean during SAGA 3, /. Geophys. Res.,
98D, 16933-16947, 1.993.
Bachelet, D., and H.U. Neue, Methane emissions from
wetland rice areas of Asia, Chemosphere, 26, 219-
237, 1993.
Bakwin, P.S., S.C. Wofsy, S.M. Fan, M. Keller, S.E.
Trumbore, and J.M. da Costa, Emissions of nitric
oxide (NO) from tropical forest soils and ex-
change of NO between the forest canopy and
atmospheric boundary layers, J. Geophys. Res.,
95, 16765-16774, 1990.
Bartlett, K.B., and R.C. Harris, Review and assessment
of methane emissions from wetlands, Chemo-
sphere, 26, 261-320, 1993.
Bartlett, K.B., P.M. Crill, J.A. Bonassi, I.E. Richey, and
R.C. Harris, Methane flux from the Amazon River
floodplain: Emissions during rising water, J. Geo-
phys. Res., 95, 16773-16788, 1990.
Baughcum, S.L., M. Metwally, R.K. Seals, and D.J.
Wuebbles, Emissions scenarios development:
Complete scenario database (Subchapter 3-4), in
The Atmospheric Effects of Stratospheric Aircraft:
A Third Program Report, edited by R.S. Stolarski
and H.L. Wesoky, NASA Ref. Publ. 1313, 185-
208, 1993.
2.27
-------
SOURCE GASES
Beck, J.P., C.E. Reeves, F.A.A.M. de Leeuw, and S.A.
Penkett, The effect of akcraft emissions on tropo-
spheric ozone in the northern hemisphere, Atmos.
Environ., 26A, 17-29,1992.
Beck, L.L., A global methane emissions program for
landfills, coal mines, and natural gas systems,
Chemosphere, 26, 447-452,1993.
Bekki, S., and J.A. Pyle, Two-dimensional assessment of
the impact of aircraft sulphur emissions on the
stratospheric sulphate aerosol layer, /. Geophys.
Res., 97, 15839-15847,1992.
Berges, M.G.M., R.M. Hofmann, D. Scharffe, and P.J.
Crutzen, Nitrous oxide emissions from motor ve-
hicles in tunnels and their global extrapolation, /.
Geophys. Res., 98D, 18527-18531, 1993.
Bloomfield, P., Inferred lifetimes, Chapter 3 in NASA
Report on Concentrations, Lifetimes and Trends of
CFCs, Halons and Related Species, edited by J.
Kaye, S. Penkett and F. Osmond, NASA Refer-
ence Publication 1339, Washington D.C.,
3.1-3.23, 1994.
Blunier,.T., J. Chappellaz, J. Schwander, J. Bamola, T.
Desperts, B. Stauffer, and D. Raynaud, Atmo-
spheric methane record from a Greenland ice core
over the past 1000 years, Geophys. Res. Lett., 20,
2219-2222, 1993.
Bonsang, B., M. Kanakidou, G. Lambert, and P. Mon-
fray, The marine source of C2-Ce aliphatic
hydrocarbons, /. Atmos. Chem., 6, 3-20, 1988.
Bonsang, B., G. Lambert, and C. Boissard, Light hydro-
carbons emissions from African savannah burning,
in Global Biomass Burning; Atmospheric, Climat-
ic and Biospheric Implications, edited by J.S.
Levine,MJT Press, 155-161, 1991.
Bouwman, A.F., Report of the third workshop of the
Global Emission Inventory Activity (GEIA).
RIVM Report 481507002, RIVM, Bilthoven, The
Netherlands, 1993.
Butler, J.H., J.W. Elkins, T.M. Thompson, B.D. Hall, and
V. Koropalov, Oceanic consumption of CH^CCl^.
Implications for tropospheric OH, J. Geophys.
Res., 96, 22347-22355, 1991.
Butler, J.H., J.W. Elkins, B.D. Hall, S.O. Cummings, and
S.A. Montzka, A decrease in the growth rates of
atmospheric halon concentrations, Nature, 359,
403-405, 1992.
Butler J.H., J.W. Elkins, T.M. Thompson, B.D. Hall,
J.M. Lobert, T.H. Swanson, and V. Koropalov, A
significant oceanic sink for atmospheric CCLt, The
Oceanography Society, Third Scientific Meeting,
April 13-16, Seattle, USA. Abstracts, p.55,1993.
Butler, J., J. Elkins, B. Hall, S. Montzka, S. Cummings,
P. Fraser, and L. Porter, Recent trends in the global
atmospheric mixing ratios of halon-1301 and ha-
lon-1211, in Baseline 91, edited by C. Dick and J.
Gras, Bureau of Meteorology—Commonwealth
Science and Industrial Research Organization
(CSIRO), in press, 1994.
Carroll, M.A., and A.M. Thompson, NOX in the non-ur-
ban troposphere, in Current Problems and
Progress in Atmospheric Chemistry, edited by J.R.
Barker, Advances in Physical Chemistry, in press,
1994. :
Chappellaz, J., T. Blunier, D. Raynaud, J.M. Barnola, J.
Schwander, and B. Stauffer, Synchronous changes
in atmospheric CHt and Greenland climate be-
tween 40 and 8 kyr BP, Nature, 366, 443-445,
1993.
CIAB, Global Methane Emissions from the Coal Indus-
try, Coal Industry Advisory Board, Global Climate
Committee, October, 1992.
Cicerone, R.J., Atmospheric carbon tetrafluoride: A
nearly inert gas, Science, 206, 59-61, 1979.
Codispoti, L., J. Elkins, T. Yoshinari, G. Frederich, C.
Sakamoto, and T. Packard, On the nitrous oxide
flux from productive regions that contain low oxy-
gen waters, in Oceanography of the Indian Ocean,
edited by B. Desai, 271-284, Oxford & IBH Pub-
lishing Co., New Delhi, India, 1992. :
Gofer, W.R., in, R.J. Bendura, D.I. Sebacher, G.L. Pel-
let, G.L. Gregory, and G.L. Maddrea, Jr., Airborne
measurements of Space Shuttle exhaust constitu-
ents, AIAA J., 23, 283-287, 1985.
Crutzen, P.J., and M.O. Andreae, Biomass burning in the
tropics: Impact on atmospheric chemistry and bio-
geochemical cycles, Science, 250, 1669-1679,
1990. ;-
Crutzen, P.J., and P.H. Zimmerman, The changing pho-
tochemistry of the troposphere, Tellus, 43AB,
136-151, 1991. ;
Cullis, C.F., and M.M. Hirschler, Man's emissions of
carbon monoxide and hydrocarbons into the atmo-
sphere, Atmos. Environ., 23, 1195-1203,11989.
2.25
-------
SOURCE GASES
Cunnold, D., P. Eraser, R. Weiss, R. Prinn, P. Siitnmonds,
B. Miller, F. Alyea, and A. Crawford,. Global
trends and annual releases of CC13F ancl'CC!2F2
estimated from ALE/GAGE and other measure-
ments from July 1978 to June 1991, J. Geophys.
Res., 99, 1107-1126, 1994. :
Dasch, J.M., Nitrous oxide emissions from veliicles, J.
Air Waste Manage. Assoc., 42, 63-67, 1992.
Davidson, E.A., Fluxes of nitrous oxide and nitric oxide
from terrestrial ecosystems, in Microbial Produc-
tion and Consumption of Greenhouse Gases:
Methane, Nitrogen Oxides, and Halomethanes,
edited by J.E. Rogers and W.B. Whitman^ Wash-
ington, D.C., 219-235, 1991. '>
Delwiche, C.C., and RJ. Cicerone, Factors affecting
methane production under rice, Global Bio-
geochem. Cycles, 7, 143-155, 1993. :
Diaz, S., J.P. Grime, J. Harris, and E. McPherson, Evi-
dence of a feedback mechanism limiting plant
response to elevated carbon dioxide, Nature, 364
616-617, 1993. : '
Dignon, J., NOX and SOX emissions from fossil liels: A
global distribution, Atmos. Environ., 26A: 1157-
1163, 1992.
Dignon, J., C.S. Atherton, J.E. Penner, and JJ. Walton,
NOX pollution from biomass burning: A global
study, Proceedings of the llth Conference on Fire
and Forest Meteorology, Missoula, Montana. April
16-19,1991. I
Dignon, J., and S. Hameed, Global emission of nijtrogen
and sulfur oxides from 1960 to 1980, J. Air Poll.
Control Assoc., 39, 180-186, 1989. '•''
Dignon, J., and J.E. Penner, Biomass burning: A source
of nitrogen oxides in the atmosphere, in global
Biomass Burning.-Atmospheric, Climatic aiid Bio-
spheric Implications, edited by J.S. Levine, MIT
Press, 370-375, 1991.,
Dignon, J., J.E. Penner, C.S. Atherton, and JJ. Walton,
Atmospheric reactive nitrogen: A model study of
natural and anthropogenic sources and the role of
microbial soil emissions, in Proceedings of
CHEMRAWN VII World Conference on the Chem-
istry of the Atmosphere: Its Impact on Global
Change, Baltimore, Maryland, December 2-6
1992.
Dlugokencky, E.J., L.P. Sleele, P.M. Lang, and K.A.
Masarie, The growth rate and distribution of atmo-
spheric methane, /. Geophys. Res., 99D
17021-17043, 1994a.
Dlugokencky, E.J., J.M. Harris, Y.S. Chung, P.P. Tans,
and I. Fung, The relationship between the methane
seasonal cycle and regional sources and sinks at
Tae-ahn Peninsula, Korea, Atmos. Environ., 27A
2115-2120, 1994b.
Dlugokencky, E.J., K.A. Maisarie, P.M. Lang, P.P. Tans,
L.P. Steele, and E.G. Nisbet, A dramatic decrease
in the growth rate of atmospheric methane in the
northern hemisphere during 1992, Geophys. Res.
Lett., 21, 45-48,1994c.
Donahue, N.M., and R.G. E'rinn, Non-methane hydro-
carbon chemistry in the remote marine boundary
layer,/ Geophys. Res., 95, 18387-18411, 1990.
Donoso, L., R. Santana, and 13. Sanhueza, Seasonal vari-
ation of N2O fluxes at a tropical savannah site:
Soil consumption of N2O during the dry season,
Geophys. Res. Lett., 20, 1379-1382, 1993.
Egli, R.A., Nitrogen oxide emissions from air traffic,
Chimia,44, 369-371, 1990.
Ehhalt, D.H., Concentrations and distributions of atmo-
spheric trace gases, Bet: Bunsenges, Phys. Chem.,
96, 229-240, 1992.
Ehhalt, D.H., and J. Rudolph, On the importance of light
hydrocarbons in multiphase atmospheric systems,
Berichte der Kernforschungsanlage Julich, No
1942, Institut fur Chemie 3, Atmospharische Che-
mie.Eds., 1-43, 1984.
Ehhalt, D.H., U. Schmidt, R. Zander, P. Demoulin, and
C.P. Rinsland, Seasonal cycle and secular trend of
the total and tropospheric column abundance of
ethane above the Jungfraujoch, /. Geophys. Res.,
96,4985-4994, 1991.
Ehhalt, D.H., F. Rohrer, and A. Wahner, Sources and dis-
tribution of NOX in the upper troposphere at
northern mid-latitudes, /. Geophys. Res., 97,
3725-3738, 1992.
Elkins, J., T. Thompson, T. Swanson, J. Butler, B. Hall,
S. Cummings, D: Fisher, and A. Raffo, Decrease in
the growth rates of atmospheric chlorofluorocar-
bons-11 and -12, Nature, 364, 780-783, 1993.
2.29
-------
SOURCE <3ASES
Erickson, D.J., Ocean to atmosphere carbon monoxide
flux: global inventory and climate implications,
Global Biogeochem. Cycles, 3, 305-314, 1989.
Etheridge, D.M., Peannan, G.I., and Eraser, P.J.,
Changes in tropospheric methane between 1841
and 1978 from a high accumulation rate Antarctic
ice core, Tellus, 44B, 282-294, 1992.
Fabian, P. Botcher, B.C. Kruger, and S. Lai, CF4 and
C2Fg in the atmosphere, /. Geophys. Res., 92,
9831-9835,1987.
Farmer, C.B., High resolution infrared spectroscopy of
the sun and earth's atmosphere from space, Micro-
chim.Acta (Wien) ffl, 198-214, 1987.
Fehsenfeld, E, J. Calvert, R. Fall, P. Goldan, A. Guen-
ther, C.N. Hewitt, B. Lamb, S. Liu, M. Trainer, H.
Westberg, and P. Zimmerman, Emissions of vola-
tile organic compounds from vegetation and the
implications for atmospheric chemistry, Global
Biogeochem. Cycles, 6, 389-430, 1992.
Fisher, D., T. Duafala, P. Midgley, and C. Niemi, Produc-
tion and emissions of CFCs, halons and related
molecules, Chapter 2 in NASA Report on Concen-
trations, Lifetimes and Trends of CFCs, Halons
and Related Species, edited by J. Kaye, S. Penkett,
and F. Osmond, NASA Reference Publication
1339, Washington, D.C., 2.1-2.35; 1994.
Fraser, P., and N. Derek, Halocarbons, nitrous oxide,
methane and carbon monoxide—the GAGE pro-
gram, in Baseline 91, edited by C. Dick and J.
Gras, Bureau of Meteorology-Commonwealth
Science and Industrial Research Organization, in
. press, 1994.
Fraser, P., S. Penkett, M. Gunson, R. Weiss, and F.S.
Rowland, Measurements, Chapter 1 in NASA Re-
port on Concentrations, Lifetimes and Trends of
CFCs, Halons and Related Species edited by J.
Kaye, S. Penkett, and F. Osmond, NASA Refer-
ence Publication 1339, Washington, D.C.,
1.1-1.68,1994a.
Fraser, P., N. Derek, L. Porter, D. Cunnold, F. Alyea, R.
Prinn, P. Simmonds, and R. Weiss, Atmospheric
chlorocarbons: Trends, lifetimes and emissions
from ALE/GAGE global observations, in Fourth
International Conference on Southern Hemi-
sphere Meteorology and Oceanography, preprints,
Hobart, March 1993, American Meteorology So-
ciety, 426-428, 1994b.
Fraser, P., D. Cunnold, F. Alyea, R. Prinn, P. Simonds, R.
Weiss, B. Miller, and A. Crawford, Lifetime and
emission estimates of 1,1,2 trichlorotrifluoro-
ethane (CFC-113) from ALE/GAGE; and other
measurements, 1978-1992, in preparation, 1994c.
Fung, I., J. John, J. Lerner, E. Matthews, M. Prather, L.P.
Steele, and P.J. Fraser, Three-dimensional model
synthesis of the global methane cycle, /. Geophys.
Res., 96, 13033-13065, 1991.
Fynes, G., I.S.C. Hughes, and P.W. Sage, The contribu-
tion of coal combustion to non-CO2 greenhouse
gas emissions inventories, International Sympo-
sium on Non-CO2 Greenhouse Gases, Why and
How to Control?, Maastricht, December, 1993.
Granier, C., and G. Brasseur, Impact of heterogeneous
chemistry on model predictions of ozone changes,
J. Geophys. Res., 97, 18015-18033, 1992.
Gunson, M.R., The atmospheric trace molecule spec-
troscopy (ATMOS) experiment—the Atlas-1
mission, SPBE vol. 1715, Optical Methods in At-
mospheric Chemistry, 513-521, 1992.
Gunson, M.R., C.B. Farmer, R.H. Norton, R. Zander,
C.P. Rinsland, J.H. Shaw, and B. C. Gao, Measure-
ments of CRt, N20, CO, H20 and O3 in the
middle atmosphere by the Atmospheric Trace
Molecule Spectroscopy Experiment on Spacelab
3, J. Geophys. Res., 95, 13867-13882,1990.
Gunson, M.R., M.C. Abrams, L.L. Lowes, E, Mahieu, R.
Zander, C.P. Rinsland, M.K.W. Ko, N[.D. Sze, and
O.K. Weisenstein, Increase in levels of strato-
spheric chlorine and fluorine loading between
1985 and 1992, Geophys. Res. Lett., in press,
1994.
Hameed,. S., and R. Cess, Impact of global warming on
biospheric sources of methane and its climate con-
sequences, Tellus, 35B, 1-7, 1983. ;
Hameed, S., and J. Dignon, Global emissions of nitrogen
and sulfur oxides in fossil fuel combustion 1970-
1986, J. Air Waste Manage. Assoc., 42, 159-163,
1991.
Hao, W.M., and D.E. Ward, Methane production from
global biomass burning, /. Geophys. Res., 98,
20657-20661, 1993.
2.30
-------
SOURCE GASES
Harris, R.C., and S. Frolking, The sensitivity of methane
emissions from northern freshwater' wetlands to
global change, in Global Warming and Freshwater
Ecosystems, edited by P. Firth and S. Fisher, in
press, 1992. /
Harwood, R.S., C.H. Jackman, I.L. Karol, and L.X. Qiu,
Predicted rocket and shuttle effects 00 stratospher-
ic ozone, Chapter 10 in Scientific Assessment of
Ozone Depletion: 1991, WMO GlobaJ Ozone Re-
search and Monitoring Project—Re{>ort No. 25,
pp. 10.1-10.12, World Meteorologies Organiza-
tion, Geneva, Switzerland, 1992. ;
Heidt, L.E., J.F. Vedder, W.H. Pollock, R.A. Lueb, and
B.E. Henry, Trace gases in the Antarctic atmo-
sphere, J. Geophys. Res., 94, 11599-11,611, 1989.
Hofmann, D.J., and S. Solomon, Ozone ^destruction
through heterogeneous chemistry following the
eruption of El Chichdn, /. Geophys. Res 94
5029-5041,1989. ;
Hofmann, D.J., Increase in the stratospheric background
sulfuric acid aerosol mass in the past 10 years. Sci-
ence, 248,996-1000, 1990.
Hofmann, D.J., Aircraft sulfur emissions, Nature 349
659,1991. ;
INPE, Deforestation in Brazilian Amazonia, Institute
Nacional de Pesquisas Espaciais, Sao Paulo Bra-
zil, 1992. ; '
IPCC, Intergovernmental Panel on Climate Change, R.
Watson, H. Rodhe, H. Oeschger, and U. Siegen-
thaler, Greenhouse gases and aerosols, Chapter 1
in Climate Change, The IPCC Scientific Assess-
ment, edited by J.T. Houghton, G.J. Jenkins, and
J.J. Ephroums, Cambridge University Press, Cam-
bridge, 1-40, 1990.
IPCC, Intergovernmental Panel on Climate Change, R.
Watson, L.G. Meira-Filho, E. Sanhue;ia, and A.
Janetos, Greenhouse gases: sources and sinks,
Chapter Al in Climate Change 1992, The Supple-
mentary Report to the IPCC Scientific Assessment,
edited by J.T. Houghton, B.A. Callander, and S.K.
Varney, Cambridge University Press, Cambridge
28-46,1992. h
IPCC, Intergovernmental Panel on Climate Change, M.
Prather, R. Derwent, D. Ehhalt, P. Fraser, E. San-
hueza, and X. Zhou, Other trace gases and
atmospheric chemistry, Chapter 2 in 1994 IPCC
Interim Report, Radiative Forcing of Climate
Change 1994, in preparation, 1994a.
IPCC, Intergovernmental Panel on Climate Change, D.
Schimel, I. Enting, M. Heimann, T. Wigley, D.
Raynaud, D. Alves, and U. Siegenthaler, The car-
bon cycle, Chapter 1, in 1994 IPCC Interim
Report, Radiative Forcing of Climate Change
1994, in preparation, 1994b.
IPCC, Intergovernmental Panel on Climate Change, P.R.
Jonas, R.J. Charlson, and H. Rodhe, Aerosols,
Chapter 3, in 1994 IPCC Interim Report, Radia-
tive Forcing of Climate Change 1994, in
preparation, 1994c.
Won, F.W., M. Brown, G.C. Toon, and M.R. Gunson,
Increase in atmospheric CHC1F2 (HCFC-22) over
southern California from 1985 to 1990, Geophys.
Res. Lett., 21, 1723-1726, 1994.
Jaffe, L.S., Carbon monoxide in the biosphere: Sources,
distribution, and concentrations, /. Geophys. Res
78, 5293-5305, 1973.
Jeffers, P.M., L.M. Ward, L.M. Woytowitch, and N.L.
Wolfe, Homogenous hydrolysis rate constants for
selected chlorinated methanes, ethanes, ethenes,
and propanes, Environ. Sci. Technol., 23, 965-969
1989.
Johnson, C., J. Henshaw, and G. Mclnnes, Impact of air-
craft and surface emissions of nitrogen oxides on
tropospheric ozone and global warming. Nature
355, 69-71, 1992.
Johnson D.E., T.M. Hill, G.M. Ward, K.A. Johnson,
M.E. Branine, B.R. Carmean, and D.W. Lodman,
Ruminant and other animals, in Atmospheric
Methane: Sources, Sinks, and Role in Global
Change, edited by M.A.K. Khalil, Springer-
Verlag, Berlin, 199-229, 1993.
Jouzel, J., N.I. Barkov, J.M. Barnola, M. Bender, J.
Chappellaz, C. Genthon, V.M. Kotlyakov, V.
Lipenkov, C. Lorius, J.R. Petit, D. Raynaud, G.
Raisbeck, C. Ritz, T. Sowers, M. Stievenard, F.
Yiou, and P. Yiou, Extending the Vostok ice-core
record of paleoclimate to the penultimate glacial
period, Nature, 364, 407-412, 1993.
2.31
-------
SOURCE GASES
Judd, A.G., R.H. Chalier, A. Lacroix, G. Lambert, and C.
Rowland, Minor sources of methane, in Atmo-
spheric Methane: Sources, Sinks, and Role in
Global Change, edited by M.A.K. Khalil, Spring-
er-Verlag, Berlin, 432-456, 1993.
Karol, I.L., Y.E. Ozolin, and E.V. Rozanov, Effect of
space rocket launches, on ozone and other atmo-
spheric gases, paper presented at the European
Geophys. Assoc. Conference, Wiesbaden, 1991.
Kawa, S.R., D.W. Fahey, L.E. Heidt, W.H. Pollock, S.
Solomon, D.E. Anderson, M. Loewenstein, M.H.
, Proffitt, JJ. Margitan and K.R. Chan, Photochem-
ical partitioning of the reactive nitrogen and.
chlorine reservoirs in the high-latitude strato-
sphere, J. Geophys. Res., 97, 7905-7923, 1992.
Kaye, J., S. Penkett and F. Onnond (Eds.), Report on
Concentrations, Lifetimes and Trends of CFCs,
Halons, and Related Species, NASA Reference
Publication 1339, Washington, D.C., 1994.
Kaye, J.A., A.R. Douglass, C.H. Jackman, R.S. Stolars-
ki, R. Zander and G. Roland, Two-dimensional
model calculation of fluorine-containing reservoir
species in the stratosphere, J. Geophys. Res, 96,
12865-12881,1991.
Keller, M., and P. Matson, Biosphere-atmosphere ex-
change of trace gases in the tropics: Evaluating the
effects of land use change, in Global Atmospheric-
Piospheric Chemistry: The First IGAC Scientific
' Conference, edited by R.G. Prinn, 103-117, 1994.
Keller, M., E. Veldkamp, A.M. Weis, and W.A. Reiners,
Pasture age effect on soil-atmosphere trace gas ex-
change in a deforested area of Costa Rica, Nature,
365, 244-246,1993.
Khalil, M.A.K., and R.A. Rasmussen, Atmospheric car-
bon tetrafluoride (CF4): Sources and trends,
Geophys. Res. Lett., 12, 671-672,1985.
Khalil, M.A.K., and R.A. Rasmussen, The global cycle
of carbon monoxide: Trends and mass balance,
Chemosphere, 20, 227-242, 1990.
Khalil, M.A.K., and R.A. Rasmussen, The global sourc-
es of nitrous oxide, J. Geophys. Res., 97,
14651-14660,1992.
Khalil, M.A.K., and R.A. Rasmussen, The environmen-
tal history and probable future of fluorocarbon 11,
/. Geophys. Res., 98, 23091-23106, 1993a.
Khalil, M.A.K., and R.A. Rasmussen, Decreasing trend
of methane: Unpredictability of future concentra-
tions, Chemosphere, 26, 803-814, 1993b.
Khalil, M.A.K., and R.A. Rasmussen, Global decrease
of atmospheric carbon monoxide, Nature, 370,
639-641, 1994.
Khalil, M.A.K., R.A. Rasmussen, and F. Moraes, Atmo-
spheric methane at Cape Meares: Analysis of a
high resolution data base and its environmental
implications, /. Geophys. Res., 98, 14753-14770,
1993a.
Khalil, M.A.K., R.A. Rasmussen, M.J. Shearer, S. Ge,
and J. A. Rau, Methane from coal burning, Chemo-
sphere, 26, 473-477, 1993b.
Kim, K.R. and H. Craig, Nitrogen-15 and oxygen-18
characteristics of nitrous oxide: A global perspec-
tive, Science, 262, 1855-1857, 1993. \
Kirchgessner, D.A., S.D. Piccot, and J.D. Winkler, Esti-
mate of global methane emissions from coal
mines, Chemosphere, 26, 453-472, 1993.
Kirchhoff, V.WJ.H., and E.V.A. Marinho, Surface car-
bon monoxide measurements in Amazonia, J.
Geophys. Res., 95, 16933-16943,1990.
Ko, M., N. Sze, W.-C. Wang, G. Shia, A. Goldman, F.
Murcray, D. Murcray, and C. Rinsland, Atmo-
spheric sulfur hexafluoride: Sources, sinks and
greenhouse wanning, /. Geophys. Res., 98,10499-
10507, 1993. ;
Koppmann, R., F.J. Johnen, C. Plass-Dulmer, and J. Ru-
dolph, Distribution of methylchloride,
dichloromethane, trichloroethene and tetrachloro-
ethene over the North and South Atlantic, /.
Geophys. Res., 98, 20517-20526, 1993.
Kreileman, G.J.J., and A.F. Bouwman, Computing land
use emissions of greenhouse gases, Water, Air and
Soil Poll., 76, 231-258, 1994. \
Kumer, J.B., J.L. Mergenthaler, and A.E. Roche,
CLAES CUt, N2O and CC12F2 (F12) global data,
Geophys. Res. Lett., 20, 1239-1242, 1993.
Kvenvolden, K.A., and T.D. Lorenson, Methane in per-
mafrost: Preliminary results from coring at
Fairbanks, Alaska, Chemosphere, 26, • 609-616,
1993. '
2.32
-------
SOURCE GASES
Lacroix, A.V., Unaccounted-for sources of fossil and
isotopically-enriched methane and their contribu-
tion to the emissions inventory, Chemosphere, 26,
507-557, 1993.
Lai, S., S. Venkataramani, and B.H. Subbaraya,, Methane
flux measurements from paddy fields in me tropi-
cal Indian region, Atmos. Environ., 27A,
1691-1694,1993.
Lambert, G., and S. Schmidt, Reevaluation of the ocean-
ic flux of methane: Uncertainties and long term
variations, Chemosphere, 26, 579-589, 1993.
Lashoff, D.A., The dynamic greenhouse: Feedback pro-
cesses that may influence future concentrations of
atmospheric trace gases and climate change, dim.
Change, 14, 213-242, 1989. '.,
Lassey, K.R., D.C. Lowe, C.A.M. Brenninkmeijer, and
A.J. G6mez, Atmospheric methane and its carbon
isotopes in the southern hemisphere: Their time
series and an instructive model, Chemosphere, 26,
95-109,1993. |
Law, C.S., and N.J.P. Owen, Significant flux pf atmo-
spheric nitrous oxide from the northwest Indian
Ocean, Nature, 346, 826-828, 1990. ',
Law, K.S., and J.A. Pyle, Modeling trace gas bijjdgets in
the troposphere, 2: CRf and CO, /. Geophys. Res.,
98, 18401-18412, 1993.
Leuenberger, M., and U. Siegenthaler, Ice-age atmo-
spheric concentration of nitrous oxide from an
Antarctic ice core, Nature, 360, 449-451, 1992.
Liaw, Y.P., D.L. Sisterson, and N.L. Miller, Comparison
of field, laboratory, and theoretical estimates of
global nitrogen fixation by lightning, /. Geophys.
Res., 95, 22489-22494, 1990.
Lobert, J.M., D.H. Scharffe, W.M. Hao, T.A. Kuhlbusch,
R. Seuwen, P. Warneck, and P.J. Crutzen, Biomass
burning: Experimental evaluation of biomass
burning emissions: Nitrogen and carbon contain-
ing compounds, in Global Biomass Burning:
Atmospheric, Climatic and Biospheric Implica-
tions, edited by J.S. Levine, MIT Press, 289-304,
1991.
Lobert, J.M., TJ. Baring, J.H. Butler, S.A. Montzka,
R.C. Myers, and J.W. Elkins, Ocean-atmosphere
exchange of trace compounds, 1992. Final report
to AFEAS on oceanic measurements of HCFC-22,
methylchloroform, and other selected halocar-
bons. National Oceanic and Atmospheric
Administration/Climate Monitoring and Diagnos-
tics Laboratory, Boulder, Colorado, USA, 1993.
Logan, J.A., Nitrogen oxides in the troposphere: Global
and regional budgets, /. Geophys. Res., 88, 10785-
10807, 1983.
Logan, J.A., Sources and sinks for carbon monoxide, in
Proceedings of the NATO Advanced Study Insti-
tute on Atmospheric Ozone: Its Variation and
Human Influences, Aldeia Las Acoteias, Algarre,
Portugal. October 1-13, 323-342, 1980.
Lopez-Valverde, M.A., M. Lopez-Puertas, C.J. Marks,
and F.W. Taylor, Global and seasonal variations in
middle atmosphere CO from UARS/IS AMS, Geo-
phys. Res. Lett., 20, 1247-1250, 1993.
Lowe, D., C.A.M. Brenninkmeijer, G. Brailsford, K.
Lassey, and A. Gomez, Concentration and '3C
records of atmospheric methane in New Zealand
and Antarctica: Evidence for changes in methane
sources, /. * Geophys. Res., 99D, 16913-16925,
1994.
Maiss, M., and I. Levin, Global increase of SF6 observed
in the atmosphere, Geophys. Res. Lett., 21, 569-
573, 1994.
Makide, Y., L. Chen, and T. Toninaga, Accurate mea-
surements and changing trends, of atmospheric
concentrations of CFC-11, CFC-12, CFC-113,
CFC-114, CH3CC13, CCLj, halon-1301 and halon-
1211, Proc. International Symposium on Global
Cycles of Atmospheric Greenhouse Gases, Sendai,
Japan, March, 38-51, 1994.
Mankin, W.G., and M.T. Coffey, Increased stratospheric
hydrogen chloride in the El Chich6n cloud, Na-
ture, 226, 170-174, 1984.
Mankin, W.G., M.T. Coffey, and A. Goldman, Airborne
observation of SO2, HC1, and O3 in the strato-
spheric plume of the Pinatubo volcano in July
1991, Geophys. Res. Lett., 19, 179-182, 1992.
2.33
-------
SOURCE GASES
Manley, S.L., K. Goodwin, and WJ. North, Laboratory
production of bromoform, methylene bromide,
and methyl iodide by macroalgae and distribution
in nearshore southern California waters, Limnol.
Oceanogr.,37, 1652-1659,1992.
Mand, S., and M.O. Andreae, Emission of methyl bro-
mide from biomass burning, Science, 263,
1255-1257,1994.
Martius, C, R. Wassmann, U. Them, A. Bandeira, H.
Rennenberg, W. Junk, and W. Seiler, Methane
emission from wood-feeding termites hi Amazo-
nia, Chemosphere, 26, 623-632, 1993.
McCulloch, A., Global production and emission:; of
bromochloro-difluoromethane and bromotri-
fluoromethane (halons 1211 and 1301), Atmos.
Environ., 26A, 1325-1329,1992.
Midgley, P., and D. Fisher, The production and release to
the atmosphere of chlorodifluoromethane (HCFC-
22), Atmos. Environ., 27A, 2215-2223, 1993.
Minson, D.J., Methane production by sheep and cattle in
Australia, Tellits, 45B, 86-88,1993.
Montzka, S.A., R.C. Myers, J.H. Butler, and J.W. Elkins,
Global tropospheric distribution and calibration
scale of HCFC-22, Geophys. Res. Lett., 20, 703-
706,1993.
Montzka, S.A., R.C. Myers, J.W Elkins, J.H. Butler, and
S.O. Cummings, Measurements of HCFC-141b
(1,1-dicbloro-l-fluoroethane) and HCFC-142b (1-
chloro-l,l-difluoroethane) hi ambient ah- samples
from the NOAA/CMDL global flask network,
manuscript in preparation, 1994.
Moore, R.M., and R. Tokarczyk, Volatile biogenic halo-
carbons hi the northwest Atlantic, Global
Biogeochem. Cycles, 7(1), 195-210,1993a.
Moore, R.M., and R. Tokarczyk, Chloro-iodomethane hi
N. Atlantic waters: A potentially significant source
of atmospheric iodine, Geophys. Res. Lett., 19,
1779-1782,1993b.
Moraes, R, and M.A.K. Khalil, Permafrost methane
content: 2. Modelling theory and results, Chemo-
sphere, 26, 595-607,1993.
MOller, J.F., Geographical distribution and seasonal vari-
ation of surface emissions and deposition
velocities of atmospheric trace gases, /. Geophys.
Res., 97, 3787-3804, 1992.
Nakazawa, T., T. Machida, M. Tanaka, Y. Fujii, S. Aoki,
and O. Watanabe, Differences of the atmospheric
CH4 concentration between the Arctic and Antarc-
tic regions in preindustrial/agricultural era,
Geophys. Res. Lett., 20, 943-946, 1993.
Neue, H.U. and P. A. Roger, Rice agriculture: Factor con-
trolling emissions, in Atmospheric Methane:
Sources, Sinks, and Role in Global Change, edited
by M.A.K. "Khalil, Springer-Verlag, Berlin, 254-
298,1993. ;
Novelli, P.C., K.A. Masarie, P.P. Tans, and P.M. J^ang,
Recent changes hi atmospheric carbon monoxide,
Science, 263, 1587-1590, 1994.
Ojima, D.S., D.W. Valentine, A.R. Mosier, W.J. Parton,
and D.S. Schimel, Effect of land use change on
methane oxidation hi temperate forest and grass-
land soils, Chemosphere, 26, 675-685,'1993.
Penkett S.A., J.D. Prosser, R.A. Rasmussen, and M.A.K.
Khalil, Atmospheric measurements of CF4 and
other fluorocarbons containing the CFs grouping,
/. Geophys. Res., 86, 5172-5178, 1981.
Penner, J.E., C.S. Atherton, J. Dignon, S.J. Ghan, J.J.
Walton, and S. Hameed, Tropospheric nitrogen: A
three-dimensional study of sources, distributions,
and deposition, /. Geophys. Res., 96, 959-990,
1991.
Penner, J.E., C.S. Atherton, and T.E. Graedel, Global
emissions and models of photochemically active
compounds, in Global Atmospheric-Biospheric
Chemistry: The First IGAC Scientific Conference,
edited by R.G. Prinn, in press, 1994.
Pepper, W., J. Legger, R. Swart, J. Watson, J. Edmonds,
and I. Mintzer, Emission Scenarios for the IPCC:
An Update. Assumptions, Methodology and Re-
sults, prepared for DPCC WG I, May 1992, 115
pp., 1992.
Piccot, S.D., J.J. Watson, and J.W. Jones, A global inven-
tory of volatile organic compound emissions from
anthropogenic sources, J. Geophysi Res., 97,
. 9897-9912, 1992.
Plumb, R.A., and M.K.W. Ko, Interrelationships be-
tween mixing ratios of long-lived stratospheric
constituents, /. Geophys. Res., 97, 10145-10156,
1992. ;
2.34
-------
SOURCE GASES
Pollock, W.H., L.E. Heidt, R.A. Lueb, J.F. Vender, MJ.
Mills, and S. Solomon, On the age of stratospheric
air and ozone depletion potentials in polar regions,
J. Geopkys. Res., 97, 12993-12999, 1992.
Prather, MJ., and R.T. Watson, Stratospheric 0zone de-
pletion and future levels,of atmospheric chlorine
and bromine, Nature, 344, 729-734, 1990.
Prather, MJ., and C.M. Spivakovsky, Tropospheric OH
and the lifetime of hydrochloroflucitocarbons
(HCFCs), /. Geophys. Res., 95, 18723-18729,
1990. !
Prather, M.J., and E.E. Remsberg, Eds., The Atmospher-
ic Effects of Stratospheric Aircraft: Report of the
1992 Model and Measurements Workshop, NASA
Reference Publication 1292, 1992.
Prather, MJ., M.M. Garcia, A.R. Douglas, C:H. Jack-
man, M.K.W. Ko and N.D. Sze, The space shuttle's
impact on the stratosphere, /. Geophys. Res., 95,
.18583-18590, 1990.
Prinn, R., D. Cunnold, P. Simmonds, F. Alyea, R. Boldi,
A. Crawford, P. Fraser, D. Gutzler, D. Hartley, R.
Rosen, and R. Rasmussen, Global average concen-
tration and trend for hydroxyl radicals deduced
from ALE/GAGE trichloroethane (methyl chloro-
form) data for 1978-1990, /. Geophys. Res., 97,
2445-2461, 1992.
Pyle, J.A., and A.E. Jones, An investigation of the impact
of the Ariane-5 launches on stratospheric ozone.
Report to the European Space Agency, 1991.
Rasmussen, R.A., Isoprene: Identification as a forest
type emission to the atmosphere, Environ. Sci.
Technol.,4, 667-671, 1972. \,
Rasmussen, R. A., M. A.K. Khalil, and F. Moraes, Perma-
frost methane content: 1. Experimental diata from
sites in Northern Alaska, Chemosphere, 26, 591-
594,1993.
Ravishankara, A.R., S. Solomon, A.A. Turnipseed, and
R.F. Warren, Atmospheric lifetimes of long-lived
halogenated species, Science, 259, 194-199,1993.
Reeburgh, W.S., N.T. Roulet, and B.H. Svenssbn, Ter-
restrial biosphere-atmosphere exchange in high
latitudes, in Global Atmospheric-Biospheric
Chemistry: The First IGAC Scientific Conference,
edited by R.G. Prinn, 165-178, 1994.
Rinsland, C.P., M.R. Gunson, M.C. Abrams, L.L.
Lowes, R. Zander, and E. Mahieu, ATMOS/AT-
LAS 1 measurements of sulphur hexafluoride
(SFg) in the lower stratosphere and upper tropo-
sphere,/. Geophys. Res., 98, 20491-20494, 1993.
Roulet, N., T. Moore, J. Bubier, and P. Lafleur, Northern
fens: Methane flux and climatic change, Tellus,
44B, 100-105, 1992,
Roulet, N.T., R. Ash, W. Quintan, and T. Moore, Meth-
ane flux from drained northern peatlands: Effect of
a persistent water (able lowering on flux, Bio-
geochem. Cycles, 7, 749-769, 1993.
Rowland, F.S., C. Warg, and D. Blake, Reduction hi the
rates of growth of global atmospheric concentra-
tions for CC13F (CFC-11) and CC12F2 (CFC-12),
in Proceedings International Symposium on Glo-
bal Cycles of Atmospheric Greenhouse Gases,
Sendai, Japan, 7-10 March, 38-51, 1994.
Rudolph, J., A. Khedim, T. Clarkson, and D. Wagen-
bach, Long term mŁ:asurements of light alkanes
and acetylene in the Antarctic troposphere, Tellus,
44B, 252-261,1992.
Russell, J.M. HI, A.F. Tuck, L.L. Gordley, J.H. Park,
S.R. Drayson, J.E. Harries, R.J. Cicerone, and PJ.
Crutzen, HALOE Antarctic observations in the
spring of 1991, Geophys. Res. Lett., 20, 718-721
1993.
Sanhueza, E., Biogenic emissions of NO and N2O from
tropical savanna soils, in Proceedings Internation-
al Symposium on Global Change (IGBP), Tokyo,
Japan, March 27-29,24-34, 1992.
Sanhueza, E., L. Cardenas, L. Donoso and M. Santana,
1994: Effect of plowing on COa, CO, CH4, N2O
and NO fluxes from tropical savannah soils. /.
Geophys. Res., 99D, 16429-16434, 1994.
Schauffler, S.M., L.E. Heidi, W.L. Pollock, T.M. Gilpin,
J.F. Vedder, S. Solomon, R.A. Lueb, and E.L. At-
las, Measurements of halogenated organic
compounds near the tropical tropopause, Geophys.
Res. Lett., 20, 2567-2570, 1993.
Schmidt, U., R. Bauer, A. FChedim, G. Kulessa, and C.
Schiller, Profile observations of long-lived trace
gases in the Arctic vortex, Geophys. Res. Lett., 18,
767-770, 1991.
2.35
-------
SOURCE GASES
Seller, W., and R. Conrad, Contribution of tropical eco-
systems to the global budgets of trace gases,
especially CH* H2, CO, and N2O, in The Geo-
physiology of Amazonia, edited by R. Dickenson,
Wiley, New York, 133-160, 1987.
Shearer, MJ., and M.A.K. Khalil, Rice agriculture:
Emissions, in Atmospheric Methane: Sources,
Sinks, and Role in Global Change, edited by
M.A.K. Khalil, Springer-Verlag, Berlin, 230-253,
1993.
Simmonds, P., D. Cunnold, G. Dolland, T. Davies, A.
McCulloch, and R. Derwent, Evidence of the
phase out of CFC use over Europe over the period
1987-1990, Atmos. Environ., 27A, 1397-1407,
1993.
Skole, D., and C. Tucker, Tropical deforestation and hab-
itat fragmentation in the Amazon: Satellite data
from 1978 to 1988, Science, 260, 1905-1910,
1993.
Steele, L.P., EJ. Dlugokencky, P.M. Lang, P.P. Tans,
R.C. Martin, and K.A. Masarie, Slowing down of
the global accumulation of atmospheric methane
during the 1980's, Nature, 358, 313-316, 1992.
Sturges, W.T., C.W. Sullivan, R.C. Schnell, L.E. Heidt,
W.H. Pollock, and C.E. Quincy, Bromoalkane pro-
duction by Antarctic ice algae: A possible link to
surface ozone loss, Tellus,45B, 120-126, 1993.
Subak, S., P. Raskin, and D.V. Hippel, National green-
house gases accounts: Current anthropogenic
sources and sinks. Climate Change, 25, 15-58,
1993.
Swanson, T., J.W. Elkins, J.H. Butler, S.A. Montzka,
R.C. Myers, T.M. Thompson, TJ. Baring, S.O.
Gumming, G.S. Dutton, A.H. Hayden, J.M.
Lobert, G.A. Holcomb, W.T. Sturges, and T.M.
Gilpin, Nitrous Oxide and Halocarbons Division,
Section 5 of CMDL No. 21 Summary Report
1992, edited by J.T. Peterson and R.M. Rosson,
59-75,1993.
Tabazadeh, A., and R.P. Turco, Stratospheric chlorine
injection by volcanic eruptions: HC1 scavenging
and implications for ozone, Science, 260, 1082-
1086,1993.
Tarr, M.A., W.L. Miller, and R.G. Zepp, Direct photo-
production of carbon monoxide from plant matter,
J. Geopkys. Res., submitted, 1994.
Taylor, F.W., et al, Remote sensing of atmospheric
structure and composition by pressure modulator
radiometer from space: The ISAMS experiment
on UARS, J. Geophys. .Res., 98, 10799-10814,
1993.
Thorneloe, S.A., M.A. Barlaz, R. Peer, L.C. Huff, L.
Davis, and J. Mangino, Waste management, in At-
mospheric Methane: Source, Sinks, and Role in
Global Change, edited by M.A.K. Khalil, Spring-
er-Verlag, Berlin, 362-398,1993.
Tokarczyk, R., and R.M. Moore, Production of volatile
organohalogens by phytoplankton cultures, Geo-
phys. Res. Lett., 21, 285-288, 1994.
Toon, G.C., C.B. Farmer, P.W. Schaper, L.L. Lowes, and
R.H. Norton, Composition measurements of the
1989 Arctic winter stratosphere by airborne infra-
red solar absorption spectroscopy, /. Geophys.
Res., 97, 7939-7961,1992a.
Toon, G.C., C.B. Farmer, P.W. Schaper, L.L. Lowes, and
R.H. Norton, Evidence for subsidence in the 1989
Arctic whiter stratosphere from airborne infrared
composition measurements, /. Geophys. Res., 97,
7963-7970, 1992b.
Toon, G.C., J.F. Blavier, J.N. Solario, and J.T. Szeto, Air-
borne observations of the 1992 Arctic winter
stratosphere by FTIR solar absorption spectrosco-
py, in Optical Methods in Atmospheric Chemistry,
SPJE Vol. 1715,457-467, 1992c. i
Toon, G.C., J.F. Blavier, J.N. Solario, and J.T. Szeto, Air-
borne observations of the composition of the 1992
tropical stratosphere by FTIR solar absorption
spectrometry, Geophys. Res. Lett., 20, 2503-2506,
1993. -
Tuck, A.F., S.J. Hovde, K.K. Kelly, J.M. Russell III, C.R.
Webster, and R.D. May, Intercomparison of HAL-
OE and ER-2 aircraft H2O and CO* observations
collected during the second Airborne Arctic
Stratospheric Experiment (AASE-H), Geophys.
Res. Lett., 20, 1243-1246, 1993.
Valentine, R.L., and R.G. Zepp, Formation;of carbon
monoxide from photodegradation of terrestrial
dissolved organic carbon in natural waters, Envi-
ron. ScLTechnoL, 27, 409-412,1993.
Wallace, L., and W. Livingston, The effect of the Pinatu-
bo cloud on hydrogen chloride and hydrogen
fluoride, Geophys. Res. Lett., 19, 1209-1212,
1992.
2.36
-------
SOURCE GASES
Wallace, D.W.R., P. Beinung and A. Putzka, Carbon tet-
rachloride and chlorofluorocarbons in the South
Atlantic Ocean, 19°S, submitted to /, Geophys.
Res., 1994. '
Warneck, P., Chemistry of the natural atmosphere, Int.
Geophys. Ser., 41, Academic, San Diego, Califor-
nia, 1988.
Wassmann, R., H. Papen, and H. Rennenberg, Methane
emission from rice paddies and possible mitiga-
tion strategies, Chemosphere, 26, 201-217, 1993.
Weisenstein, D.K., M.K.W. Ko, and N.D. Sze, The chlo-
rine budget of the present-day atmosphere: A
modeling study, /. Geophys. Res., 97, 2547-2559,
1992. j
Weiss, R., Changing global concentration of atmospher-
ic nitrous oxide, in Proceedings International
Symposium on Global Cycles of Atmospheric
Greenhouse Gases, Sendai, Japan, 7-10 March,
1994. :
Whalen, S.C., and W. Reeburgh, Interannual ivariations
in tundra methane emission: A 4-year time series
at fixed sites, Global Biogeochem. Cycles, 6, 139-
159,1992. I
Williams, E.J., A. Guenther, and EC. Fehsenfeld, An in-
ventory of nitric oxide emissions from soils in the
United States, J. Geophys. Res., 97, 7511-7519
1992. ,
WMO, Atmospheric Ozone 1985: Assessment of Our
Understanding of the Processes Controlling Its
Present Distribution and Change, World :Meteoro-
logical Organization, Global Ozone Research and
Monitoring Project - Report No. 16, Geneva,
Switzerland, 1986.
WMO, Scientific Assessment of Ozone Depletion: 1991,
World Meteorological Organization, Global
Ozone Research and Monitoring Project - Report
No. 25, Geneva, Switzerland, 1992. :
Woodbridge et ai. Estimates of total organic and inor-
ganic chlorine in the lower stratosphere from in
situ measurements during AASEII, submitted to /.
Geophys. Res., 1994. \''
Worthy, D., N. Trivett, K. Higuchi, J. Hopper, E. Dlugo-
kencky, M. Ernst, Jind M. Glumpak, Six years of in
situ atmosphericmethane measurements at Alert,
N.W.T., Canada, 19,88-1993: Seasonal variations,
decreasing growth rate and long range transport, in
Proceedings International Symposium on Global
Cycles of Atmospheric Greenhouse Gases, Sendai,
Japan, 7-10 March,, 130-133, 1994.
Zander, R., C.P. Rinsland, and P. Demoulin, Infrared
spectroscopic measurements of the vertical col-
umn abundance of sulphur hexafluoride, SFg,
from the ground, ,/. Geophys. Res., 96, 15447-
15454, 1991a. .
Zander, R., C.P. Rinsland, D.H. Ehhalt, J. Rudolph, and
Ph. Demoulin, Vertical column abundances and
seasonal cycle of acetylene, C2H2, above the Jung-
fraujoch station, derived from IR solar
observations, /. A.tmos. Chem., 13, 359-372
1991b.
Zander, R., M.R. Gunsom, C.B. Farmer, C.P. Rinsland,
F.W. Won, and E. Mahieu, The 1985 chlorine and
fluorine inventories in the stratosphere based on
ATMOS at 30° north latitude, /. Atmos. Chem., 15
171-186, 1992.
Zander, R., E. Mahieu, P. Demoulin, C.P. Rinsland, O.K.
Weisenstein, M.K.W. Ko, N.D. Sze, and M.R.
Gunson, Secular evolution of the vertical column
abundances of CHC1F2 (HCFC-22) in the Earth's
atmosphere inferred from ground-based IR solar
observations at Jungfraujoch and at Kitt Peak, and
comparison with model calculations, J. Atmos.
Chem., 18, 129-148, 1994a.
Zander, R., D.H. Ehhalt, C.P. Rinsland, U. Schmidt, E.
Mahieu, J. Rudolph, P. Demoulin, G. Roland, L.
Delbouille, and A.J. Sauval, Secular trend and sea-
sonal variability of the column abundance of N2O
above the Jungfraujoch station determined from
IR solar spectra, /. Geophys. Res., 99D, 16745-
16756, 1994b.
2.37
-------
SOURCE GASES
Zander, R., Ph. Demoulin, and E. Mahieu, Monitoring of
the atmospheric burdens of CHU, N2O, CO,
CHC1F2 and CF2ch above central Europe during
the last decade, in Proceedings of International
Symposium on Non-CO^ Greenhouse Gases-
Why and How to Control?, Maastricht, Dec.
13-15,1993, to appear in Environmental Monitor-
ing and Assessment, Kluwer Academic Pub.,
1994c.
Zimmerman, P.R., R.B. Chatfield, J. Fishman, P.J.
Crutzen, and P.L. Hanst, Estimates on the produc-
tion of CO and H^ from the oxidation of
hydrocarbon emissions from vegetation, Geophys.
Res. Lett., 5, 679-682, 1978.
Zolensky, M.E., D.S. McKay, and L.A. Kaczqr, A ten-
fold increase in the abundance of large solid
particles in the stratosphere, as measured over the
period 1976-1984, J. Geophys. Res., 94, 1047-
1056,1989.
2.38
-------
PART 2
ATMOSPHERIC PROCESSES RESPONSIBLE FOR THE
i
OBSERVED CHANGES IN OZONE
Tropical
Chapter 3
Polar Ozone
Chapter 4
and Midlatitude Ozone
Chapter 5
Tropospheric Ozone
-------
-------
CHAPTERS
Polar Ozone
•h-
Lead Author:
D.W. Fahey
Co-authors:
G. Braathen
D. Cariolle
Y. Kondo
W.A. Matthews
MJ. Molina
J.A. Pyle
R.B. Rood
J.M. Russell m
U. Schmidt
D.W. Toohey
J.W. Waters
C.R. Webster
S.C. Wofsy
Contributors:
T. Deshler
J.E. Dye
T.D.A. Fairlie
W.L. Grose
G.L. Manney
P.A. Newman
A. O'Neill
R.B. Pierce
W. Randel
A.E. Roche
C.R. Trepte
-------
-------
•• CHAPTERS
POLAR OZONE
Contents
SCIENTIFIC SUMMARY '.•
3.1 INTRODUCTION ..' 33
3.2 VORTEX FORMATION AND TRACER RELATIONS 3 5
3.3 PROCESSING ON AEROSOL SURFACES 3 1Q
3.3.1 Polar Stratospheric Cloud Formation and Reactivity 3\Q
3.3.2 Atmospheric Observations j. 3 14
3.3.2.1 Aerosol Measurements 3 14
3.3.2.2 Release of Active Chlorine 3 14
3.3.2.3 Changes in Reservoir Chlorine 3 jo
3.3.2.4 Active Bromine .>. 3 ««
3.3.2.5 Denitrification and Dehydration 3 20
3.3.3 Role of MtPinatubo Aerosol ,.) 324
3.3.4 Model Simulations ,;; o 2fi
3.4 DESTRUCTION OF OZONE !, _ 3 27
3.4.1 Ozone Loss: Observations and Gdculations 3 27
3.4.2 Variability : Z'IZZZZZZZZZZ 329
3.4.3 Photochemical Recovery 3 33
3.5 VORTEX ISOLATION AND EXPORT tO MTOLATITUDES : 3.34
3.5.1 Vortex Boundaries 3 34
3.5.2 Constituent Observations 3 35
3.5.3 Radiative Cooling .:. , <,,-
3.5.4 Trajectory Models „ ' 3 37
3.5.5 Three-Dimensional Models u 3 37
REFERENCES [; 3 4J
-------
-------
i POLAR PROCESSES
SCIENTIFIC SUMMARY
Substantial new results have been obtained since the last assessment in the areas of observations, laboratory
measurements, and modeling. These new results reaffirm the key role of anthropogenic halocarbons as the cause of
ozone loss in polar regions and increase confidence in the processes associated with this loss: the formation of a polar
vortex in high-latitude winter, the growth of aerosol surfaces at low temperatures characteristic of the vortex, the conver-
sion of inactive chlorine to active forms on these surfaces, the subsequent chlorine-cataly;red loss of ozone, the return of
chlorine to inactive forms in the polar regions in spring, and the breakup of the vortex and its dispersal to lower latitudes.
Ozone :
Results of observational and modeling studies since the last assessment reaffirm the role of anthropogenic halo-
carbon species in Antarctic ozone depletion. Satellite observations show a strong spatial and temporal correlation
of chlorine monoxide (CIO) abundances with ozone depletion in the Antarctic vortex. Photochemical model
calculations of ozone depletion are consistent with observed losses in the Antarctic.
I .
Chlorine- and bromine-catalyzed ozone loss has been confirmed in the Arctic winter. Consistent with expecta-
tions, these losses are smaller than those observed over Antarctica. Photochemical model calculations constrained
with in situ and satellite observations yield results consistent with the observed ozone loss.
Interannual variability in the photochemical and dynamical conditions of the vortices continues to limit reliable
predictions of future ozone changes in pqlar regions, particularly in the Northern Hemisphere.
Chlorine species
Satellite measurements show that elevated CIO concentrations cover most of both polar vortex regions during
much of the winter. This is consistent with the picture that virtually all available chlorine becomes fully activated
in both winter vortices through heterogeneous reactions that occur on aerosol particles formed at low tempera-
tures.
In situ and remote measurements show that hydrochloric acid (HC1) and chlorine nitrate (C1ONO2) concentra-
tions are markedly reduced in the vicinity of the elevated CIO concentrations. This anticorrelation is
quantitatively consistent with the picture ithat HC1 and C1ONO2 are converted to reactive chlorine. Chlorine in the
stratosphere originates largely from anthropogenic halocarbons.
Aerosols
Laboratory studies reaffirm that surface reactions on aerosol particles efficiently produce active chlorine from
inactive forms. The rate of the principal reaction of HC1 with C1ONO2 is a strong function of temperature and
relative humidity, and depends to a lesser extent on bulk aerosol composition.
Sulfate aerosol from the Mt. Pinatubo eruption reached high latitudes in the stratosphere, enhancing reactions
involving aerosol particles in and near the polar vortices. This led to chlorine activation over larger regions in the
high latitude stratosphere, especially nesif the vortex boundaries, and extended the spatial extent of halogen-
related ozone loss. ;
| !
The formation and reactivity of aerosol particles within the vortex can be simulated, in part, by microphysical
models. Two- and three-dimensional photochemical transport models confirm observations that chlorine can be
activated efficiently throughout the entire!vortex within days. i
3.1
-------
POLAR PROCESSES
Aerosol particles in the polar stratosphere are known ternary condensates of nitric acid (HNOs), sulfuric acid
(HaSCXj), and water (H2O). Important progress has been made in the characterization of these condensates in
theoretical and laboratory studies.
Satellite measurements confirm that the sequestering and removal of HNOa by aerosol particles is a predominant
feature of the Antarctic vortex for much of the winter, whereas removal in the Arctic is generally less intense and
more localized.
Despite extensive observational evidence for dehydration and denitrification, the underlying microphysical
mechanisms and necessary atmospheric conditions that control particle formation and sedimentation have not
been adequately described. This is an important limitation for reliably predicting ozone loss in polar regions,
particularly in the Northern Hemisphere.
Vortex
New satellite observations of long-lived tracers and modeling studies confirm that air within the center of the
polar winter vortices is substantially isolated from extravortical air, especially in the Antarctic.
i
Nearly all observational and modeling studies are consistent with a time scale of three to four months to replace a
substantial fraction of inner Antarctic vortex air.
Models show that most mass transport out of the vortex in the lower stratosphere occurs below about 16 km
altitude. !
Erosion by planetary and synoptic wave activity transports air from the vortex edge region to lower latitudes. Data
and model studies provide conflicting interpretations of the magnitude of this transport and its effect on lower
latitudes. There is little evidence of significant lateral mixing into the vortex except during strong wave events in
the Arctic. '
Observed correlations of nitrous oxide abundances with those of inactive chlorine species, reactive nitrogen, and
ozone over broad regions at high latitudes in the lower stratosphere have proved useful for diagnosing ozone
destruction throughout the vortex.
3.2
-------
POLAR PROCESSES
3.1 INTRODUCTION
i
Depletion of polar ozone in the winter seasons
continues to be an important scientific issue for both
hemispheres. While the "ozone hole" has become an an-
nual feature in the Southern Hemisphere, increased
losses have been noted in the Northern Hemisphere in
recent years. Increased losses at midlatitudes inay be
connected to the more intense loss processes occurring
in polar regions. The World Meteorological Organiza-
tion (WMO) Scientific Assessment of Ozone Depletion:
1991 reaffirmed halogen chemistry as the cause of se-
vere ozone depletion in the Antarctic as well as of
smaller losses in the Arctic (WMO, 1992). The Causes
for the observed year-to-year variability of such losses
and effects at midlatitudes were left as uncertain. For this
assessment, a wide variety of new evidence is available
to confirm the basic paradigm of ozone loss in polar re-
gions. This new evidence, which follows from a high
level of activity involving observational, laboratory, and
modeling studies that took place in the period J991-
1994, has better defined a number of the photochemical
and dynamical aspects of polar ozone depletion.
The principal cause of ozone loss in the polar re-
gions is photochemistry involving the halogen species,
chlorine and bromine. Long-lived halogens species, pri-
marily chlorofluorocarbons, are released in, the
troposphere from human activities. The photochemical
degradation of these organic source molecules iih the
stratosphere leads to the formation of inorganic halogen
species, of which chlorine monoxide (CIO), chlorine ni-
trate (C1ONO2), hydrochloric acid (HC1), bromine
monoxide (BrO), and bromine nitrate (BrONO:!) are
most important. The release of chlorine from the more
stable reservoirs occurs in high-latitude winter in reac-
tions on surfaces of stratospheric aerosol particles. The
formation and reactivity of these particles are enhiinced
at the low temperatures characteristic of the interior of
the polar vortices. This reactive processing maintains
high levels of active chlorine species that, along with
BrO, catalytically destroy ozone as this air encounters
sunlight With sufficient insolation and warmer tem-
peratures, chlorine is returned to its reservoir forms
during a photochemical recovery period and ozone de-
struction slows. The removal of reactive nitrogen by
aerosol particle sedimentation in the vortex, a process
defined as denitrification, strongly regulates the rate of
recovery by controlling the availability of active chlo-
rine. This paradigm, illustrated in Figure 3-1, has been
broadly supported by a wide variety of data and interpre-
tation in previous WMO assessments and has been
strengthened substantially in this assessment period.
This assessment period was marked by the launch
of the National Aeronautics and Space Administration
(NASA) Upper Atmosphere Research Satellite (UARS)
in late 1991, after more than a decade of preparation
(Reber, 1990; Reber et al., 1993). The satellite contains
four instruments for the measurement of trace species in
the stratosphere (Barath et al., 1993; Russell et al.,
1993a; Roche etal., 1993a; Taylor et al., 1993) and other
instruments for wind, solar radiation, and energetic par-
ticles. From an orbit of 600 km inclined 57° to the
equator, UARS provides broad coverage in both hemi-
spheres with a maximum latitude of 80°. The precession
of the orbit with respect to trie Sun provides measure-
ments during all local solar times over a month-long
period. Of particular importance for this assessment are
the UARS observations at high latitudes of the chlorine
reservoir species C1ONO2 and HC1, active chlorine in
the form of CIO, the reactive nitrogen species nitric acid
(HNO3), water vapor, aerosol extinction, and the long-
lived tracers nitrous oxide (N;>O), methane (CKt), and
hydrofluoric acid (HF). In addition, ozone measure-
ments show the distribution and evolution of ozone loss
in the polar regions. New aspects of the transport of air in
and near the vortex are evident from the observations of
long-lived tracers. The interpretation of UARS data will
remain an active research area as the data set continues
to grow.
The body of in situ observations in the strato-
sphere was greatly increased with aircraft and balloon
measurements made during the European Arctic Strato-
spheric Ozone Experiment (EASOE) (Pyle et al., 1994)
and the NASA Airborne Arctic Stratospheric Expedition
II (AASE II) (Anderson and Toon, 1993), which were
both held during the Northern Hemisphere winter of
1991/92. Each included measurements of reactive nitro-
gen and chlorine species, long-lived tracers and reservoir
species, and aerosols, combined with modeling studies
of observed photochemical and dynamical changes. The
observation period extended from pre-vortex conditions
in fall, through the lowest temperature conditions
marked by chlorine activation, a nd into the photochemi-
cal recovery period in early spring. The breadth of
3.3
-------
POLAR PROCESSES
Chlorine Reservoirs in the Polar Stratosphere
Inactive
chlorine
surface
Active
gas phase
Inactive
chlorine
Denitrification & dehydration
f >•
Surface processing
Chlorine catalyzed
ozone destruction
Inactive chlorine
Formation, cooling,
& descent
Maximum intensity
recovery
Breakup
Surface reaction threshold
Polar vortex evolution
Rgure 3-1. Schematic of the photochemical and dynamical features of the polar regions related to ozone
depletion. The upper panel represents the conversion of chlorine from inactive to active forms in winter in the
lower stratosphere and the reformation of inactive forms in spring. The partitioning between the active chlo-
rine species CI2, CIO, and CI2O2 depends on exposure to sunlight after polar stratospheric;cloud (PSC)
processing. The corresponding stages of the polar vortex are indicated in the lower panel, where the temper-
ature scale represents changes in the minimum polar temperatures in the lower stratosphere (see Figure
3-3) (adapted from Webster et a/., 1993a). .
3.4
-------
POLAR PROCESSES
instrumentation and period of measurements have re-
sulted in a unique data set for the examination of the
paradigm in Figure 3-1, In addition, ground-teised obser-
vations during EASOE and separate efforts in Antarctica
have also yielded important insights into the evolution of
reactive chlorine and nitrogen during the winter season.
Modeling studies continue to advance with im-
provements in computational facilities and algorithms
and with new atmospheric data. Specifically, photo-
chemical models that incorporate observations of
long-lived tracers, reservoir species, new kinetic data,
and meteorological conditions are now able j to make
more representative calculations of ozone loss jn the po-
lar vortex. Studies of the fluid dynamics near the vortex
now provide more detailed descriptions of ;air parcel
motion in regions of high potential vorticity (F'V) gradi-
ents, improving estimates of the transport into and out of
the vortex interior. ITie continued refinement of such
models is an essential component for future predictions
of ozone loss and its variability.
New laboratory studies have examined atspects of
the homogeneous and heterogeneous chemistry: underly-
ing the kinetics of ozone loss. Specifically, new
photolysis cross section measurements have been made
for HNO3 and C1ONO2 under stratospheric conditions.
Photolysis of HNOa »s a limiting step for photochemical
recovery in early spring in the vortex. Significant ad-
vances have been made in the understanding of the
formation and growth of aerosols and the reactivity of
aerosol surfaces in polar regions. These advances build
on the extensive effort expended jflTrecent years to devel-
op new laboratory techniques to characterize multiphase
surface growth under stratospheric conditions:: At the
same time, the understanding of the thermodynamics of
aerosol growth has progressed to explain laboratory and
atmospheric observations.
Finally, the assessment period was marked by the
eruption of Mt. Pinatubo in the Philippines in June 1991,
months before the launch of the UARS satellite land the
start of the EASOE and AASE H campaigns. The in-
creased loading of stratospheric aerosol was predicted to
cause significant changes in ozone at midlatitudes as a
result of increased heterogeneous reactivity (Krasseur
and Granier, 1992; Prather, 1992; Hofmann iind So-
lomon, 1989) (see Chapter 4). The aerosol did not reach
polar regions in abundance until the southern winter of
1992 and the northern winter of 1992-93. Observational
and modeling evidence suggests the enhancement of
volcanic aerosol near the vortex will increase ozone loss
associated with heterogeneous processes in that region.
Studies have continued as the volcanic aerosol in the
stratosphere gradually diminished over a period of sever-
al years following the eruption.
3.2 VORTEX FORMATION AND TRACER
RELATIONS
The vortex that forcas in each winter hemisphere
in the polar region sets the context of ozone depletion
(see Figures 3-1 and 3-2). The temporal as well as the
200,
90
Tracer
Zonal Wind
Figure 3-2. Schematic of the circulation and mix-
ing associated with the polar vortex in the Arctic
midwinter or Antarctic early spring periods. The ver-
tical scale is shown in altitude (km) and pressure
(mb) units. The horizontal! scale is latitude in de-
grees. Arrows indicate mixing (double) and flow
(single), with longer arrows representing larger
rates. Other features are zonal wind contours (thin
lines), jet core (J), and Icing-lived tracer isopleths
(thick lines) (Schoeberl et al., 1992)
3.5
-------
POLAR PROCESSES
SH 30 hPa
NH 30 hPa
May
Figure 3-3.' A summary of the minimum polar vortex temperatures in the period 1978 to 1994 at Ł0 tiPa, 50
hPa, and 100 hPa (1 hPa = 1 mb) in the lower stratosphere in the Northern (NH) and Southern j[SH) hemi-
spheres (National Meteorological Center analysis). The range of observations between 1978 afid 1992 is
given by the shaded region. The narrow white band is the average of the data set. The blaick dot? represent
data for 1993 in the Antarctic and 1992-93 in the Arctic winter. Lines indicate approximate temperature
thresholds for Type I (upper) and Type II (lower) PSC formation (adapted from Nagatani etal., 1990).
spatial scale of the activation of chlorine that catalytical-
ly destroys ozone is associated with the extent of low
temperatures inside the vortex. In addition, the dynami-
cal features of the vortex determine the distribution of air
from the vortex to lower latitudes and the incorporation
of lower latitude air into the vortex. Many features of
vortex formation are understood from observational and
modeling studies (Schoeberl and Hartmann, 1991;
Schoeberl etal, 1992; Dritschel and Legras, 1993; Man-
ney and Zurek, 1993; Stratum and Mahlman, 1994a, b).
After autumn equinox, increasing polar darkness and ra-
diative cooling of polar air lead to the formation of a
circumpolar wind belt. This westerly wind belt, or polar
night jet, defines the polar vortex in each;hemisphere
(see Figure 3-2). The vortex edge region is characterized
by large gradients in PV and mixing and transport prop-
erties. Large differences in the wind and temperature
fields of the vortex exist between hemispheres (see Fig-
ure 3-3) (Manney and Zurek, 1993). The yortex in the
Southern Hemisphere is stronger, develops lower tem-
peratures, and persists longer than the northern vortex.
The cause is related to differences in planetary wave ac-
tivity that modifies the temperature and dynamical
structure of the vortex. Wave activity is more frequent
and of larger amplitude in the north, owing to more dom-
inant orographic features and the greater land/sea
3.6
-------
POLAR PROCESSES
Antarctic
AAOE
Arctic
AASE
-12 • -8 -4 0 4 8 12
SKYHI Antarctic Winter
12 8 4 0-4-8
SKYHI Arctic Winter II
-12
OB
<5
Q.
E
0)
"eg
o>
o
0-
40
60
80
100
120
140
160
180
200
220
240
260
280
Q.
Q.
-12 -8-4048 ;12
Degrees Latitude from Vortex Edge
12 8 4 0-4-8
Degrees Latitude from Vortex
-12
Edge
contrast. Because ozone depletion depends cin the inter-
action of the vortex wind field with local regions of low
temperatures and the resultant chemical processing, the
temperature differences represented in Figure 3-3 under-
lie the large differences in ozone depletion observed
between the hemispheres (see Section 3.4). Thus, pre-
dictions of future ozone losses and the role of climate
change in polar processes depend directly on factors that
change the temperature and wind fields during the win-
ter seasons. ;
An important diagnostic for the formation of the
polar vortices and subsequent ozone loss is the high-lat
itude distribution of lon.g-lived trace species such as
N2O, CRt, and the chlorofluorocarbons CFC-11 and
CFC-113. All have large gradients in the stratosphere
(decreasing with altitude) resulting from photochemical
loss and transport. Air descending into the center of the
vortex reduces values of these traces species, thereby
creating horizontal gradients inside the vortex (see Fig-
ure 3-4). Balloon and aircraft measurements of N2O
beginning before vortex formation serve as a baseline for
documenting the temporal variation of the vertical struc-
ture within the vortex (Bauer et a/., 1994; Podolske et
a/., 1993). A comparison of the location of high PV from
3.7
-------
POLAR PROCESSES
HALOE
If)
tfl
o
100
-80
-64 -48
-32
-16 0 16
Latitude
32
48
64
Figure 3-5. Pressure versus latitude cross section of CH4 from the UARS Halogen Occupation Expenment
(HALOE) satellite instrument. Data are from sunset scans over the period 21 September to 15 October 1992
Analyzed S he version-17 algorithm. The pressure range corresponds to altitudes between about.16 and
65 km. Latitude is expressed in degrees, with negative latitude values corresponding to the Southern Hemi-
sphere (adapted from Russell et ai, 1993b). |
meteorological analyses and low N2O from satellite
fields shows excellent correspondence in the Arctic,
thereby increasing confidence in the analysis of vortex
structure (Manney et ai, 1994a). Simulations using a
general circulation model and an improved two-dimen-
sional model successfully reproduce important features
of the observed N2O distributions in and near the north-
ern vortex (see Figure 3-4 and Section 3.5.5) (Straihan
and Mahlman, 1994a; Garcia et ai, 1992).
Satellite observations of CH4 and HF; reveal un-
mixed vertical descent taking place at the center of the
vortex in the Antarctic (see Figure 3-5) (Russell et ai,
1993b). The lack of vertical gradient indicates that air at
lower altitudes containing larger CH4 values has not
been mixed with the descending air. Although not ob-
served before, the strong descent implied by the
observations matches earlier predictions (Dahielsen and
Houben, 1988). The observations are qualitatively simu-
3.8
-------
POLAR PROCESSES
lated with a mechanistic three-dimensional (3-D) model
(see Section 3.5.5), following many atmospheric air par-
cels as they undergo transport from the mesosphere as a
result of radiative cooling in winter and early spring
(Fisher et al., 1993). These results augment the depiction
of the vortex in Figure 3-2, further clarifying its dynam-
ical evolution. :
Observations have established that simple, com-
pact relationships exist in the lower stratosphere
between N2O and other long-lived species that also are
photochemically destroyed in the stratosphere. These re-
lationships result when photochemical lifetimes are long
compared to transport and mixing times between low-
and high-latitude regions (Plumb and Ko, 1992; Mahl-
man et al., 1980). The compactness of the relationship
allows one of the species to be predicted confidently
from a measurement of the other. The distribution of
N2O in and near the vortex is often related to (he distri-
bution of PV and potential temperature (Stnihan and
Mahlman, 1994a, b). Thus, these relationships jure useful
in predicting conditions throughout the vortex relevant
to the specific reactive processes that control ozone.
However, since the knowledge of these relationships is
based on limited data sets, assimilation of further data
must continue in order to establish the range of applica-
bility, i
The first of three important examples of these rela-
tionships is that of N2O to organic and inorganic: chlorine
reservoirs (see Figure 3-6) (Woodbridge et al., 1994;
Schmidt et al., 1991; 1994; Schauffler etal., 1993; Kawa
et al., 1992a). The principal species in the organic chlo-
rine reservoir, CCly,
CCly = CC12F2 (CFC-12) + CC13F (CFC-11) +
CC12FCC1F2(CFC-113)+ <\
CC14 (carbon tetrachloride) + !
CH3CCl3(CFC-140a) + ;
CHC1F2 (CFC-22) + CH3C1 (methyl chloride)
: (3-la)
include those species that comprise over 95 percent of
the available organic chlorine in the stratosphere. Each
species displays a compact correlation with N26, where
the slope is related to the ratio of lifetimes in the strato-
sphere (see Chapter 2). As a consequence, CCIy, as the
sum over organic species, also shows a compact relation
with N2O. The inorganic chlorine reservoir, CIy;
Northern Hemisphere 1992
100 150 200 250
IM2O (ppbv)
300
Figure 3-6. Total available chlorine (upper line)
and total inorganic chlorine (Cly) (lower line) plotted '
versus N2O from aircraft observations in the Arctic
winter of 1991/92. The vertical scale is in parts per
trillion by volume (pptv). Total organic chlorine
(CCly) is the difference between total available
chlorine and Cly. As the residence time of air in-
creases in the stratosphere, photochemical
reactions decrease N2O values in an air parcel and
convert CCIy species to Cly species. The diamond
symbol represents the rererence point for tropo-
spheric chlorine in 1991/92 of 3.67 ppbv. The
dashed lines represent estimated uncertainties
(Woodbridge et al., 1994).
Cly = Cl + 2C12 + CIO + OC1O + 2C12O2 + HOC1 +
HC1 + BrCl + C1ONO2 (3- Ib)
is produced as CCly and N2O are destroyed in the strato-
sphere. Since Cly contains CIO, an effective reactant in
ozone destruction, the distribution of Cly in polar regions
is of great interest The combination of the distribution
of N2O at high latitudes in Figure 3-4 and the compact
relations in Figure 3-6 indicates how CCly and Cly are
distributed throughout both vortices. Modeling of ozone
loss throughout the vortex cam be usefully constrained by
knowledge of these distributions (Salawitch et al.,
1993).
3.9
-------
POLAR PROCESSES
The second example is the linear relationship be-
tween N2O and the reactive nitrogen reservoir, NOy
(Fahey et aL, 1990a; Loewenstein et aL, 1993; Kondo et
al, 1994a). The primary source of NOy,
NOy = NO + NO2 + NO3 + 2N2O5 + HONO +
HO2NO2 + HNO3 + CH3C(O)OONO2 +
C1ONO2 + BrONO2 + aerosol nitrate +.... (3-2)
is the photochemical destruction of N2O in the middle
stratosphere. In the polar lower stratosphere in winter,
the sequestering of active chlorine in the form of
C1ONO2 moderates ozone destruction. The NOy/N2O
correlation has been observed to be linear before vortex
formation in the Northern Hemisphere and outside the
vortex boundary in both hemispheres. Departures from
linearity at low N2O values have been observed as ex-
pected from the photochemical destruction of NOy in the
upper stratosphere. Departures from linearity at higher
N2O values demonstrate the irreversible removal of NOy
as a result of the sedimentation of aerosol particles con-
taining NOy species. This removal of NOy greatly
enhances the potential for ozone destruction in an air
parcel located in the polar vortex in spring (Brune et al.,
1991; Salawitch etal., 1993).
The third example is the correlation of ozone with
N2O that primarily follows from the production of ozone
in regions where NaO ls photochemically destroyed. In
situ aircraft measurements, satellite observations, and
photochemical model simulations show linear correla-
tions during whiter months at mid- and high latitudes in
the absence of significant polar ozone loss (Proffitt et al.,
1990,1992,1993; Weaver etal., 1993). Since ozone also
has loss processes in the stratosphere at other latitudes
and during other seasons, deviations from a constant lin-
ear correlation cannot be attributed solely to vortex
chemistry, particularly during summer and early fall at
high latitudes (Perliski et aL, 1989; Proffitt etal., 1992).
However, during the vortex lifetime, changes hi the cor-
relation may be used to bound photochemical ozone loss
in air parcels inside or near the vortex boundary (see Fig-
ure 3-7). This is especially useful inside and outside the
Arctic vortex or outside the Antarctic vortex, where
ozone changes are generally small in comparison to the
natural variability.
3.3 PROCESSING ON AEROSOL SURFACES
3.3.1 Polar Stratospheric Cloud Formation and
Reactivity
As shown in Figure 3-1, reservoir chlorine species
are converted beginning in early winter to form the ac-
tive chlorine species such as molecular chlorine (C12)
and, ultimately, CIO and its dimer C12O2. The conver-
sion is attributed to processing of polar air by surface
reactions involving both HC1. and C1ONO2. The reac-
tions occur on sulfate aerosol particles and polar
stratospheric cloud (PSC) particles that form at the low
temperatures and constituent concentrations characteris-
tic of the interior of the winter vortices. The body of
laboratory data on the formation thermodynamics and
reactivities of these surfaces and the body of atmospher-
ic observations of stratospheric aerosols and their
constituents have continued to grow hi this assessment
period.
The basic features of the ternary condensation of
nitric acid (HNO3), sulfuric acid (H2SO4), and water
(H2O) in the stratosphere are illustrated in Figure 3-8.
With an abundance ratio in the high-latitude lower
stratosphere of these species of approximately 10 ppbv/
1 ppbm/4 ppmv, respectively, H2O is always the pre-
dominant constituent (ppbv = parts per billion by
volume, ppbm = parts per billion by mass, ppmv = parts
per million by volume). For volcanically perturbed con-
ditions, the range of H2SO4 abundance can reach 100
ppbm. Volcanic activity over the past 25 years has in-
creased the average H2SO4 abundance in the stratosphere
to near 5 ppbm. Confidence in the features of the ternary
system has been established in a wide variety of labora-
tory experiments and with the use of thermodynamical
constraints (Molina et al., 1993; Kolb et al., 1994). At
the highest temperatures, liquid aerosol particles com-
posed primarily of H2SO4 and H2O are present in the
lower stratosphere at all latitudes. At lower temperatures
(< 200 K), the H2SO4/H2O liquid increasingly takes up
HNO3. If the particles undergo freezing, HNO3 hydrates
become stable: nitric acid dihydrate (HNO3-2H2O =
NAD) and nitric acid trihydrate (HNO3-3ti2O = NAT).
Liquid or frozen particles that contain: appreciable
HNO3 at temperatures above the frost point are termed
Type I PSC particles. In the absence of HNO3, the
H2SO4/H2O liquid aerosol can freeze to form sulfuric
acid tetrahydrate (SAT) or other sulfate hydrates. Below
3.10
-------
FEE:'9, 1989
EXTERIOR FIT
x EXTERIOR DATA 10) with respect to NAT formation.
Typically, this occurs several degrees above the
frost point in the lower stratosphere (Molina et al.,
1993). The relationship of bulk solution properties
to those of stratospheric aerosols has not been de-
termined (Carslaw et al, 1994).
SAT melts at 220 to 230 K when exposed to partial
pressures of H2O that are typical of the lower
stratosphere (Middlebrqok et al., 1994; Zhang et
al., 1993a).
Both NAT and NAD may play a role in Type I PSC for-
mation when saturation ratios for HNO3 are greater than
unity. However, the phase of Type I PSCs is not certain in
this temperature range, as illustrated in Figure 3-8 (Dye
et al., 1992). Once frozen., SAT within the particles may
remain a solid well above the initial freezing tempera-
3.11
-------
POLAR PROCESSES
Polar Stratospheric Cloud Formation *
Temperature ,
-50 - -60 (°C) -70 -80 -90 ( ,t |
Vin ' ' 220 ' ' 210 (K) 200 190 180
-* 1 1 1 1 | l 1 > 1
HoSO,, / H,O Liquid solutions
1
x^J5\ Water Uptake /T/ty^
^&r \$fy l "1 i
QJS urn Sulfuric acid particles x'TTV
radus Liquid ternary solution /ijihjiji
HN03 / H2O / H2S04~^|| | | jjl
\Jr Decreasing \
H20-4ppmv. VJifitt' J <--'-'
P«50mb -*4_J^^ /\',%%
Frozen HNO3 hydrates ' ^- — •
, Type 1 PSC's
L.
A High
s>
"w
a> • ^ — — ^.
-Sg /
NAT = Nitric Acid Trihydrate ;
NAD = Nitric Acid Dihydrate
ICE
Type II PSC'S] i
•*• Denitrification & dehydration ;
-y Type II PSC's ;
**/.. — i ^-*\i
o ~ Volcano \ /x" ' ""J ' !
§ > ^V-''7 Type 1 PSC's
1 ,„„, Back9round.;><^l^--10days i
i ^ , , , ! ' i — -|^ i i * i
230 220 210 200 190 180 :
Temperature (K)
Figure 3-8. Schematic representation of the ternary condensation system for nitric acid (HNOg), sulfunc
acid (HaSCXj), and water (H2O) over a range of temperatures where growth of aerosols occurs to form Type
I and IIPSC particles in the stratosphere. The changes are represented for nominal abundances of .condens-
ing species in the lower polar stratosphere as indicated. The shading in the horizontal arrows and circular
particle diagrams represents various binary and ternary compositions as indicated. In the lower part, the
chlorine activation rate on PSCs is represented as a function of temperature (adapted from J. E. Dye, private
communication, 1994).
3.72
-------
POLAR PROCESSES
ture. The phase of the particles above the frost point af-
fects the rate of surface conversion for reactive nitrogen
and chlorine species (see Table 3-1). !
The principal heterogeneous reactions of H2SO4/
HNO3/H2O aerosols in Figure 3-8 are listed in fable 3-1.
Reaction rates are considered fast if reaction probabili-
ties are in the range 0.01-0.1 for temperatures and
reactant abundances characteristic of the stratosphere.
Reactions involving H2O are influenced by its ubiqui-
tous presence in aerosol particles throughout the
stratospheric temperature range. Reactions witlti HC1 de-
pend on the solubility of HC1 in an aerosol particle.
Laboratory studies of Reaction (3-3) reveal that the reac-
tion probability depends strongly on relative humidity
and, to a lesser extent, on aerosol composition.: Specifi-
cally, the reaction probabilities for Reaction (3-3) are
similar on Type I PSCs, SAT, and liquid sulfiiric acid
over a wide temperature range at stratospheric relative
humidity (see Figure 3-9) (Molina et al., 1993;,Hanson
and Ravishankara, 1994). The probability for Reaction
(3-5) increases exponentially as the sulfate aerosol di-
lutes with H2O near 200 K and below (Cox et aL, 1994),
as does the probability of Reaction (3-4) due to enhanced
uptake of HC1 (Hanson and Ravishankara, 1993; Luo et
al., 1994b). The increase suggests that Reactions (3-4)
and (3-5) may play a significant role in chlorine-process-
ing when temperatures are low but do not reach Type I or
Type II temperatures (Solomon et al., 1993; Hanson et
al., 1994).
The growth of the ternary aerosol system from sul-
fate aerosols to Type I and H PSCs and the surface
reactions in Table 3-1 combine effectively to release ac-
tive chlorine in the polar regions. In Figure 3-8, the rate
of chlorine activation is qualitatively noted as a function
of temperature. Some activation occurs on background
aerosol particles prior to temperatures decreasing to
Type I formation temperatures. The rate increases signif-
icantly as more surface area containing HNO3 hydrates
and ice forms. Inside the polar vortices, full activation
within an air parcel is estimated to occur within a day or
perhaps a few hours. Thus, the initial activation of the
entire vortex can occur in a matter of days (Newman et
al., 1993). When aerosol particle size and surface area
are increased by volcanic eruptions, the rate of activation
can be significantly enhanced at temperatures above
Type I formation. i
Table 3-1. Rates of heterogeneous reactions
particles.
! i
C1ONO2 + HC1 -» C12 + HNO3
HOC1 + HC1 -» C12 + H2O
C10N02 + H20 -> HOC1 + HN03
N2O5 + H2O -> 2HNO3
N2O5 + HC1 -> ClNO2 + HNO3
Ice
(Type II)
Fast
f(RH)b
H
1 Fast
f(RH)b
i
Fast
Fast :
c
r
on polar stratospheric cloud particles and sulfate aerosol
PSCs
HNO^hvdrat,
(TypeD
Fast
f(RH)b
Fast
f(RH)b
Slow
Slow
c
Sulfate Aerosols
;sa Supercooled
f(wt% H2SO4)b
• f(wt% H2SO4)b
f(wt% H2SO4)b
Fast !
c
Frozen
Fast
f(RH)b
Fast
f(RH)b
Slow
Slow
c
(3-3)
(3-4)
(3-5)
(3-6)
(3-7)
a Nitric acid trihydrate (NAT), nitric acid dihydrate (NAD)
b Rate is function of aerosol wt% H2SO4 or relative humidity (RH).
c Unlikely to be fast, but not well studied
References: Abbatt and Molina, 1992a, b; Chu et al., 1994; Fried et al., 1994; Hanson and Ravishankara, 1991 1992
1994; Kolb etal., 1994;Middlebrook^a/., 1992,1994; Molina et al., 1993; Van Doren etai, 1991; Zhang etal, 1994
3.13
-------
POLAR PROCESSES
10
H2S04 (wt%)
50 55 60
TypelPSC
H2SO4-4H2
Liquid H,SO4/H2O
190
195 200
Temperature (K)
Figure 3-9. Temperature dependence of the reac-
tion probability Y for Reaction (3-3), CIONO2 + HCI,
occurring on surfaces of sulfuric acid tetrahydrate
(H2SO4-4H2O = SAT), nitric acid trihydrate (NAT) or
Type I PSCs, and liquid sulfuric acid and water so-
lutions, H2SO4/H2O.The measurements are made
at a constant partial pressure of water vapor of 0.2
mTorr.Thus, relative humidity increases as temper-
ature decreases. The weight percent (wt%) of the
corresponding sulfuric acid/water solution is indi-
cated on the top axis (adapted from Hanson and
Ravishankara, 1994).
3.3.2 Atmospheric Observations
33.2.1 AEROSOL MEASUREMENTS
The threshold formation and growth of PSC aero-
sol particles have been observed in situ over a wide range
of conditions in both polar regions (Hofmann et al.,
1989, 1990; Hofmann and Deshler, 1989). Satellites
have made global aerosol observations using the extinc-
tion of solar illumination (Osborn et al., 1990). The data
show a persistent increase in aerosol extinction in polar
regions when temperatures fall to the range below where
Type I PSCs are expected (see Figure 3-10) (Poole and
Pitts, 1994). The observations do not allow the phase of
the aerosol to be determined. Lidar measurements in
both polar regions also detect aerosol layers where tem-
peratures reach estimated PSC thresholds (Gobbi and
Adriani, 1993; Browell et al., 1990). Lidar polarization
measurements indicate that both spherical and non-
spherical particles are present in cloud events ((Kent et
al., 1990; Adriani et al., 1994; Toon et al, 1990a). In situ
measurements with balloons show enhancements in the
size distribution for larger particles (Deshler et al.,
1994). Distinct growth begins on some particles near the
threshold for HNO3 hydrates (Dye et al., 1992) and in-
volves all pre-existing particles before decreasing
temperatures reach the frost point (Hofmann et al,
1990). Other measurements near the edges of PSCs have
been made with simultaneous constituent measurements
of reactive nitrogen and water (Kawa et a/.,; 1992b).
These measurements show definitively that the con-
densed phase includes reactive nitrogen species in the
form of HNOs, but that significant aerosol growth above
background values often requires a large supersaturation
of HNOa over the stable hydrate phases.
The systematic formation of aerosol containing
HNO3 is well documented. However, aerosol ;ineasure-
ments of concentration, size, phase, and composition
correlated with the gas phase abundance of the principal
condensing species H2SO4, HNO3, and H2O are critical-
ly absent in observational studies. In ; addition,
observations are not available to constrain important fea-
tures of the nucleation and early growth stages in an
aerosol. Without such measurements, the ability to pre-
dict the distribution of aerosol particles and their
chemical reactivity remains limited.
33.2.2 RELEASE OF ACTIVE CHLORINE i
Active chlorine is produced as a result bf the het-
erogeneous reactions in Table 3-1. The photolysis of the
C12 and hypochlorous acid (HOC1) reaction products
forms Cl, which in turn reacts with ozone to produce
CIO. CIO participates in catalytic reaction cycles that
destroy ozone (see Section 3.4.1). The activation of chlo-
rine over the winter poles has been clearly demonstrated
by in situ and remote measurements of CIO (Anderson et
al., 1991; WMO, 1992; Toohey et al, 1993; deZafra et
al, 1987). The spatial and temporal scale of CIO obser-
vations has been significantly extended by :the UARS
satellite (Waters et al, 1993a, b; Manney et al, 1994b).
Observations are available in both polar regions from
vortex formation to photochemical recovery in the 1991/
92 northern winter and the 1992 southern winter (see
Figure 3-11). In early northern winter (14 December),
infrequent PSCs keep CIO values low inside the vortex
3.14
-------
POLAR PROCESSES
14
May June July Aug Sept Oct Nov
Nov Dec Jan Feb Mar
^
AerosolMeasurement II (SAM II) satelBte data set for the ?ears 1978 to U89. TtebStom panelfSSrt
*3hŁfL?* mL^ b?<*9round aer°s°'r Th* analysis is confined to the inside of the respectiveTvortex
defined by a maximum in the geopotential! :height gradient (Poole and Pitts, 1994).
near 18 km (465 K). In early southern winter :(1 June),
lower temperatures activate sulfate aerosol and begin the
formation of PSCs, increasing CIO accordingly. In areas
of darkness inside the vortex, active chlorine is in the
form of C12, C12O2, or HOCl. When air parcels make
excursions to sunlit lower latitudes within tltie vortex
flow, CIO values increase directly from the photolysis of
C12O2 or indirectly from the photolysis of C12. As the
geographic area and frequency of PSCs continue to in-
crease due to lower temperatures (2 January/11 July),
CIO values and their extent increase substantially in both
vortices. In some areas over both poles, CIO values indi-
cate that essentially all available chlorine is in the active
form. Outside the vortex, little CIO is formed. When,
PSCs cease to exist (17 February), CIO values fall as res-
ervoir chlorine is photochemically reformed In the
southern vortex, high CIO values persist in September
because gas phase HNO3 is suppressed either due to
temperatures below the PSC threshold, which sequester
HNO3 in aerosols, or to the removal of HNO3 in denitri-
fication (see Section 3.3.2.5). This sequence for the
distribution of CIO is qualitatively and quantitatively
consistent with other in situ and remote measurements
(Toohey el al, 1993; Crewell et ai, 1994; Gerber and
Kampfer, 1994).
Features of the CIO temporal and spatial distribu-
tion are consistent with the theoretical determination of
PSC activity associated with low temperatures (Waters
et al., 1993a). The dependence of CIO on PSC activity
and, hence, temperatures within the vortex, is demon-
strated by contrasting CIO observations on 15 February
in late northern winter for two consecutive years (see
Figure 3-12). In 1993, the vortex contained temperatures
below 195 K, significantly lower than found in 1992.
Changes in available chlorine (Cly) cannot explain in-
creased active chlorine found in 1993. Instead, the
changes are attributed to increased formation and reac-
tivity of aerosols at the lower stratospheric temperatures
3.15
-------
POLAR PROCESSES
Lower Stratospheric CIO in the 1991-92 Polar Vortices
1 Jun
1.0
1.5 2.0 ppbv
Figure 3-11. (a) Observations of lower stratospheric CIO in the 1991/92 northern winter (top row) and 1992
southern winter (bottom row) from the Microwave Limb Sounder (MLS) on UARS. The color bar gives CIO
abundances in parts per billion by volume (ppbv) interpolated to the 465 K isentropic surface in the lower
stratosphere (see Figure 3-4 for altitude reference). The irregular white lines are contours of potential vortic-
ity (2.5 and 3.0 x 10-5 K m2kg-1 s*1) indicating the polar vortex boundary. Measurements poleward of the black
contour were made for solar zenith angles greater than 91° (in darkness or edge of daylight). The edge of
polar night is shown by the thin white circle concentric with the pole. No measurements are available in the
white area poleward of 80° latitude. The green contours indicate temperatures of 190 K (inner) and 195 K
(outer) (Waters et a/., 1993a, b).
II JANUARY 92 I2U
PPi.
1951
1811
1672
1533
1393
1254
1115
975
836
696
553'
418
278
139
0000
Figure 3-11. (b) CIO distribution calculated with a
three-dimensional chemistry and transport model
of the stratosphere. The CIO field for 11 January
1992 on the 465 K potential temperature surface
was mapped at all locations for local noon to
achieve better temporal coincidence with UARS
satellite measurements in (a) (note discontinuity on
both sides of date line) (Lefevre era/., 1994).
3.16
-------
POLAR PROCESSES
2.0 ppbv
satellite instrument and NMC temperatures in the
temperature in the Northern Hemisphere on 15 February 1992 and
sn«»>. NO measurements are available in the white area poleward of
in 1993 as demonstrated in a 3-D model simulation
(Chipperfield, 1994a). The 15 February data sets are rep-
resentative of the systematic differences in CIO and
temperature between 1992 and 1993 and, hence, also
demonstrate interannual variability characteristic of the
Northern Hemisphere vortex (see Section 3.4.2).
Observed changes in CIO are also consistent with
changes within the reactive nitrogen reservoir. Activa-
tion of a large fraction of available chlorine to 0O sets
an upper limit on NOX (= NO + NO2) that can bs present
in the NOy reservoir (see Equation (3-2)) to form
C1ONO2. From in situ observations near the vortex edge,
nitric oxide (NO) is suppressed wherever Clip is en-
hanced (Toohey et al., 1993; Kawa et al., 199^aj. The
same reactions that activate; chlorine (see Table 3-1) re-
duce NO and NOX through the formation of HNO3, a
longer-lived species. In addition, NOX is reduced indi-
rectly through denitrification, the irreversible removal of
NOy (see Section 3.3.2.5). Column measurements of ni-
trogen dioxide (NO2) and HNO3 arc generally consistent
with expected changes in NOy partitioning (Solomon
and Keys, 1992; Keys et al., 1993; Wanner et al., 1990a;
Koike etal., 1994).
Activation of chlorine is also indicated by increas-
es in OC1O, formed in the reaction CIO + BrO (Solomon
et al., 1989; Tung et al., 1986; Sanders et al., 1993).
OC1O has been observed in both vortices, with the larg-
est column abundances found in the Antarctic vortex
3.17
-------
POLAR PROCESSES
JAN. 20,1992
FEB. 13,1992
40 60
LATITUDE (°N)
8020 40 60 80
LATITUDE CM)
Rgure 3-13. Aircraft data from 20 January and 13 February in the 1992 Arctic winter. Values for the ichanges
in HCI and CIONO2 are noted as AHCI and ACIOMO2, respectively, where negatove values indicate cgptatnn.
Values of AHCI are determined using the observed correlation with NŁ> as a "f^^^^^SSS^
are derived in three steps. First, total inorganic chlorine is estimated along the flightfrackfrom the correlation
oftotaTorganic chlorine'with N2O (see Figure 3-6) (Kawa etal., 1992a). SecondJOOfim assumed to be
the balance in the inorganic chlorine reservoir after account is made for measured ^l, CIO, and calculated
CI202. Third, changes in CIONO2 from that calculated using the reference vatae of HCI are designated as
ACION02 The dotted vertical line indicates the vortex edge determined from the maximum zonal wind mea-
sured on board the aircraft (Webster et al., 1993a).
(Schiller et al, 1990; Sanders et al., 1993; Brandtjen et
al, 1994). The abundances are broadly consistent with
expectations from model simulations. In the Arctic and
Antarctic vortex regions following the eruption of Mt.
Pinatubo, increases in OC1O were observed before PSC
temperatures were noted in the lower stratosphere
(Solomon et al., 1993; Perner et al., 1994). Such mea-
surements are a sensitive indicator of changes in active
chlorine, especially for the low Sun conditions charac-
teristic of high-latitude winter. The activation, attributed
to enhancements in the rate of Reaction (3-5) on volca-
nic sulfate aerosols, implies additional ozone destruction
at high latitudes during periods of enhanced aerosol.
3.3.2.3 CHANGES DI RESERVOIR CHLOIUNE |
The selective conversion of the inactive chlorine
reservoirs HCI and ClONOz in surface reactions occur-
ring in the polar vortices is a fundamental aspect of the
ozone depletion process depicted in Rgure 3-1. In previ-
ous assessments, polar observations of these reservoir
species were limited to remote soundings from the
ground and aircraft in situ measurements. However, the
3.18
-------
POLAR PROCESSES
general feature of the winter conversion of the reservoirs
and their subsequent formation in spring can be found in
these observations. New observations include simulta-
neous in situ measurements of HC1 and CIO in the Arctic
region (see Figure 3-13) (Webster et al., 1993a, b). In
addition, C1ONO2 is deduced as the residual in the Cly
reservoir after account is made for HC1, CIO, and esti-
mated C12O2 (see Figure 3-6). Near-complete: removal of
HC1 was observed in some air masses at 20 km in the
vortex. Changes at the vortex edge show leases in the
reservoir species that correlate with increased CIO.
Losses in the reservoir species are comparable and
equate well with the sum of observed CIO and'calculated
C12O2, indicating stoichiometric conversion qf HCI and
C10N02 in Reaction (3-3). Before PSC processing, in
situ HC1 values are somewhat less than those of estimat-
ed C1ONO2 at mid- to high northern latitudes, conflicting
with standard photochemical models which find HC1 to
be in excess. At lower latitudes away from PSC process-
ing, the sum of the inorganic and organic chlorine
species is constant throughout the lower and upper
stratosphere, indicating that chlorine is conserved in the
conversion of chlorine to inorganic forms (Zander et al
1992). i
In remote ground-based measurements, the col-
umn abundance of HG1 over northern Sweden was
observed throughout midwinter 1991/92 (Bell et al.,
1994). The anticorrelation with column CIO clearly
shows the conversion of HC1 to active forms (see Figure
3-14). Earlier column measurements froni aircraft
showed the complete conversion of HC1 and C1ONO2
deep inside the northern vortex in January and early Feb-
ruary 1989 (Toon et al., 1992). The measurements are
consistent with complete removal of HC1 up 1:0 27 km.
Profile measurements of C1ONO2 show that the; midwin-
ter depletion extends throughout a broad vertical region
in the Arctic stratosphere (see Figure 3-15) (von Clar-
mannetal., 1993). > ;
The UARS remote measurements of HCI and .
C1ONO2 significantly extend the spatial and temporal
scale of previous observations. Inside the edge of the
Antarctic vortex in late September, significant depletion
of HC1 is found around a latitude circle near die vortex
edge (see Figure 3-16) when HC1 values are compared
with those of the long-lived tracer species CHLi :and HF.
These data sets confirm the large-scale depletion of HC1
in low-temperature regions in the Antarctic vortex. Sat-
o
o
c
I
"o
o
-40 -20 I) 20 40 60
Day relative to Jan. 1,1992
80
Figure 3-14. HCI and CIO column abundances
over Are, Sweden (63.4°N) during the EASOE
campaign in 1992. The HCI column is measured by
ground-based, infrared solar absorption spectros-
copy. The CIO column is the amount above 100 mb
(-16 km) at the same location as measured by the
UARS MLS satellite instrument (Bell et al., 1994)
35
325-
320'
15-
1.00 1.50 2.00
MIXING RAT:'0 C1ONO2 (PPBV)
2.50
Figure 3-15. Retrieved Michelson Interferometric
Passive Atmosphere Sounder-B (MIPAS-B)
CIONO2 profiles from balloon flights near Kiruna
Sweden (68°N) during the EASOE campaign in
1992. The peak of the 13 January mixing ratio pro-
file (solid curve) is at a higher altitude than the peak
of the 14/15 March profile (dashed curve). Similar
values are obtained above 25 km, but large differ-
ences between the profiles appear in the lower
stratosphere (von Clarmann et al., 1993).
3.19
-------
POLAR PROCESSES
ellite measurements also show low abundances of
C1ON02 and HNOs inside the Antarctic vortex as early
as mid-June (early winter), suggesting substantial PSC
processing (see Figure 3-17) (Santee et al, 1994; Roche
et al, 1993b, 1994). In addition, a region of high
ClONOa surrounding the vortex is noted in late winter.
C1ONO2 values will be enhanced in areas where pro-
cessing is limited or infrequent, and where sunlight is
available to produce NOa in the photolysis of HNOs and,
thereby, reform C1ONO2 in advance of HC1 (see Figures
3-1 and 3-17 and Section 3.4). This region, termed the
"collar" region as first noted in remote soundings from
aircraft (Toon et al, 1989a), is also identifiable in esti-
mates of ClONOi based on in situ observations near the
vortex edge (see Figure 3-13). In late winter, the "collar"
region extends into the sunlit vortex, as noted in Arctic
soundings which show recovery of the vertical profile of
C10NO2 (see Figure 3-15). Although the early UARS
observations are made in years of high volcanic aerosol
loading, these observations and estimates of C1ONC>2
add confidence to the role reservoir species play in the
activation of chlorine.
33.2.4 ACTIVE BROMINE
Although the bromine source gases in the strato-
sphere are less than one percent the size of chlorine
source gases, active bromine in the form of BrO plays an
important role in photochemical ozone destruction. In
situ and remote observations establish the abundance of
BrO in the range of 4 to 10 parts per trillion by volume
(pptv), corresponding to approximately half of total
available bromine (Toohey et al, 1990; Wanner et al,
1990b; Carroll et al, 1989). Observations of high levels
of OC1O also confirm the presence of BrO since OC1O is
formed in the reaction CIO + BrO (see Section 3.3.2.2)
(Salawitch et al, 1988). Since gas phase photochemistry
rapidly couples BrO with the inactive reservo'irs
(BrONO2, HBr), BrO is readily available to participate
in catalytic reaction cycles as described in detail in
Chapter 10. Calculations based on observed abundances
estimate that, depending on temperature, between 25
and 50 percent of ozone loss in the polar vortices is due
to the CIO + BrO catalytic cycle (see Section 3.4,1)
(Salawitch et al, 1993). The fractional contribution to
total ozone loss is estimated to be greater in the Arctic,
where higher temperatures reduce the effectiveness of
the CIO + CIO cycle.
33.2.5 DENTTRIFICATION AND DEHYDRATION i
PSC particles formed, at low temperatures inside
the polar vortices become large enough to sediment ap-
preciable distances in the lower stratosphere over time
periods much shorter than the winter season. As a result,
up to 90 percent of available reactive nitrogen has been
observed to be irreversibly removed from air parcels
sampled in situ in both polar vortices (Fahey et al,
1990a, b; Schlager and Arnold, 1990; Koadoetal, 1992,
1994a; Arnold et al., 1992). This irreversible .removal
defines denitrification. Removal of reactive nitrogen in
the form of HNOs helps sustain active chlorine'in an air
parcel (see Section 3.4.3). Denitrification is quantified
by using the NOy/N2O correlation observed at high lati-
tudes in the absence of PSCs (see Section 3.2). In situ
measurements indicate that the temporal and spatial ex-
tent of denitrification is substantially greater in the
Antarctic, consistent with observed lower temperatures
(see Figure 3-3). In the Arctic, at altitudes below particle
formation, the evaporation of sedimenting aerosols en-
hances NOy values (Hiibler et al, 1990). Another
example of this redistribution is provided by the compar-
ison of HNOs profile measurements and estimates of the
unperturbed NOy reservoir from the N2O tracer correla-
tion (Murcray et al, 1994; Bauer et al, 1994).
Satellite observations of HNOs at Wgh latitudes
now confirm the temporal and spatial scale of HNOs re-
moval and the contrast between the two polar regions
(Santee et al, 1994; Roche et al, 1994). In the;Southern
Hemisphere (see Figure 3-17), removal or sequestering
of HNOs in aerosol particles is observed in late fall. Se-
questering occurs when HNOsIS reversibly incorporated
into particles that do not undergo sedimentation. By
midwinter, HNO3 values less than 0.5 ppbv fill a large
fraction of the vortex where CIO values are above 1 ppbv
in the sunlit portion (see Figure 3-1 la). Values of HNOs
comparable to those expected from tracer correlations
with NOy (about 10 ppbv) surround the vortex at lower
latitudes. By late winter, after PSC temperatures cease to
occur, low HNOs values persist in the vortex, indicating
denitrification. In the Northern Hemisphere (see Figure
3-17), higher average temperatures than in thq Antarctic
(see Figure 3-3) generally limit the removal or sequester-
ing of HNOs. An example is the local minimum in
HNOs near Iceland in observations on 22 February 1993
(see Figure 3-17). Thus, sequestering and removal of
is a predominant feature of the Antarctic vortex
5.20
-------
POLAR PROCESSES
ISO
I-ongltude
th? Q?,2im w HALOEsatellie, observations of HCI (top) and CH4 (bottom) on 27 September 1992 in
the Southern Hemisphere at 66°S latitude. The data are from sunrise scans analyzed with the version-16
' The Prfssu";e ran9e corresponds to altitudes between 16 and 30 km. At low and high longitude
, the spahal gradients and low absolute values of HCI relative to CH4 indicate depletion of HCI (adapt-
3.21
-------
POLAR PROCESSES
Lower Stratospheric HNO3 in the 1992-93 Polar Vortices
28 Apr
12.5ppbv ,
Figure 3-17a. Observations of lower stratospheric HNO3 in the 1992/93 northern winter (top row) and 1992
southern winter (bottom row) from the UARS MLS satellite instrument. The color bar gives HNO3 abundanc-
es in ppbv interpolated to the 465 K isentropic surface (see Figure 3-4 for altitude reference). The; irregular
white lines are contours of potential vorticity (2.5 and 3.0 x 10-5 K m2kg-is-i) indicating the polar vortex
boundary. No measurements are available in the white area poleward of 80° latitude. Black contours indicate
temperatures of 190 K (inner) and 195 K (outer). The days were chosen to illustrate periods (1) before
temperatures fell low enough for PSC formation (26 October and 3 December in the north, 28 April in the
south), (2) when temperatures were low enough for PSC formation (22 February in the north, 2 June and 17
August in the south), and (3) after temperatures had increased above the PSC threshold (14 March in the
north, 1 November in the south) (Santee et al., 1994). ;
for much of the winter, whereas removal in the Arctic is
much less intense and more localized.
The irreversible removal of water, or dehydration,
accompanies denitrification in the Antarctic but not in
the Arctic (Fahey et al., 1990b). Dehydration requires
the sedimentation of Type II PSCs in order to effect the
removal of 50 percent of available water as observed in
the Antarctic region. Water vapor profiles in the winter
vortices show interhemispheric differences, with lower
values in the Antarctic. The differences reflect the more
frequent occurrence of low temperatures in the Antarctic
that facilitate Type II PSC formation (Kelly et al., 1989;
!
1990). Balloon and satellite observations of H2O and
CH4 in the Southern Hemisphere confirm extensive de-
hydration in the vortex and its near environment
(Hofmann and Oltmans, 1992; Tucker al., 1993; Rind et
al., 1993). Because H2O and molecular hydrogen (H2)
are produced in the oxidation of CFLj in the stratosphere
and mesosphere, changes in the quantity [2CH4 + H2O]
are a more sensitive indicator of dehydration than chang-
es in H2O alone (see Figure 3-18 and Section 3.5.2). The
large spatial and temporal scales of dehydration ob-
served over the Antarctic are not observed anywhere else
in the atmosphere (Tuck et al., 1993). The combined re-
3.22
-------
OCT-25-92
POLAR PROCESSES
MAR-14-93
OCEAN LAND
CLON02 (ppbv)
th'f IM? ,H Observations of lower stratospheric CIONO2 in the 1992/93 northern winter (top row) and in
the 993 southern winter (bottom row) from the UARS Cryogenic Limb Array Etalon Spectrometer (CLAES)
surface S^S!^ 1!^°°^ ^T C"P^°2 abundances in P&v interpolated to the 465 K isentropic
surface (see Figure 3-4 for altitude reference). The instrument does not see poleward of 80° latitude The
WH ?n nJ° "'"f ate Peri°ds (1) before temPeratures fell low enough for PSC formation (25 Octo
/oo°!mK '" ^J^' 28 APril in the s°"th). (2) when temperatures were low enough for PSC
( K lUa^ Lh6 n0rth' 12 Jurie and 17 Au9ust in the south), and (3) after temperatures had
ea 1994^°V6 ^ °4 MarCh '" ^ n°^' 2 November in the s°u h) (adapted from Roche
moval of H2O and HNO3 reduces the available condens-
able material for the formation of PSCs and, hence,
lowers the minimum formation temperature. This fea-
ture is most notable in the Antarctic between the early
and late winter periods (see Figure 3-10) (Poble and
Pitts, 1994). ' '\'. .
Despite extensive observational evidence
for de-
hydration and denitrification, ail of the underlying
microphysical mechanisms and atmospheric conditions
that control particle formation and sedimentation have
not been completely confirmed in observational or labo-
ratory studies. The overall process is complicated by the
potential roles of air parcel cooling rates and barriers to
nucleation of aerosol particles (Toon et ai, 1989b;
Wofsy et a/., I990a, b). The sedimentation process is
generally better understood (Miiller and Peter, 1992).
The combined in situ data from both vortices show that
intense denitrification (about 90-percent removal) oc-
3.23
-------
POLAR PROCESSES
HALOE
Ł
-64 -48
-32
-16 0 16
Latitude
curs with and without intense dehydration (about 50-per-
cent removal). However, intense dehydration has not
been observed without intense denitrification (Fahey et
al., 1990b). Observations do not preclude independent
processes for intense denitrification and dehydration, as
discussed in theoretical studies (Toon et al, 1990b;
Salawitch et a/.', 1989; Wofsy et al, 1990a, b). Water va-
por plays a role in denitrification due to its presence in
condensed hydrates of HNO3 (see Figure 3-8). However,
since gas phase abundances of water vapor exceed those
of HNO3 by large factors, changes in water vapor are
negligible as denitrification occurs. In addition, the anal-
ysis of the export of denitrified and dehydrated air from
the Antarctic vortex reveals a quantitative inconsistency
that may be explained by independent removal processes
(Tuck etal, 1994).
3.3.3 Role of Mt. Pinatubo Aerosol
Volcanic eruptions are potentially itaportant
sources of sulfur dioxide (SO2), HC1, and H2O for the
lower stratosphere (GRL, 1992). The eruption of Mt. Pi-
natubo in the Philippines in June 1991 is a recent large
event that affected stratospheric measurements during
this assessment period. The injection of SOi into the
lower stratosphere in the tropics exceeded that of the El
Chich6n eruption in 1982 by three times (McCormick
and Veiga, 1992). The SO2 cloud rapidly forms H2SO4,
which augments the formation and growth pf sulfate
aerosol particles in the stratosphere (Wilson et,al, 1993;
Borrmann etal., 1993). Figure 3-19 shows the evolution
of aerosol extinction from near-background conditions
before the 1991 eruption to one year later. Surface area
5.24
-------
POLAR PROCESSES
SAGE II Aerosol Observations
(b)
4/16 4/26 4/30 5/3 5/7 5/12 5/19
7/15 7/10. 7/6 7/2 6/26 6/15
S
-*-J
• «—<
60S 45S 30S 15S 0 15N30N45N60N
H++ + +++ + + + + ++++ + + 4.JH-H II Illllllllllll
9/23 9/20 9/17 9/14 9/10 9/3 8/21
60S45S30S15S 0 15N30N45N60N
5/5 4/27*4/23 4/20 * 4/16* *4/10 3/30
Total Extinction Ratio
>100
Figure 3-19. Latitude-altitude cross sections of the Stratospheric Aerosol and Gas Experiment II (SAGE II)
1-u.m extinction ratio measurements that show the effect of the eruption of Mt. F'inatubo in June 1991 on
aerosol abundance in the lower atmosphere. The specific dates of observation are indicated with crosses
below each panel for the periods: (a) 15 April to 25 May 1991 (pre-eruption); (b) 14 June to 26 July 1991
(early austral winter); (d) 20 August to 30 September 1991 (late austral winter); and (I) 29 March to 9 May
1992 (full dispersal). No data were used 2 km below the tropopause (blacked out). Small triangles indicate
truncation altitude for the SAGE II data. Lidcir data were used below this altitude. Isentropes (constant poten-
tial temperature in K) appear as white contour lines (adapted from Trepte et a/., 1993).
3.25
-------
POLAR PROCESSES
values are increased by factors up to 100 over much of
both hemispheres within the year. Since the residual cir-
culation in the stratosphere is upward in the tropics and
poleward and downward at higher latitudes, volcanic
aerosol is transported to the polar regions, where it is in-
corporated into the polar vortices. Mt. Pinatubo aerosol
did not appear in the Antarctic vortex during th& austral
winter of 1991 (see Figure 3-19d) but was present at the
South Pole following vortex breakup (Cacciani et at,
1993) and was present in the vortex during the following
austral winter (Deshler et al, 1994). In the 1991/92 bo-
real winter, some enhanced levels were observed in the
vortex (Wilson et al, 1993). The decay of volcanic aero-
sol hi the lower stratosphere occurs with a time constant
that varies with latitude and particle size, but generally
averages about one year for an integral parameter such as
particle surface area.
Although the emission of HC1 from volcanoes am
exceed the annual anthropogenic emissions of chlorine
to the atmosphere, emitted HCl is largely removed in the
troposphere before appreciable amounts can enter the
stratosphere. For the Mt. Pinatubo eruption, column
measurements of HCl before and after the eruption con-
firmed that the increase of HCl in the stratosphere was
negligible (Wallace and Livingston, 1992; Mankin et al.,
1992). The removal of HCl and H2O is expected to result
from scavenging on liquid water droplets formed in the
volcanic plume (Tabazadeh and Turco, 1993). These and
other dissolution processes reduce HCl abundances by
several orders of magnitude, thereby limiting the avail-
ability of HCl for transport to the stratosphere. In
contrast, only 0.5 to 1.5 percent of SC>2.in the plume is
removed by dissolution, thereby facilitating the transport
of SOa to the stratosphere, where it is oxidized to form
H2S04.
The principal consequence of volcanic eruptions
for the stratosphere is the enhancement of sulfate aerosol
over the globe, thereby affecting the rates of heterojge-
neous reactions that convert reactive chlorine and
nitrogen species (see Table 3-1). In midlatitudes, volca-
nic aerosol drives the conversion of dinitrogen pentoxide
(N2O5) to HNO3 (see Reaction (3-6)) to saturation
(Prather, 1992; Fahey et al., 1993; Koike et al., 1994).
Volcanic aerosol in the Antarctic is associated with an
increased frequency of PSCs and a reduction in large
particle formation within the cloud (Deshler et al.,
1994). Aerosol surface area densities found in the vortex
following the eruption of ML Pinatubo are comparable
to those in a Type IPSC formed in the absence of volca-
nic influence. Thus, increased rates of chlorine activation
above Type I PSC temperatures are expected in ,1992 and
1993 (see Figure 3-8). However, Type E PSC surface ar-
eas are still predominant at lower temperatures. In the
center of the ozone depletion region (14-18 km), chlo-
rine is fully activated in both vortices in most years and
the presence of volcanic aerosol here will not increase
the intensity of chlorine activation. However, in the 10 to
14 km region and the region above 18 Ion where chlorine
is usually not fully activated, additional surface; area pro-
vided by volcanic aerosol can result in increased
chemical processing. Furthermore, because the sulfate
aerosol is active at temperatures above the PSC forma-
tion threshold, the spatial and temporal extent of
chlorine activation will be increased, especially in the
vortex edge region. Chlorine activation there i has been
observed to be greater than that in non-volcanic periods
and is associated with enhanced ozone loss (Solomon et
al, 1993; Hofmann et al, 1992; Hofmann and Oltmans,
1993). Since the scale of this near-vortex regibn can be
comparable to or larger than the vortex ulterior, en-
hanced processing outside the vortex edge may be
especially important in ozone balance throughout the
hemispheres (see Chapter 4).
3.3.4 Model Simulations
Model simulations of the formation of P$Cs in the
vortex require detailed knowledge of both the thermody-
namics inherent in Figure 3-8 and of the nucleation and
growth features of the various aerosol particles. Several
studies have met with success in simulating the general
features of a PSC (Peter el al., 1992; Drdla and Turco,
1991; Toon et al, 1989b, 1990b). However, significant
uncertainties remain in the prediction of PSC | formation
conditions and other characteristics (Dye et.ai, 1992;
Kawa etal, 1992b). Specifically, the threshold tempera-
ture for the appearance of Type I aerosols is well below
the saturation temperature in Arctic observations. Vari-
ous explanations are possible, but remain unconfirmed at
present. In addition, details of the PSC sedimentation
process causing denitrification and dehydration are un-
certain. Specifically, uncertainty in the coupling of
denitrification and dehydration affects model simula-
tions of PSC activity as well as ozone depletion.
3.26
-------
POLAR PROCESSES
As a result of these uncertainties, model simula-
tions adopt a simplified parameterization of PSC
formation and sedimentation processes. These studies
confirm the effectiveness of the heterogeneous reactions
listed in Table 3-1 for the conversion of inactive to active
chlorine (Brasseur and Granier, 1992; Cariolle et al.,
1990; Eckman et al., 1993; Lutman et aL, 1994a; Chip-
perfield et al., 1994b, c; Newman et al, 1993; Lefevre et
al., 1994). PSCs formed in localized low-temperature
regions in the strong zonal flow of the vortex can fully
activate the vortex in the lower stratosphere in a matter
of days. Thus, predicting intense activation of chlorine in
the vortex seems not to require detailed knowledge of
PSC events. Using a 3-D transport and chemistry model,
a comparison of modeled and satellite observations of
CIO in Arctic winter shows excellent agreement (see
Figure 3-lib). In a similar study, the comparison reveals
differences in the dynamic structures that force PSC ac-
tivity at high latitudes (Douglass et al., 1993). In
addition, in situ arid satellite observations of vortex CIO
over a wide range of values can be simulated with trajec-
tory models that account for exposure.to PSCs as well as
the recovery of inactive chlorine in sunlight following a
PSC event (Lutman et al., 1994a, b; Schoeberl et al.,
1993a, b; Toohey et al., 1993).
Other model simulations are used to evaluate
ground-based measurements of OC1O in Antarctica that
were made when volcanic aerosol was present (Solomon
et al., 1993; Hanson et aL, 1994). In matching the ob-
served activation of chlorine, the simulations demonstrate
the importance of regions that have temperattures close
to, but above those required for PSC formation and mod-
est solar illumination. In these regions,': chlorine
activation on sulfate aerosols (see Table 3-1) effectively
competes with the photolysis of HNOs which (C1O)2 + M
(C10)2 + hv -> C100 + Cl
ClOO->ClfO2
2(C1 + O3 ->C1O + O2)
Net: 2O3 -> 3O2
CIO + BrO-!>Cl + Br + O
(3-8)
(3 - 8a)
(3-9)
(3-9a)
-3; BrCl + O2
BrCl + hv-3 Br + Cl
Br + O
C1 + O3 H>C1O + O2
Net: 2O3 -» 3O2
->C1+O2
C1 + O3 ->CIO + O2
Net:,O + O3 ->2O2
(3-10)
where reaction steps (3-8a) and (3-9a) do not result in
ozone destruction and where the cycles are listed in or-
der of importance for ozone: destruction inside the vortex
(Salawitch et al., 1993; Lutman et aL, 1994b; Molina
and Molina, 1987; McElroy et al., 1986; Solomon et aL,
1986; Tung et al., 1986). The rates of the homogeneous
photochemical reactions involved in chlorine catalytic
cycles follow from a wide variety of laboratory investi-
gations and are generally well understood (JPL, 1992).
However, studies continue to improve the precision of
earlier results. For example, the temperature dependence
of HNOs and C1ONO2 photolysis cross sections have
been remeasured. Those of ClONOi were found to be in
good agreement with previous recommendations for
temperatures characteristic of the lower stratosphere,
whereas those of HNOs were reduced somewhat at
stratospheric temperatures (Burkholder et al., 1994a, b).
3.27
-------
POLAR PROCESSES
o
O 3
o
CL
o
DC
O
0_
Cl* (ppb)
Ozone Loss (% / day)
Flight Days
-20
0 20 40 60
DAY OF THE YEAR (1/1/92 = 1)
80
Rgure 3-20. Calculation of CT (= CIO + 2CI2O5>) (top) and the 24-hour mean loss rate for ozone on the 470
K Dotential temperature surface (bottom) from a full diurnal photochemical model calculation. Bothare plrt-
Sd^SmSSfSpSillal vorticity (PV), in units of (10-5 K m2kg-is-i), and day of the year for, the Arct.c
vortex in 1991/92 (day 1 = 1 January 1992). The approximate mean latitudes of parcels with PV of 2 and 4
are 40° and 65°N, respectively, for this period (Salawitch et al., 1993).
A cycle involving ClONOa has also been recog-
nized to contribute to ozone depletion. After PSCs are no
longer present and the recovery period begins (see Sec-
tion 3.4.3), active chlorine forms elevated amounts of
C1ONO2. The production of Cl from ClONC^ photolysis
initiates a catalytic cycle similar to Reaction (3-8)
CToumi et al, 1993; Minton et al., 1992). In full models
of ozone destruction, C1ONO2 photolysis and associated
reactions are typically included, but the associated ozone
loss is often not distinguished from the primary catalytic
loss cycles represented in Reactions (3-8,3-9, and 3-10).
Quantitative evaluation of ozone destruction rates,
constrained by observed CIO, provides reasonable
agreement with the measured decay of ozone over the
Antarctic where ozone loss is rapid and the vortex is as-
sumed to be isolated over the measurement ;period (see
Chapter 1) (Anderson et al., 1989; Solomon, 1990;
Anderson et al., 1991). Loss rates of about one percent
per day are found there when chlorine is fully activated
in sunlight. Recent model calculations for, the Arctic
show a detailed relationship between active chlorine
abundance and ozone loss rates in the lower stratosphere
in early 1992 (see Figure 3-20) (Salawitch et al., 1993).
The model represents vortex photochemistry more com-
prehensively than in previous studies because of the use
of extensive in situ and satellite observations of reactive
3.28
-------
POLAR PROCESSES
and trace species, meteorological analyses, and recent
laboratory results for gas phase and heterogeneous reac-
tions. After parameterization, ozone loss' rates are
estimated using a full-diurnal photochemical calcula-
tion. The maximum loss rates are similar to the
Antarctic, but the rates are sustained for a shorter period,
resulting in smaller total losses. For a given chlorine lev-
el, loss rates on an isentropic surface are greater at lower
latitude (or lower PV) values where solar illumination is
greater. Cumulative losses of 15 to 20 percent in the Arc-
tic implied by Figure 3-20 are corroborated by estimates
made using in situ observations of ozone (Browell et al,
1993) and changes in the relationship between ozone
and the long-lived tracer N2O (Proffitt et al., 1993) (see
Section 3.2). Corroboration is also provided ;by model
simulations that utilize the extensive ozonespnde data
available in the 1991/92 northern winter. The data are
analyzed by using estimates of descent of polar air over
winter months and by using trajectories to identify air
parcels sampled twice in sonde measurements;' separated
in time and space (Lucic et al., 1994; von der Gathen et
al., 1994). These recent results increase confidence in
earlier estimates of ozone loss in the Arctic vortex
(Schoeberlefa/., 1990; McKenna era/., 1990; Salawitch
etal, 1990). '
With extensive observations of CIO and ozone, the
UARS satellite substantially increases the evidence that
ozone loss occurs in both polar regions and that reactions
involving CIO are the cause of this depletion (Waters et
al., 1993a; Manney et al., 1994b). Total columiii amounts
of CIO correlate well with regions depleted iin column
ozone in two consecutive years in the Antarctic! (see Fig-
ure 3-21). In addition, this correlation has been observed
in mid-August in the Antarctic (Waters etal., :L993b), in
agreement with the interpretation of in situ observations
(Proffitt et al., 1989a). In the Arctic, variability in col-
umn ozone abundances tends to obscure the smaller
Arctic losses. However, averages of CIO and iozone in
the Arctic show a negative correlation during peak CIO
values, with ozone loss rates in reasonable agreement
with calculations. Satellite N2O observations or PV
analyses are used to account for ozone changes resulting
from the transport of ozone. These results suggest that
conclusions and interpretation derived from the highly
localized in situ and ground-based data sets have rele-
vance on the vortex scale.
, fif,. A new perspective of ozone loss comes from satel-
lite observations of late-winter changes in ozone
averaged around PV contours (see Figure 3-22) (Man-
ney et al, 1993, 1994b). This approach can detect
significant changes in the 3-D distribution of ozone
without a priori assumptions about the specific role of
photochemistry or transport of ozone. With PV generally
increasing poleward, poleward transport of ozone-rich
air at upper levels and ozone loss at lower levels are both
evident in the Antarctic vortex region in each year. In
contrast, ozone increases are expected to extend to the
lowest potential temperatures in these regions without
localized, in situ photochemical loss. In the Arctic,
ozone increases are found in both 1992 and 1993, but
significant ozone loss in the Febraary-to-March time pe-
riod is found only in 1993 in the lower stratosphere. The
loss is consistent with enhanced CIO values in 1993 that
resulted from more extensive low temperatures (see Fig-
ures 3-11 and 3-12).
3.4.2 Variability
Perhaps the greatest difficulty in increasing the ac-
curacy of predictions of ozone loss in polar regions is the
interannual and intra-annual variability of the conditions
that determine loss rates. Large variability in meteoro-
logical and photochemical parameters featured in Figure
3-1 increases the difficulty of the interpretation of limit-
ed data sets and reduces their value for predicting future
changes in ozone. Variability largely follows from the
fluid mechanical features of the vortex and its environ-
ment, and the stochastic nature of the forces that act to
change the vortex and its environment. Of greatest con-
cern are changes in the spatial and temporal extent of
low temperatures and the duration of the vortex into the
spring season (Austin et at.., 1992; Austin and Butchart,
1994; Salawitch et al., 1993). Lower temperatures pro-
mote activation of chlorine, and a long-lived vortex
promotes the photochemical destruction of ozone by ac-
tive chlorine. The northern winters of 1991/92 and 1992/
93 present a striking example of interannual variability
in CIO and ozone (see Figures 3-12 and 3-22) (Larsen et
al., 1994). In general, variability in the Arctic vortex is
greater than in the Antarctic, particularly for minimum
temperatures (see Figure 3-3). Because the formation of
PSCs requires temperatures below a certain threshold,
fluctuations of a few degrees will substantially change
3.29
-------
POLAR PROCESSES
10 15 20
10" molecules/m2
2:5
140
180 220 260 300
DU above lOOhPa
Fiqure 3-21. Observations of column abundances of CIO (10™ molecules m-2) and ozone (Dobson units)
above 100 hPa (about 16 km) in the Antarctic in September 1991 and 1992 from the UARS MLS satellite
instrument (Waters et al., 1993a).
the extent of processing inside the vortex and the extent
of denitrification as sunlight returns to the vortex in
spring. Ozone destruction rates in the late vortex strong-
ly depend on the extent of denitrification (see Figure
3-23) (Brune et al., 1991; Salawitch et al., 1993). Re-
duced values of reactive nitrogen slow the formation of
the C1ONO2 reservoir and thereby maintain active chlo-
rine levels as sunlight returns to high latitudes.
The variability in both polar regions follows from
wave activity near the vortex and the interaction of
waves with tropospheric weather systems. These wave
perturbations can change the chemical evolution of the
vortex through the associated temperature changes (Far-
man et al., 1994; Gobbi and Adriani, 1993; Rood et al..
1992) or the transport into and out of the vortex, espe-
cially for the weaker Arctic vortex (Dahlberg and
Bowman, 1994; Manney et al., 1994c). Regions cooled
to PSC temperatures can process a large fraction of vor-
tex air in a relatively short period of time, Contributing
significantly to the total amount of vortex processing
(MacKenzie et al, 1994; Newman et al., 1993; Lefcvre
et al, 1994). When these low-temperature regions are
near the vortex edge, the resultant processing may influ-
ence midlatitude ozone destruction (see jChapter 4).
Wave activity also distorts the vortex from a symmetric
polar flow, thereby transporting processed air into sun-
light at lower latitudes. Because ozone loss rates
increase substantially in sunlight when chlorine is acti-
3.30
-------
POLAR PROCESSES
840
14Augto 18 Sep 1992SH
Feb to 21 Mar 1992 NH
9 Anglo 13 Sep 1993 SH. _ 10Feb to 17Mar 1993 NH
465 —
-3.4
-2.6 -l.B -1.0'
POTENTIAL VORTICITY
-0.2
0.0
0.8 1.6 2.4
POTENTIAL VORTICITY
3.2
Ozone Change (ppmv)
^
MlanddiabatLceffectsfromadiabaticandtransport««**Po^rri^rtSSSSS
h Measurements shown h,2re are the difference in ozone averaged around contours of PV
between the two dates .nd.cated in each panel. The black line gives the approximate edae of ^e vortex with
DurS0/t0 thn rl9Hht Jn Vhe N0rthem HemisPner« (NH) and to the left ff he ^^Sfe^SSS?
3.31
-------
POLAR PROCESSES
-15 0 15 30 45
DAY
60 75 90
Figure 3-23. Calculated seasonal evolution (day 1
= 1 January 1992) of CIO, HCI, NO, and ozone at
noon for an air parcel at 18 km altitude, 65°N lati-
tude, processed periodically by PSCs. Case A: No
denitrification (solid line). Case B: 90-percent deni-
trification following the first PSC event (dotted line).
Case C: No PSC processing (dashed line). Reduc-
tion in ozone during March in the absence of PSC
processing occurs because of reactions involving
NOX. Data points represent mean and standard dcj-
viation of aircraft observations during AASE II for
the 470 K potential temperature surface and poten-
tial vorticity values greater than 2.8 x 10-s K
m2kg-1s-1. Data used for CIO and NO are restricted
to daytime observations (solar zenith angle < 86°).
Concentrations of CIO, HCI, and NO have been
normalized to their respective reservoirs to remove
the influence of small-scale atmospheric gradients
(Salawitchefa/.,1993).
vated, total ozone loss may increase significantly (Brune
et al, 1991; Solomon, 1990). \
Wave activity in polar regions is also thought to be
influenced by phenomena occurring at lower latitudes.
The strongest of these is the quasi-biennial oscillation
(QBO) (van Loon and Labitzke, 1993; Angell, 1993;
Labitzke, 1992; Poole et al, 1989). The QBO refers to
changes hi the direction and magnitude of stratospheric
winds above the equator that occur with a period of
about 27 months. In years when the winds in the equato-
rial lower stratosphere are from the east, the northern
vortex is comparatively weak and warm, thereby mini-
mizing the potential for ozone depletion. In westerly
years, the vortex is colder and more intense;in both
hemispheres. El Nino/Southern Oscillation (E^SO) ef-
fects, referring to changes in sea surface temperature and
associated shifts in atmospheric mass in the South Pacif-
ic Ocean, represent a much weaker influence .(Angell,
1993; Baldwin and O'Sullivan, 1994).
Wave activity plays a more important role in sub-
seasonal variability in the Northern Hemisphere than in
the Southern Hemisphere. Specifically, major midwinter
warming events often result in the Northern Hemisphere
from strong wave activity hi the troposphere associated
with cyclones and anticyclones (Labitzke, 199|2; Man-
ney et al, 1994a). In the middle stratosphere, the polar
vortex may break apart or split during a warming, caus-
ing large amounts of lower latitude air to be transported
to high latitudes and reversing the meridional tempera-
ture gradient. Such warmings eventually mark the end of
PSC temperatures throughout the vortex and change the
effectiveness of ozone catalytic loss cycles. Wave activi-
ty also creates variability in column ozone by changing
tropopause heights and temperatures in localized regions
(Farman et al, 1994; Petzoldt et al, 1994). Ozone col-
umn amounts are reduced by convergence of ozpne-poor
air below and divergence of ozone-rich air above, and by
rapid advection of low-latitude air in the case of persis-
tent ridge formation hi the upper troposphere/lower
stratosphere (Orsolini et al, 1994). These changes do
much to obscure ozone changes due to photochemical
loss.
Volcanic eruptions are also a source of variability
in the stratosphere. In addition to chemical effects (see
Section 3.3.3), increases in stratospheric aerosol that fol-
low an eruption have direct and indirect effects on
temperature and circulation in both the stratosphere and
3.32
-------
POUR PROCESSES
troposphere (Rind a al., 1992). The direct 'effect in the
lower stratosphere is a warming in the tropics (Kinne et
al., 1992; Labitzke and McCormick, 1992) arid a cooling
in polar regions. These and other changes may influence
the vortex and the formation of PSCs. '.
As a final consideration, trends in source gas emis-
sions in the troposphere may eventually iiffect polar
ozone loss and its variability. Of greatest interest are
changes in H2O, Cft,, carbon dioxide (CO2X N2O, and
halogen-containing species (see Chapter 2),' which all
participate in establishing the meteorological and photo-
chemical context of the depletion process. Increases in
CO2 are expected to decrease temperatures in the lower
stratosphere, thereby increasing the frequency and ex-
tent of PSCs (Austin and Butchart, 1994; Austin et al.,
1992). With additional cooling caused by the subsequent
destruction of ozone, total ozone loss in the Arctic could
become comparable to that in the Antarctic. Mpre direct-
ly, a doubling of inorganic chlorine species in the
stratosphere would likely result in Arctic ozone losses
that are comparable to those in the Antarctic (Salawitch
et al., 1993). PSC frequency would also increase in re-
sponse to growth in atmospheric CFi, and to am increase
in the amount of H2O entering the stratosphere in the
tropics. A more direct source is the emission of H2O and
NOy species from aircraft operating in the upper tropo-
sphere and lower stratosphere (Peter etai, 1991).
3.4.3 Photochemical Recovery
After the cessation of PSC formation Inside the
vortex, the conversion rate of inactive reservoir chlorine
to active chlorine is reduced to pre-winter values (see
Figure 3-1). Accordingly, CIO values fall from their mid-
winter peak values throughout the vortex (see Figures
3-20 and 3-23) (Waters et al., 1993a, b; Toohey et al.,
1993; Salawitch et al., 1993). In this recovery period,
changes caused by PSCs are reversed as photochemistry'
restores reservoir chlorine to pre-winter values. In the
Northern Hemisphere, air usually experiences 1>SC tem-
peratures on only a few occasions and for only a small
fraction of time throughout midwinter (Newman et al.,
1993). Thus, recovery is ongoing throughout the winter!
in contrast to the Southern Hemisphere. Recovery pro-
ceeds with reactions involving active chlorine and
reactive nitrogen species: !
CIO + N02 + M, -> C10NO2 + M
C1 + CH4
HNO3+hv-»NO2+OH
(3-11)
(3-12)
(3-13)
(3-14)
where hv is solar radiation and OH is the hydroxyl radi-
cal. Reaction (3-14) is key to maintaining the
partitioning within the NOy reservoir in Equation (3-2).
Reaction (3-11) is predominant in the early recovery
phase because of the availability of NO2 from Reaction
(3-14). NO2 increases dramatically with the return of
sunlight to the poles when HNO3 is available (Keys et
al, 1993; Solomon and Keys, 1992). Changes in reser-
voir chlorine have been confirmed with in situ
measurements of HCland remote soundings of ClONC^
near the vortex edge and inside the vortex in the Arctic
when denitrification is low (see Figures 3-13,3-14,3.15,
3-16, and 3-17) (Lutmam et al., 1994b; Roche 'et al,
1993b, 1994). Specifically, the enhancement of ClONO^
estimates in the early recovery phase is evident in air-
craft measurements in the Arctic in February 1992 (see
Figure 3-13). As recovery progresses, more reservoir
chlorine shifts from C1ONO2 to HCI, until values present
m late fall are restored (Liu et al, 1992). When denitrifi-
cation is significant entering the recovery phase,
CIONO2 may not be formed as readily as indicated in
Figure 3-1. Instead, Reaction (3-13) dominates, restor-
ing HCI rapidly and causing HCI to exceed C1ONO2
temporarily. As reactive nitrogen is mixed back into the
air parcel, more C1ONO2 is formed and C1ONO2 and
HCI return to unperturbed values.
Ozone loss during the recovery phase depends
strongly on the extent of denitrification. With extensive
denitrification, the abundance of NO2 produced by Re-
action (3-14) is limited, thereby enhancing ozone loss
rates (see Figure 3-23) (Salawitch et al, 1993; Kondo et
al, 1994b; Brune et al, 1991). Full recovery must then
wait until breakup of the vortex facilitates mixing with
lower latitude air that has riot been denitrified. The en-
hancement of C10N02 values during recovery and
elevated temperatures mean that catalytic cycles other
than CIO + CIO contribute to ozone loss during this peri-
od (Toumi et al, 1993).
Ultimately, the importance of the recovery phase
for ozone depletion depends on details of vortex break-
up. Planetary wave activity in the spring breaks apart the
3.33
-------
POLAR PROCESSES
vortex weakened by the reduction in radiative forcing. In
the Antarctic, variability is lower, but significant interan-
nual differences.still occur in the lifetime of the vortex
(see Figure 3-3). As the area covered by PSC tempera-
tures lessens, the distortion of the vortex in a wave event:
can typically lead to a rapid breakup of the vortex.
(Krueger et al, 1992). In any year, an early breakup
phase minimizes ozone depletion. However, in the
breakup process, the vortex may distort to reach lower
latitudes, significantly increasing local ozone loss rates
(Solomon, 1990; Brune et al, 1991). After breakup, the
transport of lower latitude air to the poles displaces air
parcels depleted in ozone. At the same time, processed
air that is low in ozone, contains active chlorine, or is
potentially denitrified and dehydrated is transported to
lower latitudes (Atkinson et al., 1989; Harwood et al.,
1993). As ozone loss continues in these air parcels, mid-
latitude ozone may be significantly impacted (see
Chapter 4).
3.5 VORTEX ISOLATION AND EXPORT TO
MIDLATITUDES
Understanding the isolation of the winter polar
vortex is a key factor in understanding the budgets of
ozone and other trace constituents at high latitudes. If a
large flow exists through the region of processed air in-
side the vortex (see Figure 3-2), then photochemical loss
rates of ozone must be substantially, larger than in an iso-
lated vortex to cause observed ozone depletion
(Anderson et al., 1991). In addition, export of processed
air to lower latitudes and lower altitudes may enhance
ozone depletion in those regions (see Chapter 4) (Brune
et ai, 1991). However, even if highly isolated during
winter, processed air in the vortex has the potential to
influence lower latitudes following vortex breakup in
late winter/early spring. Significant progress has oc-
curred in this assessment period in the modeling and
interpretation of data related to the transport of air in and
near the vortex. Trace constituent observations, radiative
balance arguments, and various fluid mechanical models
of the vortex have all provided valuable insights into vor-
tex motion. In addition, the identification of a vortex
edge region and a range of definitions for the vortex
boundary have become important concepts. A large body
of those results supports a substantial isolation in winter
of an inner vortex region that is surrounded by an edge
region in which stronger mixing to midlatitudes occurs.
3.5.1 Vortex Boundaries
' [
The motion of mass into the winter polar vortex is
poleward and downward from the upper stratosphere and
mesosphere (see Figure 3-2) (Schoeberl and Hartmann,
1991; Schoeberl et al., 1992). Flow out of the vortex in
the lower stratosphere must cross through the outer
boundary or edge region or through a lower boundary or
bottom of the vortex. Since pressure increases with
depth into the vortex from above, the velocities associat-
ed with such mass flow decrease accordingly. The edge
region is denoted by the location of strong horizontal
gradients in parameters associated with the vortex.
These gradients provide definitions for a boundary of the
vortex. Choices include the maximum in the sp^ed of the
polar wind jet, the maximum latitude gradient in PV, a
large change in one or more trace constituents with lati-
tude, and a kinematic barrier as identified in ;transport
model simulations. Because of the convergence of the
meridians at high latitudes, the vortex edge region repre-
sents most of the mass of the vortex and, hence, is crucial
for the evaluation of outflow and its influence at midlat-
itudes. i
The maximum wind speed in the circumpolar flow
of the polar jet provides the most accessible definition of
the boundary (see Figure 3-2). PV gradients, though ob-
tained from highly derived quantities, are more directly
related to dynamical barriers within the flow (Schoeberl
etal., 1992). PV combines the absolute vorticity of an air
parcel with static stability expressed as the vertical gra-
dient of potential temperature (Hoskins et al., 1985). In
isentropic and frictionless flow, PV is conserved, mak-
ing it a useful diagnostic for air motion over limited
periods. Large meridional gradients of PV (generally in-
creasing poleward) form in the polar regions ,as a result
of diabatic cooling and Rossby wave breaking in the
winter season. The polar jet is a response to the tempera-
ture gradient formed by the cooling at high latitudes in
winter. A boundary defined with a change in a trace con-
stituent is often associated with processing of polar air
by PSCs formed at the low vortex temperatures (Proffitt
et al., 1989b, c). As discussed above, processing results
in chlorine activation, dehydration, denitrification, and,
ultimately, ozone loss on the scale of the vortex. Finally,
3.34
-------
POLAR PROCESSES
a kinematic barrier to large-scale isentropic flow is re-
vealed in the Lagrangian evolution of air masses 6n
isentropic surfaces (Pierce and Fairlie, 1993). The ap-
proach uses assimilated wind fields to move material
lines initialized on closed streamlines encircling the Ant-
arctic vortex. In some instances, a particuhir material
line is found which shows no irreversible deformation
for periods of days to weeks. This "separating material
line" defines a kinematic boundary to large-iscale isen-
tropic transport in the polar region. Material poleward of
this separating material line remains highly isolated
from the surrounding circulation. !
These boundary definitions are interrelated since
each is derived from or caused by features of; the wind
and temperature fields in the winter season. The kine-
matic barrier, the maximum PV gradient, arid the jet
maximum are generally located within a few degrees of
latitude of each other within the polar jet core. However,
transient distortions of the vortex caused in the lower
stratosphere by tropospheric weather systems cause an
interweaving and distortion of these boundaries within
the edge region. While circumnavigating the vortex in
the jet, an air parcel may cross the PV or jet inaximum
boundary while remaining inside the kinematic barrier
and/or outside the chemical boundary. Thus, am evalua-
tion of the vortex export of air that resides near a
boundary will, in general, be dependent on the chosen
boundary. :
In the quantification of outflow, the choice of a
vortex edge is complicated by the fact that much of the
air "outside" of the vortex remains close to the edge and
varies with the large-scale fluctuations of the vortex
(Figure 15 in Rood et al., 1992; Manney et aL, 1994c;
Waugh et al., 1994). For changes in midlatitude ozone,
the important factors are the extent to which air under-
goes horizontal transport away from the center of the
vortex or away from the edge region to lower latitudes
and the extent to which this air has undergone processing
and, perhaps, loss of ozone. A substantial amount of pro-
cessing can occur within the vortex edge region,
particularly in the Antarctic vortex (Tao and Tuck,
1994). As a result, transport within the edge region, per-
haps across a particular boundary, is of considerably less
importance. In the evaluation of ozone loss photochem-
istry within the vortex, the total loss of processed air
from the center of the vortex and edge region is the quan-
tity of interest.
At the lower boundary of the vortex region, a tran-
sition is noted below which there is a much weaker
barrier to transport out of the vortex region to lower lati-
tudes (Tuck, 1989; Loew
-------
POLAR PROCESSES
at midlatitudes. This corresponds to replacing the air in
the vortex in approximately 90 days. Subsequent revi-
sions of the satellite H2O data set (version 17)
significantly reduce the vertical and horizontal extent of
the dehydration signature at midlatitudes (see Figure
3-18) (Russell, private communication, 1994), increas-
ing the vortex replacement time to about 120 days. With
a replacement time in this range, processed air inside the
dehydrated Antarctic vortex can be characterized as
largely isolated from influencing midlatitudes.
Further study of satellite observations of H2O and
CHLt confirms the isolated character of the inner vortex
(Pierce et al, 1994). The distribution of these species
over the winter reveals sustained diabatic descent ac-
companied by dehydration in the middle of the vortex. A
gradient in dehydration is established between the center
of the vortex and the jet core region where both normal
and dehydrated air are found. Trajectory calculations
that follow air parcels sampled by satellite for 25 days in
early spring show no evidence for large-scale transport
of significantly dehydrated jet core air into midlatitudcs
on either the 425 K (16 km) or 700 K (28 km) potential
temperature surfaces. However, some Irreversible trans-
port from the edge region to lower latitudes does take
place. In addition, the observations also show descent in
the jet core region bringing down air with higher values
In the Arctic, the absence of intense and wide-
spread dehydration within the vortex makes the use of
HjO and Cftt observations to detect vortex outflow
more difficult. However, using PV as a substitute tracer
in meteorological analyses, significant outflow of pro-
cessed air from the vortex edge region was deduced for
the vortex near 18 km (475 K) (Tuck et al., 1992). This
result is not inconsistent with an isolated center of tine
vortex because the outflow is from the vortex edge re-
gion. Analysis of aircraft observations shows that the
residual motion in regions of high active chlorine inside
the vortex is poleward and downward (Proffitt et al.,
1989c, 1990, 1993). The descent rates imply significant
flow through the vortex lower boundary and large dia-
batic cooling rates. The Arctic region has also been used
as a reference state to show the existence of denitrifica-
tion and dehydration outside the Antarctic vortex (Tuck
et al, 1994). However, a quantitative inconsistency ire-
mains between the amount of denitrification and
dehydration observed outside and inside the vortex, sug-
gesting that the understanding of the respective removal
processes or vortex export processes remains incomplete
(see Section 3.3.2.5). ;
Apart from the effort to evaluate vortex ;outfiow
with the signature of dehydration, the basic observation
of a large hemispheric asymmetry in water vapor in the
lower stratosphere remains (Kelly et al., 1990). After ac-
count is made for CH* oxidation in mid- to late-winter
observations, water vapor in the Northern Hemisphere is
larger by about 1.5 ppmv. The export of dehydrated air
from the Antarctic is one explanation of the difference.
Other explanations include the role of the tropics in re-
moving water upon entry of air into the stratosphere
(Tuck, 1989; Tuck etal, 1993; Kelly et al., 1989).
3.5.3 Radiative Cooling
To provide continuity for a substantial material
flux outward through the Antarctic vortex, either a
strong vertical transport between the middle and lower
stratosphere or compensating inward horizontal trans-
port is required. To exchange the mass of the vortex
between 16 to 24 km with a 30-day time scale rjequires a
vertical velocity of-0.1 cm s-1 at 16 km, whichis equiv-
alent to a potential temperature change near 1.3 K per
day. However, both NzO trends (Hartmann et al., 1989;
Loewenstein et al., 1989; Schoeberl et al., 1991) and ra-
diative calculations (Shine, 1989; Rosenfield et al.,
1987; Schoeberl et al., 1992; Manney etal., 1994c; Stra-
han et al., 1994) give much smaller values for this
velocity, near -0.02 cm s-l (0.2 K per day). Hence, a sub-
stantial body of interpretation supports a small net flux
through the Antarctic vortex on sub-seasonal time
scales.
Using more recent satellite observations of CH^
and HF, rapid and deep descent into the Antarctic vortex
has been observed (Russell et al., 1993b; Schoeberl et
al, 1994; Fischer et al, 1993). The descent rate is con-
sistent with expected cooling rates in the upper
stratosphere (Rosenfield et al, 1994). Lowjer in the
stratosphere, the descent rate slows, with an upper limit
of 0.07 cm s-1, corresponding to a replacement time of
vortex air of about 120 days (Schoeberl et al, 1994).
This is consistent with the estimates made from the ap-
pearance of dehydrated air at midlatitudes in the satellite
observations as noted above (see Section 3.5.2)
3.36
-------
Consistent with the enhanced wave activity, the
vertical flux between the middle and lower stratosphere
in the Arctic is much larger than that found in the Antarc-
tic; mean vertical velocities in the Arctic lower
stratosphere are near -0.06 cm s-i, or 0.6 K per day
(Schoeberl et al., 1992; Strahan et al., 1994; Bjiuer et al,
1994; Manney et al., 1994c). Interannual variability in
the wave disturbances in the Arctic also create^ variabil-
ity in the vortex transport In isentropic trajectory studies
examining 14 years of meteorological data, interannual
differences were found in the predominance of inward
and outward transport across the vortex boundary (Dahl-
berg and Bowman, 1994). Thus, quantification and
prediction of interannual variability are fundamentally
more difficult in the Arctic than in the Antarctic, impact-
ing prediction of ozone changes both in the vortex and at
midlatitudes. '
3.5.4 Trajectory Models •
In trajectory models, transport is examined by cal-
culating the dispersion of an ensemble of notional air
parcels over a typical one-month period, where: the ini-
tial position of each parcel is specified. Studies aire based
on National Meteorological Center (NMC)-derived
winds (Bowman, 1993) or on United Kingdom Meteoro-
logical Office-analyzed or modeled wind fields (Chen et
al, 1994; Manney et al., 1994c; Pierce et ai, 1994;
Pierce and Fairlie, 1993). Approaches include following'
individual parcels or ensembles of parcels forming mate-
rial lines around vortex streamlines. In each case,
large-scale horizontal transport through the vortex edge
region in the Antarctic is small near 20 km (450 K isen-
tropic level) (see Figure 3-24). In the figure, parrels that
are initiated inside the vortex, as defined by column
ozone values, remain in the vortex after 30 days. Similar-
ly, the evolution of material lines in the vortex region
reveals a kinematic barrier jto large-scale isentropic flow
out of the vortex (Pierce and Fairlie, 1993). However,
substantial mixing and transport does occur across the
lower vortex boundary (16 to 20 km, or 375 to 425 K).
This transport is consistent with transport deduciid from
constituent observations (Tuck, 1989; Proffitt et al.,
1989b, 1990, 1993). However, omission of diabatic ef-
fects and inertial gravity waves in such isentropic
trajectory studies may significantly underestimate trans-
port and mixing processes at the vortex boundary (Pierce
et al., 1994).
POLAR PROCESSES
In the Arctic vortex, large episodic disruptions oc-
cur as a result of planetary and synoptic wave
disturbances. These events, which are less frequent in
the Antarctic, are associated with transport of vortex au-
to midlatitudes in the lower stratosphere in the form of
narrow tongues, or filaments, that are pulled from the
edge of the vortex (Juckes and Mclntyre, 1987; Norton,
1993; Pierce and Fairlie, 1993; Waugh et al., 1994; Man-
ney et al, 1994c). These features are simulated in
contour advection modeling in which high spatial reso-
lution is maintained in the advection of material
contours. The result is that approximately 5 to 10 percent
of the vortex area is typically transported outward, with
up to 20 percent during exceptionally large events
(Waugh et al, 1994). As sin example, the total export
from the vortex in January 1992 represents only nine
percent of the area between 30°N and the vortex edge.
There is also evidence that low-latitude air is entrained
into the vortex during large disruptions, although the
volume of air involved is probably small (Plumb et al,
1994). At the rate of one to two planetary-scale events
per month, the e-folding time for vortex exchange to
midlatitudes by this mechanism is on the order of three
to six months in the lower stratosphere, depending on the
intensity and number of such events.
3.5.5 Three-Dimensional Models
In addition to trajectory models, three-dimension-
al (3-D) chemistry transport models (CTMs), 3-D
mechanistic models, and 3-D general circulation models
(GCMs) driven by winds from meteorological data as-
similation systems support relatively limited flow
through the vortex in winter. Three-dimensional models
improve the evaluation of vortex outflow because they
include both the horizontal transport through the vortex
edge and the vertical transport connecting the lower
stratosphere with the upper stratosphere and the meso-
sphere. For example, satellite data clearly show the
descent of mesospheric air deep into the stratosphere
sometime during the winter (see Figure 3-5) (Russell et
al, 1993b). Furthermore, since outflow will likely result
from zonally asymmetric mechanisms driving transport
at the vortex edge, both planetary-scale events and syn-
optic-scale events in the lower stratosphere can be
considered in 3-D models.
3.37
-------
POLAR PROCESSES
a) 1 October 1987
b) 30 October 1987
c) 1 October 1987
d) 30 October 1987
mi-w^f^^
Figure 3-24. Evolution of air parcels on the 450 K (19 to 20 km) surface in the lower stratosphere over the
period 1 -30 October 1987 in the Antarctic. Initial locations for approximately 16,000 parcels on 10ctober are
indicated in (a) and (c) for interior and exterior vortex parcels, respectively Final locations on 30 October are
shown in (b and (d) for the same groups, respectively. In each panel, fee vertical line is the'Greenwich
meridian and the large and small circles correspond to 30° and 60° latitude, respectively. The 250 Dobson
unit (DU) contour from the TOMS satellite observations of total ozone is used to separate the two parcel
groups. Parcel motion is determined by trajectory calculations using winds derived from National Meteoro-
logical Center-analyzed height fields (Bowman, 1993).
3.38
-------
POLAR PROCESSES
In results from 3-D CTMs using winds from data
assimilation systems, relatively little flow is found from
within the vortex to midlatitudes (Rood et al., 1992).
These models incorporate diabatic and mixing effects
that have not been considered in all trajectory ;and con-
tour surgery models. In evaluations using' aircraft,
balloon, and satellite measurements, these models have
been shown to represent synoptic^ and planetary-scale
variability on seasonal time scales. These models can
simulate satellite ozone observations (e.g., Limb Infra-
red Monitor of the Stratosphere [LEMS] and Total Ozone
Mapping Spectrometer [TOMS]) for the entire winter
season equally well in vortical and non-vorticail air, sug-
gesting that the transport mechanisms are at least
qualitatively correct These CTMs also show material
peeling off the edge of the vortex into subpolar latitudes.
This transport is wave-driven, with the planetary scales
dominating the synoptic scales at altitudes above 20 mb.
The results of Rood et al. (1992) in the Northern Hemi-
sphere found that typically five percent of the air
poleward of the subtropical jet stream and outside of the
vortex had been processed by PSCs. During extreme
events, this fraction could increase to 20 percent, in
broad agreement with others (Plumb et al., 1994; Tuck et
al., 1992).
A significant uncertainty in 3-D CTMs is' whether
or not the spatial resolution is adequate to simulate vor-
tex processes. For instance, in Douglass etal. (1.^91), the
general characteristics of aircraft CIO measurements
were well simulated, but the detailed structure close to
the vortex edge was not matched. Waugh et al. (1994)
have shown that winds from the relatively coarse NMC
analyses indeed contain enough information that,
through differential advection, detailed structure can be
meaningfully simulated. Therefore, in 3-D models and
contour advection, the problem becomes one of choos-
ing the appropriate mixing scale. The ability of carefully
formulated 3-D models to perform seasonal integrations
while maintaining realistic contrast between the vortex
and midlatitudes suggests that they are reasonably
mixed. Hence, the results suggest that it is not necessary
to simulate the details of the fine structure, but it is nec-
essary to simulate a self-consistent advective cascade
with subscale mixing. Furthermore, transport studies
driven by winds from assimilation analyses are likely to
be of sufficient quality that transport across the vortex
edge can be properly evaluated. High resolution may
still be required for a quantitative evaluation of ozone
depletion that occurs as processed air originating in the
vortex is transported and mixed with lower latitude air.
Plumb et al. (1994) have also identified a discrete
event of air being transported on horizontal surfaces into
the lower vortex. Dahlbeirg and Bowman (1994) have
performed a systematic evaluation of Arctic winters and
find only limited transport into the vortex, with most of
the activity remaining on the edge. Occasional inward
transport is associated with planetary-scale blocking pat-
terns and concomitant synoptic-scale lows that are
associated with meteorological conditions in the tropo-
sphere. These studies all suggest only limited horizontal
transport of extra vortex air into the vortex throughout the
winter.
Mechanistic 3-D models are a good tool for study-
ing descent They are forcisd from observations at some
lower boundary (e.g., 100 mb), with the stratosphere al-
lowed to evolve self-consisteritly in balance with this
forcing (e.g., Fisher etal., 11993). Because of the proxim-
ity of the forcing to the lower boundary, this approach
has limited utility hi the lower stratosphere. However,
mechanistic models do provide an effective way to ad-
dress the cold-pole problem (Mahlman and Umscheid,
1987) and other biases present in GCMs. Specifically,
forcing from observations raises the polar temperature
closer to observations, affording a more accurate repre-
sentation of diabatic descent. Recent studies (e.g.,
Jackman et al., 1993; Nielsen et al., -1994) show that
mechanistic models can reproduce the descent of meso-
spheric ozone depletion and NO2 enhancement that
occurs during «olar proton events (SPEs). This winter-
time descent occurs across all stratospheric and
mesospheric altitudes and requires consistent represen-
tation of mean-meridional flow in the mesosphere. The
models do, in fact, represent the cross-equatorial trans-
port of long-lived tracers observed in the mesosphere by
satellite. Mechanistic models show unmixed descent
consistent with satellite observations (Russell et al.,
1993b; Fisher et al., 1993), Satellite data also indicate
descent with little or no large-scale mixing across the
vortex edge in the mid-stratosphere (Lahoz etal, 1993).
During midwinter, very little of the mesospheric air
leaves the vortex in the lower stratosphere. This is con-
sistent with the Stratospheric Aerosol and Gas
Experiment (SAGE) NO2 enhancements observed dur-
ing an SPE. Mixing of mesospheric air that has
3.39
-------
POLAR PROCESSES
undergone descent occurs dramatically during vortex
breakdown in the winter-to-spring seasonal transition, as
has been observed in satellite data (Harwood et al.,
1993; Lahoz et al., 1993). However, this is one-time
mixing of air that was contained in the vortex, and does
not provide a continual flow of air through the vortex.
The mechanistic models establish that, given a realistic
temperature distribution, radiative models calculate de-
scent rates that are fundamentally in agreement with
observed constituent behavior in the mid- to upjper
stratosphere. In the lower stratosphere, some uncertainty
remains in modeling the relative effects of dynamical
mixing and diabatic descent. However, Schoeberl et al.
(1994) and Strahan et aL (1994) have shown that the air-
craft NjO data are in agreement with calculated radiative
descent The uncertainty that remains in the 3-D models
will not substantially alter the arguments presented here.
GCMs provide an internally consistent, determin-
istic simulation of the atmosphere, although they cannot
be used to simulate specific events for more than a few
days, inhibiting direct day-to-day comparisons with
constituent observations. Traditionally, GCMs underes-
timate polar temperatures (cold-pole), apparently due to
a lack of dynamic activity (Mahlman and Um;ch«id,
1987). This leaves the model atmosphere too close to ra-
diative equilibrium and, subsequently, leads to weak
estimates of wintertime descent. The current generation
of GCMs that extend up to the mesosphere (Strahan and
Mahlman, 1994a, b; Boville, 1991; Cariolleera/., 1992)
has now been integrated for several seasonal cycles. In
the Northern Hemisphere, the models can produce a dis-
turbed flow in winter with the development of
stratospheric warmings associated with the amplifica-
tion of planetary waves. The model vortex is about 20 K
warmer in the Northern Hemisphere than in the Southern
Hemisphere, reasonably consistent with atmospheric ob-
servations. Comparisons with N2O data show that the
fall-to-winter descent can be simulated with consider-
able accuracy, and that the wintertime descent is
maintained at a level comparable to observations (Strah-
an et al., 1994). Transport of vortex edge air is simulated
with mixing in the midlatitudes. Deep vortical air re-
mains relatively isolated.
N2O distributions from a GCM have also been
compared with aircraft measurements (Strahan and
Mahlman, 1994a, b). These studies show that, within the
resolution constraints of the model, the processes that
produce shredding from the vortex edge are consistent
with observations. In addition, the mesoscale component
of the variance, which is linked to planetary wave break-
ing processes, is also consistent in the: Northern
Hemisphere. A separate study of the N2O aircraft obser-
vations supports only a limited outward flow near the
vortex edge (Bacmeister et al., 1992). These observa-
tions, when combined with theory and modeling results,
provide a very powerful statement about transport
through the vortex and model fidelity. ;
GCM simulations of the troposphere and strato-
sphere in the Antarctic are not as good as those in the
Arctic, because the cold-pole problem is still significant
and synoptic-scale activity is poorly represented in the
southern ocean. In data assimilation approaches, the ob-
servations in the Southern Hemisphere are not sufficient
to define many of the important waves. With less wave
activity, the model atmosphere is closer to radiative equi-
librium, resulting in less wintertime polar-night descent
and a more isolated vortex. The observations pf the Ant-
arctic vortex strongly indicate that it is closer to radiative
equilibrium than the Arctic vortex. The Antarctic tem-
peratures are lower, the vortex is larger, arid there is
significantly less wave activity perturbing the flow, fur-
ther suggesting that the Antarctic vortex is more isolated
than the Arctic vortex.
These 3-D model approaches provide a consistent
picture of dynamical processes of the polar vortex. The
mechanisms in the 3-D global models are consistent
with the barotropic models (e.g., Juckes and Mclntyre,
1987) and the contour advection models (e.g.; Waugh et
al, 1994) that have been used to isolate transport mech-
anisms. Most of the transport into and out of'the vortex
occurs along the edges, and deep vortical air is largely
isolated throughout the winter. The material that is shred
out of the vortex is spread broadly in midlatitudes, but
satellite observations and model studies (Rood et al.,
1992, 1993) suggest that the midlatitudes are not homo-
geneously mixed. There is one-time mixing of the deep
vortex air during the winter-to-spring transition, with
processed air reaching mid- to low latitudes. There is
continual circulation of midlatitude air towards the poles
at high altitudes, followed by descent as the air enters
polar night and cools. This circulation is largely on the
edge of the vortex and should not see the full impact of
polar processing. There can be substantial lojcal mixing
at low altitudes associated with dissipating synoptic
3.40
-------
scales. Given the local nature of this transput, it does not
require compensation by transport from above. In sum-
mary, given the seasonal lifetime of the vortex, the
mixing times inferred from observations and models, the
confinement of mixing to the edges, and the mixing in
the winter-to-spring transition, it seems unlikely that the
total volume of air that experiences polar chemical pro-
cessing can exceed two times the volume of the
midwinter vortex.
REFERENCES !
Abbatt, J. R D., and M. J. Molina, Heterogeneous inter-
actions of C10N02 and HC1 on nitric acid
trihydrate at 202 K, /. Phys. Chem.,96, 7674-
7679, 1992a. !
Abbatt, J. R D., and M. J. Molina, The heterogeneous
reaction HOC1 + HC1 -* CI2 + H2O on ice and ni-
tnc add trihydrate: Reaction probabilities and
stratospheric implications, Geophys. Res Lett
19, 461-464, 1992b. !
Adriani, A., T. Deshler, G. P. Gobbi, and G Di Don-
ftancesco, Polar stratospheric clouds and volcanic
aerosol during 1992 spring over McMuido Sta-
tion, Antarctica: Lidar and particle counter
comparisons, /. Geophys. Res., in press 1994
Anderson, J. G, W. H. Brune, S. A. Lloyd, D. W. Toohey,
S. P. Sander, W. L. Starr, M. Loewenstein, and J. R.'
Podolske, Kinetics of O3 destruction by 'CIO and
BrO within the Antarctic vortex: An analysis based
on in situ ER-2 data, J. Geophys. Res. 94 11480-
11520,1989. \,
Anderson, J. G., D. W. Toohey, and W. H. Brane, Free
radicals within the Antarctic vortex: The role of
CFCs in Antarctic ozone loss, Science, 251 39-46
1991. ' '
Anderson, J. G., and O. B. Toon, Airborne Arctic Strato-
spheric Expedition it: An overview, Geophys Res
Lett., 20, 2499-2502, 1993.
Angell, J., Reexamination of the relation between depth
of the Antarctic ozone hole, and equatorial QBO
and SST, 1962-1992, Geophys. Res. Lett. 20
1559-1562,1993. ',
Arnold, F.; K. Petzoldt, and E. Reimer, On the formation
and sedimentation of stratospheric nitric acid aero-
sols—Implications for polar ozone destruction,
Geophys. Res. Lett., 19, 677-680, 1992. ,.i
, e POLAR PROCESSES
Atkinson, R. J., W. A. Matthews, P. A. Newman, and R
A. Plumb, Evidence of the mid-latitude impact of
Antarctic ozone depletion, Nature, 340 290-204
1989. ' '
Austin, J. N., N. Butchait, and K. P. Shine, Possibility of
an Arctic ozone hole in a doubled-CO2 climate
Nature, 360, 221-225, 1992.
Austin, J. N., and N. Buitchait, The influence of climate
change and the timing of stratospheric warmings
on Arctic ozone depletion, J. Geophys Res 99
1127-1145,1994.
Bacmeister, J. T., M. R. Schoeberl, M. Loewenstein J
R. Podolske, S. E. Strahan, and K. R. Chan, An
estimate of the relative magnitude of small-s'cale
tracer fluxes, Geophys. Res. Lett., 19 1101-1104
1992.
Baldwin, M. P., and D. O'Sullivan, Stratospheric effects
of ENSO-related tropospheric circulation anoma-
lies, J. Climate, in press, 1994
Barath, F. T, M. C. Chavez, R. E. Cofield, D. A. Flower
M. A. Frerking, M. B. Gram, W. M. Harris J R
Holden, R. R Jarnpt, W. G. Kloezeman G J
Klose, G. K. Lau, M. S. Loo, B. J. Maddison, R J
Mattauch, R. R McKinney, G. E. Peckham, H M
Pickett, G. Siebes, F. S. Soltis, R. A. Suttie, J A
Tarsala, J. W. Waters, and W. J. Wilson, The Upper
Atmosphere Research Satellite Microwave Limb
Sounder Experiment, J. Geophys. Res 98 10751
10762, 1993. ' ' '
Bauer, R., A. Engel, H. Franken, E. Klein, G. Kulessa, C
Schiller, U. Schmidt, R. Borchers, and J. Lee,
Monitoring the vertical structure of the Arctic po-
lar vortex over northern Scandinavia during
EASOE: Regular N2Q profile observations, Geo-
phys. Res. Lett., 21, 1211 -1214, 1994.
Bell, W., N. A. Martin, T. D. Gardiner, N. R. Swann, P. T
Woods, P. F. Fogal, :and J. W. Waters, Column
measurements of stratospheric trace species over
Are, Sweden in the winter of 1991-92, Geophys.
Res. Lett., 21, 1347-1350, 1994.
Borrmann, S., J. E. Dye, D. Baumgardner, J. C. Wilson,
H. H. Jonsson, C. A. B,rock, M. Loewenstein, J R'
Podolske, G. V. Ferry, and K. S. Barr, In-situ mea-
surements of changes in stratospheric aerosol and
the N2O-aerosol relationship inside and outside of
the polar vortex, Geophys. Res. Lett., 20 2559-
2562, 1993.
3.41
-------
POLAR PROCESSES
Boville, B., Sensitivity of simulated climate to model
resolution,/. Climate, 4, 469-485, 1991.
Bowman, K. P., Large-scale isentropic mixing properties
of the Antarctic polar vortex-from analyzed winds,
/. Geophys. Res., 98, 23013-23027, 1993.
Brandtjen, R., T. Kliipfel, D. Perner, and B. M. Knudsen,
Airborne measurements during the European Arc-
tic Stratospheric Ozone Experiment: Observation
of OC1O, Geophys. Res. Lett., 21, 1363-1366,
1994.
Brasseur, G. P., and C. Granier, Mount Pinatubo aero-
sols, chlorofluorocarbons, and ozone depletion,
Science, 257, 1239-1242, 1992.
Browell, E. V., C. F. Butler, S. Ismail, P. A. Robinette, A.
F. Carter, N. S. Higdon, O. B. Toon, M. R. Schoe-
berl, and A. F. Tuck, Airborne lidar observations in
the wintertime Arctic stratosphere: Polar strato-
spheric clouds, Geophys. Res. Lett., 17, 385-388,
1990.
Browell, E. V., C. F. Butler, M. A. Fenn, W. B. Grant, S.
Ismail, M. R. Schoeberl, O. B. Toon, M. Loewen-
stein, and J. R. Podolske, Ozone and aerosol
changes during the 1991-1992 Airborne Arctic
Stratospheric Expedition, Science, 261, 1155-
1158,1993.
Brune, W. H., J. G. Anderson, D. W. Toohey, D. W. Fa-
hey, S. R. Kawa, R. L. Jones, D. S. McKenna, and
L. R. Poole, The potential for ozone depletion f.n
the Arctic polar stratosphere, Science, 252, 1260-
1266,1991.
Burkholder, J. B., R. K. Talukdar, A. R. Ravishankara,
and S. Solomon, Temperature dependence of the
HNOs UV absorption cross sections, J. Geophys.
Res., 98, 22937-22948, 1994a.
Burkholder, J. B., R. K. Talukdar, and A. R. Ravishanka-
ra, Temperature dependence of the C1ONO2 UV
absorption spectrum, Geophys. Res. Lett., 21, 585-
588,1994b.
Cacciani, M., P. Di Girolamo, A. di Sarra, G. Fiocco, and
D. Fua, Volcanic aerosol layers observed by lidar
at South Pole, September 1991-June 1992, Geo-
phys. Res. Lett., 20, 807-810, 1993.
Cariolle, D., A. Lasserre-Bigorry, and J. F. Boyer, A gen-
eral circulation model simulation of the springtime
Antarctic ozone decrease and its impact on mid-
latitudes, J. Geophys. Res., 95, 1883-1898, 1990.
Cariolle, D., M. Amodei, and P. Simon, Dynamics and
the ozone distribution in the winter stratosphere:
Modelling the inter-hemispheric difference's, J. At-
mos. Terr. Phys., 54, 627-640, 1992. j
Carroll, M. A., R. W. Sanders, S. Solomon, and A. L.
,'Schmeltekopf, Visible and near-ultraviolet spec-
troscopy at McMurdo Station, Antarctica 6.
Observations of BrO,/. Geophys. Res., 94, 16633-
16638, 1989.
Carslaw, K., B. P. Luo, S. L. Clegg, Th. Peter, P. Brim-
blecombe, and P. J. Crutzen, Stratospheric aerosol
growth and HNO3 gas phase depletion from cou-
pled HNO3 and water uptake by liquid particles,
Geophys. Res. Lett., in press, 1994.
Chen, P., J. R. Holton, A. O'Neill, and R. Swinbank,
Quasi-horizontal transport and mixing! in the
Antarctic stratosphere, J. Geophys. Res., 99,
16851-16866, 1994.
Chipperfield, M. P., A 3-D model comparison, of PSC
processing during the Arctic winters of 11991/92
and 1992/93, Annales Geophysicae, 12, 342-354,
1994a.
Chipperfield, M. P., D. Cariolle, and P. Simon, A 3D
transport model study of chlorine activation dur-
ing EASOE, Geophys. Res. Lett., 21, 1467-1470,
1994b.
Chipperfield, M. P., D. Cariolle, and P. Simon, A three-
dimensional transport model study of PSC
processing during EASOE, Geophys. Res. Lett.,
21, 1463-1466, 1994c.
Chu, L. T., M.-T. Leu, and L. F. Keyset, Heterogeneous
reactions of HOC1 + HC1 -> C12 + H2O and
C1ONO2 + HC1 -» C12 + HNO3 on ice surfaces at
polar stratospheric conditions, J. Phys. Chem., in
press, 1994.
Cox, R. A., A. R. MacKenzie, R. H. Miiller, Th. Peter,
and P. J. Crutzen, Activation of stratospheric chlo-
rine by reactions in liquid sulfuric zcid,iGeophys.
Res. Lett., 21, 1439-1442, 1994. j
Crewell, S., K. Kiinzi, H. Nett, T. Wehr, and P| Hartogh,
Aircraft measurements of CIO and HC1 during
EASOE 1991/92, Geophys. Res. Lett., 21, 1267-
1270,1994.
Dahlberg, S. P., and K. P. Bowman, Climatology of
large-scale isentropic mixing in the Arctic winter
stratosphere from analyzed winds, /. Geophys.
Res., 99, 20585-20599, 1994. '•
3.42
-------
POLAR PROCESSES
Danielsen, E. F., and H. Houben, Dynamics of the^Ant-
arctic stratosphere and implications for the ozone
hole, Anthropogene Beeinflussung der
Ozonschicht, 191-242, DECHEMA, Frankfurt-
am-Main, 1988.
Deshler, T., B. J. Johnson, and W. R. Rozier, Changes in
the character of polar stratospheric clouds over
Antarctica in 1992 due to the Pinatubo volcanic
aerosol, Geophys. Res. Lett., 21, 273-276, 1994.
deZafra, R. L., M. Jaramillo, A. Parrish, P. Solomon, R.
Connor, and J. Barrett, High concentrations of
chlorine monoxide at low altitudes in the Antarctic
spring stratosphere: Diurnal variation; Nature,
525,408-411,1987. :
Douglass, A. R., R. B. Rood, J. A. Kaye, R. S. Stolarski,
D. J. Allen, and E. M. Larson, The influence of
polar heterogeneous processes on reactive chlo-
rine at middle latitudes: Three-dimensional model
implications, Geophys. Res. Lett., 18,25-28,1991.
Douglass, A. R., R. B. Rood, J. W. Waters, lJ Froide-
vaux, W. Read, L. S. Elson, M. Geller, Y. Chi, M.
C. Cerniglia, and S. Steenrod, A 3D simulation of
the early winter distribution of reactive chlorine in
the north polar vortex, Geophys. Res. Lett., 20,
1271-1274, 1993.
Drdla, K., and R. P. Turco, Denitrification through PSC
formation: A 1-D model incorporating tempera-
ture oscillations, /. Atmos. Chem., 12, 319, 1991.
Dritschel, D. G., and B. Legras, Modeling oceanic and
atmospheric vortices, Pkys. Today, 44-51, March
1993.
Dye, J. E., D. Baumgardner, B. W. Gandrad, S. R. Kawa,
K. K. Kelly, M. Loewenstein, G. V. Fen:y, K. R.
Chan, and B. L. Gary, Particle size distributions in
Arctic polar stratospheric clouds, growth and
freezing of sulfuric acid droplets, and implications
for cloud formation, J. Geophys. Res., 97, 8015-
8034,1992.
Eckman, R. S., R. E. Turner, W. T. Blackshear, T D. A.
Fairlie, and W. L. Grose, Some aspects of the inter-
action between chemical and dynamic processes
relating to the Antarctic ozone hole, Adv. Space \
Res., 75,311-319, 1993. I:
Fahey, D. W., S. Solomon, S. R. Kawa, M. Loewenstein,
J. R. Podolske, S. E. Strahan, and K. R. Chan, A
diagnostic for denitrification in the winter polar
stratospheres, Nature, 345, 698-702, 1990a.
Fahey, D. W., K. K. Kelly, S. R. Kawa, A. F. Tuck, M.
Loewenstein, K. R. Chan, and L. E. Heidt, Obser-
vations of denitrification and dehydration in the
winter polar stratospheres, Nature, 344, 321-324,
1990b. ;
Fahey, D. W., S. R. Kawa, E. L. Woodbridge, P. Tin, J. C.
Wilson, H. H. Jonsspn, J. E. Dye, D. Baumgard-
ner, S. Borrmann, D'. W. Toohey, L. M. Avallone,
M. H. Proffitt, J. J. Margitan, M. Loewenstein, J.
R. Podolske, R. J. Salawitch, S. C. Wofsy, M. K.
W. Ko, D. E. Anderson, M. R. Schoeberl, and K.
R. Chan, In situ m
-------
POLAR PROCESSES
Hanson, D. R., and A. R. Ravishankara, Investigation of
the reactive and nonreactive processes involving
C1ONO2 and HC1 on water and nitric acid doped
ice, J. Phys. Chem., 96, 2682-2691, 1992.
Hanson, D. R., and A. R. Ravishankara, Uptake of HC1
and HOC1 onto sulfuric acid: Solubilities, diffusivi-
ties, and reaction, J. Phys. Chem^ 97, 12309-12319,
1993.
Hanson, D. R., and A. R. Ravishankara, The reaction
C1ONO2 + HC1 on NAT, NAD, and frozen sulfuric
acid and the hydrolysis of ^05 and C1ONO2 on
frozen sulfuric acid, J. Geophys. Res., 98, 22931-
22936,1994.
Hanson, D. R., A. R. Ravishankara, and S. Solomoiti,
Heterogeneous reactions in sulfuric acid aerosols:
A framework for model calculations, J. Geophys.
Res., 99,3615-3629, 1994.
Hartmann, D. L., L. E. Heidt, M. Loewenstein, J. R. Pod-
olske, J. R Vedder, W. L. Starr, and S. E. Strahan,
Transport into the south polar Vortex in early
spring,/. Geophys. Res., 94, 16779-16795, 1989.
HarwocKf, R. S., E. S. Carr, L. Froidevaux, R. F,. Jarnot,
W. A. Lahoz, C. L. Lau, G. E. Peckham, W. (3.
Read, P. D. Ricaud, R. A. Suttie, and J. W. Wateis,
Springtime stratospheric water vapour in the
Southern Hemisphere as measured by MLS, Geo-
phys. Res. Lett., 20, 1235-1238, 1993.
Hofmann, D. J., and T. Deshler, Comparison of strato-
spheric clouds in the Antarctic and Arctic,
Geophys. Res. Lett., 16, 1429-1432, 1989.
Hofmann, D. J., and S. Solomon, Ozone destruction
through heterogeneous chemistry following the
eruption of El Chich6n, J. Geophys. Res., 94,
5029-5041, 1989.
Hofmann, D. J., J. M. Rosen, J. W. Harder, and J. V.
Hereford, Balloon-borne measurements of aero-
sol, condensation nuclei, and cloud particles in the
stratosphere at McMurdo Station, Antarctica, dur-
ing the spring of 1987, /. Geophys. Res., 94,
11253-11269,1989.
Hofmann, D. J., T. Deshler, F. Arnold, and H. Schlager,
Balloon observations of nitric acid aerosol forma-
tion in the Arctic stratosphere: II. Aerosol,
Geophys. Res. Lett., 17, 1279-1282, 1990.
Hofmann, D. J., and S. J. Oltmans, The effect of strato-
spheric water vapor on the heterogeneous reaction
rate of C1ONO2 and H2O for sulfuric acid aerosol,
Geophys Res. Lett., 19, 2211-2214,1992!
Hofmann, D. J., S. J. Oltmans, J. M. Harris, S. Solomon,
T. Deshler, and B. J. Johnson, Observation and
possible causes of new ozone depletion in Antarc-
tica in 1991, Nature, 359, 283-287,1992.
Hofmann, D. J., and S. J. Oltmans, Anomalous Antarctic
ozone during 1992: Evidence for Pinatubo volca-
nic aerosol effects, J. Geophys. Res., 98,
18555-18561, 1993.
Hoskins, B. J., M. E. Mclntyre, and A. W. Robertson, On
the use and significance of isentropic [potential
vorticity maps, Quart. J. Roy. MeteoroL Soc., Ill,
877-946, 1985. \
Hiibler, G., D. W. Fahey, K. K. Kelly, D. D. Montzka, M.
A. Carroll, A. F. Tuck, L. E. Heidt, W. H. Pollock,
G. L. Gregory, and J. F. Vedder, Redistribution of
reactive odd nitrogen in the lower Arctic strato-
sphere, Geophys. Res. Lett., 17, 453-456^ 1990.
Jackman, C. H., R. D. McPeters, A. R. Douglass, J. E.
Nielsen, D. J. Allen, M. C. Cemiglia, R. JB. Rood,
and J. M. Zawodny, Polar stratospheric downward
transport during northern winter observed by
SAGE If and simulated using a three-dimensional
model after the October 1989 solar proton events,
EOS Transactions, 74, 178, 1993.
JPL, Chemical Kinetics and Photochemical Data for
Use in Stratospheric Modeling, Evaluation No. 10,
Jet Propulsion Laboratory, JPL Publication 92-20,
1992. '
Juckes, M. N., and M. E. Mclntyre, A high resolution
one-layer model of breaking planetary waves in
the stratosphere, Nature, 328, 590-596, 1987.
Kawa, S. R., D. W. Fahey, L. E. Heidt, W. H. Ppllock, S.
Solomon, D. E. Anderson, M. Loewenste'in, M. H.
Proffitt, J. J. Margitan, and K. R. Chan, Photo-
chemical partitioning of the reactive nitrogen and
chlorine reservoirs in the high-latitude strato-
sphere, J. Geophys. Res., 97,1905-1923', 1992a.
Kawa, S. R., D. W. Fahey, K. K. Kelly, J. E; Dye, D.
Baumgardner, B. W. Gandrud, M. Loewenstein,
G. V. Ferry, and K. R. Chan, The Arctic polar
stratospheric cloud aerosol: Aircraft measure-
ments of reactive nitrogen, total water, and
particles, J. Geophys. Res., 97, 7925-7938,1992b.
3.44
-------
POLAR PROCESSES
Kelly, K. K., A. F. Tuck, D, M. Murphy, M."H. Proffitt,
D. W. Fahey, R. L. Jones, D. S. McKenaa, M. Loe-
wenstein, J. R. Podolske, S. E. Strahan, G. V.
Ferry, K. R. Chan, J. F. Vedder, G. L. Qregory, W.
D. Hypes, M. P. McCormick, E. V. Browell, and L.
E. Heidt, Dehydration in the lower '• Antarctic
stratosphere during late winter and early spring,
1987,/. Geophys. Res., 94, 11317-11357,1989.
Kelly, K. K., A. F. Tuck, L. E. Heidt, M. Loev/enstein, J.
R. Podolske, S. E. Strahan, and J. F. Vedder, A
comparison of ER-2 measurements of stratospher-
ic water vapor between the 1987 Antarctic and
1989 Arctic Airborne Missions, Geophys. Res.
Lett., 17, 465-468, 1990.
Kent, G. S., L. R. Poole, M. P. McCormick, S. F.
Schaffner, W. H. Hunt, and M. T. Osbora, Optical
backscatter characteristics of Arctic polar strato-
spheric clouds, Geophys. Res. Lett., 17, :377-380
1990.
Keys, J. G., P. V. Johnston, R. D. Blatherwick, and F. J.
Murcray, Evidence for heterogeneous reactions in
the Antarctic autumn stratosphere, Nature, 361
49-51, 1993.
Kinne, S., O. B. Toon, and M. J. Prather, Buffering of
stratospheric circulation by changing amounts of
tropical ozone: A Pinatubo case study, Geophys.
Res. Lett., 19, 1927-1930, 1992.
Koike, M., N. B. Jones, W. A. Matthews, P. V. Johnston,
R. L. McKenzie, D. Kinnison, and J. Rodriguez,
Impact of Pinatubo aerosols on the partitioning
between NO2 and HNO3, Geophys. Res. Lett., 21
597-600, 1994. \
Kolb, C. E., D. R. Worsnop, M. S. Zahniser, >. Davi-
dovits, L. F. Keyser, M.-T. Leu, M. J. Molina,
D. R. Hanson, A. R. Ravishankara, L. R. Williams,
and M. R. Tolbert, Laboratory studies of atmo-
spheric heterogeneous chemistry, . in Current
Problems in Atmospheric Chemistry, edited by J.
R. Barker, in press, 1994. ;
Kondo, Y, P. Aimedieu, M. Koike, Y Iwasalca, P. A.
Newman, U. Schmidt, W. A. Matthews, >M. Ha-
yashi, and W. R. Sheldon, Reactive nitrogen,
ozone, and nitrate aerosols observed in the Arctic
stratosphere in January 1990,/. Geophys. Res., 97
13025-13038,1992. . ;
Kondo, Y., U. Schmidt, T. Sugita, P. Aimedieu, M.
Koike, H. Ziereis, and Y Iwasaka, Total reactive
nitrogen, N2O, and ozone in the winter Arctic
stratosphere, Geophys. Res. Lett., 21, 1247-1250
1994a.
Kondo, Y, W. A. Matthews, S. Solomon, M. Koike, M.
Hayashi, K. Yamaizaki, H. Nakajima, and' K.
Tsukui, Ground based measurements of column
amounts of NO2 ove:r Syowa Station, Antarctica, /
Geophys. Res., 99, 14535-14548, 1994b.
Krueger, A., M. R. Schoeiberl, P. A. Newman, and R. S.
Stolarski, The 1991 Antarctic ozone hole: TOMS
observations, Geophys. Res. Lett., 19 1215-1218
1992. !
Labitzke, K., On the variability of the stratosphere in the
Arctic regions in winter, Ber. Bunsenges. Phys
Chem., 96, 496-501, 1992.
Labitzke, K., and M. P. McCormick, Stratospheric tem-
perature increases due to Pinatubo aerosols,
Geophys. Res. Lett., 19, 207-210, 1992.
Lahoz, W. A., E. S. Carr, L. Froidevaux, R. S. Harwood,
J. B. Kumer, J. L. Mergenthaler, G. E. Peckham,
W. G. Read, P. D. Ricaud, A. E. Roche, and J. W.
Waters, Northern Hemisphere mid-stratosphere
vortex processes diagnosed from H2O, N2O, and
potential vorticity, Geophys. Res. Lett., 20, 2671-
2674, 1993. \
Larsen, N., B. Knudsen, I. S. Mikkelsen, T. S. J0rgensen,
and P. Eriksen, Ozone depletion in the Arctic
stratosphere in early 1993, Geophys. Res. Lett., 21
1611-1614,1994.
Lefevre, R, G. P. Brasseur, I. Folkins, A. K. Smith, and
P. Simon, The chemisliy of the 1991-92 stratospher-
ic winter Three-dimensional model simulations, /.
Geophys. Res., 99, 8183-8195, 1994.
Liu, X., R. D. Blatherwick, F. J. Murcray, J. G. Keys, and
S. Solomon, Measurements and model calcula-
tions of HC1 column amounts and related
parameters over McMurdo during the Austral
spring in 1989, J. Geophys. Res., 97, 20795-
20804, 1992. ;
Loewenstein, M., J. R. Podolske, K. R. Chan, and S. E.
Strahan, Nitrous oxide as a dynamical tracer in the
1987 Airborne Antarctic Ozone Experiment, /.
Geophys. Res., 94, 11589-11598, 1989.
3.45
-------
POLAR PROCESSES
Loewenstein, M., J. R. Podolske, D. W. Fahey, E. L.
Woodbridge, P. Tin, A. Weaver, P. A. Newman, S.
E. Strahan, S. R. Kawa, M. R. Schoeberl, and L. R.
Lait, New observations of the NOy/^O correla-
tion in the lower stratosphere, Geophys. Res. Lett.,
20, 2531-2534, 1993.
Lucic, D., N. R. P. Harris, J. A. Pyle, and R. L. Jones,
Diagnosing ozone loss in the Northern Hemi-
sphere for the 1991/92 winter, /. Geophys. Res., in
press, 1994.
Luo, B. P., Th. Peter, and P. J. Crutzen, Freezing of
stratospheric aerosol droplets, Geophys. Res. Lett,,
21, 1447-1450,1994a.
Luo, B. P., Th. Peter, and P. J. Crutzen, HC1 solubility
and liquid diffusion in aqueous sulfuric acid under
stratospheric conditions, Geophys. Res. Lett., 21,
49-52,1994b.
Lutman, E. R., J. A. Pyle, R. L. Jones, D. J. Lary, A. R.
MacKenzie, I. Kilbane-Dawe, N. Larsen, and B.
M. Knudsen, Trajectory model studies of C1OX ac-
tivation during the 1991/92 Northern Hemispheric
winter, Geophys. Res. Lett., 21,1419-1422,1994a.
Lutman, E. R., R. Toumi, R. L. Jones, D. J. Lary, and J.
A. Pyle, Box model studies of CIOX deactivation
and ozone loss during the 1991/92 Northern
Hemispheric winter, Geophys. Res. Lett., 21,
1415-1418, 1994b.
MacKenzie, A. R., B. M. Knudsen, R. L. Jones, and E. R.
Lutman, The spatial and temporal extent of chlo-
rine activation by polar stratospheric clouds in die
Northern Hemisphere winters of 1988/89 and
1991/92, Geophys. Res., Lett., 21, 1423-1426,
1994.
Mahlman, J. D., H. Levy H, and W. J. Moxim, Three-
dimensional tracer structure and behavior as
simulated in two ozone precursor experiments, /.
Atmos. Sci., 37, 655-685, 1980.
Mahlman, J. D., and L. Umscheid, Comprehensive mod-
eling of die middle atmosphere: The influence of
horizontal resolution, in Transport Processes in
the Middle Atmosphere, edited by G. Visconti and
R. Garcia, D. Reidel, Hingham, Massachusetts,
251-266,1987.
Mankin, M. G., M. T. Coffey, and A. Goldman, Airborne
observations of SO* HC1, and O3 in the strato-
spheric plume of die Pinatubo volcano in July
1991, Geophys. Res. Lett., 19, 179-182, 1992.
Manney, G. L., and R. W. Zurek, Intel-hemispheric com-
parison of the development of the stratospheric
polar vortex during fall: A 3-dimensional perspec-
tive for 1991-1992, Geophys. Res. Lett., 20,
1275-1278, 1993.
Manney, G. L., L. Froidevaux, J. W. Waters, L. S. Elson,
E. F. Fishbein, R. W. Zurek, R. S. Harwood, and
W. A. Lahoz, The evolution of ozone observed by
UARS MLS in die 1992 late winter southern polar
vortex, Geophys. Res. Lett., 20, 1279-128J2, 1993.
Manney, G. L., R. W. Zurek, A. O'Neill, R. Swinbank,
J. B. Kumer, J. L. Mergentiialer, and A. E. Roche,
Stratospheric warmings during February and
March 1993, Geophys. Res. Lett., 21, 813-816,
1994a. . ;
Manney, G. L., L. Froidevaux, J. W. Waters, R. W. Zurek,
W. G. Read, L. S. Elson, J. B. Kumer, J.!L. Mer-
genthaler, A. E. Roche, A. O'Neill, R. S. Harwood,
I. MacKenzie, and R. Swinbank, Chemical deple-
tion of lower stratospheric ozone in me 1992-1993
northern winter vortex, Nature, 370, 429-434,
1994b. >
Manney, G. L., R. W. Zurek, A. O'Neill, and R. Swin-
bank, On die motion of air through die
stratospheric polar vortex, /. Atmos. Sci., 51,2973-
2994, 1994c. |
McCormick, M. P., and R. E. Veiga, SAGE H measure-
ments of early Pinatubo aerosols, Geophys. Res.
Lett., 19, 155-158, 1992. '.
McElroy, M. B., R. J. Salawitch, S. C. Wofsy, and J. A.
Logan, Reduction of Antarctic ozone duetto syner-
gistic interactions of chlorine and bromine,
Nature, 321, 759-762, 1986. j
McKenna, D. S., R. L. Jones, L. R. Poole, S. Solomon,
D. W. Fahey, K. K. Kelly, M. H. Proffttt, W. H.
Brune, M. Loewenstein, and K. R. Chan, Calcula-
tions of ozone destruction during the! 1988/89
Arctic winter, Geophys. Res. Lett., 17, '553-556,
1990. :
Middlebrook, A. M., B. G. Koehler, L. S. McNeill, and
M. A. Tolbert, Formation of model polar strato-
spheric cloud films, Geophys. Res. Lett., 24,
2417-2420, 1992. ;
3.46
-------
POLAR PROCESSES
Middlebrook, A. M., L. T. Traci, L. S. McNeill, B. G.
Koehler, O. W. Saastad, and M. A. Tolbert, FUR
studies of thin H2SO4/H2O films: Formation, wa-
ter uptake, and solid-liquid phase changes, /.
Geophys. Res., in press, 1994.
Minton, T. K., C. M. Nelson, T. A. Moore, and M. Oka-
mu'ra, Direct observation of CIO from chlorine
nitrate photolysis, Nature, 328, 1342-1345, 1992.
Molina, L. T., and M. J. Molina, Production of Cl2O2
from the self-reaction of the CIO radical, J. Phys.
Chem., 97,433-436, 1987. !
Molina, M. J., R. Zhang, P. J. Wooldridge, J. R. McMa-
hon, J. E. Kim., H. Y. Chang, and IK. Beyer,
Physical chemistry of the H2SO4/HN03/H2O sys-
tem: Implications for the formation of polar
stratospheric clouds and heterogeneous chemistry,
Science, 261, 1418-1423, 1993. !
Miiller, R., and T. Peter, The numerical modeling of the
sedimentation of polar stratospheric cloud parti-
cles, Ber. Bunsenges. Phys. Chem., 96, 353-361
1992.
Murcray, F. J., J. R. Starkey, W. J. Williams, W A. Mat-
thews, U. Schmidt, P. Aimedieu,, and C.
Camy-Peyret, HNO3 profiles obtained during the
EASOE campaign, Geophys. Res. Lett.,. 21, 1223-
1226, 1994.
Nagatani, R. M., A. J. Miller, M. E. Gelman,and P. A.
Newman, A comparison of Arctic lower strato-
spheric winter temperatures for 1988-89 with
temperatures since 1964, Geophys. Res.' Lett., 17,
333-336,1990.
Newman, P. A., L. R. Lait, M. R. Schoeberl, Eii R. Nash,
K. K. Kelly, D. W. Fahey, R. M. Nagatani, D. W.
Toohey, L. M. Avallone, and J. G. Anderson,
Stratospheric meteorological conditions in the
Arctic polar vortex, 1991 to 1992, Science, 261,
1143-1146,1993. !:
Nielsen, J. E., R. B. Roo'd, A. R. Douglass, M. C.
Cerniglia, D. J. Allen, and J. E. Rosenfidd, Tracer-
evolution in winds generated by a global spectral
mechanistic model, J. Geophys. Res., 99, 5399-
5420, 1994. ; '
Norton, W. A., Breaking Rossby waves in a model strato-
sphere diagnosed by a vortex following system
and contour advection technique, /. Atmos. Sci
51, 654-673, 1993.
Ohtake, T, Freezing points of H2SO4 aqueous solutions
and formation of stratospheric ice clouds, Tellus
45B, 138-144, 1993.
Orsolini, Y. D., D. Cariolle, and M. Deque, A GCM
study of the late January 1992 'mini-hole' event
observed during EASOE, Geophys Res. Lett., 21
1459-1462, 1994.
Osborn, M. T., M. C. Pitts, K. A. Powell, and M. P. Mc-
Cormick, SAM n aerosol measurements during
• the 1989 AASE, Geophys. Res. Lett., 17, 397-400
1990. . ;
Perliski, L. M., S. Solomon, and J. London, On the inter-
pretation of the seasonal variations of stratospheric
ozone, Planet. Space Sci., 37, 1527-1538, 1989.
Perner, D., A. Roth, and T. Klupfel, Groundbased mea-
surements of stratospheric OC1O, NO2, and 03 at
S0ndre Str0mfjord in winter 1991/92, Geophys.
Res. Lett., 21, 1367-1370, 1994.
Peter, Th., C. Briihl, and P. J. Crutzen, Increase in the
PSC-formation probability caused by high-flying
aircraft, Geophys. Res. Lett., 18, 1465-1468,1991.
Peter, Th., R. Miiller, K. Drdla, K. Petzoldt, and E. Rei-
mer, A micro-physical box model for EASOE:
Preliminary results for the January/February 1990
PSC event over Kiiruna, Ber. Bunsenges. Phys.
Chem., 96, 362-367, 1992.
Petzoldt, K., B. Naujokat, and K. Neugebohren, Correla-
tion between stratospheric temperature, total
ozone, and tropospheric weather systems, Geo-
phys. Res. Lett., 21, 1203-1206, 1994.
Pierce, R. B., and T. D. A. Fairlie, Chaotic advection in
the stratosphere: Implications for the dispersal of
chemically perturbed air from the polar vortex, /.
Geophys. Res., 98, 18589-18595, 1993.
Pierce, R. B., W. L. Grose, J. M. Russell III, and A. F.
/ Tuck, Spring dehydration in the Antarctic vortex
observed by HALOE, / Atmos. Sci., 51, 2931-
2941, 1994.
Plumb, R. A., and M. K. W, Ko. Interrelationships be-
tween mixing ratio;; of long-lived stratospheric
constituents, J. Geophys. Res., 97, 10145-10156,
1992. I
3.47
-------
POLAR PROCESSES
Plumb, R. A., D. W. Waugh, R. J. Atkinson, P. A. New-
man, L. R. Lait, M. R. Schbeberl, E. V, Browell, A.
J. Simmons, and M. Loewenstein, Intrusions into
the lower stratospheric Arctic vortex during the
winter of 1991/92, /. Geophys. Res., 99, 1089-
1105,1994.
Podolske, J. R., M. Loewenstein, A. Weaver, S. E. Stra-
han, and K. R. Chan, Northern Hemisphere nitrous
oxide morphology during the 1989 AASE and the
1991-92 AASEII campaigns, Geophys. Res. Lett.,
20, 2535-2538,1993.
Poole, L. R., S. Solomon, M. P. McCormick, and M. C.
Pitts, The interannual variability of polar strato-
spheric clouds and related parameters in
Antarctica during September and October, Geo-
phys. Res. Lett., 16, 1157-1160, 1989.
Poole, L. R., and M. C. Pitts, Polar stratospheric cloud
climatology based on Stratospheric Aerosol Mea-
surement n observations from 1978 to 1989, /.
Geophys. Res., 99, 13083-13089, 1994.
Prather, M. J., Catastrophic loss of ozone in dense volca-
nic clouds, /. Geophys. Res., 97, 10187-10191,
1992.
Proffitt, M. H., D. W. Fahey, K. K. Kelly, and A. F. Tuck,
High-latitude ozone loss outside the Antarctic
ozone hole, Nature, 342, 233-237, 1989a.
Proffitt, M. H., K. K. Kelly, J. A. Powell, B. L. Gary, M.
Loewenstein, J. R. Podolske, S. E. Strahan, and K.
R. Chan, Evidence for diabatic cooling and pole-
ward transport within and around the 1987
Antarctic ozone hole,/. Geophys. Res., 94,16797-
16813, 1989b.
Proffitt, M. H., J. A. Powell, A. F. Tuck, D. W. Fahey, K.
K. Kelly, A. J. Krueger, M. R. Schoeberl, B. L.
Gary, J. J. Margitan, K. R. Chan, M. Loewenstein,
and J. R. Podolske, A chemical definition of the
boundary of the Antarctic ozone hole, J. Geophys.
Res., 94, 11437-11448,1989c.
Proffitt, M. H., J. J. Margitan, K. K. Kelly, M. Loewen-
stein, J. R. Podolske, and K. R. Chan, Ozone loss
in the Arctic polar vortex inferred from high-alti-
tude aircraft measurements, Nature, 347, 31-36,
1990.
Proffitt, M. H., S. Solomon, and M. Loewenstein, Com-
parison of 2-D model simulations of ozone and
nitrous oxide at high latitudes with stratospheric
measurements, J. Geophys. Res., 97, 939-944,
1992. j
Proffitt, M. H., K. Aikin, J. J. Margitan, M. Loewenstein,
J. R. Podolske, S. E. Strahan, and A. (Weaver,
Ozone loss inside the northern polar vortex during
the 1991-1992 winter, Science, 261, 1150-1154,
1993. '
Pyle, J. A., N. R. P. Harris, J. C. Farman, F. Ajmold, G.
Braathen, R. A. Cox, P. Faucon, R. L. Jones, G.
Megie, A. O'Neill, U. Platt, J.-P. Pommereau, U.
Schmidt, and F. Stordal, An overview of the
EASOE campaign, Geophys. Res. Lett., 21, 1191-
1194,1994. |
Reber, C. A., The Upper Atmosphere Research; Satellite,
EOS Transactions, 71, 1867-1878, 1990.
Reber, C. A., C. E. Trevathan, R. J. McNeal, and M. R.
Luther, The Upper Atmosphere Research Satellite
(UARS) mission, /. Geophys. Res., 98, 10643-
10647,1993. i
Reihs, C. M., D. M. Golden, and M. A. Tolbert, Nitric
acid uptake by sulfuric acid solutions under strato-
spheric conditions: Determination of Henry's law
solubility, /. Geophys. Res., 95, 16545-16550,
1990.
Rind, D., N. K. Balachandran, and R. Suozzo, Climate
change and the middle atmosphere. Part II: The
impact of volcanic aerosols, /. Climate, 5, 189-
208, 1992.
Rind, D., E.-W. Chiou, W. Chu, S. Oltmans, J.: Lerner, J.
Larsen, M. P. McCormick, and L. McMaster,
Overview of the Stratospheric Aerosol and Gas
Experiment II water vapor observations; Method,
validation, and data characteristics, J., Geophys.
Res., 98, 4835-4856, 1993. '
Roche, A. E., J. B. Kumer, J. L. Mergenthaler, G. A. Ely,
W. G. Uplinger, J. F. Potter, T. C. James; and L. W.
Sterritt, The Cryogenic Limb Array Etalon Spec-
trometer (CLAES) on UARS: Experiment
description and performance, /. Geophys. Res., 98,
10763-10775, 1993a. ,
3.48
-------
Roche, A. E., J. B. Kumer, and J. L. Me rgenthaler,
CLAES observations of ClONOa and HNO3 in the
Antarctic stratosphere, between June 15 and Sep-
tember 17, 1992, Geophys. Res. Lett., 20,
1223-1226, 1993b. ;;
Roche, A. E., J. B. Kumer, J. L. Mergenthaler, R. W.
Nightingale, W. G. Uplinger, G. A. Ely, J. R Potter,
D. J. Wuebbles, P. S. Cornell, and D. E: Kinnison,
Observations of lower stratospheric ClONOa,
HNO3, and aerosol by the UARS CLAES experi-
ment between January 1992 and April 1993, J.
Atmos. Set., 51, 2877-2902, 1994. [
Rood, R. B., J. E. Nelson, R. S. Stolarski, A. !R. Doug-
lass, J. A. Kaye, and D. J. Allen, Episodic total
ozone minima and associated effects on heteroge-
neous chemistry and lower stratospheric transport,
J. Geophys. Res., 97, 7979-7996, 1992.
Rood, R. B., A. R. Douglass, J. A. Kaye, and D. B. Con-
sidine, Characteristics of wintertime arid autumn
nitric acid chemistry as defined by Limb Infrared
Monitor of the Stratosphere (LIMS) data, J. Geo-
phys. Res., 98, 18533-18545, 1993.
Rosenfield, J. E., M. R. Schoeberl, and M. A. Geller, A
computation of the stratospheric diabatic circula-
tion using an accurate radiative transfer model, J.
Atmos. Sci., 44, 859-876, 1987. \'
Rosenfield, J. E., P. A. Newman, and M. R. Schoeberl,
Computations of diabatic descent in the strato-
spheric polar vortex, /. Geophys. Res., 99
16677-16689,1994. r
Russell, J. M., in, L. L. Gordley, J. H. Park, S. JR. Dray-
son, W. D. Hesketh, R. J. Cicerone, A. F. Tuck, J.
E. Frederick, J. E. Harries, and P. J. Crutzen, The
Halogen Occultation Experiment, /. Geophys.
Res., 98, 10777-10797, 1993a. ;
Russell, J. M., ffl, A. F. Tuck, L. L. Gordley, J, H. Park,
S. R. Drayson, J. E. Harries, R. J. Cicerone, and P.
J. Crutzen, HALOE Antarctic observations in the
spring of 1991, Geophys. Res. Lett., 20, 719-722,
1993b.
Salawiteh, R. J., S. C. Wofsy, M. B. McElroy, Chemistry
of OCIO in the Antarctic stratosphere: Implica-
tions for bromine, Planet. Space Sci., 36, 213-224
1988.
POLAR PROCESSES
Salawiteh, R. J., G. P. Gobbi, S. C. Wofsy, and M. B.
McElroy, Denitrification in the Antarctic strato-
sphere, Nature, 339, 525-527, 1989.
Salawiteh, R. J., M. B. McElroy, J. H. Yatteau, S. C.
Wofsy, M. R. Schoeberl, L. R. Lait, P. A. Newman,
K. R. Chan, M. Loewenstein, J. R. Podolske, S. E.
Strahan, and M. H. Proffitt, Loss of ozone in the
Arctic vortex for the winter of 1989, Geophys.
Res. Lett., 17,561-564, 1990.
Salawiteh, R. J., S. C. Wofsy, E. Gottlieb, L. R. Lait, P. A.
Newman, M. R. Schoeberl, M. Loewenstein, J. R.
Podolske, S. E. Strahan, M. H. Proffitt, C. R. Web-
ster, R. D. May, D. W. Fahey, D. Baumgardner, J.
E. Dye, J. C. Wilson, K. K. Kelly, J. W. Elkins, K.
R. Chan, and J. G.:.Anderson, Chemical loss of
ozone in the Arctic ;polar vortex in the winter of
1991-1992, Science, 261, 1146-1149, 1993.
Sanders, R. W, S. Solomon, J. P. Smith, L. Perliski, H.
L. Miller, G. H. Mount, J. G. Keys, and A. L.
t Schmeltekopf, Visible and near-ultraviolet spec-
troscopy at McMurdo Station, Antarctica 9.
Observations of OCIO from April to October
1991, /. Geophys. Res., 98, 7219-7228, 1993.
Santee, M. L., W. G. Read, J. W. Waters, L. Froidevaux,
G. L. Manney, D. A. Rower, R. F. Jarnot, R. S.
Harwood, and G. E. Peckham, Interhemispheric
differences in polar stratospheric HNO3, H2O,
CIO, and O3 from UARS MLS, submitted to Sci-
ence, 1994. i
Schauffler, S. M., L. E. Heidf, W. H. Pollock, T. M.
Gilpin, J. F. Vedder, S. Solomon, R. A. Lueb, and
E. L. Atlas, Measurements of halogenated organic
compounds near the tropical tropopause, Geophys.
Res. Lett., 20, 2567-2570, 1993.
Schiller, C., A. Wahner, U. Platt, H.-P. Dorn, J. Callies,
and D. H. Ehhalt, Near UV atmospheric absorp-
tion measurements of column abundances during
Airborne Arctic Stratospheric Expedition, Janu-
ary-February 1989: 2. OCIO observations,
Geophys. Res. Lett., 17, 501-504, 1990.
Schlager, H., and F. Arnold, Measurements of strato-
spheric gaseous nitric acid in the winter Arctic
vortex using a novel rocket-borne mass spectro-
metric method, Geophys. Res. Lett., 17, 433-436,
1990. :
3.49
-------
POLAR PROCESSES
Schmidt, U., R. Bauer, A. Khedim, E. Klein, G. Kulessa,
and C. Schiller, Profile observations of long-lived
trace gases in the Arctic vortex, Geophys. Res.
Lett., 18, 767-770, 1991.
Schmidt, U., R. Bauer, A. Engel, R. Borchers, and J. Lee,
The variation of available chlorine, Cly, in the Arc-
tic polar vortex during EASOE, Geophys. Res.
Lett., 21, 1215-1218,1994.
Schoeberl, M. R., M. H. Proffitt, K. K. Kelly, L. R. Lait,
P. A. Newman, J. E. Rosenfield, M. Loewenstein,
J. R. Podolske, S. E. Strahan, and K. R. Chan,
Stratospheric constituent trends from ER-2 profile
data, Geophys. Res. Lett., 17, 469-472, 1990.
Schoeberl, M. R., and D. L. Hartmann, The dynamics of
the stratospheric polar vortex and its relation to
springtime ozone depletions, Science, 251, 46-52,
1991.
Schoeberl, M. R., L. R. Lait, P. A. Newman, and J. E.
Rosenfield, The structure of the polar vortex, J.
Geophys. Res., 97, 7859-7882, 1992.
Schoeberl, M. R., A. R. Douglass, R. S. Stolarski, P. A.
Newman, L. R. Lait; D. W. Toohey, L. M. Aval-
lone, J. G. Anderson, W. H. Brune, D. W. Fahey,
and K. K. Kelly, The evolution of CIO and NO
along air parcel trajectories, Geophys. Res. Lett.,
20,2511-2514,1993a.
Schoeberl, M. R., R. S. Stolarski, A. R. Douglass, P. A.
Newman, L. R. Lait, J. W. Waters, L. Froidevaux,
and W. G. Read, MLS CIO observations and Arctic
polar vortex temperatures, Geophys. Res. Lett., 20,
2861-2864,1993b.
Schoeberl, M. R., M. Luo, and J. E. Rosenfield, An anal-
ysis of the Antarctic . HALOE trace ,gas
observations, J. Geophys. Res., in press, 1994.
Shine, K. P., Sources and sinks of zonal momentum in
the middle atmosphere diagnosed using the diabat-
ic circulation. Quart. J. Roy. Meteorol. Soc., 115,
265-292,1989.
Solomon, S., Progress towards a quantitative under-
standing of Antarctic ozone depletion, Nature,
347, 347-354,1990.
Solomon, S., R. R. Garcia, F. S. Rowland, and D. J.
Wuebbles, On the depletion of Antarctic ozone,
Nature, 321, 755-758, 1986.
Solomon, S., R. W. Sanders, M. A. Carroll, and A. L.
Schmeltekopf, Visible and near-ultraviolet spec-
troscopy at McMurdo Station, Antarctica 5.
Observations of the diurnal variations of BrO and
OC10,/. Geophys. Res., 94, 11393-11403, 1989.
Solomon, S., and J. G. Keys, Seasonal variations in Ant-
arctic NOX chemistry, J. Geophys. Res., 97,
7971-7978,1992. |
Solomon, S., R. W. Sanders, R. R. Garcia, and J. G.
Keys, Increased chlorine dioxide over Antarctica
caused by volcanic aerosols from Mount Pinatubo,
Nature, 363, 245-248, 1993. ;
Strahan, S. E., and J. D. Mahlman, Evaluation of the
SKYHI general circulation model using aircraft
N2O measurements: 1. Polar winter stratospheric
meteorology and tracer morphology, J.'Geophys.
Res., 99, 10305-10318, 1994a.
Strahan, S. E., and J. D. Mahlman, Evaluation of the
SKYHI general circulation model using aircraft
N2O measurements: 2. Tracer variability and dia-
batic meridional circulation, /. Geophys. Res., 99,
10319-10332, 1994b. I
Strahan, S. E., J. E. Rosenfield, M. Loewenstein, J. R.
Podolske, and A. Weaver, Evolution of; the 1991-
1992 Arctic vortex and comparison; with the
GFDL SKYHI general circulation model, J. Geo-
phys. Res., 99, 20713-20723, 1994.
Tabazadeh, A., and R. P. Turco, Stratospheric chlorine
injection by volcanic eruptions: HC1 scavenging
and implications for ozone, Sc.ience, 260, 1082-
1086, 1993. ;
Tabazadeh, A., R. P. Turco, and M. Z. Jacobson, A model
for studying the composition and chemical effects
of stratospheric aerosols, J. Geophys. Res., 99,
12897-12914, 1994.
Tao, X., and A. F. Tuck, On the distribution 'of cold air
near the vortex edge in the lower stratosphere, J.
Geophys. Res., 99, 3431-3450, 1994.
Taylor, F. W., C. D. Rodgers, J. G. Whitney, S.T. Werrett,
J. J. Barnett, G. D. Peskett, P. Venters, J. Ballard,
C. W. P. Palmer, R. J. Knight, P. Morris, T. Night-
ingale, and A. Dudhia, Remote sensing of
atmospheric structure and composition by pres-
sure modulator radiometry from space: The
ISAMS experiment on UARS, /. Geophys. Res.,
98, 10799-10814, 1993. ;
3.50
-------
POLAR PROCESSES
Toohey, D. W., J. G. Anderson, W. H. Brune, and ;K. R.
Chan, In situ measurements of BrO in the Arctic
stratosphere, Geophys. Res. Lett., 17, 513-516,
1990.
Toohey, D. W., L. M. Avallone, L. R. Lait, P. A. Newman,
M. R. Schoeberl, D. W. Fahey, E. L. Woodbridge,
and J. G. Anderson, The seasonal evolution of re-
active chlorine in the Northern Hemisphere
stratosphere, Science, 261, 1134-1136,1993.
Toon, G. C, C. B. Farmer, L. L. Lowes, P. W. Schaper,
J.-F. Blavier, and R. H. Norton, Infrared aircraft
measurements of stratospheric composition over
Antarctica during September 1987, J. Geophys.
Res., 94, 16571-16596, 1989a.
Toon, G. C, C. B. Farmer, P. W. Schaper, L. L. L,owes,
and R. H. Norton, Composition measurements of
the 1989 Arctic winter stratosphere by airborne in-
frared solar absorption spectroscopy, /. Geophys.
Res., 97, 7939-7962, 1992.
Toon, O. B., R. P. Turco, J. Jordan, J. Goodman, zmd G.
Ferry, Physical processes in polar stratospheric ice
clouds,/ Geophys. Res., 94, 11359-11380,1989b.
Toon, O. B., E. V. Browell, S. Kinne, and J. Jordan, An
analysis of lidar observations of polar stratospher-
ic clouds, Geophys. Res., Lett., 17, 393-396,
1990a.
Toon, O. B., R. P. Turco, and P. Hamill, Denitrification
mechanisms in the polar stratospheres, Geophys.
Res., Lett., 17, 445-448, 1990b. ;
Toumi, R., R. L. Jones, and J. A. Pyle, Stratospheric
ozone depletion by ClONOa photolysis, Nature,
365, 37-39, 1993. :
Trepte, C. R., R. E. Veiga, and M. P. McCormick, The
poleward dispersal of Mount Pinatubo volcanic
aerosol,/. Geophys. Res., 98, 18563-18573,1993.
Tuck, A. F., Synoptic and chemical evolution of ttob Ant-
arctic vortex in late winter and early spring, 1987,
/. Geophys. Res., 94, 11687-11737, 1989. .
Tuck, A. F., T. Davies, S. J. Hovde, M. Noguer-Aliba, D.
W. Fahey, S. R. Kawa, K. K. Kelly, D. M. Murphy,
M. H. Proffitt, J. J. Margitan, M. Loewenstein, J.
R. Podolske, S. E. Stratum, and K. R. Chan, Polar
stratospheric cloud processed air and potential
vorticity in the Northern Hemisphere lower Sitrato-
sphere at mid-latitudes during winter, /. Geophys.
Res., 97, 7883-7904, 1992.
Tuck, A. F., J. M. Russell inland J. E. Harries, Strato-
spheric dryness: Antiphased desiccation over
Micronesia and Antarctica, Geophys. Res. Lett.,
20, 1227-1230,1993.
Tuck, A. F., D. W. Fahey, M. Loewenstein, J. R. Podol-
ske, K. K. Kelly, S. J. Hoyde, D. M. Murphy, and J.
W. Elkins, Spread of denitrification from 1987
Antarctic and 1988-1989 Arctic stratospheric vor-
tices, J. Geophys. Res., 99, 20573-20583,1994.
Tung, K. K., M. K. W. Ko, J. M. Rodriguez, and N.-D.
Sze, Are Antarctic ozone variations a manifesta-
tion of dynamics or chemistry?, Nature, 322,
811-814,1986. ;
Van Doren, J. M., L. R. Watson, P. Davidovits, D. R.
Worsnop, M. S. Zahniser, and C. E. Kolb, Uptake
of N2Os and HNQs by a2 observations,
Geophys. Res. Lett., 17, 497-500, 1990a.
Wahner, A., J. Callies, H.-P. Dom, U. Platt, and C.
Schiller, Near UV atmospheric absorption mea-
surements of column abundances during Airborne
Arctic Stratospheric Expedition, January-Febru-
ary 1989: 3. BrO observations, Geophys. Res.
Lett., 17, 517-520, 1990b.
3.51
-------
POLAR PROCESSES
Wallace, L., and W. Livingston, The effect of the Pinatu-
bo cloud on hydrogen chloride and hydrogen
fluoride, Geophys. Res. Lett., 19, 1209, 1992.
Waters, J. W., L. Froidevaux, W. G. Read, G. L. Manney,
L. S. Elson, D. A. Flower, R. F. Jarnot, and R. S.
Harwood, Stratospheric CIO and ozone from the
Microwave Limb Sounder on the Upper Atmo-
sphere Research Satellite, Nature, 362, 597-602,
1993a.
Waters, J. W, L. Froidevaux, G. L. Manney, W. G. Read,
and L. S. Elson, MLS observations of lower strato-
spheric CIO and O3 in the 1992 Southern
Hemisphere winter, Geophys. Res. Lett., 20, 1219-
1222, 1993b.
Waugh, D. W., R. A. Plumb, R. J. Atkinson, M. R.
Schoeberl, L. R. Lait, P. A. Newman, M. Loewen-
stein, D. W. Toohey, L. M. Avallone, C. R.
Webster, and R. D. May, Transport of material out
of the stratospheric Arctic vortex by Rossby wave
breaking,/. Geophys. Res., 99, 1071-1088, 1994.
Weaver, A., M. Loewenstein, J. R. Podolske, S. E. Stra-
han, M. H. Proffitt, K. Aikin, J. J. Margitan, H. H.
Jonsson, C. A. Brock, J. C. Wilson, and O. B.
Toon, Effects of Pinatubo aerosol on stratospheric
ozone at mid-latitudes, Geophys. Res. Lett., 20,
2515-2518,1993.
Webster, C. R., R. D. May, D. W. Toohey, L. M. Aval.-
lone, J. G. Anderson, P. A. Newman, L. R. Lait, M.
R. Schoeberl, J. W. Elkins, and K. R. Chan, Chlo-
rine chemistry on polar stratospheric cloud
particles in the Arctic winter, Science, 261, 1130-
1134,1993a.
Webster, C. R., R. D. May, D. W. Toohey, L. M. Aval-
lone, J. G. Anderson, and S. Solomon, In situ
measurements of the C10/HC1 ratio: Heteroge-
neous processing on sulfate aerosols and polar
stratospheric clouds, Geophys. Res. Lett., 20,
2523-2526, 1993b.
Wilson, J. C., H. H. Jonsson, C. A. Brock, D. W. Toohey,
L. M. Avallone, D. Baumgardner, J. E. Dye, L. R.
Poole, D. C. Woods, R. J. DeCoursey, M. Osbora,
M. C. Pitts, K. K. Kelly, K. R. Chan, G. V. Ferry,
M. Loewenstein, J. R. Podolske, and A. Weaver, In
situ observations of aerosol and chlorine monox-
ide after the 1991 eruption of Mount Pinatubo:
Effects of reactions on sulfate aerosol, Science,
' 261, 1140-1143, 1993.
WMO, Scientific Assessment of Ozone Depletion: 1991,
Global Ozone Research and Monitoring Project—
Report No. 25, World Meteorological Organiza-
tion, Geneva, 1992. :
Wofsy, S. C., G. P. Gobbi, R. J. Salawitch, and M. B.
McElroy, Nucleation and growth of HNQa-SHaO
particles in the polar stratosphere, /. Atmps. ScL,
47, 2004-2012, 1990a. ;
Wofsy, S. C., R. J. Salawitch, J. H. Yatteau, M. B. McEl-
roy, B. W. Gandrad, J. E. Dye, and D. Baumgardner,
Condensation of HNOs on falling ice particles:
Mechanism for denitrification of the polar strato-
sphere, Geophys. Res. Lett., 17, 449-452,1990b.
Woodbridge, E. L., J. W. Elkins, D. W. Fahey, L. E.
Heidt, S. Solomon, T. J. Baring, T. J. Gilpin, W. H.
Pollock, S. M. Schauffler, E. L. Adas, M. Loewen-
stein, J. R. Podolske, C. R. Webster, R. D. May, R.
J. Salawitch, S. C. Wofsy, J. M. Gilligan, and S. A.
Montzka, Estimates of total organic and inorganic
chlorine in the lower stratosphere from in situ
measurements during AASE IL /. Geophys. Res.,
in press, 1994.
Worsnop, D. R., L. E. Fox, M. S. Zahniser, and S. C.
Wofsy, Vapor pressures of solid hydrates of nitric
acid: Implications for polar stratospheric clouds,
Science, 259,71-74, 1993.
Zander, R., M. R. Gunson, C. B. Farmer, C. P. Rinsland,
F. W. Irion, and E. Mahieu; The 1985 chlorine and
fluorine inventories in the stratosphere based on
ATMOS observations at 30 north latitude, /. Atm.
Chem., 15, 171-186, 1992. \
Zhang, R., P. J. Wooldridge, J. P. D. Abbatt, and M. J.
Molina, Physical chemistry of the HjSO^aO bi-
nary system at low temperatures: Stratospheric
implications, J. Phys. Chem., 97, 7351-7358,
1993a. ;
Zhang, R., P. J. Wooldridge, and M. J. Molina, Vapor
pressure measurements for the H2SO4/HNO3/
H2O and H2SO4/HC1/H2O systems: Incorporation
of stratospheric acids into background sulfate
aerosols, J. Phys. Chem., 97, 8541-8548,' 1993b.
Zhang, R., J. T. Jayne, and M. J. Molina, Heterogeneous
interactions of ClONOa and HC1 with sulfiiric acid
tetrahydrate: Implications for the stratosphere, /.
Phys. Chem., in press, 1994\
5.52
-------
CHAPTER 4
Tropical and Midlatitude Ozone
Lead Author:
R.L. Jones
Co-authors:
L. Avallone.
L. Froidevaux
S. Godin
L. Gray
S. Kinne
M.E. Mclntyre
P.A. Newman
R. A. Plumb
J.A. Pyle
J.M. Russell III
M. Tolbert
R. Toumi
A.F. Tuck
P. Wennberg
Contributors:
R. Cebula
S. Chandra
E. Fleming
L. Flynn
S. Hollands worth
C. Jackman
L.R. Poole
-------
-------
:! CHAPTER 4
TROPICAL AND MIDLATITUDE OZONE 1
Contents i
SCIENTIFIC SUMMARY ............................ . !
•*•.•...................
4.1 GENERAL INTRODUCTION ........... . [[[ 4 3
i t
I. CHEMICAL PROCESSES INFLUENCING MIDDLE LATITUDE AND TROPICAL OZONE ..................... 4.3
4.2 INTRODUCTION ............................... :.j. [[[ 4 3
4.2. 1 Laboratory Studies of Photochemistry and Gas Phase Kinetics .................. : .......... 43
4.2.2 Heterogeneous Processes ........... i. [[[ 45
4.2.3 Atmospheric Observations ........ '.'. ...................... . ........................................... ; 47
4.2.3.1 NOx/NOy Ratio ........... ;; .................................... ; ..................... IZIZZZZ'Z" '" ............ 47
4.2.3.2 Partitioning of Radical Species ............................................... .. '. 47
4.3 ERUPTION OF MOUNT PINATUBO [[[ . 4 12
4.3.1 Effects on Chemical Composition ......... . [[[ ; _ 4 13
4.3.2 Implications for the Normal State of the Atmosphere ................................. ; 4 15
4.4 PHOTOCHEMICAL OZONE LOSS PROCESSES AT MIDLATITUDES ........... . ................................... . 4.15
4.5 THE SOLAR CYCLE AND QUASI-BieJNIALOSOLLAnONCQBO) EFFECT ON TOTAL OZOhffi ........... 4.16
4.5.1 Solar Ultraviolet Variability and Total Ozone ............................................... 4 jg
4.5.2 The Quasi-Biennial Oscillation and Total Ozone .......................................... 4 17
II. TRANSPORT PROCESSES LINKING THE TROPICS, MIDDLE, AND HIGH LATITUDES ................... 4.18
4.6 INTRODUCTION ................... : ............ ''•}. [[[ [ _ 4 18
4.6.1 Transport of Air from the Tropics to Middle Latitudes ................................. >. 4 lg
4.6.2 The Mount Pinatubo Eruption: Implications for Understanding of Transport Processes ................. '.. 4.19
4.6;2. 1 Tropical Latitudes ......... .. ........... ; [[[ ; __ : 4 \g
4.6.2.2 Middle and High Latitudes ........ . [[[ . 4 20
4.6.3 Circulation-Induced Ozone Changes Resulting from the Mount Pinatubo Eruption .......................... 4.21
4.6.3. 1 Radiative Effects of Stratospheric Aerosol ..................................... ; .................................... 4 2 1
4.6.3.2 Heating by Mount Pinatubo Aerosols ............................................. 1 ...................................... 4 22
4.6.3.3 Aerosol Heating and Induced Response [[[ -. 4 22
i f 5
4.7 TRANSPORT OF A/R FROM POLAR REGIONS TO MIDDLE LATITUDES ... ' .................... . ................. 4.23
- 4.7.1 Transport of Air from High Latitudes: Possible Influence on Midlatitude Ozone Loss ..................... 4.23
4.7.2 Fluid-Dynamical Considerations \ [[[ : ...................................... 423
4.7.3 Observational Studies Relating to Transport through the Vortex ................... : ........................... 4.25
-------
-------
TROPICAL/MIDLATITUDE PROCESSES
SCIENTIFIC SUMMARY
Since the last Assessment, much new information has been obtained about the photochemical and dynamical
processes that influence ozone concentrations at middle latitudes. Measurements in the lower stratosphere have signif-
icantly increased our confidence in the basic gas phase and heterogeneous processes affecting ozone at middle latitudes,
although some discrepancies still exist. Laboratory studies have provided data that have led to an improved quantifica-
tion of the photochemical processes that affect ozone at middle latitudes. Understanding of the dynamical factors
influencing middle latitudes has improved, although significant uncertainties remain. The relative contributions of
these different processes to the ozone trends observed middle latitude are still poorly understood and important uncer-
tainties are still outstanding.
The major new findings are: ; • :
Photochemical Processes ;
Observations, coupled with photocheihical model calculations, have established with little doubt the role of the
heterogeneous hydrolysis of ^65 on the sulfate aerosol layer. However, there are instances where discrepancies
still arise, and it is unclear whether these reflect deficiencies in modeling known chemistry (e.g., imperfect knowl-
. edge of aerosol surface areas, photolysis rates, etc.) or, more profoundly, missing chemical processes.
• Measurements of radical species in the low stratosphere have provided direct confirmation that in situ photochem-
ical ozone loss in the lower stratosphere at midlatitudes is dominated by HOX and (man-made) halogen chemistry,
and not by (largely natural) NOX chemistry. Nevertheless, NOX chemistry exerts an important control on the
effectiveness of the halogen loss cycles. Current photochemical models can reproduce observed radical concen-
tration changes and coupling between different chemical families, provided that heterogeneous reactions are
incorporated and that the source gases !are suitably constrained by observations. '
• Satellite and in situ measurements of chlorine monoxide (CIO) concentrations iri the low stratosphere at middle
latitudes in both hemispheres show the existence of a seasonal cycle with maximum CIO during winter months.
This variation appears to be broadly consistent with changes in NOX due to in situ heterogeneous processes but
does not appear consistent with the timing of springtime vortex dilution or wintertime flow through the vortex.
i '
• There is evidence that the hydrolysis of chlorine nitrate (C1ONO2) on sulfate aerosols can occur at low tempera-
tures and may be important in middle latitudes under high aerosol loading conditions.
i '
• There are unresolved discrepancies beitween models and observations regarding the partitioning between reser-
voir and reactive species, notably the ratio of C1OX to HC1. Even when constrained by observed source gas fields
and radical species, photochemical models in the low stratosphere significantly overestimate observed HC1
amounts. In the uppei stratosphere models overestimate the C1O/HC1 ratio. - »
(
Laboratory Studies
;
• Recent studies have confirmed that N^iOg hydrolysis on sulfate aerosol surfaces is fast and occurs readily under
most stratospheric conditions, .while ructions that lead directly to chlorine activation depend strongly on atmo-
spheric temperature and humidity.
• The rate of the reaction of BrO with HO2 has been revised upwards by a factor of 6, implying a much larger
bromine-catalyzed ozone loss in the low stratosphere. !
4.1
-------
TROPICAL/MIDLATITUDE PROCESSES ;
Quenching rates for vibrationally excited O2 appear to be faster than previously thought, reducing [the likely
importance of the photolysis of vibrationally excited O2 as a source of ozone in the upper stratosphere. The
discrepancy between observed and modeled ozone in this region still persists.
Dynamical Processes i
• The transport of air from polar regions has the potential to influence ozone concentrations at middle; latitudes.
While there are uncertainties about the relative contributions of transport and in situ chemistry for niidlatitude
ozone loss, both processes directly involve ozone destruction by bromine- and chlorine-catalyzed reactions.
• Observations and models indicate that, above about 16 km in winter, air at midlatitudes is mixed relatively effi-
ciently and that influx of air from the tropics and from the interior of the polar vortex is weak. However, the
importance of the erosion of ah- from the edge of the polar vortex relative to in situ chemical effects for midlati-
tude ozone loss is poorly known.
• Below 16 km, air is more readily transported between polar regions and midlatitudes. The influence of this
transport on midlatitude ozone loss has not beera quantified.
Eruption of the Mt. Pinatubo Volcano
• The eruption of Mt. Pinatubo in 1991 led to a massive increase in sulfate aerosol in the lower stratosphere. There
is compelling evidence that this led to significant, but temporary, changes in the partitioning of NO|X, reactive
halogen compounds, and abundances of HOX in the low and mid-stratosphere at middle latitudes in such a way as
to accelerate photochemical ozone loss. However, there is also evidence that circulation changes associated with
heating on Mt. Pinatubo aerosols led to significant changes in the distribution of ozone in the tropics and middle
latitudes. Changes in photolysis rates arising directly from the presence of volcanic aerosols are also 'thought to
have affected ozone amounts. . ;
4.2
-------
TROPICAL/MIDLATITUDE PROCESSES
4.1 GENERAL INTRODUCTION
In the light of the observed trends in ozone away
from polar regions, a wide range of observational and
modeling studies have been focused on the midlatitude
lower stratosphere. A large number of dynamical and
photochemical mechanisms that can influence the con-
centrations of stratospheric ozone have been identified.
Those processes that are now thought to be the more
important for midlatitude ozone loss are shown schemat-
ically in Figure 4-1. Assessments of the importance of
chemical and dynamical processes are given in Sections
I and n respectively.
I. CHEMICAL PROCESSES INFLUENCING MIDDLE
LATITUDE AND TROPICAL OZONE
4.2 INTRODUCTION
The main photochemical processes that are
thought to be important in midlatitude oz:pne photo-
chemistry are shown schematically in Figure 4-1. The
diagram is intended to show the winter months when the
polar vortex is well established. ;
Ozone is produced by the photolysis of O2 at
wavelengths shorter than 242 nm to give oxygen atoms,
followed by recombination (1). Variations in the solar
output, for example during the 11-year solar cycle, lead
directly to small changes in the photolysis of O2 and thus
to a correlated change in ozone amounts (2)j Catalytic
ozone loss occurs through a range of gas phase chemical
cycles (3), those currently thought to be most important
in die low stratosphere at midlatitudes being shown in
the figure. It is known dial at middle latitudes the hy-
drolysis of N2O5 (4) can proceed effectively on sulfate
aerosols, reducing the available NOX and indirectly in-
creasing the degree of chlorine activation. At the lower
temperatures present at higher latitudes, the hydrolysis
of chlorine nitrate (C1ONO2) can occur (5), leading di-
rectly to increased chlorine activation. In ,the colder
polar regions, chlorine activation on polar stratospheric
clouds may also occur (6). Processes (4) and ;(5) are de-
pendent on the aerosol loading in the stratosphere, and
thus on the level of volcanic activity.
Processes (4) and (5) have the effect of altering the
balance of photochemical ozone loss by the different
chemical cycles shown, reducing the effectiveness of the
NOx-pnly cycles in the low stratosphere in favor of the
HOx-only and coupled HOx-halogen cycles. Informa-
tion leading to this picture is discussed below.
4.2.1 Laboratory Studies of Photochemistry
and Gas Phase Kinetics
Several new laboratory measurements of rate pa-
rameters and absorption cross sections (DeMore et ai,
1992) are of direct consequence for understanding ozone
loss in the tropics and midlatitudes.
The photolysis of nitric acid (HNOs) in the atmo-
sphere is important because it affects NOX concentrations
and thus, indirectly, C1O,; amounts. The temperature de-
pendence of the HNC>3 absorption cross section (Ratti-
gan et al., 1992; Burkholder et ai, 1993) and the
wavelength dependence :of the hydroxyl radical (OH)
quantum yield from HNO3 photolysis (Turnipseed et ai,
1992; Schiffman et ai, 1993) have now been measured.
The OH yield was confiirmed to be nearly unity at long
wavelengths, but other (products, such as HONO, be-
come more important at wavelengths shorter than 250
nm. The absorption cross section shows a temperature
dependence (smaller at lower temperatures) that is stron-
gest at wavelengths longer than 300 nm. As a result, the
greatest effect on the calculated photolysis rate occurs at
altitudes below about 28 km (Burkholder et ai., 1993).
These new data will yield more accurate calculations of
the HNOs photolysis rate in photochemical models, but
the magnitude of the effect will depend on the prior for-
mulation used by each model.
The product yield from C1ONO2 photolysis at 193
and 248 nm has also been investigated (Minton et al.,
1992), indicating that die products are split roughly
evenly between CIO + NO2 and Cl + NO3. In contrast to
prior measurements, no evidence was found for O-atom
formation. However, it must be recognized that most
C1ONO2 photolysis takes place at wavelengths longer
than -280 nm, where different products may form. If a
similar effect were to be present at longer wavelengths,
the result would be a reduction in the efficiency of ozone
destruction from C1ONO2 photolysis (Toumi et al.,
1993c). In addition, the quantum yields for NOs photol-
ysis between 570 and 635 nm have recently been
remeasured (Orlando et al., 1993) and give photolysis
rates in reasonable agreement with the currently recom-
4.3
-------
TROPICAUMIDLATITUDE PROCESSES
s
N,O5 hydrolysis
(leading indirectly
chlorine activation
'!"
S
o "c
C10NO,
hydrolysis
(leading directly
chlorine activatio
'•>
in
Partial barrier
to horizontal
transport
ANA
A
4S
0) CD
N C
O .O
0> t)
•o o>
c
JO CO
Ett
Ł CO
•B 03
8§
.Ł Q.
W "5
lo
2 =6
~ CO
J= >_
g>0
O O)
JZ C
«'§
-n
o X
S cB
"5 •*
a c
-C
-------
TROPICAUMIDLATITUDE PROCESSES
mended values (DeMore et al., 1992) but that are proba-
bly temperature dependent. :
Increases in calculated photolysis rates and, hence,
reductions in calculated lifetimes for species such as
chlorofluorocarbons (CFCs) and nitrous oxide (N2O)
may be expected following adoption of more accurate
cross sections for oxygen in the Schumann-Rurige bands
calculated using line-by-line methods (Minschwaner et
al., 1992;ToumiandBekki, 1994). Recent estimates for
N2O, CF2C12 (CFC-12), and CFC13 (CFC-11) are 123,
116, and 44 years, respectively (Minschwaner et al.,
1993).
A recent measurement has shown that the room-
temperature rate constant for the reaction of BrO with
HO2 is a factor of 6 larger than previously determined
(Poulet et al., 1992). Combined with an estimated tem-
perature dependence based on the HO2 + CIO reaction
(DeMore et al., 1992), this new determination dramati-
cally increases the importance of bromine-catalyzed
ozone loss, particularly in the 15 to 20 km region. The
magnitude of possible HBr production from this reaction
is currently under scrutiny. However, atmospheric HBr
observations by Traub et al. (1992) suggest that only 2 ±
2 pptv of HBr is present- at 32 km, implying that this
channel must be slow.
Some unresolved discrepancies between observa-
tions and models exist for the partitioning of inorganic
chlorine species in the stratosphere that could impact
model predictions of ozone trends.
In the upper stratosphere there are uncertainties
regarding the C1O/HC1 ratio. The first simultaneous
measurements of these species were reported by Stach-
nik et al. (1992) and supported the earlier1 assertion
(McElroy and Salawitch, 1989) that models overesti-
mate this ratio in the upper stratosphere. Calculations of
the C1O/HC1 ratio can be considerably improved by in-
cluding a minor channel (approximately 5%) for the
reaction of OH with CIO; to give HC1. This reaction
channel is unobserved to date, however, a channel of this
magnitude would be within the upper limit suggested by
laboratory studies (DeMore et al., 1992). Addition of
this channel to photochemical model calculations im-
proves agreement with Atmospheric Trace Molecule
Spectroscopy (ATMOS) data (Natarajan and Callis,
1991), with the annual amplitude of 03 changes (Chan-
dra et al., 1993), and with the ozone trend iruthe upper
stratosphere (Toumi and Bekki, 1993a). :
• jlnthe lower stratosphere, there are indications of
.outstanding problems in the ratio of HC1 to Cly (see Sec-
tion 4.2.3.2).
The underestimation of upper stratospheric ozone
remains unresolved (Minschwaner et al., 1993), al-
though the deficit now appears to be less than 20%.
Calculations that incorporate the photolysis of vibra-
tionally excited oxygen as. a potential additional source
of ozone gave promising results (Toumi et al., 1991;
Toumi, 1992; Minschwaner et al., 1993; Eluszkiewicz
and Allen, 1993), but recent laboratory measurements of
the quenching rates for vibrationally excited O2 imply
they are more rapid than previously thought (Price et al.,
1993), suggesting that this mechanism is unlikely to be
important.
4.2.2 Heterogeneous Processes '
Five heterogeneous reactions on stratospheric sul-
furic acid aerosols have been identified that could play
important roles in the midilatitude ozone balance:
N2O5 + H2O -
CION02 + H20
C1ONO2 + HC1
HOC1 + HC1 -»
N2O5 + HC1 ->
2HNO3
-> HOC1 + HNO3
» C12 + HNO3
C12 , + H2O
C1NO2 + HNO3
(1)
(2)
(3)
(4)
(5)
These reactions all activate chlorine, either directly by
converting reservoir species to photochemically active
forms (2-5), or indirectly by reducing NOX, which regu-
lates CIO via formation of ClONO2 (reaction 1).
Recent laboratory results suggest that the rates of
heterogeneous reactions (1-5) on stratospheric sulfuric
acid aerosol particles (SSAs) depend strongly on the
chemical composition and phase of the aerosols. SSAs
are thought to be composed primarily of sulfuric acid
and water, but at temperatures lower than about 205 K
they may take up significant amounts of HNO3 (Molina
et al., 1993; Zhang et al., 1993; Tabazadeh et al., 1993)
and may eventually freeze, with uncertain effects on the
rates of heterogeneous reactions.
The hydrolysis of N;jOs (reaction 1) occurs rapidly
on all liquid SSAs, with very little temperature depen-
dence (see Tolbert, 1993, for a review of these results).
In contrast, the hydrolysis of C1ONO2 (reaction 2) is a
strong function of aerosol composition, occurring faster
4.5
-------
TROPICAL/MIDLATITUDE PROCESSES
for more dilute aerosols (Hanson et al., 1994, and refer-
ences therein). In the stratosphere, this property
manifests itself as a strong temperature dependence,
with an increasing reaction rate at low temperatures; at
which SSAs are most dilute.
At present, it is difficult to assess the importance
of reactions 3-5, as there are few relevant measurements;
values for HC1 solubility vary by an order of magnitude
(Zhang et al., 1993; Williams and Golden, 1993; Hanson
and Ravishankara, 1993b; Luo et al., 1994). There are
no direct measurements of diffusion coefficients and
very few second-order rate constants have been deter-
mined. In general, however, these reactions appear to be
limited by the availability of HC1 in solution. Because
HC1 solubility increases rapidly with decreasing temper-
ature and decreasing H2SO4 concentration, the rates of
reactions 3-5 should behave similarly with temperature
to that of reaction 2.
A reactive uptake model has been used to investi-
gate the differences between reaction probabilities in
small particles and hi the bulk liquid (Hanson et al.,
1994). The differences are illustrated in Figure 4-2,
which shows calculated reaction probabilities (Y) as
functions of weight percent sulfuric acid (and tempera-
ture) for reactions 2 and 3 on 0.5 um particles, together
with the measured bulk rate for reaction 2.
At very low temperatures, possibly after the for-
mation of polar stratospheric clouds (PSCs), sulfuric
acid aerosol particles are likely to freeze as sulfuric acid
tetrahydrate (SAT). Once frozen, SAT are expected to
remain solid until they warm to above 210 to 215 K
(Middlebrookefa/., 1993). Although there are relatively
few studies of heterogeneous reactions on frozen SSAs,
some results are available. Reaction 1, fast on all liquid
SSAs, appears to be quite slow on SAT, even at high rel-
ative humidity (Hanson and Ravishankara, 1993a).
Reaction 2 also appears to be slower on SAT than on liq-
uid SSAs, although there is uncertainty in the measured
value of Y (Hanson and Ravishankara, 1993a; Zhang et
al., 1994). In contrast, reaction 3 occurs readily on SAT
surfaces at high relative humidity. Like its counterpart
on type I PSCs, the rate of reaction 3 on SAT decreases
as the relative humidity decreases. Reactions 4 and 5
have not yet been studied on SAT surfaces.
Laboratory work also shows that several species in
the HOX family, for example, OH and HO2 (Hanson et
al., 1992) and CH2O (Tolbert et al., 1993), are readily
45 50
wt% H2S04
55
60
65
10"
10'
,-2
10
10
,-3
190
195
200
205
Atmospheric Temperature (K)
Figure 4-2. The uptake coefficients (r) fo'r CIONO2
onto small liquid sulfuric acid droplets due to reac-
tion with HCI (solid curve) and with HaQ (dashed
curve) are shown here. These values are calculat-
ed with parameters obtained from laboratory
measurements over bulk liquid surfaces jusing the
methodology presented in Hanson et al. (1994).
The calculation was made for a partial pressure of
water equal to 2x10"4 mTorr, equivalent to 5 ppmv
at 50 hpa. The approximate H2SO4 content of the
droplets is shown at the top of the figure.! The dot-
ted curve is the laboratory measured Y for ClONOa
with HaO in the absence of HCI. The reaction prob-
ability for HOCI + HCI is similar to that for CIONO2 +
HCI. The reactive loss coefficient for NjOs on 60
wt% H2SC>4 at low temperatures is -0.1, and prob-
ably does not vary greatly from this value over the
range of acid content shown in this figure. (Adapt-
ed from Hanson et al., 1994.) ;
taken up by SSAs. The competing gas phase reactions of
OH and HO2 are so rapid that heterogeneous loss does
not significantly perturb the HOX budget or partitioning
(Hanson et al., 1994). However, condensed phase reac-
tions of OH or HO2 and uptake of HOX reservoirs such as
CH2O may impact the chemistry of other radical fami-
lies.
A number of studies (Abbatt, 1994; Hanson and
Ravishankara, 1994) have shown that heterogeneous re-
actions of bromine compounds (HBr, HOBr, and
BrONO2) occur on sulfate aerosol and may be important
sources of halogen atoms.
4.6
-------
TROPICAUMIDLATITUDE PROCESSES
Finally it should be noted that heterogeneous reac-
tions also occur very readily on polar stratospheric
clouds. These processes, which may have an impact on
midlatitude chemistry (see Section 4.7), are discussed in
Chapters.
4.2.3 Atmospheric Observations
Since the last Assessment, measurements from a
variety of sources including the Stratospheric Photo-
chemistry, Aerosols and Dynamics Expedition, (SPADE)
and the second Airborne Arctic Stratospheric Expedition
(AASEII) campaigns, from the Upper Atmosphere Re-
search Satellite (UARS) and ATMOS instruments, and
from ground-based instruments have all provided new
information that bears directly on the issue of midlati-
tude ozone loss. Details of these advances are given
below.
4.23.1 NO3Ł/NOY RATIO
Many new measurements indicate that incorpora-
tion of reaction 1 (see above) into photochemical models
results in better agreement between theory and measure-
ments. A variety of observations of nitrogen oxide
species show a lower-than-gas-phase NOx/NOy ratio
including in situ (Fahey et al., 1993; Webster et al.,
1994a), column measurements (Keys et al., 1993; Koike
et al., 1993), and ATMOS data (McElroy et al., 1992;
Toumi et al., 1993b). Indirect measurements (e.g., the
balloon-borne CIO profiles measured by Aval lone et al.,
1993a) also support inclusion of N2Os hydrolysis in
models, in order to more accurately reproduce observa-
tions. Figure 4-3 illustrates a comparison between data
and models from that study. !
Observations obtained during the SPADE cam-
paign showed that models that neglect heterogeneous
chemistry provide a completely inadequate description
of the observed radicals, but that inclusion of the hetero-
geneous hydrolysis of N2Os and C1ONO2 at the
recommended rates resulted in better agreement be-
tween observation and theory. The modeled partitioning
between NOX and NOy generally shows good (30 per-
cent) agreement with the measured ratio when the
hydrolysis of N2Os and the temperature dependence of
the nitric acid cross sections (Burkholder et al., 1993)
are used (e.g., Salawitch et al., 1994a, b; Wennberg et
al., 1994).
L
a.
(A)
10
CIO
10 * 0.0 1.2
A (umW3)
Figure 4-3. Curve A: A 0.5-km average of mea-
sured CIO, shown as solid circles. The dashed line
represents gas-phase-only model results and the
heavy solid line shows the calculation with addition
of N2O5 hydrolysis. The dotted lines depict the
range of uncertainty in .calculated CIO for the heter-
ogeneous model resulting from the reported
uncertainty in ozone, v/hich was used to initialize
the trace gases in the model. The model is unable
to reproduce CIO at the lowest altitudes, possibly
due to inaccurate partitioning of HCI and CIONO2-
Curve B: Surface area density used for the hetero-
geneous model calculations. (From Avallone et al.,
1993a.) i
Systematic differences between model and obser-
vations suggest, however, that our knowledge of N2Os
chemistry may still be incomplete. For example, Toumi
et al. (1993c) conclude that the currently recommended Y
for N2Os hydrolysis is too fast to be consistent with the
ATMOS observations. Other specific anomalies remain,
for example anomalous NOx/NOy ratios (Fahey et al.,
1994). However, it is unclear whether these reflect defi-
ciencies in modeling known chemistry (e.g., imperfect
knowledge of aerosol surface areas, photolysis rates,
etc.), or more profoundly, missing chemical or transport
processes. j
4.23.2 PARTITIONING OF RADICAL SPECIES
During the SPADE campaign (November 1992 to
April/May 1993), measurements of the concentrations
of the free radicals NO2, NO, CIO, HO2, and OH were
obtained, together with those of ozone and a number of
tracers and reservoir species (CO2, H2O, N2O, CHLj,
HCI). The main results from this campaign that have
implications for ozone photochemistry are summarized
in the following paragraphs.
4.7
-------
TROPICAL/MIDLATITUDE PROCESSES
Modeled OH and HO2 concentrations are in rea-
sonable agreement with the measurements, although
usually systematically lower by 10-20% (Salawitch e.t
al, 1994a, b; Wennberg et ai, 1994). While this is well
within the uncertainty of the measurements, there were
at times more serious discrepancies: OH and HO2 con-
centrations at high solar zenith angles are much higher
(as much as 10 times at 90° SZA) than expected. This is
most pronounced in the sunrise data (Wennberg et al.,
1994). It is unclear what process is responsible for this
HOX production (Michelsen et al, 1994), although,
since there is a simultaneous increase in NO, the pho-
tolysis of HONO formed by the heterogeneous
decomposition of HNO4 has been suggested (Wennberg
eta/., 1994).
The partitioning between OH and HO2 agrees well
(15 percent) with that expected based on a simple steady
state model using the measured concentrations of NO
and Oa (Cohen et al, 1994). This result is a confirma-
tion of our understanding of the coupling between the
HOX and NOX families and the ozone reaction chemistory
with OH and HO2.
The measured partitioning between NOa and NO
is usually in reasonable (30%) agreement with the ex-
pected steady-state relationship:
NO2/NO = {kNO+o3(63) + kNO+cio(C10)} / JNOy
although disagreements of more than a factor of two are
occasionally observed (JaegUS et al, 1994).
In combination with the earlier AASEII observa-
tions (King et al, 1991; Avallone et al, 1993b), the
SPADE measurements demonstrate the role that aero-
sols play in enhancing CIO (Salawitch et al, 1994a, b).
The ratio of CIO to the available inorganic chlorine was
observed to be strongly anticorrelated with the available
NOX (Wennberg et al, 1994; Stimpfie et al, 1994), and
it was observed that CIO concentrations dropped be-
tween the fall and spring flights^ consistent with a direct
response to observed NOX enhancements.
However, balancing the chlorine budget in me
lower stratosphere remains problematic. Calculations of
the C1O/HC1 ratio from ER-2-based measurements
(Webster era/., 1993) show that this ratio is not accurate-
ly represented by a model that includes die
heterogeneous hydrolysis of N2O5, as shown in Figure
4-4. This conclusion was also drawn in the work of
Avallone et al (1993a), in which the model was unable
to reproduce the measured values of CIO below |about 18
km altitude. However, as discussed below, provided the
ratio of NOX to NOy is modeled correctly, accurate simu-
lations of observed CIO are obtained, implying that the
modeled HC1 concentrations are in error, but not the
modeled CIO concentrations. Models predict much
higher (1.5 to 3 times) HC1 than was measured (Webster
et al. 1994b; Salawitch et al, 1994b). While the dis-
agreement observed during SPADE was smaller than
that during the AASE II campaign, large differences
remain.
If, however, die inorganic chlorine unaccounted
for is taken to be chlorine nitrate, the observed. CIO and
NO concentrations would imply that the photolysis rate
of chlorine nitrate must be approximately 1/3 of the
recommended value (Webster et al, 1994b)j While
simultaneous measurements of chlorine nitrate pre a pre-
requisite to resolving this problem, the reasons for this
discrepancy, and the implication for ozone losfc, remain
unclear.
Figure 4-5 shows measurements of the diurnal de-
, pendencies of stratospheric free radicals obtained on the
flights of May 11 (sunrise) and May 12 (sunset) at 37°N
and 63 hPa (18.8 km). Making certain assumptions (see
caption), Salawitch et al (1994a), using a dataiassimila-
tion photochemical model constrained by the observed
source gas fields, obtained very good agreement with the
observations (see Figure 4-5), implying a good under-
standing of the basic controlling processes. :
Further confirmation of our generally good under-
standing of fast photochemistry over a range of
conditions in the low stratosphere was provided by the
SPADE survey flights, which were made from 15-60°N
with altitude profiles (15-21 km) made approximately
every 10 degrees of latitude. Figure 4-6 show^ data ob-
tained during the SPADE ER-2 flights of May 14 and
May 18, 1993, compared with the data-assimilation
model of Salawitch et al. (1994b) constrained by ob-
served source gas fields. Observed changes in aerosol
surface area along the flight track of between -|5 and -15
um2cm~3 are also included in the calculations. j Details of
the model calculations are given in die figure caption.
Salawitch et al conclude that inclusion of! heteroge-
neous processes is essential if radical concentrations are
to be modeled correctly, although discrepancies remain,
notably in the modeled OH/HO2 and NO/NO2 ratios.
4.8
-------
TROPICAL/MIDLATITUDE PROCESSES
OCTOBER 1991 H> MARCH 1992
LATITUDES 26 -»90°N
100
0.00
0.24 0.48 0.72
I HCŁ(ppbv)
0.96
1.120
Figure 4-4. Scatter plot of CIO versus HCI data from instruments aboard the NASA ER-2, taken on the
flights of Oct. 14,1991, Feb. 13,1992, and Mar. 15 and 22,1992, covering latitudes between 26° and 90°N.
The data included are limited to CIO mixing ratios less than 100 pptv, and to solar zenith angles less than or
equal to 80°. Also plotted are results from the 2-D model of Solomon and Garcia using either gas-phase-only
photochemistry, or including the heterogeneous hydrolysis of NaOs on two levels of sulfate aerosol surface
area that bracketed the ER-2 observations. The figure illustrates that observed HCI concentrations for a
given CIO amount are approximately a factor of 2 lower than model calculations including heterogeneous
chemistry would imply. (From Webster et al., 1993.) ;
These discrepancies can be reduced, but not eliminated,
with further refinements.
Measurements from the Microwave Limb Sounder
(MLS) have allowed the global behavior of ClOi concen-
trations in the low stratosphere to be determined. In
Figures 4-7 (left and right panels) (Froidevaux et al.,
1994) are shown monthly mean zonal average CIO mix-
ing ratios at 22 hPa, 46 hPa, and 100 hPa averaged over
30°N-50°N and 30°S-50°S^ respectively. The data ex-
tend from September 1991 through to the end of 1993.
The MLS data reveal a distinct seasonal: cycle in
both hemispheres, with maximum CIO mixing ratios
seen during midwinter months. This variation appears to
be qualitatively consistent with expected changes in
NOX as discussed above.- In agreement with studies
mentioned above, lower stratospheric midlatitude CIO
values of 0.1 to 0.2 ppbv, as measured by MLS, cannot
be explained with gas phase chemistry alone (Froide-
vaux et al., 1994). The MLS data show that CIO mixing
ratios at 46 hPa and 22 hPa iare closely comparable in the
respective seasons in the two hemispheres.
Differences would.be expected in the extent of air
exposed to heterogeneous processes in the polar regions
of the Southern and Northe:rn Hemispheres, and indeed
interannual differences would be present, particularly in
the North (Jones and Kilbjine-Dawe, 1994). Thus, the
consistent phasing of the observed CIO maxima and,
after adjustment for season, the comparable CIO
amounts in the two hemispheres, suggest that in. situ
chemistry rather than the processing of air from the polar
vortex is the main factor c
ontrolling these midlatitude
CIO concentrations (Froidevaux etal, 1994). However,
it should be noted that the MLS instrument has very
limited sensitivity in the lowest region of the strato-
4.9
-------
TROPlCAUMIDLATrrUDE PROCESSES
1.0 -
_
O
O 0.01
O
0.0
90 60 30 30 60 90 ,
SOLAR ZENITH ANGLE (deg) ;
Figure 4-5. Measurements (dots) of the diurnal variations of stratospheric free radicals NO2, NO, HO2, OH
(crosses and dots represent data from the JPL and NOAA instruments, respectively), and CIO from two ER-
2 flights of May 11 (sunrise) and May 12 (sunset), both near 37°N and 63 hPa and [N2O] between;240 and
260 ppbv, plotted as a function of solar zenith angle. Also shown are results from a constrained data assim-
ilation model (Salawitch et al., 1994a). Three'calculations are shown. Dark dotted curve: gas phase
reactions only, using rate constants and cross sections of DeMore era/. (1992). Curve 1, dark solid line: as
for above, except including also the heterogeneous hydrolysis of N2O5 and CIONO2. Curve 2, gray line: as
for curve 1, except including the heterogeneous decomposition of HNO4 to form HONO, the O(1 D) quantum
yield of Michelsen era/. (1994), and the temperature-dependent cross sections of HNO3 from Burkholder et
al. (1993). (From Salawitch et al., 1994a.) ;
4.10
-------
2 -
TROPICAL/MIDLATITUDE PROCESSES
30 40
LATITUDE (N)
50
60
:
Figure 4-6. Measurements (points) of NO2, NO, CIO, HO2, OH, and HCI obtained on May 14 and 18, 1993,
during which the ER-2 flew from 15 - 55°N. Also shown are calculations from the data assimilation model of
Salawitch era/., 1994b. The individual calculations are as for Figure 4-5. As can be seen, all three calcula-
tions significantly overestimate HCI concentrations. (From Salawitch etal., 1994b.)
4.11
-------
TROPICAL/MIDLATITUDE PROCESSES
0.3
I"
3
O.O
0.3 f
|o,
o '
0.0
7 6 « 10 1
23458789101234567B910
7% 9 10 123 45 6 7 B » 10 1 2 3 « S 6
M
0.2
«
T 0.1
8
0.0
eg
2=
o.a •
0.1
s
§
749 10 123458789 10 12 3<56789 10
3 4 S 6 7 B 9 10 123*56789 10
T'a's'lO 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10
0.3
6.1
0.0
r I
n\-
J
I I I |i ' I
4l|l-
-Uir.J
7 8 9 10 1 2 3
1991
Q 0.2 r
S'O'M'B J P U'A'M'J'J1
from Froidevaux era/., 1994.)
sphere (-100 hPa). This is an important limitation be-
cause the bulk of the ozone column at midlatitudes
resides in this region, and any hemispheric differences in
CIO at these low altitudes might not be detected.
Detailed comparisons of HC1 data from the NASA
ER-2 instrument of Webster et al. (e.g., 1993) and the
Halogen Occupation Experiment (HALOE) (Russell et
ai, 1993b) are ongoing, C1ONO2 measurements from
the Cryogenic Array Etalon Spectrometer (CLAEiS)
(Roche et al., 1993) will also augment this data set to
provide a nearly complete inorganic chlorine budget for
the lower stratosphere.
4.3 ERUPTION OF MT. PINATUBO i
The eruption of Mt. Pinatubo, located in the Phil-
ippines (15°N, 120°E), culminated in an enormous
explosion on June 14-15, 1991. The plume reached alti-
tudes in excess of 30 km, depositing 15 to |20 Mt of
sulfur dioxide (SO2) into the stratosphere (Bluth et al.,
1992; McPeters, 1993; Read etal., 1993), nearly 3 times
as much as the El Chichon eruption in 1982. Qonversion
of SO2 into sulfuric acid (H2SO4) occurred rjapidly, re-
sulting in sulfate aerosol surface areas as l^rge as 85
|j.m2cnr3 over Northern midlatitudes (Deshler et al.,
1992, 1993) at some altitudes. This huge perturbation
has allowed a test of many aspects of our understanding
of heterogeneous chemical processes in the stratosphere,
building on the earlier work of Hofmann and Solomon
(1989). ;
4.12
-------
TROPICAL/MIIDLATITUDE PROCESSES
4.3.1 Effects on Chemical Composition
In the first few months following the ML Pinatubo
eruption, low ozone amounts were detected in the trop-
ics, roughly coincident with the region of largest aerosol
loading (e.g., Grant et al., 1992, 1994; Schoeberl et al,
1993a). On a longer time scale, ozone reductions were
observed at midlatitudes by satellite (Gleason et al.,
1993; Waters et al, 1993), ground-based (Bojkov et al,
1993; Kerr et al, 1993), and in situ instrumentation
(Weaver et al, 1993; Hofmann et al, 1994). Details of
this anomalous ozone behavior are given in Chapter 1.
The short-term tropical ozone decline has been at-
tributed both to dynamical effects (see Section 4.6.3),
and to a reduction of O2 photolysis, and hence ozone
production, due to absorption of solar ultraviolet radia-
tion by SOa (Bekki et al, 1993). SO2 has also been
shown to be capable of catalyzing ozone production via
the following mechanism (Crutzen and Schmailzl, 1983;
Bekki etal, 1993):
SO2 + hv ->• SO + O(X>220nm)
SO + O2-> SO2+ O I
2(O+O2 + M -> O3 + M) '!
net: 3O2 -» 2O3 !..
I >
The longer-term ozone decrease is likely to be the
result of a combination of enhanced heterogeneous
chemistry resulting from the large increase in sulfate
aerosol surface area, changes in the radiation field, and
altered stratospheric dynamics (see, for example, Bras-
seur and Granier, 1992; Michelangeli etal, 1992; Pitari
and Rizi, 1993; Tie et al, 1994).
Once the initial SO2 plume is converted to aerosol
particles, enhanced absorption of terrestrial emission
and backscattering of solar radiation is expected espe-
cially in the tropics, leading to changed photolysis rates.
The enhanced backscatter reduces the photolysis of all
molecules below the cloud, but for molecules that absorb
radiation at wavelengths longer than 300 nm, which can
penetrate to the low stratosphere (for example, O3 and
NO2), photolysis rates are enhanced above the cloud-
top. The net effect is to accelerate photochemical ozone
loss, leading to reductions in column ozone of several
percent in the vicinity of the cloud (Pitari and Rizi, 1993;
Tie etal., 1994). . ".
17 September 1991 22 March 1992
X
o
5
ox
x
1
o
0.20
0.15
0.10
0.05
0
0.08
0.06
0.04
0.02
0
3 >~~
j
i j»
i
j
i
t-
i
V-
%:'
J'.^^Ł
1 \ *i —
J "...
!
1
j
1 i
1 —
afc*"'
1
— r; 1 1 1
i
';
^f
' jJr&Vl
0*^S
j „ ~_
\
\
1
,
: ( Vf •
} &&
*Ł;
\
\ l 1 1 , -
— r; 1 1 1 •
( -4 :
. i
i ' i i ' i
0 5 10 i 15 20 25
Aerosol surface area (urn* cm-3)
30
Figure 4-8. Scatter plot of NOx/NOy and CIO/CL
data from the NASA ER-2 with observed aerosol
surface area (solid circles) in high and low aerosol
conditions. Gas phase only (open circles) and het-
erogeneous case (crosses) model calculations are
included (using observed aerosol surface areas
with reaction 1). The surface area scale has no
meaning for the gas phase case, except to sepa-
rate data from the two flights. The vertical dashed
line represents background aerosol surface area.
The curved lines represent the dependence on sur-
face area in the model heterogeneous case for the
average conditions in September (solid) and March
(dashed) data sets. Also shown are the corre-
sponding ClO/Cly obseivations. (From Fahey et
al., 1993.) ;
A variety of chemical changes thought to be the
results of heterogeneous reactions on the Mt. Pinatubo
aerosol cloud have been observed. Fahey et al (1993)
and Kawa et al. (1993) showed dramatic reductions in
the NOx/NOy ratio as sulfate surface area increased (see
Figure 4-8). In response to this change, the'amount of
active chlorine (ClO/Cly) was observed to increase, as
expected (Wilson et al, 1993; Avallone et al, 1993b;
4.13
-------
TROPICAL/MIDLATITUDE PROCESSES
observation
Model (LNNL)
Model (AER)
Dec
Figure 4-9. Percentage changes in HNO3 and
NOa column amounts above Lauder, New Zealand,
(45°S) following the arrival of the Mt. Pinatubo
aerosol. The Lawrence Livermore National Labo-
ratory (LLNL) results are for 42.5°S and the
Atmospheric Environmental Research, Inc. (AER)
results are for 47°S. Heterogeneous chemistry is
included in the calculations based on the observed
aerosol field from SAGE II. (From Koike et a/.,
1994.)
and Wennberg et al, 1994). In addition, several mea-
surements of column NO2 at middle and high latitudes
showed substantial decreases (25 to 50%) in comparison
to previous years (Johnston et al, 1992; Koike et al,
1994; Mills et al, 1993; Coffey and Mankin, 1993; So-
lomon et a/., 1994a).
The hydrolysis of N2Og is expected to saturate at
moderate values of surface area (Prather, 1992), but the
hydrolysis of C1ONO2 may become increasingly impor-
tant as surface area grows, as in the case of the Mt.
Pinatubo aerosol. This saturation effect is evident in the
NOx/NOy measurements of Fahey et al. (1993) (Figure
4-8) and further confirmed by the lack of major CIO en-
hancements at mid- to high Northern latitudes (Avallone
et al., 1993a; Dessler et al, 1993) and seen in the MLS
data shown in Figure 4-7. Further qualitative support for
N2Os hydrolysis comes from Koike et al. (1994), who
have observed the effects of Mt. Pinatubo aerosol on
NO2 and HNO3 over New Zealand. In Figure 4-9 are
shown percent changes in HNO3 and NO2 colunins over
Lauder (45°S) from June 1990 to December 1993. The
data show a reduction in NO2 columns as the Mt.| Pinatu-
bo cloud reached Lauder, and a simultaneous increase in
column HNO3. Model calculations using the observed
aerosol field from SAGE II (Stratospheric Aerosol and
Gas Experiment II) (Kent and McCormick, 1993) as in-
put show good qualitative agreement wjith the
observations, although the magnitude of the changes is
underestimated in the models.
A number of studies have provided evidence for
the heterogeneous hydrolysis of C1ONO2 on sulfate
aerosols, particularly during periods of volcanic|activity.
Solomon et al. (1993) argue that the observation of en-
hancements to the OC1O column in the austral fall of
1992 are due to the hydrolysis of C1ONO2 on sulfate
aerosols. The formation of substantial OC1O amounts
requires enhanced CIO in addition to moderate BrO con-
centrations, and had in the past only been .detected
following the appearance of PSCs. However, in 1992,
OC1O was detected earlier and in larger quantities than
in previous years, suggesting that chlorine had been acti-
vated to some degree on sulfate aerosols. This
conclusion is predicated on the absence of PSC process-
ing prior to the observation of OC1O. ,
Column reductions of C1ONO2, HC1, and HNO3
were also observed from the NASA DC-8 during transit
below a very cold region of volcanically enhanced aero-
sol (O. Toon et'al, 1993). The heterogeneous reaction
probability Y for C1ONO2 hydrolysis calculated from
these observations, taking into account the history of the
air parcels, is very close to laboratory values. Dessler et
al. (1993) attempt a similar calculation based qn balloon
observations of CIO and NO and determine a jvalue of Y
again consistent with laboratory experiments. (However,
the possible influence of PSC processing at some earlier
time cannot be entirely ruled out as influencing the ob-
served concentrations of CIO and OC1O. ;
Calculations using 2- and 3-dimension|al models
suggest that once the aerosol cloud is dispersed from the
tropics and the aerosol loading begins to increase at mid-
and high latitudes, significant (several percent) column
ozone reductions arising ffom accelerated: heteroge-
neous -chemistry are likely to be widespread, with
maximum reductions (up to'~10%) at midlatitudes in
winter where the photolysis of HNO3 is slow (Pitari and
Rizi, 1993; Tie et al., 1994). This is discussed further in
4.14
-------
TROPICAL/MIIDLATITUDE PROCESSES
Section 4.6.3. However, there are several outstanding
anomalies. For example, despite the more rapid move-
ment of the Mt. Pinatubo aerosol cloud to the Southern
Hemisphere (Trepte et al., 1993), ozone trend;;; were ap-
parently smaller in the South compared to the North
following the Mt. Pinatubo eruption. This is discussed
further in Section 4.4.
4.3.2 Implications for the Normal State of the
Atmosphere
Observations of chemical constituents in tfie pres-
ence of enhanced sulfate aerosol surface area and over a
wide temperature range have shown that the heteroge-
neous hydrolysis of C1ONO2 should be considered in
addition to the hydrolysis of N2O5 to more accurately
simulate ozone loss in the stratosphere. The area of larg-
est debate regarding heterogeneous chemistry is an
accurate quantification of the actual rates of reaction in
the stratosphere. Laboratory determination of rate pa-
rameters and uptake coefficients for a variety of species
is essential, but improved understanding of the composi-
tion and physical characteristics of the stratospheric
aerosol layer at temperatures less than about. 210 K when
ternary (H2O/HNO3/H2SO4) solutions may exist is
equally important. Understanding of the potential role
of heterogeneous processes other than the hydrolysis of
N2O5 and C1ONO2 is expected to improve as a result of
measurements made during periods of highly perturbed
surface area, as reaction rates are expected to be large
enough to cause an observable effect (Hanson et al.,
1994). It is unlikely that all of these processes will be
important under "background" surface area conditions,
but the possible continued emission of sulfur from cur-
rent subsonic aircraft and a proposed supersomic fleet
may significantly increase the sulfate aerosol loading of
the stratosphere (Bekki and Pyle, 1992). Under such a
scenario, heterogeneous reduction of NOX and subse-
quent enhancement of active^chlorine may have a serious
effect on the ozone balance in the tropical and midlati-
tude stratosphere; \
4.4 PHOTOCHEMICAL OZONE LOSS !
PROCESSES AT MIDLATITUDES \
There is now a much clearer understanding of the
relative importance of different photochemical destruc-
tion cycles to ozone loss in, the low stratosphere, support-
ed, as discussed above, by a comprehensive range of
atmospheric measurements, laboratory studies, and
model calculations. However, knowledge of the abso-
lute rate of photochemical ozone loss still remains
uncertain, primarily because of limitations in our ability
to model accurately the distributions of source gases,
but also because of uncertainties in heterogeneous
chemistry. |
A number of recent studies have provided a rela-
tively consistent picture of the relative importance of
different ozone destruction cycles (e.g., Avallone et al.,
. 1993a; Rodriguez et al, 1994; Garcia and Solomon,
1994; Wennberg et al., 1994). In Figure 4-10 are shown
calculations of the contributions of different photochem-
ical cycles to ozone loss between 13 and 23 km, for
32°-63°N (fromRodriguez,etai, 1994). Panels show: a)
background aerosol conditions, b) volcanically en-
hanced aerosol with hydrolysis of N2O5 only, and c)
volcanically enhanced aerosol with hydrolysis of both
N2Os and C1ONO2. The broad picture is of reactions
involving HO2 being responsible for over half the photo-
chemical destruction of ozone in the low stratosphere,
while halogen (chlorine and bromine) chemistry ac-
counts for a further third. Although catalytic destruction
by NOX accounts for less thsin 20% of the photochemical
ozone loss, NO and NO2 are vital in regulating the abun-
dance of hydrogen and halogen radicals and thus the
total photochemical ozone destruction rate. The effect
of increased aerosol loading is to enhance the HOX and
halogen destruction cycles at the expense of the NOX,
with a net increase of-20% in the ozone loss rate at peak
aerosol loading. Figure 4-1 i shows loss rates as a func- "
tion of altitude from the model of Garcia and Solomon
(1994) for 40°N in March for background aerosol load-
ing. Below 22 km, reactions involving HOX dominate,
while between 23 and 40 Ion, NOX cycles dominate.
Bromine and chlorine loss cycles are important in the
low stratosphere and chlorine becomes dominant near 40
km. Reductions in NOX above about 22 km, where it
represents the dominant photochemical loss mechanism,
would therefore result in local ozone increases (Tie et
al., 1994). i
This provides a possible explanation of the appar-
ent absence of an ozone reduction in the Southern
Hemisphere following Mt. Pinatubo (Hofmann et al.,
1994). The altitudes at which die Mt. Pinatubo cloud
4.15
-------
TROPICAL/MIDLATITUDE PROCESSES
«?
o
g-
V)
3
ra
o
in
a
o
ir
a>
in
vt
O
•a
u
&
3/91
3/92 3/93 3/94 3/95
Figure 4-10. Calculated total average loss frequen-
cies and relative contributions of different catalytic
loss cycles from March 1991 to March 1995 showing
the estimated effect of the Mt. Pinatubo eruption
Average loss frequencies are defined as the total
loss rate of ozone between 13 and 23 km, and 32°
and 63°N, divided by the total ozone content in this
region. The relative contribution of each catalytic cy-
cle is indicated by the different shadings: solid (NOX
cycles); dense diagonal (Clx cycles); large dot (Clx-
Brx); shaded (Brx); white (HOx); and diagonal (O +
O3) Panel a) shows loss frequencies for a back-
ground aerosol case; panels b) and c) are for
volcanically enhanced aerosol and show, respective-
ly the effect of N2O5 hydrolysis alone, and the effects
of both N2O5 and CIONO2 hydrolysis. (From Rod-
riguez et al., 1994.)
penetrated the two hemispheres differed markedly, with
peak concentrations near or above 22 km in the South,
but at lower altitudes in the North (Trepte et al., 1993).
In the Southern Hemisphere, ozone losses below 22 km
would have been compensated for by slowed destruction
above, leading to little net change in the ozoije column.
However, in the North, the absence of aeroso^ at higher
altitudes meant that little or no compensating slowing of
the ozone destruction occurred at the higher altitude,
leading to significant overall declines in the column.
Finally, the suggestion has been made that iodine
compounds (primarily CH3D can reach the low strato-
sphere in sufficient quantities to perturb significantly the
ozone photochemical balance (Solomon et al., 1994b).
The relevant reactive iodine compounds have yet to be
detected in the stratosphere. However, if confirmed, this
process would have a significant impact on;our under-
standing of photochemical ozone loss in the low
stratosphere at midlatitudes. • ;
!
4.5 THE SOLAR CYCLE AND QUASI-BIENNIAL
OSCILLATION (QBO) EFFECTS ON
TOTAL OZONE '.
The largest depletions of ozone noted in Chapter 1
of this document have occurred in the lowest part of the
stratosphere and are systematic from year to jyear. While
such changes are qualitatively consistent .with either
local chemical removal by HOX and halogen cycles
(Section 4.4) or the transport of ozone-depleted air from
polar regions (Section 4.7), they are not, according to our
best understanding, compatible with either changes in
solar output or QBO effects. Nevertheless; solar cycle
and QBO influences on total ozone must be removed if
ozone trends are to be quantified reliably, ;
4.5.1 Solar Ultraviolet Variability and Total
Ozone
Although solar radiation at wavelengths less than
300 nm accounts for only about l% of the tbtal radiative
output of the sun, it is the principal energy source at
altitudes between the tropopause and the lower thermo-
sphere. It both drives the photochemistry! of the upper
atmosphere and is a source of heating, thus affecting the
circulation of the upper atmosphere. Variations of the
. solar ultraviolet (UV) flux can affect column ozone
4.16
-------
TROPICALyMlDLATITUDE PROCESSES
10
icr
Ox Loss Rate {mixing ratio/sec)
2K(CIO)(BrO)
2K(0)(03)
Total HOX-related
Total CIOx-related
Total NOX-related
Total Brpx-reloted
Figure 4-11. Calculated 24-hour averaged Ox loss
rates (mixing ratio/sec) from various chemical cy-
cles for 40°N in March for low (i.e., non-volcanic)
sulfate aerosol loading. In these circumstances the
dominant ozone loss below 22 km is due: to reac-
tions involving OH and HO2, with NOX dominating
between 23 and 40 km. Under higher aerosol load-
ing conditions, coupled HOX - halogen cycles
become more significant. (From Garcia and So-
lomon, 1994.)
amounts and profiles, with the largest changes occurring
in the upper stratosphere (Hood et al., 1993; Brasseur,
1993; Fleming et al., 1994). (
Most solar UV variation occurs with time: scales of
about 11 years (e.g., Cebula et al., 1992) and 27 days.
Over the-11-year cycle, Lyman alpha (121.6 nm) radia-
tion varies by about a factor of two (Lean, 1991). The
mid-UV (200 - 300 nm) strength varied by about 9% be-
tween the 1986 solar minimum and the 1990 solar
maximum. Figure 4-12 displays the F10.7 index (a mea-
sure of the solar UV flux [e.g., Donnelly, 1988])
superposed on the SBUV/SBUV2 (Solar Backscatter
Ultraviolet spectrometer) total ozone that has had the
QBO signal, the seasonal signal, and the trend removed
by statistical methods (see Stolarski et al., 1991, and
Section 4.5.2 below). The figure shows that global aver-
age total ozone (40°S to 40°N) changes are correlated
with UV flux variations, | changing by about 1.5% (4.5
Dobson units, DU) from !solar maximum to solar mini-
mum. These changes are in reasonable agreement with
calculations using 2-D models (Fleming et al., 1994;
Garcia etal., 1984; Brasseur, 1.993; Huang and Brasseur,
1993; Wuebbles et al, 1991).
4.5.2 The Quasi-Bieninial Oscillation and Total
Ozone
I
Variability in the equatorial lower stratosphere is
dominated by the presence of an oscillation in equatorial
winds, with a period of approximately 27 months,
known as the quasi-biennial oscillation (QBO). The
oscillation affects not only the winds but also the thermal
structure and the distribution of ozone and other minor
constituents at all latitudes (e.g., Chipperfield et al.,
1994a, and references therein). Despite the magnitude
of the ozone QBO being relatively small (approx. 5-10
DU at the equator; up to about 20 DU at high latitudes) it
is nevertheless significant in ozone trend studies and
must be characterized and removed. Ozone trend analy-
82
84 ; 86 88
Time (Years)
90
92
Figure 4-12. Response of SBUV/SBUV2 40°N-
40°S average column ozone to the solar cycle as
determined by a linear regression model,after sub-
traction of the seasonal cycle, trend, and QBO
(clashed curve). Also shown is the 10.7 cm radio
flux (solid curve), which is a proxy for the solar out-
put. The figure shows that changes in global
column ozone of the order of 1.5% (4.5 DU) are to
be expected during the 11-year cycle in solar out-
put, mostly at higher altitudes. (From P. Newman,
personal communication, 1994.)
4.17
-------
TROPICAUMIDLATITUDE PROCESSES
ses (e.g., Stolarski et a/., 1991) use linear regression
techniques to isolate and remove the QBO signal. Ob-
served equatorial wind data (e.g., at 30 hPa) are
employed as the reference time series, with the possibil-
ity of a time lag to take into account the observed
variations of the QBO signal with latitude. However, ob-
servations of the ozone QBO show a strong seasonal and
hemispheric asymmetry and the period of the observed
ozone QBO at mid- and high latitudes is also not identi-
cal to that at the equator, often being closer to two yearc
(Gray and Dunkerton, 1990). The use of equatorial wind
data in ozone trend analyses to characterize the QBO
signal at all latitudes is therefore not ideal.
n. TRANSPORT PROCESSES LINKING THE
TROPICS, MIDDLE, AND HIGH
LATITUDES
4.6 INTRODUCTION
The structure of the lower stratosphere in winter,
the period when observed declines in ozone at middle
latitudes are largest, is shown schematically in Figure
4-1. The diagram is intended to show the winter hemi-
sphere when the polar vortex is well established. While
the processes described below are known to occur to
some extent at least, their magnitudes and relative contri-
butions to the observed ozone declines have in many
cases not been quantified reliably. While different in
detail, both hemispheres correspond broadly to this
picture.
In an altitude or log(pressure) framework, isen-
tropes rise both in the tropics and in polar regions,
indicative of the lower temperatures in both regions.
Mixing along these isentropes can be rapid, on a time
scale of days to weeks, except where potential vortidty
(PV) gradients exist. In these regions, mixing is inhibit-
ed by a combination of horizontal wind shearing and
dynamical "Rossby wave elasticity." The midlatitude
region is bounded by a flexible PV barrier on its pole-
ward side (a), and a similar but less distinct tropical
barrier to transport (b) at -20 degrees. Mixing along
isentropes is relatively rapid in middle latitudes in the so
called "surf zone" (c). Both barriers undergo epis
-------
TROPICA17MIDLATITUDE PROCESSES
recently following the El Chichon and Mt Pinatubo
eruptions (Trepte et al., 1993; Hofmann et al., 1994).
The existence of at least a partial subtropical transport
barrier, at the equatorward edge of the winter midlati-
tude "surf zone," has also been deduced from theoretical
arguments and numerical models (Mclntyre 1990;
Norton 1994;Polvaniera/., 1994). Recent observations
of the tracers N2O and CC>2 in the low stratosphere
(Boering et al, 1994) provide direct observational sup-
port for the relatively short mixing times in the "surf
zone" region.
Analyses of data from the Limb Infrared Monitor
of the Stratosphere (LIMS) instrument on Nimbus 7,
from in situ aircraft data, and from instruments on the
Upper Atmosphere Research Satellite (UARS) have all
shown strong gradients of tracers and of potential vortic-
ity in the sub-tropics, with occasional fiiahients of
tropical material being entrained poleward (Leovy et al.,
1985; Murphy et al., 1993; Randel et al., 1993); this be-
havior is also reproduced in dynamical models (Boville
et al., 1991; Norton, 1994; Pierce et al., 1993; Rood
et al., 1992; Waugh, 1993a; Chen and Holton, 1994;
Polvani et al., 1994; Bithell et al., 1994). i
The tropical lower stratosphere is also strongly in-
fluenced by the quasi-biennial oscillation (QBO), which
has a significant impact on the meridional circulation
(Plumb and Bell, 1982). The QBO affects meridional
transport of ozone and other trace species by a modula-
tion of planetary (Rossby) wave transport. When the
Rossby wave amplitude increases sufficiently, the waves
"break," resulting in irreversible transport in' midlati-
tudes. The latitudinal region in which the waves break
(the "surf zone") is affected by the background winds in
equatorial regions, particularly in the case of strongly
nonlinear waves (O'Sullivan and Young, 1992)'.: Easterly
equatorial winds confine the Rossby waves further pole-
wards than westerly winds, resulting in enhanced
meridional exchange of air ^between the subtropical and
higher latitudes (see, e.g., Baldwin and Dunkerton,
1990; Garcia, 1991; Dunkerton and Baldwin, 1992).
Extensive observations — ground-based, in situ,
and satellite-based — of the formation, dispersion, and
decay of stratospheric aerosol produced by the eruption
of Mt. Pinatubo (15°N) in June 1991 have provided
much insight into the processes of transport out of the
tropics. These observations and their implications are
described in the following. r
4.6.2 The Mt. Pinatubo Eruption:
Implications for Understanding of
Transport Processes
Prior to the June 1991 eruption, the total strato-
spheric aerosol mass (as inferred using SAGE
observations) was approximately 1 Mt, but by the end of
1991 the estimated mass had increased to -r30 Mt. The
total mass has since decreased to approximately 10 Mt
by mid-1993 and to around 3 Mt by mid-1994 (see Fig-
ure 4-13). The formation of the Mt. Pinatubo aerosol
cloud in the stratosphere and its subsequent dispersal
around the globe, monitored from the ground and satel-
lites, have provided useful tests of our understanding of
transport processes. \
4.6.2.1 TROPICAL LATITUDES
In many respects, the temporal development of the
Mt. Pinatubo aerosol distribution was similar to that ob-
served following other high altitude tropical injections.
A distinguishing characteristic of this eruption was the
rapid movement of volcanic material across the equator
within two weeks of the eruption (Bluth et al., 1992;
McCormick and Veiga, 1992). Young et al. (1993) re-
SAGE U-Estimated Stratospheric Aerosol Mass
1991
1995
Figure 4-13. SAGE II estimated stratospheric
aerosol mass, showing the near exponential decay
on a time scale of -11 months following the erup-
tion in mid-1991. (Thomason and Poole, private
communication.) !
4.19
-------
TROPICAUMIDLATITUDE PROCESSES
ported that this drift was induced by local aerosol heat-
ing. The heating was also sufficient to cause large
increases in tropical stratospheric temperatures (Labitz-
ke and McCormick, 1992) and may have contributed to
the upward transport of aerosols to above 35 km by Oc-
tober 1991 (Trepte et al., 1993). The tropical aerosol
reservoir has gradually diminished in magnitude since
the eruption as aerosols became dispersed poleward arid
were removed by sedimentation.
It is now appreciated that the detrainment of tropi-
cal air to midlatitudes occurs in episodic events when the
polar vortex becomes displaced from the pole and inter-
acts with the subtropical flow (e.g., Randel et al., 1993;
Trepte era/., 1993;Waugh, 1993a).
Transport from the tropics takes place in two re-
gimes, at different altitudes. In the lower transport
regime (about a scale height above the tropopause) ziir
moves rapidly poleward and downward (Fujiwara et al.,
1982; Kent and McCormick, 1984). This transport is
most effective during winter, especially in the Northern
Hemisphere. Poleward spreading of aerosols is also ob-
served during summer associated with tropospheiic
monsoon circulations. Early appearances of aerosol
above Japan and Germany, amongst other places, were
associated with this circulation (Hayashida, 1991; Jaie-
ger, 1992). The dispersion rate of the main aerosol cloud
was estimated from shipboard lidar measurements to be
around 5 degrees latitude per month in the region 8°N to
22°N, during the period July 11 to September 21 (Nardi
era/., 1993).
In the upper regime (above 22 km), aerosols are
redistributed by motions associated with the QEIO
(Trepte and Hitchmann, 1992). During the descending
QBO westerly wind shear, anomalous subsidence (rela-
tive to the climatological upwelling) takes place over the
equator, transporting aerosols downward and poleward
below the shear layer. However, during the descending
QBO easterly shear, enhanced lofting of aerosol occurs
over the equator, with poleward flow above the shear
layer.
Consistent with this picture, within the upper re-
gime at altitudes where zonal mean westerlies existed,
strong meridional gradients of volcanic aerosol, indica-
tive of rapid poleward mixing, were present near 20°N
and S (Browell et al., 1993; Trepte et al., 1993), while at
heights where easterlies lay over the equator, the sub-
tropical gradients were diminished or absent, with great-
er mixing taking place at higher altitudes. '
4.6.2.2 MEDDLE AND HIGH LATITUDES •
Some aerosol spread rapidly to middle [and high
Northern latitudes in the low stratosphere, being first
observed two weeks after the eruption in midlatitudes
(Jaeger, 1992), in early August at Andoya (69°N) and on
August 11 at Ny-Alesund (79°N) (Neuber et al., 1994).
However, the main part of the cloud did not reach Haute-
Provence and Garmisch-Partenkirchen (48°N) until
mid-October, with the highest backscatter ratios being
observed in December 1992. In the Arctic, values of in-
tegrated backscatter at Spitzbergen were generally lower
than at midlatitudes (Jaeger, 1993). In addition to satel-
lite observations, extensive aerosol measurements from
ground-based lidars in middle and high latitudes, and
from airborne lidar and in situ instruments jvere per-
formed during the European Arctic Stratospheric Ozone
Experiment (EASOE) and NASA Airborne Arctic
Stratospheric Expedition (AASE II) campaigns of win-
ter 1991-92. Together, these data indicate efficient
latitudinal transport below about 400-450K (15-19 km)
but a largely isolated vortex, with little aerosol penetra-
tion, at higher altitudes (Tuck, 1989; Brock et al., 1993;
Browell et al, 1993; Neuber et al., 1994;'Pitts and
Thomason, 1993), although the differences in aerosol
characteristics between inside and outside the vortex are
less apparent when referenced to NaO (Borrmann et al.,
1993; Prpffitt et al, 1993). Also, occasional .intrusions
of aerosol-rich midlatitude air into polar regions have
been documented (Rosen et al, 1992; Plumb et al,
1994). Lidar measurements performed during the same
period of time at Dumont d'Urville (68°S) showed that
the aerosol did not penetrate the Antarctic vortex in 1991
(Godin et al, 1992). In contrast, the smaller volcanic
cloud of Cerro Hudson (46°S), which was injected into
the lower stratosphere (around 12 km) in August 1991,
spread rapidly into polar regions, again revealing effi-
cient transport beneath the vortex (Godin et al, 1992;
Pitts and Thomason, 1993; Schoeberl et al,, I993b).
(Refer to Figure 4-14.) |
4.20
-------
TROPICAL/MIDLATITUDE PROCESSES
- (f)
-70
-60 -50 ,; -40 -30 -80 -70 -60 -50 -40 -30
Latitude Latitude f
>1CO
I
Figure 4-14. Latitude-altitude median cross sections of SAGE II and Stratospheric Aerosol Measurement
(SAM II) 1-nm aerosol extinction ratio for six periods shown in each panel. The crosses and circles indicate
the daily mean latitudes of the SAGE II aind SAM II observations, respectively. The main region of high
extinction ratio at 20 - 25 km is due to the Mt. Pinatubo aerosol cloud, while the band at 10 km originates from
the Mt. Cerro Hudson eruption. The higher altitude cloud does not penetrate the established polar vortex,
while that at lower altitudes progresses poleward more readily. (From Pitts and Thomason, 1993.)
4.6.3 Circulation-Induced Ozone Changes
Resulting from the Mt. Pinatubo
Eruption
4.63.1 RADIATIVE EFFECTS OF STRATOSPHERIC !
AEROSOL
| !
Changes in stratospheric aerosol loading alter the
radiative properties of the atmosphere, and so have the
potential to not only modify local temperatures; but also
to alter the stratospheric circulation. In general, changes
both to the absorption and emission of infrared radiation
and to the solar heating rate must be considered.
In the infrared, the effects of small aerosol parti-
cles (radius less than -6.1 u,m) can generally be
neglected, as their extinction is insignificant. Infrared
absorption and emission become increasingly important
at larger aerosol sizes. The strength of infrared heating
also depends on the difference between the aerosol tem-
perature and the radiative temperature of the troposphere
below. Thus, the largest differential infrared heating by
4.21
-------
TROPICAL/MIDLATITUDE PROCESSES
aerosol particles of a given size would occur over warm
surfaces, such as are found in the tropical troposphere.
The absorption of solar radiation by sulfuric acid
particles (of any size) is small. This contrasts with, for
example, ash particles, which would be expected to ab-
sorb solar radiation strongly.
4.63.2 HEATING BY MT.PINATOBO AEROSOLS
The Mt Pinatubo eruption injected huge amounts
of SO2 into the stratosphere. Ash particles also injected
may have caused transitory local heating before falling
back into the troposphere. The subsequent conversion of
SO2 into H2SO4/H2O particles (-75% sulfuric acid solu-
tion droplets at normal stratospheric temperatures, e.g.,
. Sheridan et al., 1992) generated, within several weeks,
sharp increases not only in aerosol abundance and but
also aerosol size (e.g., Valero and Pilewskie, 1992).
Infrared radiative effects dominated in the pres-
ence of the sulfuric acid aerosols particles, which were
transparent to solar radiation. Heating in the stratospher-
ic aerosol layer was strongest over tropical latitudes, not
only because most of the aerosol was initially confined
to that region, but also because there, the tropospheric
radiative temperatures were highest.
Reductions in aerosol optical depth occurred in the
tropics as aerosol was transported to higher latitudes arid
sedimentation took place (Stowe et al., 1992; Mer-
genthaler et aL, 1993). The resulting reduction in
differential infrared heating was partially compensate
by growth of aerosols to larger sizes (Dutton et al,
1994). Thus, calculated infrared heating anomalies in
the tropical lower stratosphere, which reached a maxi-
mum of about 0.4 K/day immediately following the
eruption (e.g., Brasseur and Granier, 1992), remained at
above 0.2 K/day for at least another year.
4.633 AEROSOL HEATING AND INDUCED RESPONSE
The relationship between aerosol heating and the
induced circulation is complex. Locally, temperatures
are determined both by local heating and by the remote-
ly-forced meridional circulation. The stratospheric
meridional circulation is, in turn, controlled not only by
radiation but also by midlatitude wave driving (Hayrnes
et al., 1991), although the control by the latter may be
less complete in the tropics than in midlatitudes. Model-
ing studies (Pitari, 1993) established that changes in (he
Brewer-Dobson circulation and in planetary wave be-
havior would occur in response to tropical temperature
changes resulting from the increased aerosol loading.
There are also a number of feedback effects, many
of which are negative, implying that the actual response
would be weaker than radiative calculations alone would
suggest For example, there is a negative ozone feed-
back effect: enhanced upward motion in the tropical
lower stratosphere would reduce ozone concentrations,
resulting in smaller (mainly solar but also infrared)
ozone heating in the aerosol layer. Also, local warming
would be expected to reduce directly the infrared radia-
tive heating rate. j
Following the eruption, temperature anomalies of
2-3 K were observed in the tropical lower stratosphere
(LabitzkeandMcConnick, 1992). Brasseur and Granier
(1992) and Pitari (1993), using 2-D and 3-D models, re-
spectively, have calculated radiative heating anomalies
of around 0^2 K/day in that region and anomalous up-
welling of around 0.05 mm/s through much, of the
tropical stratosphere during the Northern winter. Kinne
et al. (1992) deduced a stronger circulation response in a
calculation that did not include dynamical feedbacks.
Tracer observations confirmed this picture, indicating
enhanced upward motion over the central tropics (e.g.,
G. Toon et al., 1993). ;
Ozone concentrations in the tropical lower strato-
sphere were reduced well into the second year after the
Mt. Pinatubo eruption (Grant et al., 1992). The ozone
reductions immediately following the eruption may be
explained almost entirely by aerosol-induced upwelling
(Kinne et al., 1992). For longer-term changes,' chemical
as well as dynamical effects must be considered. Calcu-
lations of ozone reduction arising from circulation
changes have been made in 2-D (Brasseur and Granier,
1992; Tie et al., 1994) and 3-D (Pitari, 1993; Pitari and
Rizi, 1993) models. '
In the tropics, models calculate column'ozone re-
ductions of the order of 5%. In the model of Pitari and
Rizi (1993), this reduction was attributed largely to
changes in photolysis rates, with the direct effect of cir-
culation changes being small (0-2%). ; However,
Brasseur and Granier (1992) and Tie et al. (1994) sug-
gest that circulation changes led to somewhat larger
reductions (up to -5%) in column ozone in the months
immediately following the eruption. Loss of tropical
4.22
-------
ozone in the tropics by heterogeneous chemical pro-
cesses was, in all instances, found to be smaiil.
At midlatitudes, circulation changes were found to
lead to generally small reductions in ozone in the South-
ern (summer) Hemisphere, but to significant increases
(2-8%) at middle and high latitudes in the Northern
(winter) Hemisphere in the year following the eruption.
However, in the winter hemisphere, the column ozone
increases due to transport were more than offset by large
(>10%) losses due to heterogeneous chemistry that was
more effective largely because of reduced winter photol-
ysis rates, leaving widespread net ozone reductions of
5-10%. !
Overall, the calculations suggest that as a result of
the Mt. Pinatubo eruption, chemical effects, through het-
erogeneous reactions and changes in photolysis rates,
appear to be the major factors leading to ozone changes
globally; changes in atmospheric transport are likely to
have produced significant regional effects.
4.7 TRANSPORT OF AIR FROM POLAR
REGIONS TO MIDDLE LATITUDES
4.7.1 Transport of Air from High Latitudes:
Possible Influence on Midlatitude Ozone
Loss
There are differences of view regarding the impor-
tance of the transport of air through polar regions to
middle latitudes and its impact on midlatitude ozone
loss. One view is that containment of air within the polar
vortex is, to a good approximation, complete during win-
ter, and that virtually all transport occurs as the polar
vortex breaks down during spring. During this process,
air from within the polar vortex, in which ozone may
have been depleted, mixes with low-latitude air and re-
duces the midlatitude ozone column purely by dilution.
There is then a clear uppeolimit on ozone loss: no more
ozone can be destroyed than the amount contained with-
in the polar vortex when it first forms in early winter.
An alternative view is that expressed as the "flow-
ing processor hypothesis," namely, that the air in polar
regions is not well contained and that a substantial vol-
ume of air passes through those regions to middle
latitudes throughout the winter months. If vortex tem-
peratures are low enough, then polar stratospheric
clouds (PSCs) will form within the vortex and heteroge-
TROPICAMVHDLATITUDE PROCESSES
I.
neous chemistry will cause reactive chlorine concentra-
tions to rise. Denitrification, which allows active
chlorine compounds to persist for longer, may also occur
(as may dehydration). ; Large amounts of air passing
through the vortex to middle latitudes could thus be
chemically primed for ozone loss. In such a situation,
although temperatures in middle latitudes may never
have reached the threshold for PSC formation, the ef-
fects of heterogeneous PSC chemistry (and dehydration)
would still be apparent.: Midlatitude ozone loss could
then proceed, initiated by the polar air. Because the vol-
ume of lower-stratospiheric air exposed to PSC
chemistry could be substantially greater than the instan-
taneous volume of the polar vortex, the potential for
ozone loss would be significantly enhanced over simple
dilution. Two main transport pathways have been pro-
posed: that air from the polar vortex spreads outwards
throughout the hemisphere during the winter at altitudes
up to -35 km (10 hpa); or that air descends rapidly
through the lower boundary of the vortex at about 0 =
400 K to the sub-vortex region, where it can be transport-
ed to lower latitudes. i
It is also possible that chlorine activation is not
confined to the polar vortex, but could occur on PSCs or
on sulfate aerosol outside, the vortex, either in the region
surrounding the vortex edge or in the very low strato-
sphere below about 6 = 400 K where the polar vortex is
less well defined. PSC formation in both regions is cer-
tainly possible, and indecxl is likely at the temperatures
observed in winter. >
These different scenarios have very different im-
plications for understanding and predicting midlatitude
ozone loss. j
4.7.2 Fluid-Dynamica) Considerations
The present perception of polar-vortex dynamics
is that the vortex, above a "transition isentrope" at about
400K, is very likely to behave in a manner similar to that
of idealized polar vortices in single-layer, high-resolu-
tion models (e.g., Legras'and Dritschel, 1993; Norton,
1994 and references therein), in laboratory experiments
(&g., Sommeriaera/., 1989), and in the multi-layer, low-
resolution models now being run by several groups.
From these studies, it appears mat the edge of the vortex
acts as a flexible barrier, 'strongly inhibiting the large-
scale, quasi-horizontal eddy transport. However, none
of the models suggests that this inhibition is complete.
4.23
-------
TROPICAL/MIDLATITUDE PROCESSES
Fluid-dynamical evidence points not to mean out-
flow but to weak mean inflow. It is just such inflow that
creates the vortex on the seasonal time scale. Total pole-
ward parcel displacements of a few degrees in latitude
are enough to create the vortex; these displacements aire
of the same order as would be given by a simplistic angu-
lar momentum budget for a frictionless, exactly circular
vortex.
Some additional inflow, of a similar order of mag-
nitude, is required to maintain the vortex against Rossby
and gravity wave drag. Conversely, if a strong mean out-
flow were to exist in the real wintertime polar lower
stratosphere, then a strong eastward force of unknown
origin would have to be acting to maintain the vortex.
An outflow strong enough to conform to the flowing
processor hypothesis in its extreme form, i.e., an outflow
strong enough to export, say, three vortex masses per
winter, would, in the absence of an eastward force, oblit-
erate the vortex on a short time scale of the order of 5
days (Mclntyre, 1994).
Eddy-induced erosion of the vortex can act against
the mean inflow. Outward eddy transport is limited by
the rate at which the vortex edge can be eroded by epi-
sodic disturbances and associated midlatitude stirring.
There is remarkable consistency with which strong
eddy-transport inhibition is predicted over a big range of
artificial model diffusivities, and of numerical resolu-
tions from effective grid sizes from about 10 degrees of
latitude (Pierce and Fairlie, 1993) through 1 degree of
latitude (Juckes and Mclntyre, 1987) to effectively infi-
nite resolution obtained via adaptive Lagrangian
numerics (e.g., Legras and Dritschel, 1993; Dritschel
and Saravanan, 1994). Model studies that use either
Eulerian techniques or high-resolution Lagrangian
advection techniques (contour advection or many-
particle) on model-generated wind fields or on meteoro-
logically analyzed wind fields have also been performed
(e.g., Pierce and Fairlie, 1993; Manney et al, 1994;
Norton, 1994; Rood et al., 1992; Fisher et al., 1993;
Waugh, 1994b; Waugh et al, 1994; Chen et al., 1994).
All give weak erosion rates, in the sense that the mass
transported is, conservatively, no more than about a third
of a vortex mass per month on average, regardless of the
ambiguity in defining the vortex edge (due to its filamen-
tary fine structure) and" regardless of the very wide range
of model resolutions and artificial model eddy diffusivi-
However, the possible roles of unresolvable mo-
tions such as breaking inertia-gravity waves in the lower
stratospheric vortex edge have yet to be quantified.
Several other transport mechanisms should be
considered. For example, if there is significant descent
of vortex air into the sub-vortex region, this could be rap-
idly dispersed throughout the hemisphere. However, in
order to sustain a large enough transport through the vor-
tex, diabatic descent rates within the lower-stratospheric
vortex would need to be much greater than seems com-
patible with observed temperatures and with very
extensive studies in atmospheric radiation, whose phys-
ics is fundamentally well understood (e.g., Schoeberl et
al., 1992). ;
Another possibility is that the sub-vortex region
below the transition isentrope around 400-425 K could
itself act as a "flowing processor." The transition isen-
trope exists because of stirring by anticyclones and other
synoptic-scale meteorological disturbances underneath
the vortex and the upward-evanescent character of the
relevant waves. There is less inhibition of quasi-hori-
zontal eddy transport at these lower levels: large eddy
exchanges of midlatitude air with the sub-vortex region
are thus expected (Tuck, 1989; Mclntyre, 1994). The
sub-vortex region is also cold enough, in the late Antarc-
tic winter at least, to produce chlorine activation,
dehydration, and denitrification (Jones and Kilbane-
Dawe, 1994).
It is also possible that chlorine activatiqn can take
place in air parcels that are above the transition isentrope
but are outside the vortex (see for example, Jones et al.,
1990; MacKenzie et al, 1994; and Pyle et al, 1994).
There is evidence that all three mechanisms
(shown schematically in Figure 4-1) are realized to some
degree, and presumably in different proportions in the
South and North. Large transport rates of air from with-
in the vortex seem likely to occur only when, the vortex
breaks down. However, the above arguments suggest
that while transport due to vortex erosion may play a no-
ticeable role in midlatitude ozone loss, it would not
appear to be the dominant one.
ties.
4.24
-------
TROPICAMWIIDLATITUDE PROCESSES
4.7.3 Observational Studies Relating to
Transport through the Vortex
4.7.3.1 EXCHANGE OF AIR ACROSS THE VORTEX
BOUNDARY
The appearance near the vortex edge of filamen-
tary structure in many species (see, e.g., Murphy et al,
1989; Tuck et al., 1992; and below), and features such as
laminae in ozone profiles (Reid et al., 1993)! is highly
suggestive that some irreversible exchange of air is oc-
curring across the boundary of the vortex. A number of
studies have attempted to quantify outflow rates.
Using data from the Halogen Occupation Experi-
ment (HALOE), Russell et al. (1993a) deduced a time
constant for the-'replacement of air between 15 and 20
km by horizontal transfer of less than a month in October
1991. Tuck et al. (1993), using HALOE data/have also
suggested that dehydration originating from within the
Antarctic polar vortex spread over the entire; Southern
Hemisphere up to the 10-20 hPa region. The; extent of
the dehydration implied, they argued, that vortex air was
being flushed out "several times" during the winter
months. However, subsequent revisions of these satellite
data based on an improved retrieval (Chapter 3; Figure
3-18) have markedly reduced the vertical and latitudinal
extents of the dehydration apparent in the data/implying
significantly lower outflow rates than these early studies
suggested.
Several other studies have suggested relatively
rapid exchange. Tao and Tuck (1994) examimsd the dis-
tribution of temperatures with respect to the vortex edge
in the Southern winter of 1987 and the Northern winter
of 1988-1989. They find that there is evidence of air
chemically unprocessed by PSCs being dynamically re-
supplied to the vortex, they argue by mixing and descent.
Tuck et al. (1994) used ER-2 observations of NOy from
the airborne missions in 1987, 1988-1989, and 1991-
1992 to attempt to quantify" vortex outflow rates. From
the appearance of hemispheric and interanniial differ-
ences in midlatitude NOy, they concluded that (he vortex
must have been flushed more than once during the peri-
od of denitrification and dehydration.
The latter study is hard to reconcile with a number
of other studies. Proffitt et al. (1989) argue that NOy and
N2O observations obtained during the 1987 airborne
mission point rather to significant inflow and descent
(see Section 4.7.3.2). However, the extent of inflow pro-
posed by Proffitt et al. is!probably inconsistent with the
angular momentum budget (e.g., Plumb, 1990). Jones
and MacKenzie (1994) also used the observed relation-
ship between NOy and N2O concentrations to attempt to
quantify the transport of air from the polar regions to
midlatitudes. In that study some instances when recent-
ly denitrified air was found outside the vortex were
observed, arguing against complete containment, but
these features were small in scale. However, they found
no evidence of large-scale outflow of air from the polar
vortices above 6 = 400 K,
4.73.2 DESCENT OF AIR THROUGH THE LOWER
BOUNDARY OF THE VORTEX
Proffitt et al. (1989.J 1990, 1993) and Tuck (1989)
have argued that the descent of ozone-depleted air
through the lower boundary of the polar vortex, where it
can be dispersed to lower latitudes, can significantly re-
duce midlatitude ozone ; amounts. Using statistical
relationships between O3 ^d N2O, Proffitt et al. (1993)
deduced altitude-dependent changes in ozone during the
Northern winter of 1991-1992, with decreases at the bot-
tom of the vortex and increases at the highest altitudes
accessible to the ER-2 aircraft. The increase aloft was
attributed to ozone-rich iair entering the vortex from
above, while the reduction lower down was taken to be
the result of chlorine-catalyzed loss during descent
through the region of PSC 'formation. Basing the rate of
downward motion of air oii a cooling rate of 1 K/day (in
6), ozone-depleted air released from the bottom of the
vortex in 1991-1992 was.jthey argued, sufficient to re-
duce significantly the ozone column in middle latitudes.
Using the same methodology, Collins et al. (1993) made
measurements of the N2O-O3 correlation from the DC-8,
which they interpret as showing descent of vortex air
over the Arctic to levels just above the tropopause during
the winter of 1991-1992; they also show 20% ozone re-
ductions in March relative to January and February.
However, there is considerable debate about the
importance of the descent cjf air through the vortex lower
boundary for midlatitude ozone. The efficacy of such a
process will depend on the irate of descent of air, and thus
on diabatic cooling rates. For example, the cooling rate
used in Proffitt et al., 1993:; (1 K/day) is a factor of 2-10
larger than other published estimates of diabatic cooling
rates (e.g., Schoeberl era/.,.; 1992).
4.25
-------
TROPICAL/MIDLATITUDE PROCESSES
4.733 TRANSPORT IN THE SUB-VORTEX REGION
There is less inhibition of quasi-horizontal eddy
transport at the levels below the vortex (6 = 400-425 K)
and large eddy exchanges of midlatitude air with the
sub-vortex region are thus expected (Tuck, 1989; Mc-
Intyre, 1994). Jones and Kilbane-Dawe (1994) have
investigated the extent to which ozone can be reduced in
this region by in situ chemical processes rather than by
transport of ozone-depleted air from the vortex above.
They pointed out that temperatures in this region fall sig-
nificantly below the threshold for polar stratospheric
cloud formation, and thus for chlorine activation, for
much of the Southern winter and, although more vari-
able, the same is frequently seen in the North. Using
ozonesonde measurements made during the 1987 South-
em winter, Jones and Kilbane-Dawe (1994) identified
significant reductions of ozone in the sub-vortex region
(-350K) extending to ~55°S. As these reductions oc-
curred at a time when temperatures were cold enough for
chlorine activation, and when the ozone vertical gradient
was such that diabatic descent would have increased
ozone mixing ratios, they attributed these reductions to
in situ photochemical loss. The total in situ ozone loss in
the sub-vortex region was, in 1987, a significant fraction
of the overall hemispheric reduction. They also argue
that in the Northern Hemisphere in some winters (eg.,
the winter of 1993-4) the sub-vortex region may allow a
significant fraction of lower latitude air to become chlo-
rine-activated, and may, in some years, be more
important than at higher altitudes.
4.7.4 Model Studies Relating to Transport
through the Vortex
Since the last WMO/UNEP assessment, a number
of new modeling studies have been carried out to investi-
gate the extent to which air is mixed between the polar
vortex and middle latitudes. Many of these have concen-
trated on the Arctic vortex, studied extensively in the two
polar campaigns, EASOE and AASEII, in the winters of
1991/92. Studies using UARS data have also appeared.
The studies all show mixing, to a greater or lesser
extent, but most support the idea of only limited flow
through the polar vortex. However, it should be pointed
out that chemical tracer measurements suggest that
structure exists on scales so far unresolved by even the
highest resolution analyses but, as indicated above.
whether this represents a fatal flaw in model studies is
not clear.
In separate studies of the AASE and Airborne Ant-
arctic Ozone Experiment (AAOE) data, invojving the
reconstruction of the chemical constituent fields in PV-6
space followed by an estimation of the meridional circu-
lation and eddy diffusivities, Schoeberl et al. (1992)
reached similar conclusions to the earlier study of Hart-
mann et al. (1989): that the center of the vortexj is highly
isolated but that exchange of trace gases do|es occur,
principally at the vortex edge, by erosional wavfc activity.
Consistent with earlier studies, Rood et al. (1992) con-
clude that intense cyclonic activity close to tjie vortex
edge and large planetary-scale events are the major
mechanisms of extra-vortex transport. Nevertheless, in
their study of a disturbed period in January and February
1989, only a small amount of vortex air was found at
lower latitudes. j
Erosion at the vortex edge has been demonstrated
in greater detail in a number of new studies using the
technique of contour advection with surgery (Norton,
1994; Waugh, 1994b; Plumb et al., 1994; Waiigh et al.,
1994). Results from these studies show thin; filaments
being dragged around the vortex edge and being carried
into middle latitudes. An example during a disturbed
period in January 1992 is shown in Figure 4-15. The fine
structure evident in the figure is consistent with potential
vorticity, but reveals structure on scales not resolvable in
the PV maps. Estimates of the outflow of vortex air to
midlatitudes by Waugh et al, 1994 (see Table 4-1) sug-
gest, however, that while major erosion event's do occur
(e.g., Jan. 16-28), the net outflow of air from* the vortex
appears small, at least above 400 K. Similar conclusions
are drawn by Pierce and Fairlie (1993) in a study of the
evolution of material lines; by Strahan and Mahlman
(1994), who compared high resolution general circula-
tion model results with N2O. observations near the vortex
edge; and by Dahlberg and Bowman (1994), who carried
out isentropic trajectory studies for nine Northern Hemi-
sphere winters. j
A number of studies have attempted to model the
chemical effects relating to ozone loss. Chipperfield et
al. (1994b) and Chipperfield (1994) studied the Arctic
winters of 1991/92 and 1922/93. Figure 4-16 shows the
distribution of a tracer indicating that air has experienced
the conditions (low temperatures and sunlijght) neces-
sary for rapid ozone loss. Most of the tracer is well
4.26
-------
TROPICA17MIDLATITUDE PROCESSES
Figure 4-15. High resolution evolu-
tion of the vortex on the 450 K isen-
tropic surface, 16-28 January 1992,
as determined using contour advec-
tion with surgery. Model contours
were initialized on 16 January on
potential vorticity contours from the
NMC analysis. Subsequently the
contours were advected with the daily
analyzed 450 K balanced winds.
Note the transport of a significant
volume of air to midlatitudes near
165°E. (F:rom Plumb etal., 1994.)
.°ffthir transP°rted out °f the ™rtex expressed as a percentage of the vortex area
and 30°N (P2) during selected periods during the
or
Period
Dec. 6 - 18, 1991
Dec. 16 - 26, 1991
Dec. 23, 1991 to Jan. 2, 1992
Jan. 1-11, 1992
Jan. 7 - 17, 1992
Jan. 16-28, 1992
Feb. 2-11, 1992 J
Feb. 9 - 19, 1992
Feb. 19 - 28, 1992
1
1
5
7
3
31
2
0
5
0
0
1
2
1
7
0
0
3
4.27
-------
TROPICAL/MIDLATITUDE PROCESSES
Figure 4-16. Modeled distribution of
a tracer showing chlorine activation
and exposure to sunlight on the 475
K potential temperature surface for
January 20,1992. In this case, the
tracer is the number of hours of
ozone destruction. As in Figure 4-
15, significant transport outside the
vortex is seen near 165°E. High
CIOX concentrations are calculated
to be present in the same region.
(From Chipperfield etal., 1994b.)
Figure 4-17. Trajectory endpoints
for 28 January 1992. The trajecto-
ries were initialized on the 475 K po-
tential temperature surface 10 days
earlier close to the edge of the polar
vortex in a region favorable for PSC
formation and in which chlorine acti-
vation was expected to have oc-
curred. While the majority of the
trajectories remained on the vortex
edge, a significant number became
detached from the vortex: In the lat-
ter group (shown near the Black
Sea) O3 losses of -1% per day were
calculated. (From Pyle etal., 1994.)
4.28
-------
TROPICAL/MIDLATITUDE PROCESSES
within the polar vortex, but there is an activated region
moving away from the vortex edge at 165°E. The level
of consistency between these results and the contour ad-
vection results of Plumb etal. (1994) for the same period
(Figure 4-15) is striking, particularly given their differ-
ences of approach.
A number of recent studies using trajectories con-
firm the notion that chlorine activation can occur at the
vortex edge (see, e.g., Jones etal., 1990). MacKenzie et
al. (1994), using ensembles of isentropic trajectories,
found examples of air that had been chlorine-activated
by PSCs outside the vortex. However, in these cases the
relevant low temperatures had been encountered at PV
values characteristic of the vortex edge. Such a conclu-
sion is consistent with the work of Tao and Tuck (1994).
However, MacKenzie et al. (1994) found no evidence of
air having been ejected from the center of the vortex.
Broadly similar results were obtained by Pierce et al.
(1994), who performed trajectory studies to analyze the
HALOE data in the Southern Hemisphere. In an explicit
calculation of photochemical ozone loss, Pyle et al.
(1994) used trajectories to show that a region of high PV
chlorine-activated air had been eroded from the vortex to
45°N in January 1992 (see Figure 4-17). Ozoiie deple-
tions of around 0.1% per day were calculated for those
trajectories staying close to the vortex edge. However,
for the trajectories that moved to lower latitudes., ;a deple-
tion of the order of 1% per day was calculated.
Thus, while it appears that the processes discussed
in Section 4.7.1 are indeed affecting midlatitude ozone
amounts, these modeling studies suggest that their im-
pact is modest.
REFERENCES
Ab'batt, J.P.D., Heterogeneous reaction of HOBr with
HBr and HC1 on ice surfaces at 228 K, Geophys.
Res. Lett., 21, 665-668, 1994.
Avallone, L.M., D.W. Toohey, W.H. Brune, R.J. Sala-
witch, A.E. Dessler, and J.G. Anderson,
Balloon-borne in situ measurements of CIO and
ozone: Implications for heterogeneous chemistry
and mid-latitude ozone loss, Geophys. Res. Lett.,
20, 1794-1797, 1993a.
Avallone, L.M., D.W. Toohey, M.H. Proffitt, J.J.
Margitan, K.R. Chan, and J.G. Anderson, In situ
measurements of CIO at mid-latitudes: Is there an
effect from Mt. Pinatubo?, Geophys. Res. Lett., 20,
2519-2522, 1993b. |
Baldwin, M.P., and T.J.; Dunkerton, Quasi-biennial
oscillation above 10,mb, Geophys. Res. Lett, 18,
1205-1208, 1990. j
Bekki, S., and J. A. Pyle, 2-5 assessment of the impact of
aircraft sulphur emissions on the stratospheric
sulphate aerosol layer, J. Geophys. Res., 97,
15839-15847, 1992. [ ' '
Bekki, S., R. Toumi, and J.A. Pyle, Role of sulphur
photochemistry in tropical ozone changes after the
eruption of Mt. Pinatubo, Nature, 362, 331-333
1993. • |
Bithel, M., LJ. Gray, J.E. Harries, J.M. Russell, and A.F.
Tuck, Synoptic interpretation of measurements
from HALOE, / Afmos. ScL, 51, 2942-2956,
1994. j
Bluth, G.J.S., S.D. Doiron, A.J. Krueger, I.S. Walter, and
C.C. Schnetzler, Global tracking of the SO2 cloud
from the June 1991 Nit. Pinatubo eruptions, Geo-
phys. Res. Lett., 19, 151-154, 1992.
Boering, K.A., B.C. Daub, S.C. Wofsy, M. Loewenstein,
J.R. Podolske, and E.R. Keim, Tracer-tracer rela-
tionships and lower stratospheric dynamics: CO2
and N2O correlations during SPADE, Geophys.
Res. Lett., 21, 2567-2570, 1994.
Bojkov, R.D., C.S. Zerefos, D.S. Balis, I.C. Ziomas, and
A.F. Bais, Record lowjtotal ozone during Northern
winters of 1992 and 1993, Geophys. Res. Lett., 20,
1351-1354, 1993. i
Borrmann, S., et al.. In situ measurements of changes in
stratospheric aerosol ;and the N2O-aerosol rela-
tionship inside and outside of the polar vortex,
Geophys. Res. Lett., 20, 2559-2562, 1993.
Boville, B.A., J.R. Holton, and P.W. Mote, Simulation of
the Pinatubo aerosol cloud in general circulation
model, Geophys. Res.'Lett., 18, 2281-2284, 1991.
Brasseur, G.P., The response of the middle atmosphere
to long-term and short-term solar variability: A
two-dimensional model, / Geophys. Res., 98,
23079-23090, 1993. j
Brasseur, G.P., and C. Granier, Mt. Pinatubo aerosols,
chlorofluorocarbons, and ozone depletion, Science,'
257, 1239-1242, 1992J
4.29
-------
TROPICAL/MIDLATITUDE PROCESSES
Brock, C.A., H.H. Johnson, J.C. Wilson, I.E. Dye, D.
Baumgardner, S. Borrmann, M.C. Pitts, M.T.
Osbom, D.J. DeCoursy, and D.C. Woods, Rela-
tionships between optical extinction, backscatter
and aerosol surface and volume in the stratosphere
following the eruption of Mt. Pinatubo, Geophys.
Res. Lett., 20, 2555-2558, 1993.
Browell, E.V., C.F. Butler, M.A. Fenn, W.B. Grant, S.
Ismail, M.R. Schoeberl, O.B. Toon, M. Loewen-
stein, and J.R. Podolske, Ozone and aerosol
changes during the 1991-1992 Airborne Arctic
Stratosphere Expedition, Science, 261, 1155-
115B, 1993.
Burkholder, J.B., R.K. Talukdar, A.R. Ravishankara, and
S. Solomon, Temperature dependence of the
HNOa UV absorption cross sections, J. Geophys.
Res., 98,22937-22948, 1993.
Cebula, R., and E. Hilsenrath, Ultraviolet solar irradi-
ance measurements from the SSBUV-1 amd
SSBUV-2 missions, in Proceedings of the Work-
shop of the Solar Electromagnetic Radiation Study
for Solar Cycle 22, edited by RJF. Donnelly,
NOAA Environmental Research Laboratori.es,
Boulder, Colorado, 1992.
Chandra, S., C.H. Jackman, A.R. Douglass, E.L. Flem-
ing, and D.B. Considine, Chlorine-catalyzed
destruction of ozone: Implications for ozone vari-
ability in the upper stratosphere, Geophys. Res.
Lett., 20, 351-354, 1993.
Chen, P., and J.R. Holton, Mass exchange between the
tropics and extratropics in the stratosphere, /.
Atmos. Sci., submitted, 1994.
Chen, P., J.R. Holton, A. O'Neill, and R. Swinbank,
Quasi-horizontal transport and mixing in the
Antarctic stratosphere, /. Geophys. Res., 99,
16851-16866, 1994.
Chipperfield, M.P., A 3-D model comparison of PSC
processing during the Arctic winters of 1991/92
and 1992/93, Ann. Geophys., 12, 342-354, 1994.
Chipperfield, M.P., L.J. Gray, J.S. Kinnersley, and J.
Zawodny, A two-dimensional model study of the
QBO signal in SAGE II NO2 and O3, Geophys.
Res. Lett., in press, 1994a.
Chipperfield, M.P., D. Cariolle, and P. Simon, A three-
dimensional transport model study of PSC
processing during EASOE, Geophys. Res. Utt.,
21, 1463-1466, 1994b.
Coffey, M.T., and W.G. Mankin, Observations pf the loss
of stratospheric NO2 following volcanic eruptions,
Geophys. Res. Lett., 20, 2873-2876, 1993.
Cohen, R.C., et al, Odd nitrogen in the loWer strato-
sphere: New constraints from the; in situ
measurement of hydroxyl radicals, submitted to
Geophys. Res. Lett., 1994. ;
Collins, J.E., G.W. Sachse, B.E. Anderson, A.J. Wein-
heimer, J.G. Walega, and B.A. Ridley, AASE-H
in-situ tracer correlations of methane, nitrous-
oxide, and ozone as observed aboard the DC-8,
Geophys. Res. Lett., 20,2543-2546, 1993.
Crutzen, P.J., and U. Schmailzl, Chemical budgets of the
stratosphere, Planet. Space Sci., 31, 1009-1032,
1983.
Dahlberg, S.P., and K.P. Bowman, Climatology of large
scale isentropic mixing in the Arctic winter strato-
sphere, submitted to J. Geophys. Res., 1994.
DeMore, W.B., S.P. Sander, D.M. Golden, R.F. Hamp-
son, MJ. Kurylo, C.J. Howard, A.R.; Ravishan-
kara, C.E. Kolb, and MJ. Molina, Chemical
Kinetics and Photochemical Data for Use in
Stratospheric Modeling, Evaluation No,. 9, NASA
JPL Publication 92-20, Jet Propulsion Laboratory,
Pasadena, California, 1992. ;
Deshler, T., D.J. Hofmann, B.J. Johnson, and W.R. Ro-
zier, Balloonborne measurements of the Pinatubo
aerosol size distribution and volatility at Laramie,
Wyoming during the summer of 1991, Geophys.
Res. Lett., 19, 199-202, 1992. '•
Deshler, T, B.J. Johnson, and W.R. Rozier, Balloon-
borne measurements of Pinatubo aerosol during
1991 and 1992 at 41°N: Vertical prbfiles, size
distribution and volatility, Geophys. Res. Lett., 20,
1435-1438, 1993. ;
Dessler, A.E., R.M. Stimpfle, B.C. Daube,( R.J. Sala-
witch, E.M. Weinstock, D.M. Judah, J.D. Burley,
J.W. Munger, S.C. Wofsy, J.G. Anderson, M.P.
McCormick, and W.P. Chu, Balloonborne mea-
surements of CIO, NO and O3 in a volcanic cloud:
An analysis of heterogeneous chemistry between
20 and 30 km, Geophys. Res. Lett., 20,2527-2530,
1993. :
Donnelly, R.F., Uniformity in solar UV flux variations
important to the stratosphere, Ann. Geophys., 6,
417-424, 1988. |
4.30
-------
TROPICAL/MIIDLATITUDE PROCESSES
Dritschel, D.G., and R. Saravanan, Three-dimensional
quasi-geostrophic contour dynamics, with an
application to stratospheric vortex Dynamics,
Quart. J. Roy. Meteorol. Soc., in press, 1994.
Dunkerton, T.J., and M.P. Baldwin, Modes of interannual
variability in the stratosphere, Geophys.Res. Lett.,
79,49-52,1992.
Dutton, E.G., P. Ready, S. Ryan, and J.J. DeLuisi, Fea-
tures and effects of aerosol optical depth observed
at Mauna Loa, Hawaii: 1982-1992, J. Geophys.
Res., 99, 8295-8301, 1994.
Dyer, A.J., and B.B. Hicks, Global spread of the volcanic
dust from the Bali eruption of 1963, Quart. J. Roy.
Meteorol. Soc., 94, 545-554, 1968.
Eluszkiewicz, J., and M. Allen, A global analysis of the
ozone deficit in the upper stratosphere and meso-
sphere, /. Geophys. Res., 98, 1069-1081, 1993.
Fahey, D.W., S.R. Kawa, E.L. Woodbridge, P. Tin, J.C.
Wilson, H.H. Jonsson, J.E. Dye, D. Baumgardner,
S. Borrmann, D.W. Toohey, L.M. Avallone, M.H.
Proffitt, J. Margitan, M. Loewenstein, J.R. Podol-
ske, R.J. Salawitch, S.C. Wofsy, M.K.W. Ko, D.E.
Anderson, M.R. Schoeberl, and K.R. Clian, In situ
measurements constraining the role of sulphate
aerosols in mid-latitude ozone depletion, Nature,
363, 509-514, 1993. '.
Fahey, D. W., et al.. The removal of NO near the tropo-
pause at mid-latitudes: A photochemical anomaly,
NASA High Speed Research Program, Annual
Meeting, Virginia Beach, Virginia, 1994.
Feely, H.W., and H. Spar, Tungsten-185 from nuclear
bomb tests as a tracer for stratospheric! meteoro-
logy. Nature, 188, 1062-1064, 1960. ;
Fisher, M., A. O'Neill, and R. Sutton, Rapid descent of
mesospheric air into the stratospheric polar vortex,
Geophys. Res. Lett., 20,. 1267-1270, 1993.
Fleming, E., S. Chandra, C. Jackman, D. Considine, and
A. Douglass, The middle atmospheric response to
short- and long-term solar UV variations: Analysis
of observations and 2-D model results, J. Atmos.
Terr. Phys., in press, 1994.
Froidevaux, L., J.W. Waters, W.G. Read, L.S. Elson,
D.A. Flower, and R.F. Jarnot, Global ozone obser-
vations from the UARS MLS: An overview of
zonal-mean results, /. Atmos. ScL, 51, 2846-2866,
1994.
Fujiwara, M., T. Shibata.land M. Hirono, Lidar observa-
tions of a sudden increase of aerosols in the
stratosphere caused by volcanic injections, II.
Sierra Negra event,'/. Atmos. Terr. Phys., 44, 811-
818, 1982. |
Garcia, R.R., Parameterization of planetary wave break-
ing in the middle atmosphere, J. Atmos. Sci., 48,
1405-1419, 1991. !
Garcia, R.R., S. Solomon, R.G. Roble, and D.W. Rusch,
A numerical study of the response of the middle
atmosphere to the1 11-year solar cycle, Planet.
Space ScL, 32, 411-423, 1984.
Garcia, R.R., and S. Solomon, A new numerical model
of the middle atmosphere 2. Ozone and related
species,/ Geophys. Res., 99, 12937-12951,1994.
Gleason, J.F., P.K. Bharti.i, J.R. Herman, R. McPeters, P.
Newman, R.S. Stolarski, L. Flynn, G. Labow, D.
Larko, C. Seftor, C. Wellemeyer, W.D. Komhyr,
A.J. Miller, and W. Planet, Record low global
ozone in 1992, Science, 260,523-526, 1993.
Godin, S., C. David, A. Sarkissiah, F. Goutal, P. Aime-
dieu, and E. Le Bouar, Systematic measurements
at the Antarctic station of Dumont d'Urville, in
Proceedings of the Quadrennial Ozone Confer-
ence, Charlottesville, Virginia, 1992.
Grant, W.B., J. Fishman, E.V. Browell, V.G. Brackett, D.
Nganga, A. Minga!, B. Cros, R.E. Veiga, C.F.
Butler, M.A. Fenn,;and G.D. Nowicki, Observa-
tions of reduced ozone concentrations in the
tropical stratosphere after the eruption of Mt.
Pinatubo, Geophys, Res. Lett., 19, 1109-1112,
1992. j
Grant, W.B., E.V. Browell, J. Fishman, V.G. Brackett,
, R.E. Veiga, D. Nganga, A. Minga, B. Cros, C.F.
Butler, M.A. Fenn, C.S. Long, and L.L. Stowe,
Aerosol-associated j changes in tropical strato-
spheric ozone following -the eruption of Mt.
Pinatubo, /. Geophys. Res., in press, 1994.
Gray, L.J., and T.J. Dunkeirton, The role of the seasonal
cycle in the quasi-biennial oscillation in ozone, /.
Atmos. ScL, 47, 2429-2451, 1990.
Hanson, D.R., J.B. Burkholder, C.J. Howard, and A.R.
Ravishankara, Measurement of OH and HO2 radi-
cal uptake coefficients on water and sulfuric acid
surfaces,/ Phys. Chem., 96, 4979-4985, 1992.
4.31
-------
TROPICAL/MIDLATITUDE PROCESSES
Hanson, D.R., and A.R. Ravishankara, The reaction
C1ON02 + HC1 on NAT, NAD and frozen sulfuric
acid and the hydrolysis of N2O5 and C1ONO2 o:n
frozen sulfuric acid, /. Geophys. Res., 98, 22931-
22936, 1993a.
Hanson, D.R., and A.R. Ravishankara, Uptake of HC1
and HOC1 onto sulfuric acid—Solubilities, diffu-
sivities and reaction, /. Phys. Chem., 97,
12309-12319,1993b.
Hanson, D.R., A.R. Ravishankara, and S. Solomon,
Heterogeneous reactions in sulfuric acid aerosols:
A framework for model calculations, J. Geophys.
Res., 99,3615-3629, 1994.
Hanson, D.R., and A.R. Ravishankara, Heterogeneous
•chemistry of bromine species in sulfuric acid
under stratospheric conditions, Geophys. Res.
Lett., in press, 1994.
Hartmann, D.L., L.E. Heidt, M. Loewenstein, J.R. Pod-
olske, J. Vedder, W.L. Starr, and S.E. Strahan,
Transport into the south polar vortex in early
Spring,/. Geophys.Res., 94, 16779-16795, 1989.
Hayashida, S., Atmospheric effects, in Global Volcanism
Network Bull., 16, 1991.
Haynes, P.H., C.J. Marks, M.E. Mclntyre, T.G. Shep-
herd, and K.P. Shine, On the downward control of
extratropical diabatic circulations by eddy-induced
mean zonal forces, /. Atmos. Sci., 48, 651-679,
1991.
Hofmann, D.J., and S. Solomon, Ozone destruction
through heterogeneous chemistry following the
eruption of El Chich6n, J. Geophys. Res., 94,
5029-5041, 1989.
Hofmann, D.J., SJ. Oltmans, W.D. Komhyr, J.M. Har-
ris, J.A. Lathrop, A.O. Langford, T. Deshler, El.J.
Johnson, A. Torres, and W.A. Matthews, Ozone
loss in the lower stratosphere over the United
States in 1992-1993: Evidence for heterogeneous
chemistry on the Pinatubo aerosol, Geophys. Res.
Lett., 21, 65-68, 1994.
Hood, L.L., R.D. McPeters, J.P. McCormack, L.E.
Flynn, S.M. Hollaridswdrth, and J.F. Gleason,
Altitude dependence of stratospheric ozone trends
based on Nimbus-7 SBUV data, Geophys. Res.'
Lett., 20,2667-2670, 1993.
Huang, T.Y.W., and G.P. Brasseur, The effect of long-
term solar variability in a two-dimensional
interactive model of the middle atmosphere, /.
Geophys. Res., 98, 20413-20427, 1993.
Jaeger, H., The Pinatubo eruption cloud observed by
lidar at Garmisch-Partenkirchen, Geophys. Res.
Lett., 19, 191-194, 1992.
Jaeger, H., ESMOSII Final Report, CEC, DG XII, 1993.
Jaegle, L., et al., In situ measurements of the NO2/NO
ratio for testing atmospheric photochemical mod-
els, Geophys. Res. Lett., 21, 2555-2558, \994.
Johnston, P.V., R.L. MacKenzie, J.G. Keys, and W.A.
Matthews, Observations of depleted stratospheric
NO2 following the Pinatubo volcanic ;eruption,
Geophys. Res. Lett., 19, 211-213, 1992. ;
Jones, R.L., D.S. McKenna, L.R. Poole, and S. So-
lomon, On the influence of polar stratospheric
cloud formation on chemical composition during
the 1988/89 Arctic winter, Geophys. Resi Lett., 17,
545-548, 1990. :
Jones, R.L., and I. Kilbane-Dawe, The influence of the
stratospheric sub-vortex region on mid-latitude
composition, submitted to Geophys. Res. Lett.,
1994. i
Jones, R.L., and A.R. MacKenzie, Observational studies
of the role of polar regions in mid-latitude ozone
loss, submitted to Geophys. Res. Lett., 1994.
Juckes, M.N., and M.E. Mclntyre, A high resolution,
one-layer model of breaking planetary; waves in
the stratosphere, Nature, 328,590-596, -1987.
Kawa, S.R., D.W. Fahey, J.C. Wilson, M.R. Schoeberl,
A.R. Douglass, R.S. Stolarski, E.L. Woodbridge,
H. Jonsson, L.R. Lait, PA. Newman, M.H. Prof-
fitt, D.E. Anderson, M. Loewenstein, K.R. Chan,
C.R. Webster, R.D. May, and K.K. Kelly, Interpre-
tation of NOx/NOy observations from AASE-II
using a model of chemistry along trajectories,
Geophys. Res. Lett., 20, 2507-2510, 1993.
Kent, G.S., and M.P. McCormick, SAGE arid SAM II
measurements of global stratospheric aerosol
optical depth and mass loading, J. Geophys. Res.,
89, 5303-5314, 1984. ;
Kent, G.S., and M.P. McCormick, SAGE and SAM II
measurements of global stratospheric aerosol
optical depth and mass loading, J. Geophys. Res.,
95,5303-5314,1993.
4.32
-------
TROPICAL/MIDLATITUDE PROCESSES
Kerr, J.B., D.I. Wardle, and D.W. Tarrasick, Record low
ozone values over Canada in early 1993, Geophys.
Res. Lett., 20, 1979-1982, 1993. !
Keys, J.G., P.V. Johnston, R.D. Blatherwick, and F.J.
Murcray, Evidence for heterogeneous reactions in
the Antarctic autumn stratosphere, Nature, 361,
49-51, 1993.
King, J.C., et al., Measurements of CIO and O3 from
21°N to 61°N in the lower stratosphere during
February 1988: Implications for heterogeneous
chemistry, Geophys. Res. Lett., 18, 2273-2276,
1991.
Kinne, S., O.B. Toon, and M.J. Prather, Buffering of
stratospheric circulation by changing amounts of
tropical ozone—a Pinatubo case study, Geophys.
Res. Lett., 19, 1927-1930, 1992.
Koike, M, Y. Kondd, W.A. Matthews, P.V. Johnston, and
K. Yamazaki, Decrease of stratospheric NOa at
44°N caused by Pinatubo volcanic aerosols,
Geophys. Res. Lett., 20, 1975-1978, 1993.
Koike, M.Y., N.B. Jones, W.A. Matthews, P.V. Johnston,
R.L. MacKenzie, D. Kinnison, and J. Rodriguez,
Impact of Pinatubo aerosols on the partitioning
between NO2 and HNO3, Geophys. Res. l.ett., 21,
597-600, 1994. '.]
Labitzke, K., and M.P. McCormick, Stratospheric tem-
perature increases due to Pinatubo aerosols,
Geophys. Res. Lett., 19, 207-210, 1992.
Lean, J., Variations in the sun's radiative output, Rev.
Geophys., 29, 505-535, 1991.
Legras, B., and .D. Dritschel, Vortex stripping and the
generation of high vorticity gradients in two-di-
mensional flows, in Advances in Turbulence IV,
Applied Scientific Research Vol. 51, edited by
F.T.M. Nieuwstadt, Kluwer, Dordrecht, The Neth-
erlands, 445-455, 1993.
Leovy, C.B., C.R. Sun, M.H. Hitchmann, E.E. Rems-
berg, J.M. Russell III, L.L. Gordely, J.C. Gille, and
L.V. Lyjak, Transport of ozone in the middle atmo-
sphere: Evidence for planetary wave breaking, J.
Atmos. Sci., 42, 230-244, 1985.
Luo, B.P., S.L. Clegg, T. Peter, R. Mueller, and P.J.
Crutzen, HC1 solubility and liquid diffusion in
aqueous sulfuric acid under stratospheric Condi-
tions, Geophys. Res. Lett., 21, 49-52, 1994.
MacKenzie, A.R., B. Knuds.cn, R.L. Jones, and E.R. Lut-
man, The spatial and. temporal extent of chlorine
activation by polar Stratospheric clouds in the
Northern-Hemisphere winters of 1988/89 and
1991/92, Geophys. Res. Lett., 21, 1423-1426,
1994. !
Manney, G.L., J.D. Farrarajand C.R. Mechoso, Simula-
tions of the February 1979 stratospheric sudden
warming: Model comparisons and three-dimen-
sional evolution, Monthly Weather Rev., in press,
1994. j
McCormick, M.P., and R.E. Veiga, SAGE n measure-
ments of early Pinatubo aerosols, Geophys. Res.
Lett., 19, 155-158, 1992.
McElroy, M.B., and R.J. Sallawitch, Changing composi-
tion of the global stratosphere, Science, 243,
763-770, 1989. j
McElroy, M.B., RJ. Salawitch, and K. Minschwaner,
The changing stratosphere, Planet. Space Sci., 40,
373-401, 1992. j
Mclntyre, M.E., Middle atmospheric dynamics and
transport: Some current challenges to our under-
standing, in Dynamics, Transport and Photo-
chemistry in the Middle Atmosphere of the Southern
Hemisphere, San Francisco NATO Workshop, edit-
ed by A. O'Neill, Kluwer, Dordrecht, The Nether-
lands, 1-18, 1990. j
Mclntyre, M.E., The stratospheric polar vortex and sub-
vortex: Fluid dynamics and midlatitude ozone
loss, to be submitted: to Phil. Trans. Roy. Soc.
London, 1994. !
McPeters, R.D., The atmospheric SO2 budget for Pina-
tubo derived from NOAA-11 SBUV/2 spectral
data, Geophys. Res. Lett., 20, 1971-1974, 1993.
Mergenthaler, J.L., J.B. Kumer, and A.E. Roche,
CLAES south-looking aerosol observations for
1992, Geophys. Res. Lett., 20, 1295-1298, 1993.
Micheiangeli, D.V., M. Alleiji, Y.L. Yung, R.L. Shia, D.
Crisp, and J. Eluszkiewicz, Enhancement of atmo-
spheric radiation by ani aerosol layer, J. Geophys.
Res., 97, 865-874, 1992.
Michelsen, H.A., RJ. Salawitch, P.O. Wennberg, and
J.G. Anderson, Production of O('D) from photo-
lysis of O3, Geophys. ties. Lett., 21, 2227-2230,
1994. I
4.33
-------
TROPICAL/MIDLATITUDE PROCESSES
Middlebrook, A.M., L.T. Iraci, L.S. McNeill, B.C.
Koehler, O.W. Saastad, M.A. Tolbert, and D.R.
Hanson, FTIR studies of thin H2S04/H2O films:
Formation, water uptake and solid-liquid phase
changes, J. Geophys. Res., 98, 20473-20481,
1993.
Mills, MJ., A.O. Langford, T.J. O'Leary, K. Arpag, H.L.
Miller, M.H. Proffitt, R.W. Sanders, and S. So-
lomon, On the relationship between stratospheric
aerosols and nitrogen dioxide, Geophys. Res. Lett.,
20, 1187-1190,1993.
Minschwaner, K., G.P. Anderson, L.A. Hall, and K.
Yoshino, Polynomial coefficients for calculating
O2 Schumann-Runge cross sections at 0.5 cm"1
resolution, /. Geophys. Res., 97, 10103-10108,
1992.
Minschwaner, K., RJ. Salawitch, and M.B. McElroy,
Absorption of solar radiation by O2: Implications
for 03 and lifetimes of N2O, CFC13 and CF2C12, /.
Geophys. Res.. 98, 10543-10561, 1993.
Minton, T.K., C.M. Nelson, T.A. Moore, and M. Oku-
mura. Direct observation of CIO from chlorine
nitrate photolysis, Science, 258,1342-1345, 1992.
Molina, M.J., R. Zhang, P.J. Wooldridge, J.R. Mc-
Mahon, I.E. Kim, H.Y. Chang, and K.D. Beyer,
Physical chemistry of the H2SO4/HNO3/H2O sys-
tem: Implications for polar stratospheric clouds,
Science. 261, 1418-1423, 1993.
Murphy, D.M., A.F. Tuck, K.K. Kelly, K.R. Chan, M.
Loewenstein, J.R. Podolske, M.H. Proffitt, and
S.E. Strahan, Indicators of transport and vertical
motion from correlations between in situ measure-
ments in the Airborne Antarctic Ozone
Experiment, /. Geophys. Res., 94, 11669-11685,
1989.
Murphy, D.M., et al., Reactive nitrogen and its correla-
tion with ozone in the lower stratosphere and
upper troposphere, J. Geophys. Res., 98, 8751-
8773, 1993.
Nardi, B., M.-L. Chanin, A. Hauchecome, S.I. Avdy-
ushin, G.F. Tulinov, B.N. Kuzmenko, and I.R.
Mezhuev, Morphology and dynamics of the
Pinatubo aerosol layer in the northern hemisphere
as detected from a ship-borne lidar, Geophys. Res.
Lett.. 18, 1967-1970, 1993.
Natarajan, M., and L.B. Callis, Stratospheric photo-
chemical studies with Atmospheric Trace
Molecule Spectroscopy (ATMOS) measurements,
/. Geophys. Res., 96,9361-9370, 1991.
Neuber, R., G. Beyerie, G. Fiocco, A. di Sarra, K.H.
Fricke, C. David, S. Godin, B.M. Khusden, L.
Stefanutti, G. Vaughan, and J.-P. Wolf, Latitudinal
distribution of stratospheric aerosols during the
EASOE winter 1991/92, Geophys. Res. Lett., 21,
1283-1286, 1994. '-.
Norton, W.A., Breaking Rossby waves in a model strato-
sphere diagnosed by a vortex-following coordi-
nate system and a technique for advectihg material
contours, /. Atmos. ScL, 51, 654-673, 1994.
O'Sullivan, D., and R.E. Young, Modeling the quasi-
biennial oscillation's effect on the winter strato-
spheric circulation, J. Atmos. ScL, 49,2437-2448,
1992.
Orlando, J.J., G.S. Tyndall, O.K. Moortgat and J.G. Cal-
vert. Quantum yields for NO3 photolysis between
570 and 635 nm, / Phys. Chem., 97, 10996-
11000,1993. I
Pierce, B.R., andT.D.A. Fairlie, Chaotic advection in the
stratosphere: Implications for the dispersal of
chemically perturbed air from the polar vortex, J.
Geophys. Res., 98, 18589-18595, 1993.
Pierce, R.B., W.L. Grose, J.M. Russell III, and A.F.
Tuck, Evolution of southern hemisphere spring air
masses observed by HALOE, Geophy's. Res. Lett.,
27,213-216,1994. i
Pitari, G., A numerical study of the possible perturbation
of stratospheric dynamics due to Pinatubo aero-
sols: Implications for tracer transport, /. Atmos.
ScL, 50, 2443-2461, 1993.
Pitari, G., and V. Rizi, An estimate of the chemical and
radiative perturbation of stratospheric ozone fol-
lowing the eruption of Mt. Pinatubo, /. Atmos.
ScL, 50, 3260-3276, 1993. !
Pitts, M.C., and L.W. Thomason, The impact of the erup-
tions of Mt. Pinatubo and Cerro; Hudson on
Antarctic aerosol levels during the 1991 austral
spring, Geophys. Res. Lett., 20,245U2454, 1993.
Plumb, R. A., Atmospheric dynamics — ozone depletion
in the Arctic, Nature, 347, 20-21, 1990.
Plumb, R.A., and R.C. Bell, A model of the quasi-bien-
nial oscillation on an equatorial beta-plane. Quart.
J. Roy. Meteorol. Soc., 108, 335-352, 1982.
4.34
-------
TROPICAL/MIDLATITUDE PROCESSES
Plumb, R.A., D.W. Waugh, R.J. Atkinson, M.R. Schoe-
berl, L.R. Lait, P.A. Newman, E.V. Browell, A.J.
Simmons, and M. Loewenstein, Intrusions into the
lower stratospheric Arctic vortex during the winter
of 1991/92, J. Geophys. Res., 99, 1089-1106,
1994.
Polvani, L.M., D.W. Waugh, and R.A. Plumb, On the
subtropical edge of the stratospheric surf zone,
submitted to J. Atmos. ScL, 1994. .
Poulet, G., M. Pirre, F. Maguin, R. Ramaroson, and G.
LeBras, Role of the BrO + HO2 reaclion in the
stratospheric chemistry of bromine, Geophys. Res.
Lett., 19, 2305-2308, 1992.
Prather, M., Catastrophic loss of stratospheric ozone in
dense volcanic clouds, /. Geophys. Res., 97,
10187-10191,1992.
Price, J.M., J.A. Mack, C.A. Rogaski, and A.M. Wodtke,
Vibrational-state-specific self-relaxation rate con-
stant—Measurements of highly vibrationally
excited O2 (v = 19-28), J. Chem. Phys., 175, 83-
98, 1993.
Proffitt, M.H..K.K. Kelly, J.A. Powell, B.L. Xjary, M.
Loewenstein, J.R. Podolske, S.E. Strahan, and
K.R. Chan, Evidence for diabatic cooling and
poleward transport within and around the 1987
Antarctic ozone hole, J. Geophys. Res., 94, 16797-
16813, 1989. ';
Proffitt, M.H., J.J. Margitan, K.K. Kelly, M.!Loewen-
stein, J.R. Podolske, and K.R. Chan, Ozone loss in
the arctic polar vortex inferred from high-altitude
aircraft measurements, Nature, 347, 31-36, 1990.
Proffitt, M.H., K. Aikin, J.J. Margitan, M. Loewenstein,
J.R. Podolske, S.E. Strahan, andA. Weaver, Ozone
loss inside the northern polar vortex during the
1991-1992 winter; Science, 261,1150-1154,1993.
Pyle, J.A., G.D. Carver, and U. Schmidt, Some case
studies of chlorine activation during the EASOE
campaign, Geophys.j Res. Lett., 21, 1431-1434,
1994. ;
Randel, W.J., J.C. Gille, A.E. Roche, J.B. Kiimer, J.L.
Mergenthaler, J.W. Waters, E.F. Fishbein, and
W.A. Lahoz, Stratospheric transport from the
tropics to middle latitudes by planetary-wave
mixing, Nature, 365, 533-535, 1993.
Rattigan, O., E.R. Lutman, R.L. Jones, and R.A. Cox,
Temperature dependent absorption cross-sections
and atmospheric photolysis rates of nitric acid,
Ber. Bunsenges. Phys. Chem., 96, 399-404, 1992.
Read, W.G., L. FroidevauX, and J.W. Waters, Microwave
Limb Sounder measurement of stratospheric SO2
from the Mt. Pinatubo volcano, Geophys. Res.
Lett., 20, 1299-1302, 1993.
Reid, S.J., G. Vaughan, and E. Kyro, Occurrence of
ozone laminae near the boundary of the strato-
spheric polar vortex, /. Geophys. Res., 98,
8883-8890, 1993. |
Roche, A.E., J.B. Kumer, and J.L. Mergenthaler,
CLAES observations of ClONOa and HNO3 in the
Antarctic stratosphere, between June 15 and
September 17,- 1992, Geophys. Res. Lett., 20,
1223-1226, 1993. j
Rodriguez, J., M.K.W. Ko, N.-D. Sze, and C.W. Heisey,
Ozone response to enhanced heterogeneous pro-
cessing after the eruption of Mt. Pinatubo,
Geophys. Res. Lett., 21, 209-212, 1994.
Rood, R., A. Douglass, and C. Weaver, Tracer exchange
between tropics and middle latitudes, Geophys.
Res. Lett., 19, 805-808, 1992.
Rosen, L.M., N.T. Kjome, and H. Fast, Penetration of
Mt. Pinatubo aerosols into the north polar vortex,
Geophys. Res. Lett., 19, 1751-1754, 1992.
Russell, J.M., HI, A.F. Tuck, L.L. Gordley, J.H. Park,
S.R. Drayson, J.E. Harries, R.J. Cicerone, and P.J.
Crutzen, HALOE >\ntarctic observations in the
spring of 1991, Geophys. Res. Lett., 20, 719-722,
1993a. |
Russell, J.M., HI, L.L. Gordley, J.H. Park, S.R. Drayson,
• W.D. Hesketh, R:J. Cicerone, A.F. Tuck, J.E. Fred-
erick, J.E. Harries, and P.J. Crutzen, The Halogen
Occultation Experiment, /. Geophys. Res., 98,
10777-10797, 1993b.
Salawitch, R.J., et al., The diurnal variation of hydrogen,
nitrogen, and chlorine radicals: Implications for
the heterogeneous production of HNO2, Geophys.
Res. Lett., 21, 2551-2554, 1994a.
Salawitch, R.J. et al.. The distribution of hydrogen, ni-
trogen, and chlorine radicals in the lower
stratosphere: Implications for changes in Oj due
to emission of NOy from supersonic .aircraft,
Geophys. Res. Lett., 21, 2547-2550, 1994b.
4.35
-------
TROPICAL/MIDLATITUDE PROCESSES
Schiffman, A., D.D. Nelson, and DJ. Nesbitt, Quantum
yields for OH production from 193 and 248 nm
photolysis of HNO3, /. Chem. Phys., 98, 6935-
6946,1993.
Schoeberl, M.R., 'L.R. Lait, P.A. Newman, and J.E.
Rosenfield, The structure of the polar vortex, ,/.
Geophys. Res., 97,7859-7882, 1992.
Schoeberl, M.R., P.K. Barthia, E. Hilsenrath, and O.
Torres, Tropical ozone loss following the eruption
of Mt. Pinatubo, Geophys. Res. Lett., 20, 29-32,
1993a.
Schoeberl, M.R., S.D. Doiron, L.R. Lait, P.A. Newman,
and A.J. Krueger, A simulation of the Cerro Hud-
son S02 cloud, J. Geophys. Res., 98, 2949-2955,
1993b.
Sheridan, P.J., R.C. Schnell, DJ. Hofmann, and T.
Deshler, Electron microscope studies of Mt.
Pinatubo aerosol layers over Laramie, Wyoming,
during summer 1991, Geophys. Res. Lett., j!9,
203-206, 1992.
Solomon, S., R.W. Sanders, R.R. Garcia, and J.G. Keys,
Increased chlorine dioxide over Antarctica caused
by volcanic aerosols from Mt. Pinatubo, Nature,
363,245-248,1993.
Solomon, S., R.W. Sanders, R.O. Jakoubek, K.H. Arpag,
S.L. Stephens, J.G. Keys, and R.R. Garcia, Visible
and near-ultraviolet spectroscopy at McMurdo
Station, Antarctica 10. Reductions of stratospheric
NO2 due to Pinatubo aerosols, /. Geophys. Res.,
99,3509-3516,1994a.
Solomon, S., R.R. Garcia, and A.R. Ravishankara, On
the role of iodine in ozone depletion, /. Geophys.
Res., 99, 20491-20499, 1994b.
Sommeria, J., S.D. Meyers, and H.L. Swinney, Labora-
tory model of a planetary eastward jet, Nature,
337,58-61, 1989.
Stachnik, R.A., J.C. Hardy, J.A. Tarsala, J.W. Waters,
and N.R. Erickson, Submillimeterwave hetero-
dyne measurements of stratospheric CIO, HC1, Oa
and.HO2: First results, Geophys. Res. Lett., 19,
1931-1934, 1992.
Stimpfle, R.M., et al.. The response of CIO radical
concentrations to variations in NO2 radical con-
centrations in the lower stratosphere, Geophys.
Res. Lett., 21, 2543-2546, 1994.
Stolarski, R.S., P. Bloomfield, R.D. McPeters,;and J.R.
Herman, Total ozone trends deduced from Nimbus
7TOMS data, Geophys. Res. Lett., 18,1015-1018,
1991.
Stowe, L., R.M. Carey, and P.P. Pelagrino, Monitoring
the Mt. Pinatubo aerosol layer with NOAA/11
AVHRR data, Geophys. Res. Lett., 19, i 159-162,
1992. ',
Strahan, S.E., and J.D. Mahlman, Evaluation of the
SKIHI general circulation model using aircraft
N2O measurements. 2. Tracer variability and dia-
batic meridional circulation, J. Geophys- Res., 99,
10319-10332,1994.
Tabazadeh, A., R.P. Turco, and M.Z. Jacobson;, A model
for studying the composition and chemical effects
of stratospheric aerosols, J. Geophys.]Res., 99,
12897-12914, 1993. '
Tao, X., and A.F. Tuck, On the distribution of cold air
near the vortex edge in the lower stratosphere, J.
Geophys. Res., 99, 3431-3450, 1994. |
Thomason, L.W., and L.R. Poole, NASA' Langley
Research Center, private communication, 1994.
Tie, X.X., G.P. Brasseur, B. Briegleb, and G. Granier,
Two-dimensional -simulation of Pinatubo aerosol
and its effect on stratospheric ozone, /: Geophys.
Res., 99, 20545-20562, 1994.
Tolbert, M.A., Heterogeneous chemistry in ;the strato-
sphere: A review of laboratory measurements, in
The Atmospheric Effects of Stratospheric Aircraft:
A Second Program Report, NASA Reference Pub-
lication 1293,1993. ;
Tolbert, M.A., J. Pfaff, I. Jayaweera, and M.J. Prather,
Uptake of formaldehyde by sulfuric acid solu-
tions: Impact on stratospheric ozone, /. Geophys.
Res., 98, 2957-2962, 1993. i
Toon, G.C., J.-F. Blavier, J.N. Solario, andiJ.T. Szeto,
Airborne observations of the composition of the
1992 tropical stratosphere by FTIR solar absorp-
tion spectrometry, Geophys. Res. Lett,, 20, 2503-
2506, 1993. ',
Toon, O., E. Browell, B. Gary, L. Lait, J. Livingston, P.
Newman, R. Pueschel, P. Russell, Mi Schoeberl,
G. Toon, W. Traub, F. Valero, H. Selkirk, and J.
Jordan, Heterogeneous reaction probabilities, sol-
ubilities and physical state of cold volcanic
aerosols, Science, 261, 1136-1140, 1993.
4.36
-------
TROPICAL/MIDLATITUDE PROCESSES
Toumi, R., An evaluation of autocatalytic ozone produc-
tion from vibrationally excited oxygen in the
middle atmosphere, J. Atmos. Chetn., 15, 69-77,
1992.
Toumi, R., R.J. Kerridge, and J.A. Pyle, Highly vibra-
tionally excited oxygen as a potential \ source of
ozone in the upper stratosphere and mesosphere,
Nature, 351, 217-219, 1991.
Toumi, R., and S. Bekki, The importance of the reactions
between OH and CIO for stratospheric ozone,
Geophys. Res. Lett., 20, 2447-2450, 1993a.
Toumi, R., S. Bekki, and R. Cox, A model study of
ATMOS observations of the heterogeneous loss of
N2Os on the sulphate aerosol layer, J. Atmos.
Chem., 16, 135-144, 1993b.
Toumi, R., R.L. Jones, and J.A. Pyle, Stratospheric
ozone depletion by C1ONO2 photolysis, Nature,
365, 37-39, 1993c.
Toumi, R., and S. Bekki, Sensitivity of stratospheric
composition to oxygen absorption of solar radia-
tion (175-210 nm), J. Atmos. Chem., M, 57-73,
1994. ;
Traub, W.A., D.G. Johnson, K.W. Jucks, and K.V.
Chance, Upper limit for stratospheric HBr using
far-infrared thermal emission spectroscopy,
Geophys. Res. Lett., 19, 1651-1654, 1992.
Trepte, C.R., and M.H. Hitchmann, The stratospheric
tropical circulation deduced from satellite aerosol
data, Nature, 355, 626-628, 1992. |
Trepte, C.R., R.E. Veiga, and M.P. McCorrnick, The
poleward dispersal of Mt. Pinatubo volcanic aero-
sol,/. Geophys. Res., 98, 18562-18573, 1993.
Tuck, A.F., Synoptic and chemical evolution of the
Antarctic vortex in late winter and early spring,
1987,/. Geophys. Res., 94, 11687-11737, 1989.
Tuck, A.F., et al., Polar stratospheric cloud-processed air
and potential vorticity in the northern hemisphere
lower stratosphere at mid-latitudes during winter,
J. Geophys. Res., 97, 7883-7904, 1992.
Tuck, A.F., J.M. Russell III, and J.E. Harries, Strato-
spheric dryness: Antiphased desiccation over Micro-
nesia and Antarctica, Geophys. Res. Lett., 20,
1227-1230, 1993.
Tuck, A.F., D.W. Fancy, M. Loewenstein, J.R. Podolske,
K.K. Kelly, S.J. Hovde, D.M. Murphy, and J.W.
Elkins, Spread of clenitrification from 1987 Ant-
arctic and 1988-1989 Arctic stratospheric vortices,
/. Geophys. Res., 99, 20573-20583, 1994.
Turnipseed, A.A., G.L. Vaghjiani, J.E. Thompson, and
A.R. Ravishankara,;Photodissociation of HNOa at
193, 222 and 248 nm — Products and quantum
yields, / Chem. Phys., 96, 5887-5895, 1992.
Valero, F.P.J., and P. Pilewskie, Latitudinal survey of
spectral optical depths of the Pinatubo volcanic
cloud, derived particle sizes, columnar mass load-
ings, and effects on planetary albedo, Geophys.
Res. Lett., 19, 163-166, 1992.
Waters, J.W., L. Frpidevaux, W.G. Read, G.L. Manney,
L.S. Elson, D.A. Flower, R.F. Jarnot, and R.S.
Harwood, Stratospheric CIO and ozone from the
Microwave Limb Sounder on the Upper Atmo-
spheric Research Satellite, Nature, 362, 597-602,
1993. !
Waugh, D.W., Subtropical'stratospheric mixing linked to
disturbances of the;polar vortices, Nature, 365,
535-537, 1993a.
Waugh, D.W., Contour surgery simulation of a forced
polar vortex, /. Atmos. Sci., 50, 714-730, 1993b.
Waugh, D.W, R.A. Plumci, R.J. Atkinson, M.R. Schoe-
berl, L.R. Lait, P.A. Newman, M. Loewenstein,
D.W. Toohey, L.M. Avallone, C.R. Webster, and
R.D. May, Transport of material out of the strato-
spheric Arctic vortex! by Rossby wave breaking, /.
Geophys. Res., 99, 1071-1088, 1994.
Weaver, A., M. Loewenstein, J.R. Podolske, S.E. Stra-
han, M.H. Proffitt, 1C. Aikin, J.J. Margitan, H.H.
Jonsson, C.A. Brock; J.C. Wilson, and O.B. Toon,
Effects of Pinatubo a'erosol on stratospheric ozone
at mid-latitudes, Geophys. Res. Lett., 20, 2515-
2518,1993. |
Webster, C.R., R.D. May, D.W Toohey, L.M. Avallone,
J.G. Anderson, and S. Solomon, In situ measure-
ments of the C10/HC1 ratio: Heterogeneous
processing on sulfate aerosols and polar stratospher-
ic clouds, Geophys.'Res. Lett., 20, 2523-2526,
1993. j
4.37
-------
TROPICAL/MIDLATITUDE PROCESSES
Webster, C.R., R.D. May, M. Allen, L. Jaegle, and M.P.
McCormick, Balloon profiles of stratospheric
NC>2 and HNOs for testing the heterogeneous
hydrolysis of N2Os on sulfate aerosols, Geophys.
Res. Lett., 21,53-56, 1994a.
Webster, C.R., el al, Hydrochloric acid and the chlorine
budget of the lower stratosphere, Geophys. Res.
Lett., 21, 2575-2578, 1994b.
Wennberg, P.O., et al, The removal of stratospheric O3
by radicals: In situ measurements of OH, HO2,
NO, NO2, CIO, and BrO, Science, 266, 398-404,
1994.
Williams, L.R., and D.M. Golden, Solubility of HC1
in sulfuric acid at stratospheric temperatures,
Geophys. Res. Lett., 20, 2227-2230, 1993.
Wilson, J.C., H. Jonsson, C.A. Brock, D.W. Toohey,
L.M. Avallone, D. Baumgardner, J.E. Dye, L.R.
Poole, D.C. Woods, R.J. DeCoursey, M. Osborn,
M.C. Pitts, K.K. Kelly, K.R. Chan, G.V. Ferry, M.
Loewenstein, J.R. Podolske, and A. Weaver, In situ
observations of aerosol and chlorine monoxide
after the 1991 eruption of Mt. Pinatubo—Effect of
reactions on sulfate aerosol, Science, 261, 1140-
1143,1993.
Wuebbles, D.J., D.E. Kinnison, K.E. Grant, and J. Lean,
The effect of solar flux variations and (race gas
emissions on recent trends in stratospheric ozone
and temperature, J. Geotnagn. Geoelectr.,43,709-
718, 1991. ;
Young, R.E, H. Houben, and O.B. Toon, Simulation of
the dispersal of the Mt. Pinatubo volcanic aerosol
cloud in the stratosphere immediately following
the eruption, Geophys. Res. Lett., to appear, 1993.
Zhang, R., P.J. Wooldridge, and M.J. Molina, Vapor
pressure measurements for the H2S04/HNO3/
H2O and H2SO4/HC1/H2O systems: Incorporation
of stratospheric acids into background sulfate
aerosols, J. Phys. Chem., 97, 8541-8548J 1993.
Zhang, R., J.T. Jayne, and M.J. Molina, Heterogeneous
interactions of C1ONO2 and HC1 with sulfuric acid
tetrahydrate: Implications for the stratosphere, J.
Phys. Chem., in press, 1994.
'- 4.38
-------
CHAPTERS
i
Tropospheric Ozone
Lead Authors:
A. Volz-Thomas
B.A. Ridley
Co-authors:
M.O. Andreae
W.L. Chameides
R.G. Derwent
I.E. Galbally
J. Lelieveld
S.A. Penkett
M.O. Rodgers
M. Trainer
G. Vaughan
X.J. Zhou
Contributors:
E. Atlas
C.A.M. Brenninkmeijer
D.H. Ehhalt
J. Fishman
F. Flocke
D.Jacob
J.M. Prospero
F. Rohrer
R. Schmitt
H.G.J. Smit
A.M. Thompson
-------
-------
CHAPTER 5
TROPOSPHERfC OZONE
Contents
SCIENTIFIC SUMMARY ;,
5.1 INTRODUCTION J
5.2 REVIEW OF FACTORS THAT INFLUENCE TROPOSPHERIC OZONE CONC ENTRATIONS
5.2.1 Stratosphere-Troposphere Exchange.
5.2.2 The Photochemical Balance of Ozone in the Troposphere
..5.1
.5.3
.5.3
.5.4
.5.5
5.3 INSIGHTS FROM FIELD OBSERVATIONS: PHOTOCHEMISTRY AND TRANSPORT <.
5.3.1 Urban and Near-Urban Regions..'. '
5.3.2 Biomass Burning Regions '1 " f
5.3.3 Remote Atmosphere and Free Troposphere " ^"12
5.4 FEEDBACK BETWEEN TROPOSPHERIC OZONE AND LONG-LIVED GREENHOUSE GASES.
REFERENCES I
..5.20
.5.21
-------
-------
TROPOSPHERIC OZONE PROCESSES
SCIENTIFIC SUMMARY
Although representing only 10 percent of the total ozone column, tropospheric ozone, is important because it can
influence climate, as it is a greenhouse gas itself,' and because its photolysis by UV radiation in the presence of water
vapor is the primary source for hydroxyl radicals (OH). Hydroxyl radicals are responsible for the oxidative removal of
many trace gases, such as methane (CFLt), hydrofiuorocarbons (HFCs), and hydrochlorofl'uorocarbons (HCFCs), that
influence climate and/or are important for the stratospheric ozone layer. !
Tropospheric ozone arises from two processes: downward flux from the stratosphere,; and in situ photochemical
production from the oxidation of hydrocarbons and carbon monoxide (CO) in the presence of NOX (NO + NO2). Ozone
is removed from the troposphere by in situ chemistry and by uptake at the Earth's surface. The role of photochemistry
in the local ozone balance depends strongly on Ihe concentration of NOX. Human impact Jon tropospheric ozone and
hydroxyl occurs through the emission of precursors, e.g., NOX, CO, and hydrocarbons. In the case of free tropospheric
ozone, this is brought about by the export of both ozone and its precursors, in particular NOX, from source regions.
While substantial uncertainties remain in assessing the global budget of tropospheric ozone, recent studies have
led to significant advances in understanding the local balance of ozone in some regions of lihe atmosphere.
• Recent measurements of the NOy/O3 ratio have basically confirmed earlier estimates of the flux of ozone from the
stratosphere to be in the range of 240-820 Tg(O3)/yr, which is in reasonable agreement with results from general
circulation models (GCMs). j
• The observed correlation between ozone arid alkyl nitrates suggests a natural ozone concentration of 20-30 ppb in
the upper planetary boundary layer (at about 1 km altitude), which agrees well with the estimate from the few
reliable historic data (cf. Figure 1-16, Chapter 1). ;
i
• Measurements of the gross ozone production rate yielded values as high as several tens of ppb per hour in the
polluted troposphere over populated regions, in good agreement with theoretical predictions. Likewise, the effi-
ciency of NOX in ozone formation in moderately polluted air masses was found to be in. reasonable agreement
with theory. . [
• Direct measurements of hydroxyl and peroxy radicals have become available. While they do not serve to estab-
lish a global climatology of OH, they do provide a test of our understanding of the fast photochemistry. To date,
theoretical predictions of OH concentrations (from measured trace gas concentrations and photolysis rates) tend
to be higher than the measurements by up to a factor of two. ,
• Measurements of peroxy radical concentrations in the remote free troposphere are treasonable agreement with
theory; however, significant misunderstanding exists with regard to the partitioning of odd nitrogen and the bud-
get of formaldehyde. ; i
' I
• Measurements have shown that export of ozone produced from anthropogenic precursors over North America is
a significant source for the North Atlantic region during summer. It has also been shown that biomass burning is
a significant source for ozone in the tropics during the dry season. These findings sliow the influence of human
activities on the global ozone balance. ,: !
5.1
-------
TROPOSPHERIC OZONE PROCESSES |
|
. Photochemical net ozone destruction in the remote atmosphere has been identified in several experiments. It is
likely to occur over large parts of the troposphere with rates of up to several ppb per day. Consequently, an
increase in UV-B radiation (e.g., from stratospheric ozone loss) is expected to decrease tropospheric ozone in the
remote atmosphere but in some cases will increase production of ozone in and transport from the more polluted
regions. The integrated effect on hydroxyl concentrations and climate is uncertain.
Uncertainties in the global tropospheric ozone budget, particularly in the free troposphere, are mainly associated
with uncertainties in the global distribution of ozone itself and its photochemical precursors, especially CO and NOX.
These distributions are strongly affected by dynamics, by the magnitude and spatiaVtemporal distribution of sources,
particularly those for NOX to the middle and upper troposphere from the stratosphere, lightning, aircraft, and cbnvective
systems, and by the partitioning and removal of NOy constituents. The role of heterogeneous processes Deluding
multiphase chemistry in the troposphere is not well characterized, and the catalytic efficiency of NOX in catalyzing
ozone formation in the free troposphere has not been confirmed by measurements.
5.2
-------
5.1 INTRODUCTION
As is outlined in more detail in Chapter 1, there is
some, albeit limited, evidence to suggest that ozone
concentrations in the troposphere of the Northern Hemi-
sphere have increased by a factor of two or more over the
past 100 years, with most of the increase having oc-
curred since 1950. This conclusion is consistent with
ozone data gathered continuously since the 1970s at a
series of remote and in some cases high altitude stations.
It is interesting to note that all stations north of about
20°N exhibit a positive trend in ozone over the past two
decades that is significant to the 95% confidence level.
During the same time, a statistically significant negative
trend of about 0.5%/yr is observed at the South Pole.
For the most part, the trends appear to fall more or
less along a straight line that extends from -0.5%/yr at
90°S to +0.8%/yr at 70°N. Somewhat anomalous are the
large positive trends observed at the high elevation sites
in Southern Germany (1-2%/yr); these large trends per-
haps reflect a regional influence above and beyond the
smaller global trend (Volz-Thomas, 1993). It should be
noted, however, that the average positive trends observed
in the Northern Hemisphere are largely due to the rela-
tively rapid ozone increase that occurred in the seventies.
Over the last decade, no or little ozone increase has oc-
curred in the free troposphere except over Southern
Germany and Switzerland. Indeed, ozone concentra-
tions at some locations in the polluted planetary
boundary layer (PBL) over Europe have decreased over
the last decade (Guicherit, 1988; Low et al, 1992).
It is important that we understand the causes of the
apparent increase in tropospheric ozone concentrations
in the Northern Hemisphere because of ozone's central
role in global biogeochemistry, its effectiveness as a
greenhouse gas, especially in the upper troposphere, and
its toxicity to living organisms. It is equally important,
however, to understand the causes for the decrease ob-
served at high latitudes in the Southern Hemisphere
because of the influence of ozone on the concentration of
hydroxyl radicals and, hence, the oxidizing capacity of
the atmosphere, which controls the budgets of many
long-lived greenhouse gases. This, in turn, requires a
quantitative understanding of the chemical and meteoro-
logical processes that determine the budget of
tropospheric ozone. :
5.2 REVIEW OF FACTO RS THAT
TROPOSPHERIC 02 ONE
TROPOSPHEF 1C OZONE PROCESSES
INFLUENCE
CONCENTRATIONS
The presence of ozone in the troposphere is under-
stood to arise from two basic processes: (1) tropospheric/
stratospheric exchange that causes the transport of strato-
spheric air, rich in ozone, into the troposphere, and (2)
production of ozone from photochemical reactions oc-
curring within the troposphere. Similarly, removal of
tropospheric ozone is accomplished through two com-
peting processes: (1) transport to and removal at the
Earth's surface, and (2) in situ chemical destruction. For
the past two decades, research on tropospheric ozone has
largely focused on understanding the relative roles of
these processes in controlling the abundance and distri-
bution of ozone in the troposphere. The basic chemical
mechanisms that control the local ozone budget are now
reasonably well understood., except for the role of heter-
ogeneous processes. The: situation is not as good
concerning our quantitative understanding of the natural
sources of ozone and its precursors. Transport of ozone
and NOy (see Figure 5-1) from the stratosphere and pro-
duction of NOX (= NO + N0'2) through lightning must be
known to a better degree in order to assess the role of
anthropogenic influences, siich as air traffic (see Chapter
11). Likewise, a better understanding is needed of the
atmospheric transport processes that redistribute ozone
and its precursors between the polluted continental re-
gions and the remote atmosphere and between the
planetary boundary layer and the free troposphere.
Boundary layer processes, including large-scale
eddy mixing and smaller scale turbulence, control the
rate at which sources of NO,; and hydrocarbon emissions
can combine to begin ozone production chemistry. Be-
cause of the nonlinear dependence of photochemical
ozone formation on the precursor concentrations, mod-
els that assume instantaneous mixing over large spatial
grids may significantly overestimate ozone production
rates. Vertical transport of ozone and its precursors be-
tween the boundary layer and higher altitudes (together
with exchange of air with the stratosphere) has a strong
influence on ozone distributions in the troposphere due
to the longer lifetimes of ozone and precursors in the free
troposphere. On the other hand, the upward flow in con-
vective systems must bei balanced by downward
mesoscale flow, which then carries ozone and odd nitro-
gen species from the free troposphere into the planetary
5.3
-------
TROPOSPHERIC OZONE PROCESSES
boundary layer, where they are destroyed more rapidly
(see Lelieveld and Crutzen, 1994). Observations of 03
in convective systems suggest that both mechanisms are
in effect (Dickerson era/., 1987; Pickering et al, 1992b),
but their relative magnitude has not been evaluated ex-
perimentally. Lastly, long-range horizontal advection
influences ozone distributions by transport of both ozone
and its precursors from source areas into other regions,
including the marine environment. This type of long-
range transport has been shown to be an important factor
in the generation of large regional-scale events of elevat-
ed ozone (see, for example, Fishman et al., 1985;
Vukovich et al, 1985; Logan, 1989; Sillman et al.,
1990).
5.2.1 Stratosphere-Troposphere Exchange
Following the elucidation by Haynes et al. (1991)
of the control exercised on the diabatic circulation in the
stratosphere by waves propagating up from the tropo-
sphere (the so-called Downward Control Principle), a
clearer picture of stratosphere-troposphere exchange
processes has emerged. Trace species such as ozone and
NOy with sources in the middle stratosphere are fed into
the lower stratosphere by the diabatic circulation at a rate
determined by the dissipation of planetary and gravity
wave fields in the stratosphere and mesosphere. Tide
lower stratosphere (especially in midlatitudes) is subject
to efficient isentropic mixing, which maintains a close
correlation between trace species (Plumb and Ko, 1992).
The lower levels of the stratosphere also exchange air
with the troposphere.
Estimates of fluxes across the tropopause remain
uncertain. For example, the net downward flux of air
estimated by Holton (1990) and Rosenlof and Holton
(1993), which was based on the concepts described
above, is a factor of 2-3 larger than the lower limit of the
upward flux derived by Follows (1992) from the growth
of CFC-11 in the troposphere. Deriving an analogous
estimate for the trace species is even more difficult be-
cause their distributions must be accurately known.
However, the very close correlation between nitrous ox-
ide (N2O), NOy, and ozone in the lower stratosphere
(Fancy et al., 1990; Murphy et al., 1993) offers the pos-
sibility of deriving the flux of trace gases from the N;>O
budget. Murphy and Fahey (1994) used an annual de-
struction rate of N2O in the stratosphere of 8-17 Tg(N)/
yr to infer a transport of 0.28-0.6 Tg(N)/yr of;NOy and
240-820 Tg/yr of ozone into the troposphere. This corre-
sponds to a flux of (2-6)xlO'° molecules cm"2 s"1 of
ozone, which is slightly less than the earlier ;estimates
made from observations of tropopause folding events
(Danielsen and Mohnen, 1977) and is comparable to the
fluxes derived from general circulation models (e.g.,
Gidel and Shapiro, 1980; Levy et al., 1985).
The most active regions of stratosphere-tropo-
sphere exchange are in cyclonic regions of the upper
troposphere, near jet streams, troughs, and cutoff lows.
The contribution of tropopause folds to the exchange has
been well documented (e.g., WMO, 1986) and has been
confirmed by recent work (Ancellet et al., 1991, 1994;
Wakamatsuefa/., 1989). Potential vorticity (PV) analy-
ses on isentropic surfaces near the tropopause show long
streamers of elevated PV curving anticyclonically from
high latitudes, corresponding to narrow streaks and a
low tropopause. These streaks are clearly revealed by
Meteosat water vapor images (Appenzeller and Davies,
1992), but their contribution to stratosphere-troposphere
exchange has yet to be assessed. The contribution of cut-
off lows, formed by cyclonically-curving PV streamers
(Thorncroftera/., 1993), is better understood. These are
preferentially found in particular regions of the world,
e.g., Europe (Price and Vaughan, 1992), and can promote
exchange by vigorous convective mixing as wejl as shear
instabilities near jet streams (Price and Vaughan, 1993;
Lamarque and Hess, 1994). Recently, the contribution
of mesoscale convective systems to stratosphere-tropo-
sphere exchange was shown to be of j potential
importance (Poulida, 1993; Alaartef al, 1994).
There have been no studies of trends in strato-
sphere-troposphere exchange, so the contribution of the
stratospheric source to the observed trend in tropospheric
ozone remains an open question. A better understanding is
also required of transport between the lower stratosphere
and the troposphere, and of links between theiopposing
ozone trends in these two regions of the atmosphere.
Decreasing ozone concentrations in the lower strato-
sphere would, at first approximation, imply a decreasing
flux into the troposphere. However, this effect could be
offset by changes in the meridional circulation in the
stratosphere. Following the Downward Control Princi-
ple and assuming the primary source of ozone in the
stratosphere to have remained constant, changes in
downward flux would have to be forced by changes in
5.4 .
-------
gravity wave dissipation. Therefore, changes in climate
could well have led to changes in the ozone flux from the
stratosphere. As noted by WMO (1992), however, there
have not been enough studies of trends in stratospheric
temperatures and transport to deduce trends in the ozone
flux into the troposphere.
5.2.2 The Photochemical Balance of Ozone in
the Troposphere
The production of ozone in the troposphere is ac-
complished through a complex series of reactions
referred to as the "photochemical smog mechanism."
The basics of this mechanism were originally identified
by Haagen-Smit (1952) as being responsible for the rise
of air pollution in Los Angeles in the 1950s. As is out-
lined in Figure 5-1, this well-known mechanism (see
NRC, 1991) involves the photo-oxidation of volatile or-
ganic compounds (VOC) and carbon monoxide (CO) in
the presence of NOX (= NO + NO2). Typical of this
mechanism are reactions (R5-1) through (R5-7):
(R5-1)
(R5-2)
(R5-3a)
(R5-4)
(R5-5)
2 x (R5-6)
2 x (R5-7)
RH + OH
R + O2 + M
RO2 + NO
R0 + 02
HO2 + NO
NO2 + hv
O + O2 + M
-4 R + H
-» RO2n
-> RO +
[2Q ;|
hM ;
N02
-» HO2 + R'CHO:
-> OH +
-^ NO +
-> O3 + l
No2 ;
o ;
Net: RH + 4O2 + 2hv -> R'CHO + H2O + 2O3
where an initial reaction between a hydrocarbon (RH)
and a hydroxyi radical (OH) results in the production of
two O3 molecules and an aldehyde R'CHO or a ketone.
Additional ozone molecules can then be produced from
the degradation of R'CHO. In addition to the oxidation
of hydrocarbons, ozone can be generated from CO oxi-
dation via (R5-8) and (R5-9) followed by j(R5-5),
(R5-6), and (R5-7). J >'. '
(R5-8)
(R5-9)
CO + OH -»
+ O2 + M ->
CO2
HO2
H
Hydrocarbons and CO provide the fuel for the pro-
duction of tropospheric ozone and are consumed in the
process. In remote areas of the troposphere, CO and
methane typically provide the fuel for ozone production
(Seiler and Fishman, 1981). In urban locations, reactive
TROPOSPHERIC OZONE PROCESSES
olefinic and aromatic hydrocarbons (often but not exclu-
sively of anthropogenic origin) are usually the dominant
fuel, while in more rural environments reactive biogenic
VOC such as isoprene often dominate (Trainer et al,
1987; Chameides et al., 19'88).
In contrast to hydrocarbons and CO, NOX is con-
served in the process of O2:one production and thus acts
as a catalyst in ozone formation. The conversion of NO
to NO2 by peroxy radicals (HO2 and RO2) is the crucial
step, since the rapid photolysis of NO2 yields the oxygen
atom required to produce ozone (R5-7). Indeed, the in
situ rate of formation of ozone is given by
P(03) = [NO] • {k5 • [H02j + Zk3ai • [R02]j}
\
\
As catalysis continues until NOX is permanently
removed by physical processes (deposition) or is trans-
formed to other NOy compounds that act as temporary or
almost permanent reservoirs, the catalytic production ef-
ficiency of NOX can, at first approximation, be defined as
the ratio of the rate at which NO molecules are converted
to NO2 by reaction with peroxy radicals to the rate of
transformation or removal of NOX. The lifetime of NOX
varies from a few hours in the boundary layer to at least
several days in the upper troposphere. Thus the catalytic
production efficiency of NOX can vary considerably and
nonlinearly over the more than three orders of magnitude
range of concentrations (see Figure 5-9) typically found
between remote and polluted regions of the troposphere
(Liu etal., 1987; Lin etal., !1988; Hov, 1989).
As is seen in Figure 5-1, the conversion of NO to
NO2 occurs to a large extent through reaction with O3
itself: !
(R5-10)
NO + O3|
NO2 + O2
This process constitutes only a temporary loss, be-
cause O3 (and NO) are regenerated in the photolysis of
NO2 (R5-6) followed by R5-7. The cycle adjusts the
photostationary state between O3, NO, and NO2, and
therefore, the NO/NO2 ratio;(Leighton, 1961). Through
this, reaction R5-10 influences the catalytic efficiency of
NOX in ozone formation since it decreases the fraction of
NOX that is responsible for O3 production via R5-3a and
R5-5 and, at the same time, increases the fraction that is
responsible for the loss of Npx.
Because of the rapid interconversion between NO
and NO2 during daylight, the quantity NOX = NO + NO2
was defined. Similarly, it was found to be useful to
5.5
-------
TROPOSPHERIC OZONE PROCESSES
RONO, HNOJ NA PANs
; Deposition
Wash Out
Deposition
Figure 5-1 a. Schematic view of the cycles of NOX and NOy and their relation to the chemical ozone balance.
The quantity NOZ is defined as NOy - NOX and represents the sum of all oxidation products of NOX.
Figure 5-1 b. Primary formation of OH from O3 photolysis and the HOX cycle in the absence of NOX. It leads
to formation of hydrogen peroxide and net destruction of ozone.
define the quantity Ox = O3 + NO2, in order' to account
for temporary losses of O3 in highly polluted environ-
ments (Guicherit, 1988; Kley et ai, 1994). It is a better
measure of the time-integrated ozone production than
ozone itself (Volz-Thomas et ai, 1993a).
Photochemical loss of tropospheric ozone is ac-
complished through photolysis followed by reaction of
the O('D) atom with water vapor, (R5-11) and (R5-12).
Additional losses occur through reaction of the HO2 rad-
ical formed in (R5-9) with O3 via (R5-13) and (to a
lesser extent) through reaction of OH with O3 (R5- 14):
(R5-11)
(R5-12)
(R5-13)
(R5-14)
O3 + hv -
O('D) + H2O -
HO2 + O3 -
OH + O3 -
•> 0('D) + O2
4 2 OH
* OH + 2O2
4 HO2 + O2
The photochemical rate of ozone loss is approxi-
mated by
L(03) = [03] • {J|T FO'D + kn ' [H°2l + kH * [OH]}
where FoiD is the fraction of excited oxygerj atoms that
react with water vapor. This expression is only approxi-
mate and is more appropriate to the remote free
troposphere, since it neglects important loss processes
that can occur in the continental boundary layer, such as
dry deposition and reactions with unsaturated hydrocar-
bons. It also neglects potential losses that have been
suggested to occur in cloud droplets and nighttime or
wintertime losses through nitrate radical (Np3) chemis-
try. As such, it is a lower limit for the loss rate.
Ultimately, the budget of ozone in a given region is
governed by transport of ozone into or out of the region
and the net rate of ozone formation, P(O3) -[ L(O3). Ex-
cept for urban regions, where NO2 is the predominant
sink for OH radicals and, hence, limits the formation of
RO2 radicals, the rate of O3 production is most often lim-
ited by the availability of NOX even in the boundary layer
over the European and North American continents. In
the remote atmosphere, not only is the production rate of
O3 limited by the availability of NOX, the concentration
5.6
-------
TROPOSPHERIC OZONE PROCESSES
of NOX can be so small that L(C>3) exceeds P(C>3). These
regions thus act as a buffer against any excess ozone im-
ported from areas having higher production rates or from
the stratosphere. ;:
A coarse estimate for the "critical" NO concentra-
tion at which local 63 production and loss rates are
equivalent was given by Crutzen (1979) by simply
equating the rate of R5-13 to the rate of R5-5. The criti-
cal daytime NO concentration thus derived is within a
factor of two of 10 ppt, depending on the actual ozone
concentration and other factors. However, this is a lower
limit because other loss processes have been neglected
and the term Jn'Fo'o is me dominant contribution to
L(O3) in the remote lower-to-middle troposphere. For
example, from experimental observations made at 3.4
km in the mid-North Pacific Ocean region, tliis term ac-
counted for nearly 50% of the total loss rate when
averaged over 24 hours (Ridley et al., 1992).
Nevertheless, any non-zero concentration of NOX
contributes to 03 production, compensates the loss rate,
and increases the lifetime of O3. Since P(O3) is so sensi-
tive to the NOX abundance and L(O3) is, to first
approximation, insensitive to NOX in the remote atmo-
sphere, possible trends in tropospheric 03 are intimately
dependent upon trends in NOX concentrations. Clearly,
assessing the contribution of photochemical processes to
trends in global and regional ozone relies on a good
knowledge of the distribution of O3, the fuels (CO, CH4,
NMHC), and especially the distribution of NOX. Reac-
tions R5-11 and R5-12 not only constitute an!important
03 loss rate but also initiate the oxidation cycles via OH
radicals (see Section 5.4) and therefore link stratospheric
03 change to tropospheric photochemistry through the
sensitive dependence of Jn on the overhead column of
°3-
The depletion of stratospheric ozone during the
last decade has led to increased ultraviolet radiation of
wavelength 290-320 nm penetrating to the troposphere
(see Chapter 9). Liu and Trainer (1988) studied the in-
fluence pf enhanced UV radiation on tropospheric ozone
with a simple photochemical model and fourid that the
net effect depended on ambient NOX levels. To first or-
der, an increase in the ultraviolet flux essentially
accelerates the already-existing production and destruc-
tion processes. For this reason, a positive trend in UV
radiation will, most likely, cause a negative trend in tro-
pospheric ozone in regions where the net photochemical
balance is negative, that is, over large areas of the South-
ern Hemisphere and the> remote oceanic regions of the
Northern Hemisphere (Section 5.3). The long-term ob-
servations at the South' Pole (Schnell et al., 1991;
Thompson, 1991) indicate mat some enhanced net de-
struction of tropospheric ozone may already be
occurring in association with the large stratospheric
ozone losses in that region. On the other hand, a long-
term increase in UV radiation will likely contribute to an
increase in photochemical ozone formation in the NOX-
rich continental regions and possibly in large-scale
plumes downwind of these or areas of biomass burning.
Removal or conversion of NOX to longer-lived res-
ervoirs clearly decreases the local catalytic efficiency of
03 production. During daytime, losses of NOX proceed
through the reaction of NO2 with OH radicals (R5-15)
and the formation of perpxyacetylnitrate (PAN) and its
homologues(R5-16): •
NO2 + OH in M -> HNO3 + M
RC(O)O2 <-» RC(O)O2NO2
(R5- 15)
(R5-16)
While nitric acid (HNC^), at least in the planetary
boundary layer, provides -an effective sink for NOX, the
thermally unstable compound PAN provides only a tem-
porary reservoir for NO2: Most important, the lifetime
of PAN becomes long enough at the colder temperatures
of the middle and upper troposphere that it can be trans-
ported over long distances and serve as a carrier of NOX
into remote regions. NO, is also removed by the forma-
tion of alkyl nitrates (RONO2) that are formed in the
alternative reaction path of RO2 with NO (R5-3b) (At-
kinson et aL, 1982): -
(R5-3b) RO2 + NO [-» RONO2
Similar to PAN, alkyl nitrates could provide a source of
NOX to more remote regions via photolysis or through
reaction with OH following transport (Atlas, 1988).
An important loss process for NOX that was not
included in earlier model Studies (Liu et al., 1987) is the
bxidation of NO2 by ozone itself The NO3 radical
formed in reaction (R5- 17) is extremely sensitive to pho-
tolysis but can build up Jat night to concentrations of
several hundred ppt (Plait et al., 1981; Wayne et al.,
1991). Because of the thermal equilibrium (R5-18) that
is established between NO3, NO2, and N2O5, heteroge-
neous losses of N2Os or NOs in addition to reactions of
with some hydrocarbons provide a sink for NOX in
5.7'
-------
TROPOSPHERIC OZONE PROCESSES
addition to the reaction with OH (R5-15). For example,
(R5-19) constitutes a non-photochemical conversion of
active NOX to long-lived aerosol nitrate (or HNO3 in
case of evaporation of the droplets). According to model
calculations, this mechanism could provide a significant
sink for NOX on a global scale (Dentener and Crutzen,
1993). Observations of the chemical lifetime of NO3
(Platt et al, 1 984) indicate that in the boundary layer, the
initial reaction R5-17 is often the rate-limiting step for
the removal of NOX by these processes.
(R5-17)
(R5-18)
(R5-19)
NO2 + NO3
N03 + 02
N2O5
2H+ + 2N03~
The occurrence of clouds changes the chemical
processing in an air mass significantly (Chameides and
Davis, 1982). Although the volume fraction of liquid
water in clouds is only of the order of 10"6 or less, some
gases are so soluble that they largely go into the aqueous
phase. This has several consequences: (1) the soluble
gases are concentrated in a relatively small volume,
which can enhance reaction rates, and (2) soluble gases
are separated from insoluble ones, so that some reaction
rates are significantly reduced. An important example is
reaction R5-5, which almost ceases within clouds lie-
cause HO2 is very soluble, whereas NO remains in 'the
interstitial air. Furthermore, the dissociation of dis-
solved HO2 yields O2~, which destroys O3 in the
aqueous phase. The production of HO2 in the droplets
results to a large extent from the oxidation of dissolved
formaldehyde. A radical reaction cycle is thus initiated
in which both formaldehyde and O3 are destroyed.
The estimated effect of cloud chemistry is that the
photochemical O3 production rate in the lower tropo-
sphere (where most clouds occur) is reduced by 30-40%,
while O3 destruction reactions are enhanced by up to a
factor 2 (Lelieveld and Crutzen, 1990). The net effect of
cloud processes on the O3 burden in the troposphere is
estimated to be much smaller, however, since these pro-
cesses compete with dry deposition (Dentener et al.,
1993). Model simulations suggest a 10-30% lower tro-
pospheric ozone burden as compared to a cloud-free
atmosphere (Johnson and Isaksen, 1993; Dentener et al,
1993).
5.3 INSIGHTS FROM FIELD OBSERVATIONS:
PHOTOCHEMISTRY AND TRANSPORT
During the summer months, elevated and poten-
tially harmful levels of ozone are commonly observed in
urban and rural areas of North America and Europe (Cox
et al., 1975; Logan, 1985). Slow-moving hign pressure
systems with predominantly, clear skies and elevated
temperatures set the stage for the photochemical forma-
tion and accumulation of ozone and other oxidants over
wide regions during episodes that last several days
(Guicherit and van Dop, 1977; Vukovich et al., 1977).
There is substantial evidence from field measurements
and model calculations that most of this ozone is being
produced photochemically from ozone precursors emit-
ted within the region. The export of O3 and its precursors
from the urban to regional and global scales represents
the greatest potential Impact on trends in global ozone by
anthropogenic activities.
5.3.1 Urban and Near-Urban Regions
High ozone levels in and downwind of urban re-
gions remain an important air quality problem
throughout the world. While most industrialized coun-
tries have made significant progress in lowering peak
ozone concentrations over the last two decades, un-
healthy levels of ozone persist in and around many larger
cities. In particular, in many developing countries, the
absence or ineffectiveness of emissions control efforts
can result in extremely high ozone concentrations.
The limited atmospheric chemical measurements
from urban areas in developing countries suggest that
conditions in many of these areas essentially mimic con-
ditions observed in the Organization for 'Economic
Cooperation and Development (OECD) countries during
the 1960s before implementation of large-scale emis-
sions control programs. For example, observations of
individual hydrocarbon ratios in Mexico City, Mexico,
during 1992 (Seila et al., 1993) were similar to those
observed in Los Angeles, California, during! the 1960s
and are consistent with motor vehicles as [the major
source of hydrocarbon emissions in this area. \ Motor ve-
hicle emissions have also been demonstrate^ to be the
major source of hydrocarbons in Athens, Greece; Rio de
Janeiro, Brazil; and Beijing, China (see Tang et al,
1993; Xiuli etal, 1994).
5.5
-------
TROPOSPHERIC OZONE PROCESSES
The impact of anthropogenic NOX and VOC emis-
sions on regional and global ozone levels depends on the
rate of ozone formation, the amount of ozone that is
formed per precursor, and the rate and the pathway of
transport out of the source regions. Direct and indirect
measurements of peroxy radical concentrations that
were made at several.rural sites indicate concentrations
of up to several hundred ppt at noontime on clear sum-
mer days (Parrishetal., l986;Volzetai, 1988;:Mihelcic
etal., 1990; Mihelcic and Vote-Thomas, 1993; Cantrell
et al., 1993). When combined with concurrent NO mea-
surements, these RO2 radical concentrations indicate
substantial in situ ozone production rates of several tens
of ppb/h at rural locations in the vicinity of industrialized
regions (Volz et al., 1988; Trainer etal., 1991; Cantrell et
al., 1993). Such measurements can be used to determine
the relative roles of UV radiation, NOX, and VOC for in
situ ozone production. The observed ozone increase is
usually much smaller than the gross production rate de-
rived from the RO2 measurements, which indicates that
the losses through dry deposition or reactions ^ with un-
saturated VOCs such as terpenes and by dilution must be
of similar magnitude as the production rate. This indi-
cates that the characteristic lifetime of ozone in polluted
air masses is rather small, e.g., less than one day.
In photochemically aged air in summer, ©3 was
found to increase with increasing NOy concentration,
from a background value of 30-40 ppb O3 at NOy mixing
ratios below 1 ppb to values between 70 to 100 ppb at
NOy levels of 10-20 ppb (Fahey et al., 1986). As is ex-
pected from photochemical theory, an even better
correlation is observed between ozone and the products
of the NOX oxidation (Trainer et al., 1993; Volz-Thomas
et al., 1993a). Figure 5-2 shows the results from mea-
surements made during summertime at several rural
locations in the U.S. and Canada, and at Schauinsland in
Europe. The slope of the correlation provides, at first
approximation, experimental information on the ozone
production efficiency, e.g., the number of ozone mole-
cules produced by each NOX molecule before oxidation
to more stable products such as HNOs and peroxyacetyl
nitrate (see Section 5.2.2). The increments in the indi-
vidual data sets in Figure 5-2 range from 4 to 10 and
suggest a somewhat smaller production efficiency than
what has been predicted by models for NOX levels typi-
cally encountered in rural regions of the industrialized
countries (Liuetal., 1987; Lin etal., 1988; Hov, 1989).
140 -1
120
— 100
-Q
Ł
Ą 80'
s
° 60-
40 -
C
140 n
120 •
"5
Ł 100 •
«...
o
M 60 •
40-
(
r
i
! O
i
a ,
*> ;
.-' a . ° j a Scotia
/i^*"*" I * Bondville
vdfgp 1 x Whitetop
jg * i o EQbert
?
5 Hi 15 20 '25
Products of NO, oxidation [ppb]
{•
i
-' -~ JJEV"lŁ^ "
-^^S^^\ ' -
• 'f^^K-^Pr^fart* m- * " " Schauinsland
fiP*t!*-c'f?- ?•" •". Ox/NOz-fit
*^T.-' ' "" ':" — — Orthog.-regr.
) 5 10 15 20 25
NO, .. NO, •• NO, [ppb]
j
Figure 5-2a. Ozone versus the concentration of
NOX oxidation products (e.g., NOZ in Figure 5-1), as
measured at four sites in the eastern United States
and Canada during summer 1988 and the results
from a model calculation (based upon Trainer et al.,
1993). |
Figure 5-2b. Same relation as 5-2a measured at
Schauinsland, Germany, during summer 1990 in air
masses advected from! the Rhine Valley (based
upon Volz-Thomas etal,, 1993a). The quantity Ox
= 03 + NOg is used to 'account for titration of 03
under high NOX conditions (R5-10 in Section 5.2.2).
The data also indicate a significantly lower production
efficiency for the air masses encountered at Schauin-
sland in Europe. \
The role of hydrocarbons and nitrogen oxides for
ozone formation on the urban / sub-urban scale was stud-
ied by Hess et al., (1992a, b, c) in an outdoor smog
chamber using a synthetic gas blend that closely resem-
bled that of automobile exhaust. The most important
finding was that the initial; rate of ozone formation de-
pended on the mix of hydrocarbons used and, of course,
on the availability of UY light. However, the final
5.9
-------
TROPOSPHERIC OZONE PROCESSES
amount of 63 produced during one day depended mainly
on the availability of NOX. To some extent, the latter
depends on the hydrocarbon mix, specifically on the ex-
istence of NOX sinks in the chemistry through formation
of organic nitrates (Carter and Atkinson, 1989).
Insight into the chemical breakdown of hydrocar-
bons and their role in ozone formation can be obtained
from field measurements of alkyl nitrates (RONOi),
since these species are formed as a by-product in reac-
tion (R5-3), which is rate-limiting in ozone formation.
From an extensive series of measurements made at
Schauinsland, a mountain site in Southern Germany, in
summer, a linear relation was found between ozone arid
alkyl nitrate concentrations, which is shown in Figure
5-3 (Flocke et al, 1991, 1993). The high degree of cor-
relation found in air masses that originate in the Rhine
. Valley, and thus represent a relatively uniform mix of
hydrocarbons, clearly points out that most of the ozone
observed at Schauinsland in summer (70 ppb average
and peak values of 130 ppb) is formed in situ from an-
thropogenic precursors emitted within the region. By
extrapolation to RONO2 concentrations of zero, an
estimate of 20-30 ppb is obtained for today's non-photo-
chemical background mixing ratio of ozone in the
continental boundary layer in summer (Flocke, 1992;
Flocke et al., 1993; Volz-Thomas et al., 1993b). This
finding supports the conclusions drawn by Volz and Kley
(1988) and by Staehelin et al., (1994) from historic mea-
surements (see Chapter 1) and proves the predominant
anthropogenic influence on ozone levels in some rural
areas today. Since alkyl nitrates are not removed by raiin-
out, they are better suited for such an extrapolation than
either NOy or NOZ (= NOy - NOX), since the latter con-.
tain soluble HNOs as a major constituent.
The European studies also led to the conclusion
that about one ozone molecule per carbon atom is
formed from the oxidation of hydrocarbons in these air
masses (Flocke, 1992). Furthermore, the relative abun-
dance of the different alkyl nitrates indicates that most, of
the smaller RO2 radicals are not formed from the oxida-
tion of the respective parent hydrocarbons but by
decomposition of larger alkoxy radicals. This finding is
in agreement with results from laboratory studies (Atkin-
son et al., 1992) and RO2 production from the
decomposition of RO radicals is now a common feature in
detailed chemical mechanisms used in urban airshed mod-
els (Carter, 1990; Atkinson 1990). The finding is also
=
oc
0.22x-i-26
0 100 200 300 400 ; 500
Allcylnitrate Mixing Ratio [ppt] •
Figure 5-3. Correlation of Ox = Oa + NC<2 concen-
trations with those of alkyl nitrates (RQNO2) as
observed at Schauinsland, Germany, in summer
under polluted conditions (based upon Flocke et
a/., 1992). Ox and RONOa emerge from the same
reaction (R5-3).
consistent with the fact that measured ratios of organic
peroxy radicals to HC>2 are significantly larger than those
predicted by models that do not include this mechanism
(Mihelcic and Volz-Thomas, 1993). The conclusion is
that the rate of production of RC>2 radicals is greater than
originally assumed in these models.
Carbon monoxide is an anthropogenic pollutant
that has a relatively long photochemical lifetime (1
month in summer) and is not affected by raino|ut. Thus,
it is a suitable tracer of anthropogenic pollution on long-
er time scales (Fishman and Setter, 1983). Parrish et al.
(1993) observed a strong correlation between ozone and
CO with a consistent slope AOs/ACO = 0.3 'at several
island sites in eastern Canada (Figure 5-4). :The sites
were located at approximately 500-km intervals down-
wind of the northeastern urban corridor of the United
States, and covered approximately one-third of the dis-
tance from Boston to Ireland. By scaling the observed
slope to a CO emission inventory, they inferred a net ex-
port of 5 Tg anthropogenic 03 out of the eastern U.S. in
summer. Chin et al. (1994) successfully simulated the
observed O^-CO relationship in a continental-scale
three-dimensional (3-D) model and concluded that the
correlation slope of 0.3 is a general characteristic of aged
polluted air in the U.S.- The model allowed iri particular
to correct for the effect of O^ deposition. From this cal-
culation, Chin et al. (1994) estimated that^ export of
eastern North American pollution contributes 7 Tg of 03
5.10
-------
TROPOSPHEI 1C OZONE PROCESSES
100
80
60
a
a.
40
20
slope = 0.27
R2 = 0.74
slope = 0.29
R2 = 0.71
...'_>•••
•?'•' ;'-h'
•'-:".'.'• '%'.-•%
. ..",
•/•%Ł••.,
' '
Seal Island
slcpe = 0.22
= 0.46
»:; iii-:
Ms-
Sable Island
. . i . . .
Cape Rac<
100 200
i
0 100 0
CO (ppbv)
100
Figure 5-4. Relation between O3 and CO observed at three island sites in the I
during summer 1992 (based upon Parrish etal., 1993).
in summer. Jacob et al. (1993) used the same model to
estimate that pollution from all of North America con-
tributes 30 Tg of O3 to the Northern Hemisphere in
summer, of which 15 Tg is due to direct export and 15 Tg
is due to export of NOX leading to 03 production in the
remote troposphere. This anthropogenic source of 63 is
about one-third of the estimated cross-tropopause trans-
port of 63 in the Northern Hemisphere in summer.
Considering that the U.S. accounts for about 30% of fos-
sil fuel NOX emissions in the Northern Hemisphere, it
can be concluded that anthropogenic sources make a
major contribution to tropo^pheric ozone on the hemi-
spheric scale, of magnitude comparable to influx from
the stratosphere.
While the summertime measurements show a
strong positive correlation of ozone with anthropogenic
tracers such as NOy and CO, a negative correlation was
observed during winter. A decrease in the ©3 concentra-
tion with increasing CO concentration is observed at a
number of locations in North America and Europe (Poul-
ida et al., 1991; Parrish etal., 1993; Scheel etaL, 1993;
00 300
North Atlantic west of Canada
Simmonds, 1993; Derwent etal., 1994). Derwente/a/.
conclude from their analysis of the air masses that arrive
at Mace Head, Ireland, that; the European continent is a
net source of ozone in summer, but is a net sink in winter.
This estimate, however, is only valid for the planetary
boundary layer and does aot include the influence of
NOX export on the net photochemical balance of ozone.
A seasonal trend is als:o apparent in the correlation
of ozone with NOy and NOZ (Fahey et al., 1986; Volz-
Thomas et al., 1993a). The wintertime measurements of
03, NOX, and NOy at Schaiiinsland indicate a decrease
of ozone with increasing concentrations of the products
of the NOX oxidation and, heince, support the importance
of nighttime chemistry in the oxidation of NOX at the
expense of ozone in polluted air masses.
Since.anthropogenic NOX emissions do not have a
strong seasonal variation, Gal vert et al. (1985) argued
that the absence of a seasonal cycle in nitrate deposition
rates provided evidence for the importance of NO3
chemistry in the removal of NOX. However, more recent
data from the National Aci:d Deposition Program and
5.11
-------
TROPOSPHERIC OZONE PROCESSES
SEASONAL DEPICTIONS OF
TROPOSPHERIC OZONE DISTRIBUTION
December - February
-180 -120 -60 0 60 120 180
June - August
-180 -120 -60 0 60 120 180
Longitude
-180 -120 -60
-180
120 180
September - November
-120 -60 0 60
Longitude
120 180
<20 25
35 45
Dobson Units
55 60>
other North American sites do indeed show a seasonal
cycle in bulk nitrate deposition rates, with larger values
in summer (Correll et ai, 1987; Doddridge et ai, 1992),
as would be expected if OH radicals played the dominant
role in controlling the NOX budget. Some further sup-
port for the dominance of OH in controlling the removal
of NOX is provided by the observation of larger NOX. lev-
els in the Arctic winter (Dickerson, 1985; Honrath and
Jaffe, 1992).
5.3.2 Biomass Burning Regions ;
Biomass burning takes many forms; among them,
forest and savanna fires, burning of agricultural wastes,
and the use of biomass fuels as a domestic energy source
are the most important. Biomass fires release a mixture
of gases containing the same ozone precursors emitted
from fossil fuel combustion: NOX, CO, CH4, and non-
methane hydrocarbons (NMHC), including a large
proportion of alkenes. Ozone production in aged bio-
mass-burning plumes has been shown by numerous
investigators (Delany et at., 1985; Andreaelef ai, 1988,
1992; Cros et ai, 1988; Kirchhoff et ai, J1989. 1992;
5.12
-------
TROPOSPHE
KirchhoffandMarinho, 1994). The global emissions of
ozone precursors from biomass burning haves been esti-
mated in a recent review by Andreae (1993) to be
comparable in magnitude to the emissions from fossil
fuel burning. Evidence for the importance of tropo-
spheric ozone production frompyrogenic precursors has
been obtained from the analysis of satellite data (Figure
5-5; Fishman et ai, 1991), which show a substantial en-
hancement of tropospheric ozone downwind from the
biomass burning regions in South America and Africa.
Ozonesonde measurements in Africa and on Ascension
Island in the central South Atlantic do indeed confirm
the persistence of high ozone levels in the mid-tropo-
sphere during the burning season (Cros et ai, 1992;
Fishman etal., 1992). Aircraft measurements have dem-
onstrated the origin of these ozone-enriched air masses
from biomass burning (Marenco etal., 1990; v^ndreae et
al., 1988, 1992, 1994a). Compelling evidence was also
collected in more recent aircraft campaigns that docu-
mented the distribution of ozone and its pyrogenic
precursors in a region extending from South America
across the Atlantic Ocean to southern Africa (Andreae et
a/., 1994b).
Due to the dispersed nature of biomass burning
and the relatively small number of field investigations on
ozone production from pyrogenic precursors, it is still
difficult to provide a quantitative estimate of ozone pro-
duction from this source. The observed ratio of ozone to
CO enhancements in aged burning plumes varies from
near zero in some tundra fire emissions (Wofsy et al.,
1992) to almost one in some aged savanna fire plumes
(Andreae etal., 1994a). As shown in Figure 5-6, these
differences appear to be related to the ratio of NOX to CO
(and consequently NMHC) in the emissions. By using
an average OyCO-ratio of 0.3 and a CO emission of 300
Tg C/yr from biomass burning, Andreae (1993), estimat-
ed a global gross O3 production of ca. 400 Tg O^/yr from
biomass burning, with an uncertainty of at least a factor
of two. A recent model study estimated a similar gross
rate of 540 Tg/yr; however, the net production of ozone
from biomass burning was found to be only 100 Tg/yr
(Lelieveld and Crutzen, 1994). This large difference
emphasizes, as already discussed for the northern mid-
latitudes above, the crucial role of transport processes in
distributing the ozone between the PEL, where it is de-
stroyed rapidly, and the free troposphere, where its
chemical lifetime is long enough for it to be dispersed
1.0
O
U
0.5
RIC OZONE PROCESSES
005
0.10
dNOy/ACO
FHgure 5-6. Ratio of O3 to CO in aged biomass
burning plumes as a function of the NOy/CO ratio.
The increase seen in the data clearly indicates the
important role of NOX for ozone formation in the
plumes (based upon Andreae et al., 1994b). The
straight line represents a fit to the majority of the
data. It is consistent with an average Oa/NOy ratio
of 4-5, quite similar to the ratios observed in sub-
urban air masses over Eiurope.
hemisphere-wide. The accurate description of these
transport processes probably represents the largest diffi-
culty in current global models, as is discussed in more
detail in Chapter 7. I
The secular trends of biomass burning are highly
uncertain. Obviously, fire has been present on Earth
since the evolution of land plants, and human activity
has resulted in large fires in the savannas of Africa and
South America since the advent of human beings. How-
ever, other types of biomass burning have clearly been
increasing over the last century, especially deforestation
fires and domestic biomass fuel use. These types of bio-
mass burning have especially high emission factors for
ozone precursors. Andreaei(.1994) estimated that the re-
lease of trace gases from biomass burning has increased
by about a factor of two or three since 1850. Semi-quan-
titative measurements made during the last century,
albeit not considered sufficiently reliable for an indepen-
dent quantitative assessment (Kley et al., 1988), would
indeed support a secular increase in tropospheric ozone
concentrations in the Southern Hemisphere (Sandroni et
al., 1992). No trend is seendn the surface ozone records
obtained over the last two decades at Cape Point, South
5.13
-------
TROPOSPHERIC OZONE PROCESSES
Africa (Scheel et ai, 1990) and American Samoa, Pacif-
ic Ocean (Oltmans and Levy, 1994).
5.3.3 Remote Atmosphere and
Free Troposphere
While recent work has provided major advances in
our understanding of the ozone budget over continental
regions relatively close to the centers of precursor emis-
sions and clearly demonstrated the anthropogenic
perturbation of tropospheric ozone on a regional scale,
the system is still far from being understood on a hemi-
spheric or global scale. Compounding issues are: (1) the
strong, if not overriding, influence of transport from both
the stratosphere and the continental source regions,
transport that is episodic rather than steady; (2) the large
difference in chemical time constants between the
boundary layer and the free troposphere; (3) the uncer-
tainty in the production rate for NOX by lightning and its
distribution; (4) the extremely low NOX and NOy constit-
uent concentrations, which represent a real measurement
challenge; and (5) the large volume that must be covered
to establish a climatology. Concerns have also been
raised about the validity of NO2 and NOy measurements
in the troposphere obtained by commonly used tech-
niques (see Davis et, al., 1993; Crosley, 1994).
Nevertheless, the large number of field experiments per-
formed over the last years (see Carroll and Thompson,
1994) has led to a better understanding of the 63 budget
on a global scale. Even greater insight is expected to
come out of the recently completed or ongoing experi-
ments that have included direct measurements of OH
and ROi radicals and the seasonal variations of active
nitrogen compounds in the remote atmosphere.
Plumes of "pollution" from biomass burning re-
gions and from the industrialized regions of North
America, Europe, and Asia have been identified from
satellite observations (Fishman et ai, 1990). Being
downwind of continental source regions, measurements
made in the marine boundary layer at Barbados, West
Indies, show that large variations in O3 concentrations
can be associated with changes in long-range transport
patterns. There is a pronounced seasonal cycle for O3 at
Barbados (Oltmans and Levy, 1992, 1994). During the
winter and spring, daily averaged values are typically in
the range of 20-35 ppb, while during the summer, values
are typically 10-20 ppb. During the winter-spring period
there are often large changes in 03 concentration; these
changes are strongly anticorrelated with a number of
aerosol species, including NO3~ (Savoie et al., 1992).
The changes in O3 are driven by changing transport pat-
terns over the North Atlantic as opposed to ;chemical
reactions involving O3 and nitrogen species in the atmo-
sphere. Analyses of isentropic trajectories clearly show
that high O3 and low NO3~ are associated with; transport
from the middle and high latitudes and from relatively
high altitudes in the free troposphere. Conversely, high
NO3" and relatively low O3 are associated with transport
from Africa. The lack of association of high O3 with
ground-level sources is supported by the strong anticor-
relation of O3 with 210-Pb; conversely, the strong
correlation of NO3" and 210-Pb (and a weaker correla-
tion with Saharan dust) indicates that NO3 is derived
principally from continental surface sources, probably in
Europe and North Africa. These associations suggest
that African biomass burning could be a significant
source of NO3", but appears to be a minor source for O3
at Barbados. Although substantial amounts pf O3 may
have been produced as a consequence of the burning, a
substantial fraction must have been destroyed in transit
in the marine PBL.
The importance of transport processes for the glo-
bal ozone distribution is also emphasized in studies
made at the Spanish Meteorological observatory at Iza-
na, Tenerife. The station is located at an elevation of 2.4
km, above the top of the marine inversion most of the
time. At Izana, ozone concentrations have a well-de-
fined seasonal cycle, with monthly means of about 40
ppb in winter, about 55 ppb in spring, and about 50 ppb
in July (Schmitt et al., 1988). The concentrations in
summer are much higher on average than those observed
at Mauna Loa, Hawaii,' and exhibit a bimodal distribu-
tion. Low mixing ratios of -20 ppb are advected from
the open ocean and the Saharan desert, and high values
of up to 100 ppb generally result from relatively rapid
transport from northern latitudes (Schmitt and Hanson,
1993). It has been suggested that the persistence of high
ozone concentrations in the summer could be due to the
transport of ozone from Europe, based on isentropic
back-trajectories and the correlation of high: ozone epi-
sodes with increased concentrations of tracers of
anthropogenic origin such as CH4, PAN, VOC, and CO
(Schmitt era/., 1988, 1993; Volz-Thomas et'al., 1993c).
The seasonal cycle of ozone at Izana is similar to that for
5.14
-------
TROPOSPHERIC OZONE PROCESSES
-60°S -40° -20° 0 20° 40° 60°N
LATITUDE
500
-60°S -40° -20° 0 20° 40° 60°N
LATITUiDE
Figure 5-7. NO concentrations measured in the free troposphere during STRATOZ III and TROPOZ II
(based upon Ehhalt et al., 1992; Wanner et al., 1994). The flight track was similar in both missions and
extended over the North and South Atlantic and the west coast of South Americal
non-seasalt sulfate (nss-SO4~) and NOj~, which could
be interpreted as supporting an anthropogenic source for
ozone. However, on a day-tg-day basis, ozone is strong-
ly anticorrelated with aerosol NC>3~ and nss-SC>4=
(Prospero et al., 1993). This and the coherence between
ozone and 7-Be, which is produced from cosniic rays in
the upper troposphere and lower stratosphere (see Brost
et al., 1991), could imply a major contribution of ozone
from the stratosphere or effective losses of aerosol ni-
trate and sulfate during convective transport from the
planetary boundary layer into the free troposphere. In
this regard, the results obtained at Izana are similar to
those obtained in the marine boundary layer at Barba-
dos, i
Evidence that stratospheric input is an important
component of the upper tropospheric ozone budget, es-
pecially in spring and early summer, was presented by
Beekmann et al. (1994), based on the correlation be-
tween ozone mixing ratio and potential vorticity, and by
Smit etal. (1993), based on a time series of ozonesonde
measurements in the upper and lower troposphere. The
poleward increase in upper tropospheric ozone suggests
5.15
-------
TROPOSPHERIC OZONE PROCESSES
AASE1
NOx (pptv)
12-
11-
ID-
S'
7'
6-
12-
11-
10-
•g 8-
3
5 7-
6-
5-
35 40 45 50 55 6ff 65 7Q 75 80 85
Latitude .
AASE2
NOx (pptv)
-60=-
35 40 45 50 55 60 65 70 75 80 85
Latitude
Figure 5-8. Summary of NOX = NO + NO2 concen-
trations in the free troposphere measured in the
Northern Hemisphere during the AASEI and AASE
II missions (based upon Carroll et al., 1990a and
Weinheimer et ai, 1994).
that this component is even more important at high lati-
tudes. Evidence for stratospheric input to the Arctic
troposphere was presented by Shapiro et al. (1987) and
Oltmans et al. (1989). Furthermore, airborne lidar mea-
surements made over the Arctic region in summer found
that stratospheric intrusions dominated the ozone budget
in the free troposphere (Browell et al, 1992; Gregory et
al., 1992). There is also a suggestion in ozonesonde data
from the South Pole (Gruzdev and Sitnov, 1993) that
ozone depletion in the Antarctic polar vorte;x extends
into the upper troposphere.
An example of progress in determining large-scale;
reactive nitrogen distributions over the complete trppo-
spheric altitude regime is shown in Figure-5=7;.which
contrasts the seasonal distribution.of NO from aircraft
measurements made during the Tropospheric Ozone II
(TROPOZ II) mission in January 1991 (Wanner el al,
1994) and the Stratospheric Ozone ffl (STRATOZ III)
mission in June 1984 (Drummonder aL, 1988; Ehhalt et
al, 19.92)i The mixing ratios are considerably higher in
the Northern Hermsphese, particularly at high latitudes
in winter, and at 20-50°Nr at high altitudes in summer.
Vertical gradients are strongest in June north of 20°S.
The high mixing ratios of NO at northern midlatitudes
are attributed to stratospheric input, aircraft emissions,
and convective transport from the "polluted" boundary
layer (Ehhalt etal, 1992).
The NO concentrations observed during TROPOZ
II are much larger than what has been observed by other
investigators at similar latitudes and seasons. Figure 5-8
shows NOX concentrations observed during the Arctic
Airborne Stratospheric Expedition (AASE) I and II mis-
sions (Carroll et al, 1990a; Weinheimer et al, 1994).
While these flights were made during the same season as
TROPOZ II and at overlapping latitudes, they show
much lower NOX (=NO+NO2) concentrations than the
NO concentrations alone that were observed during
TROPOZ II. The AASE measurements are in general
agreement although separated by a three-year period.
The difference may be due to the shorter measurement
period of the TROPOZ program, and an unustial synop-
tic event, compared to the longer-term AASEjprograms.
Barring unexpected measurement uncertainty, the differ-
ences demonstrate the difficulty in ascertaining a
climatology of a short-lived species like NOX pver larger
scales.
5.16
-------
TROPOSPHIERIC OZONE PROCESSES
10.000
1.000
.a
a.
Q.
X
O
0,100
0.010-
0.001
Ground Data
O Flight Data
0.010
0.100
1.000
NOy [ppb]
10.000
100.000
Figure 5-9. Summary of NOX and NOy concentrations in the PBL and free troposphere (from Prather etal
1994, based on Carroll and Thompson, 1994). The majority of the airborne measurements shows NOX
concentrations that are too small to sustain net ozone production. The letters and numbers within the sym-
bols refer to the following measurement campaigns (see Appendix for acronym definitions): a = ABLESa- A =
AASE; b = ABLESb; B = Barrow, Alaska; H = Harvard Forest; K = Kinterbush, Alabama; M = MLOPEX-' n =
NACNEMS; N = Niwot Ridge, Colorado; P = Point Arena, California; s = SOS/SONIA, S = Scotia Pennsylva-
nia; T = TOR; 2 = CITE2; 3 = CITE3. i
Murphy et al. (1993) have measured vertical distri-
butions of NOy and O3 into the stratosphere. Although a
strong correlation between NOy and O3 was found in the
stratosphere, they observed only weak to no correlation
between these constituents in the troposphere, i.e., the
tropospheric NOy/O3 ratio can be larger and more vari-
able, a reflection of the variety of sources, sinks, and
transport processes of NOy and O3 in the troposphere. In
contrast, Wofsy et al (1992) and Hiibler et al. (1992a, b)
reported a significant positive correlation when the data
are averaged over a large number of observations. The
observed slope was much steeper than that derived from
continental boundary layer studies (Section 5.3.2) and
approached that found in the stratosphere. The large de-
crease in the NOy/C>3 ratio between the continental
surface studies and the remote free atmosphere is be-
lieved to largely reflect the shorter lifetime of NOy
compared to O3 in the free: troposphere, mixing, and in-
put from the stratosphere. 1
A summary of tropospheric NOX and NOy concen-
trations from Prather et.al. (1994) is shown in Figure 5-9.
It is based on the compilation of Carroll and Thompson,
(1994) of measurements made by various groups in the
lower and middle troposphlere over the U.S. and Europe.
Although very high concentrations from urban areas are
excluded, the concentrations of NOX span a range of
three orders of magnitude.'. On the average, a correlation
between NOX and NOy is sieen. However, the individual
data sets clearly show that the shorter-lived NOX can still
vary over an order of magnitude for a given NOy concen-
tration. From this and the differences in NOX observations
in the upper troposphere a.t northern latitudes discussed
above, it is clear that present measurements are insufficient
to reasonably describe a meaningful climatology.
5.17
-------
TROPOSPHERIC OZONE PROCESSES
Aircraft programs have continued to strengthen the
role of PAN as a reservoir for NOX, at least in the 3-6 km
altitude range over continental regions (Singh et al.,
1992,1994), where PAN decomposition was able to ac-
count for much of the observed NOX, a result that
emphasizes the role of transport of odd nitrogen reser-
voirs. Very high PAN concentrations of up to 200 ppt
were also observed in long-range transport events at Iza-
na during spring, whereas PAN concentrations remained
below 20 ppt at the higher temperatures of summer
(Schmitt and Hanson, 1993). Other studies have shown
that the importance of PAN as a NOX reservoir is not glo-
bal. Measurements made in the Northern Hemisphere
upper troposphere mostly over the Atlantic Ocean have
generally shown smaller mixing ratios than observed in
the middle troposphere over continental regions, and
Southern Hemisphere mixing ratios were very small
throughout the troposphere (Rudolph et al, 1987; Per-
ros, 1994). Similarly, during studies at the Mauna Loa
Observatory experiment, PAN was not a major constitu-
ent. HNOs was the dominant reservoir (median of 43%
of NOy), followed by NOX (14%), paniculate nitrate
(5%), PAN (5%), and alkyl nitrates (2%) (Atlas et al.,
1992).
The role of the remote marine PEL as a strong net
sink for ozone has been clearly identified in a large num-
ber of investigations, a finding first reported by Liu et al.
(1983). For example, a clear anticorrelation in the diur-
nal and seasonal variation of 03 and H2O2 was observed
by Ayers et al. (1992) in marine air at Cape Grim, Tas-
mania (Figure 5-10). As is seen in Figure 5-1, HO2
radical recombination leads to formation of H2O2, which
can thus be utilized as a tracer for photochemical active
ty. The results are consistent, with net photochemical
destruction of Os in a very low NOX atmosphere. Net
photochemical destruction of O3 in the tropical PEL of
up to 25%/day was also inferred from the data gathered
during several ship cruises (Thompson et al., 1993; Smit
et al., 1989; Smit and Kley 1993; Harris et al., 1992).
The photochemical buffer regions are not confined
to the remote maritime lower atmosphere. Aircraft
flights covering Alaska, northern Ontario and Quebec,
and Labrador have concluded that the surface layer, es-
pecially the boreal forest, was an efficient sink for O3
and NOy (Gregory et al., 1992; Jacob et al., 1992; Bak-
win et al., 1992). In some regions of these flights, NOX
was nearly independent of altitude up to 6 km with a
median mixing ratio of only 25 ppt, insufficient to over-
come average net photochemical destructio'n of Oj
(Sandholm et al, 1992). Earlier studies over the conti-
nental U.S. by Carroll et al. (1990b) found that air
masses between the boundary layer and 5-6 km, were
nearly equipartitioned between net loss, approximate
balance, and net production of 03. :
An extremely interesting finding that yet awaits
complete explanation is the occurrence of nearly com-
plete 03 depletion in the Arctic surface layer in spring
(Barrie etal., 1988; Bottenheim et al, 1990; McConnell
et al, 1992; Fan and Jacob, 1992). A recent analysis of
the ratios of different hydrocarbons provides evidence
for bromine chemistry being responsible for the ozone
removal (Jobson et al, 1994), although Platt and Haus-
mann (1994) argue that the measured BrO concentrations
were too small to explain the complete ozone depletion
on the short time scales implied by the observations.
Net ozone loss of 0.5 ppb/day, or -1%/day, was
also found in the free troposphere near 3.4 km from ob-
servations at the Mauna Loa Observatory (Ridley et al,
1992). The concentrations of peroxy radicals and the
rate of ozone formation, P(O3), were derived from the
photostationary state of NOX (Figure 5-11) and the loss
rate, L(Os), was inferred from model calculations based
on the measured concentrations of all relevant parame-
ters. It is noteworthy that both the total concentration of
HO2 and RO2 determined during this study, as well as
the modeled HO2/RO2 ratio, are in good agreement with
recent direct measurements made by matrix isolation
and ESR spectroscopy at Izana, Tenerife (D. Mihelcic,
private communication).
The net destruction rate found in spring at Mauna
Loa in the free troposphere is slow enough that vertical
exchange with the marine boundary layer can overrule in
situ chemistry. Vertical soundings made frpm a ship
cruise in the Pacific clearly demonstrate the importance
of convective transport for the ozone balance of the free
troposphere. Extremely low ozone concentrations, that
had their origin in the marine boundary layer, were
found in the upper troposphere (Smit and Kley, 1993).
These observations contrast those made or modeled over
continental regions, where an emphasis has been on the
role of convection of boundary layer precursors in aug-
menting O3 production in the middle .and upper
troposphere (Dickerson et al., 1987; Pickering et al,
1992a, b; Thompson et al, 1994). As was suggested by
5.18
-------
TROPOSPHERIC OZONE PROCESSES
Peroxide and Ozone Average Diurnal-Cycles at Cape Grim
15
14.5 •
1.3.5
<§ 13
12.5
12
Hour of Day
18
03
H202
r 1100
• 1000 >
• 900 s
700 S
w
600 <->
!JOO
24
Peroxidn and Ozone Seasonal Cycles at Cape Grim
35
30
25
20
15
10
5 •
•O3
•H202
11400
,1200 T
11000
1800
600
,400 c
•r s
Figure 5-1 Oa. Average diurnal cycles for peroxide and ozone in background air at Cape Grim for January
1992 (based upon Ayersef a/., 1992). . I
Figure 5-1 Ob. Seasonal cycles of peroxide and ozone in background air at Cape Grim (based upon Ayers et
31., 1992)."
modeling studies (Lelieveld and Crutzen, 1994), down-
ward mesoscale flow in the cloud environment can carry
Os to the Earth's surface, where it is destroyed more rap-
idly. Although these model studies yet 'await
confirmation by experimental data, it is likely that deep
convection tends to increase free tropospheric ozone lev-
els downwind of continental source areas but may
i
reduce tropospheric O3 in regions that are removed from
polluted areas. !
Intensive studies at N/Iauna Loa have suggested
some possible discrepancies'in our understanding of the
atmospheric oxidizing capacity. Programs completed
more recently may help to determine whether these re-
sults are more universal in the remote troposphere. First,
the abundance of formaldehyde (HCHO) predicted from
5.19
-------
TROPOSPHERIC OZONE PROCESSES
4 6 8 10 12 14 16 18 20
HAWAII STANDARD TIME
Figure 5-11. Average diurnal variation of peroxy
radical mixing ratios derived from measurements of
trace gases and photolysis rates during the Mauna
Loa Observatory Photochemistry Experiment (Rid-
ley et al., 1992). The bars give the mean and
standard deviation of the total peroxy radical mixing
ratio estimated from the photostationary state of
NOX. The solid lines are model predictions for the
mixing ratios of (HOa + CHaOa) and of Cr-I^Oa.
respectively.
a model (Liu et al., 1992) was three times larger than the
observed median (Heikes, 1992). Since measurements
of a variety of hydrocarbons (Greenberg et al, 1992)
showed that CH4 oxidation was the dominant source of
HCHO, the results implied that model abundance of OH
was too high or, more likely, that other HCHO removal
processes not included in the model were important.
Second, the observed HNO3/NOX ratio was also in poor
agreement with the photochemical model, unless the re-
moval rate for HNO3 was increased equivalent to a 3-5
day lifetime. More limited aircraft measurements have
also indicated a smaller-than-expected ratio (Huebert et
al, 1990). The model used by Ehhalt et al (1992) to
describe the aircraft observations of NO also implied a
very short average lifetime of NOy, on the order of a few
days. If the HNO3 reservoir is indeed removed faster
than commonly described in models, the increased
efficiency of O3 production in the remote atmosphere is
weakened compared to that modeled previously. How-
ever, simply decreasing the model lifetime of |HNO3 will
eventually cause significant discrepancies in; simulating
the mixing ratios of NOX in the remote atmosphere, since
NOX is ultimately lost through HNO3 formation. Clear-
ly, more systematic investigations of 03, NOX, and other
NOy species and suitable tracers for transport need to be
conducted in remote locations in order to better under-
stand the interplay between transport and chemistry in .
determining the "global" budget of ozone and its poten-
tial for future increase.
5.4 FEEDBACK BETWEEN TROPOSPHERIC
OZONE AND LONG-LIVED GREENHOUSE
GASES
The concentrations of many trace gases that con-
tribute to the greenhouse effect of the atmosphere or are
involved in the budget of ozone in the stratosphere or the
troposphere, i.e., CH4, CO, NMHC, NOX, methyl bro-
mide (CH3Br), HFCs, and HCFCs, are mediated through
oxidation by OH radicals. Reaction R5-11 followed by
R5-12 provides the major source for OH in the unpollut-
ed troposphere. Therefore, OH concentrations are
strongly linked to the UV flux below 320 nm[(UV-B) and
the concentrations of water vapor and ozone itself. In
addition, OH is affected by other trace gases. For this
reason, rising levels of CH4, CO, and NOX jnay lead to
changes in the oxidizing capacity of the troposphere (see
Thompson and Cicerone, 1986), which in1 turn influ-
ences the concentrations of gases relevant to global
warming and/or stratospheric ozone depletion.
On short time scales, increases in UV flux, H2O,
and O3 lead to increases in OH, as is clearly borne out by
the good correlation found between OH concentrations
and the photolysis frequency of ozone (f*latt et al,
1988). On longer time scales, however, the net effect of
enhanced UV radiation and H2O concentrations on OH
depends on the net photochemical balance of ozone
P(O3) - L(O3) in the particular region of the atmosphere,
that is, on the NOX concentration, and advective trans-
port of ozone from other regions. \
Besides being required for ozone maintenance,
NOX increases change the partitioning between OH and
HO2 to favor OH via reaction R5-5. Thus, increasing
NOX will lead to an increase in OH, at least for NOX con-
centrations below 1 ppb (Hameed et al., 1979; Logan et
5.20
-------
TROPOSPHERIC OZONE PROCESSES
al., 1981). At higher concentrations, reaction (R5-15)
becomes the major loss process for OH (and HOX = OH
+ HO2) and a further increase of NOX will tend to reduce
OH concentrations. At very low NOX levels, e.g., below
a few tens of ppt, recycling of OH occurs via reaction
(R5-13). In this sense, the dependence of OH on NOX in
the remote atmosphere and on long time scales is much
stronger than implied by models that use fixed ozone
concentration fields.
The exact concentration of NOX at which the influ-
ence of NOX upon OH changes sign depends on the
concentrations of ozone (see above) and those of other
trace gases such as CO, CH4, and NMHC. The latter
gases change the HOX partitioning in favor of HO2 (R5-1
to R5-4 and R5-8 to R5-9) and thereby serve to reduce
OH. However, this negative influence is not a linear one
because of the reduction in HOX losses that proceed via
OH reactions, Le., R5-15.
Although attempts to measure OH were made in
the early seventies (see'Wang et al, 1976; Pemer et al,
1976), direct measurements are still extremely sparse
(Pemeretal., 1987;Platter at., 1988;FeItonefa/., 1988;
Dorn et al., 1988; Mount and Eisele, 1992; Eisele et al.,
1994), in particular in the remote atmosphere. One rea-
son for this is the experimental difficulty involved given
the extremely low concentrations of OH radicals due to
their reactivity. In addition, because of the fast/response
of OH to changes in the controlling boundary condi-
tions, these measurements will not and cannot produce a
global field of OH concentrations. They can, however,
lead to improvements in understanding the chemical
budget when accompanied by measurements of the con-
trolling factors (Ehhalt et al, 1991) and, hence, help to
calibrate the photochemical models used to derive global
OH fields (Ehhalt et al., 1991; Poppe et al., 1994), in
particular with the recent advances in measurement ca-
pability for OH. However, any attempt in modeling the
global OH field and, in particular, its secular trend, for
example that induced by the increase in methane con-
centrations (see Chapter 7) or UV radiation, relies on an
accurate knowledge of the distribution and trends of
ozone, water vapor, and a number of other parameters,
but most importantly, on the distribution in space and
time of NOX.
Average figures on global OH concentrations have
been derived from the concentrations of tracers that are
removed from the atmosphere preferentially by OH.
Among these are CH3CGI3, with an atmospheric turn-
over time of about 6 years, and 14CO, with a turnover
time of a few months. A detailed discussion of these
indirect attempts is given in Chapter 7.
More recently, other potentially important oxi-
dants in the troposphere have been suggested in addition
to OH. Among these are chlorine atoms (Pszenny et al.,
1993), which may be formed in the marine boundary
layer from reactions of ^65 with aerosol chloride (Fin-
laysen-Pitts et al, 1989; Zetsch and Behnke, 1992).
Penkett et al (1993) concluded, from the measured ratios
of iso- to normal alkanes in the atmosphere that NO3
radicals could play a significant role in the atmospheric
oxidation of NMHC on larger regional scales, in particu-
lar at higher latitudes. [ The importance of these
additional oxidizing reagents is, however, still in the hy-
pothesis stage. While it hiis been suggested that atomic
chlorine and bromine could play a role in certain regions
of the troposphere, for example during spring in the Arc-
tic (Fan and Jacob, 1S|92; Jobson et al, 1994),
concentrations larger than |1% of the average global OH
concentration seem to be inconsistent with the budgets
of some trace gases (J. Rudolph, private communica-
tion). '
REFERENCES i
Alaart, M., H. Kelder, and L.C.G. Heijboer, On the trans-
port of trace gases by extra-tropical cyclones, in
Proc. Quadr. Ozone Symp., Charlottesville, Virginia,
June 4-13, 1992, inpiress, 1994.
Ancellet, G., J. Pelon, M. Beekrnann, A. Papayannis, and
G, Megie, Ground-beised lidar studies of ozone ex-
changes between the stratosphere and the
troposphere, /. Geophys. Res., 96, 22401-22421,
1991. i
i
Ancellet, G., M. Beekmann, and A. Papayannis, Impact
of a cut-off low development on downward trans-
port of ozone in the troposphere, J. Geophys. Res.,
99,3451-3468, 1994,!
I
Andreae, M.O., The influence of tropical biomass bum-
ing on climate and the atmospheric environment,
in Biogeochemistry of Global Change: Radiatively
Active Trace Gases\ edited by R.S. Oremland,
Chapman & Hall, New York, 113-150, 1993.
5.21
-------
TROPOSPHERIC OZONE PROCESSES
Andreae, M.O., Climatic effects of changing atmospher-
ic aerosol levels, in World Survey of Climatology,
Vol. 16: Future Climates of the World, edited by A.
Henderson-Sellers, Elsevier, Amsterdam, in press,
1994.
Andreae, M.O., E.V. Browell, M. Garstang, G.L. Grego-
ry, R.C. Harriss, G.F. Hill, D.J. Jacob, M.C.
Pereira, G.W. Sachse, A.W. Setzer, P.L. Silva Dias,
R.W. Talbot, A.L. Torres, and S.C. Wofsy, Bio-
mass-buming emissions and associated haze
layers over Amazonia,/. Geophys. Res., 93,1509-
1527,1988.
Andreae, M.O., A. Chapuis, B. Cros, J. Fontan, G. Helas,
C. Justice, Y.J. Kaufman, A. Minga, and D. Ngari-
ga, Ozone and Aitken nuclei over equatorial
Africa: Airborne observations during DECAFE
88,/. Geophys. Res., 97, 6137-6148, 1992.
Andreae, M.O., B.E. Anderson, D.R. Blake, J.D. Bracl-
shaw, J.E. Collins, G.L. Gregory, G.W. Sachse,
and M.C. Shipham, Influence of plumes from bio-
mass burning on atmospheric chemistry over the
equatorial Atlantic during CITE-3, /. Geophys.
Res., 99, 12793-12809, 1994a.
Andreae, M.O., J. Fishman, M. Garstang, J.G. Goldam-
mer, C.O. Justice, J.S. Levine, RJ. Scholes, B.J.
Stocks, A.M. Thompson, B. van Wilgen, and the
STARE/TRACE-A/SAFARI Science Team, Bio-
mass burning in the global environment: First
results from the IGAC/BIBEX field campaign
STARE/TRACE-A/SAFARI-92, in Global Atmo-
spheric-Biospheric Chemistry: The First IGAC
Scientific Conference, edited by R.G. Prinn, Ple-
num, New York, in press, 1994b.
Appenzeller, C., and H.C. Davies, Structure of strato-
spheric intrusions into the troposphere, Nature,
358, 570-572, 1992.
Atkinson, R., Gas-phase tropospheric chemistry of or-
ganic compounds: A review, Atmos. Environ., 24A,
1-41, 1990.
Atkinson, R., S.M. Aschmann, W.P.L. Carter, A.M. Win-
er, and J.N. Pitts, Jr., Alkyl nitrate formation from
the N0x-air photooxidations of C2-Cg n-alkanes,
/. Phys. Chem., 86,4563-4569, 1982.
Atkinson, R., D.L. Baulch, R.A. Cox, R.F. Hampson, Jr.,
J.A. Kerr, and J. Troe, Evaluated kinetic and pho-
tochemical data for atmospheric chemistry:
Supplement IV, Atmos. Environ., 26, 1187-1230,
1992. ;
Atlas, E.L, Evidence for >Cs alkyl nitrates in rural and
remote atmospheres, Nature, 331, 426-428, 1988.
Atlas, E., et al. Partitioning and budget of NOy species
during the Mauna Loa Observatory Experiment, J.
Geophys. Res., 97, 10449-10462, 1992. \
Ayers, G.P., S.A. Penkett, R.W. Gillett, B. Bandy, I.E.
Galbally, C.P. Meyer, C.M. Elsworth, S.TJBentley,
and B.W. Forgan, Evidence for photochemical
control of ozone concentrations in unpolluted ma-
rine air, Nature, 360, 446-448, 1992. :
Bakwin, P.S., et al., Measurements of NOX and NOy over
Arctic tundra, J. Geophys. Res., 97, 16545-16557,
1992. i
Barrie, L.A., J.W. Bottenheim, R.C. Schnell, P.J.
Crutzen, and R.A. Rasmussen, Ozone destruction
and photochemical reactions at polar sunrise in the
lower Arctic atmosphere, Nature, 334, 138-140,
1988. !
Beekmann, M., G. Ancellet, and G. Megie, Climatology
of tropospheric ozone in southern Europe and its
relation to potential vorticity, J. Geophys: Res., 99,
12841-12854, 1994. [
Bottenheim, J.W., et al., Depletion of lower tropospheric
ozone during Arctic spring: The Polar Sunrise
Experiment 1988, /. Geophys. Res., 95, 18555-
18568, 1990. ;
Brost, R.A., J. Feichter, and M. Heimann, Three-dimen-
sional simulation of 7Be in a global climate model,
J. Geophys. Res., 96, 22423-22445, 1991.
Browell, E.V., C.F. Butler, S.A. Kooi, M.A. Fenn, R.C.
Harris, and G.L. Gregory, Large-scale variability
of ozone and aerosols in the summertime Arctic
and sub-Arctic troposphere, J. Geophys. Res., 97,
16433-16450, 1992. j
Calvert, J.G., A. Lazrus, G.L. Kok, B.G. Heikes, J.G.
Walega, J. Lind, and C.A. Cantrell, : Chemical
mechanisms of acid generation in the troposphere,
Nature, 317, 27-35, 1985.
5.22
-------
TROPOSPHERIC OZONE PROCESSES
Cantrell, C.A., et al., Peroxy radicals as measured in
ROSE and estimated from photostationary state
deviations, J. Geophys. Res., 98, 18355-18367,
1993.
Carroll, M.A., and A.M. Thompson, NOX in the non-
urban troposphere, in Current Problems and
Progress in Atmospheric Chemistry, edited by J.
Barker, World Pub. Company, to be published,
1994.
Carroll, M.A., D.D. Montzka, G. Hubler, K.K. Kelly,
and G.L. Gregory, In situ measurements of NOX in
the Airborne Arctic Stratospheric Expedition,
Geophys. Res. Lett., 17, 493-496, 1990a.
Carroll, M.A., etai, Aircraft measurements of NOX over
the eastern Pacific and continental United States
and implications for ozone production, / Geo-
phys. Res., 95, 10205-10233, 1990b.
Carter, W.P.L., and R. Atkinson, Alkyl nitrates formation
from the atmospheric photooxidation of alkanes:
A revised estimation method, / Atm. Chem., 8,
165-173, 1989.
Carter, W.P.L., A detailed mechanism for the gas-phase
atmospheric reactions of organic compounds, At-
mos. Environ., 24A, 481-518, 1990.
Chameides, W.L., and D.D. Davis, The free radical
chemistry of cloud droplets and its impact upon
the composition of rain, J. Geophys. Res., 87,
4863-4877, 1982.
Chameides, W.L., R.W. Lindsay, J. Richardson, and C.S.
Kiang, The role of biogenic hydrocarbons in urban
photochemical smog: Atlanta as a case study, Sci-
ence, 241, 1473-1475, 1988.
Chin, M., D.J. Jacobi J.W. Munger, D.D. Parrish, and
E.G. Doddridge, Relationship of ozone and car-
bon monoxide over North America, J. Geophys.
Res., 99, 14565-14573, 1994.
Correll, D.L., J.J. Miklas, A.H. Hines, and J.J. Schafer,
Chemical and biological trends associated with
acidic atmospheric deposition in the Rhode River
watershed and estuary, Water Air & Soil Pollut.,
35,63-86,1987.
Cox, R.A., A.E.J. Eggleton, R.G. Derwent, J.E. Love-
lock, and D.E. Pack, Long-range transport of
photochemical ozone in Northwestern Europe,
Nature, 255, 118-121, 1975.
Cros, B., R. Delmas, D. Nganga, B. Clairac, and J. Fon-
tan, Seasonal trends of ozone in equatorial Africa:
Experimental evidence of photochemical forma-
tion,/ Geophys. Res., 93, 8355-8366, 1988.
Cros, B., D. Nganga, A. i Minga, J. Fishman, and V.
Brackett, Distribution of tropospheric ozone at
Brazzaville, Congo, determined from ozonesonde
measurements, J. Geophys. Res., 97, 12869-
12875, 1992. '
Crosley, D.R., Issues in (he measurement of reactive
nitrogen compounds :in the atmosphere, SRI Inter-
national Report MP94-035, March 1994.
Crutzen, P.J., The role of NO and NO2 in the chemistry
of the troposphere and stratosphere, Ann. Rev.
Earth Planet. Sci., 7,'443-472, 1979.
Danielsen, E.F., and V.A. Mohnen, Project Dustorm re-
port: Ozone transport, in situ measurements and
meteorological analyses of tropopause folding, J.
Geophys. Res., 82, 5867-5877, 1977.
Davis, D.D., et al., A photostationary state analysis of
the NO2-NO system' based on airborne observa-
tions from the subtropical/tropical North and
South Atlantic, / Geophys. Res., 98, 23501-
23523, 1993. . j
Delany, A.C., P. Haagensen, S. Walters, A.F. Wartburg,
and P.J. Crutzen, : Photochemically produced
ozone in the emission from large-scale tropical
vegetation fires, / Geophys. Res., 90, 2425-2429,
1985. j
Dentener, F.J., and P.J. Ciutzen, Reaction of NaOs on
tropospheric aerosols: Impact on global distribu-
tions of NOX, 03 and Ohio, J. Geophys. Res., 98,
7149-7163, 1993. j
Dentener, F.J., J. Lelieveld, and P.J. Crutzen, Heteroge-
neous reactions in clouds: Consequences for the
global budget of 03, in Proceedings CEC/EU-
ROTRAC Symposium in Varese, Italy, 18-22 Oct.,
1993. ; .
Derwent, R.G., P.O. Simmonds, and W.J. Collins, Ozone
and carbon monoxide measurements at a remote
maritime location, [Mace Head, Ireland, from
1990-1992, Atmos. Environ., 28, 2623-2637, 1994.
i
Dickerson, R.R., Reactive; nitrogen compounds in the
Arctic,/ Geophys. Res., 90, 10739-10743, 1985.
Dickerson, R.R., et al.. Thunderstorms: An important
mechanism in the transport of pollutants, Science,
235,460-465, 1987.' .
5.2J
-------
TROPOSPHERIC OZONE PROCESSES
Doddridge, B.G., et al., Trace gas concentrations and
meteorology in rural Virginia 2. Reactive nitrogen
compounds, 7. Geophys. Res., 97, 20631-20646,
1992.
Dom, H.-R, J. Callies, U. Platt, and D.H. Ehhalt, Mea-
surement of tropospheric OH concentrations by
laser long-path absorption spectroscopy, Tellus,
40B, 437-445, 1988.
Drummond, J., D.H. Ehhalt, and A. Volz, Measurements
of nitric oxide between 0-12 km altitude and 67°N
and 60°S latitude obtained during STRATOZ III,
J. Geophys. Res., 93, 15831-15849, 1988.
Ehhalt, D.H., H.R Dom, and D. Poppe, The chemistry of
the hydroxyl radical in the troposphere, Proc. Roy.
Soc. Edinburgh, 97B, 17-34, 1991.
Ehhalt, D.H., F. Rohrer, and A. Wanner, Sources and dis-
tribution of NOX in the upper troposphere at
northern mid-latitudes, 7. Geophys. Res., 97,
3725-3738, 1992.
Eisele. F.L., G.H. Mount, F.C. Fehsenfeld, J. Harder, E.
Marovich, D.D. Parrish, J. Roberts, M. Trainer,
and D. Tanner, Intercomparison of tropospheric
OH and ancillary trace gas measurements'at Fritz
Peak Observatory, Colorado, J. Geophys. Res., 99,
18605-18626, 1994.
Fahey, D.W., et al., Reactive nitrogen species in the tro-
posphere: Measurements of NO, NOj, HNOs,
paniculate nitrate, peroxyacetyl nitrate (PAN), 03,
and total reactive odd nitrogen (NOy) at Niwot
Ridge, Colorado, J. Geophys. Res., 91, 9781-
9793, 1986.
Fahey, D.W., S. Solomon, S.R. Kawa, M. Loewenstein,
J.R. Podolske, S.E. Strahan, and K.R. Chan, A di-
agnostic for denitrification in the winter polar
stratosphere, Nature, 345, 698-702, 1990.
Fan, S.-M., and D.J. Jacob, Surface ozone depletion in
Arctic spring sustained by aerosol bromine reac-
tions, Nature, 359, 522-524, 1992.
Felton, C.C., J.C. Sheppard, and M.J. Campbell, Mea-
surements of the diurnal OH cycle by a 14C-tracer
method, Nature, 335, 53-55, 1988.
Finnlayson-Pitts, B.J., M.J. Ezell, and J.N. Pitts, Jr., For-
mation of chemically active chlorine compounds
by reactions of atmospheric NaCl particles with
gaseous N2O5 and C1ONO2, Nature, 337, 241-
244, 1989.
Fishman, J., and W. Seiler, Correlative nature !of ozone
and carbon monoxide in the troposphere: Implica-
tions for the tropospheric ozone budget, 7.
Geophys. Res., 88, 3662-3670, 1983. j
Fishman, J., P.M. Vukovich, and E.V. Browell, The pho-
tochemistry of synoptic-scale ozone synthesis:
Implications for the global tropospheric ozone
budget, 7. Atmos. Chem., 3, 299-320, 1985.
Fishman, J., C.E. Watson, J.C. Larsen, and J.A. Logan,
Distribution of tropospheric ozone determined
from satellite data, 7. Geophys. Res., 95, 3599-
3617, 1990. '
Fishman, J., K. Fakhruzzaman, B. Cros, and D. Nganga,
Identification of widespread pollution in the
southern hemisphere deduced from satellite analy-
ses, Science, 252, 1693-1696, 1991. :
Fishman, J., V.G. Brackett, and K. Falchruzzaman, Dis-
tribution of tropospheric ozone in the tropics from
satellite and ozonesonde measurements, J. Atmos.
Terrestrial Phys., 54, 589-597, 1992. ;
Flocke, F., Messungen von Alkylnitraten (Cl-C8) am
Schauinsland im Schwarzwald. Bin Beitrag zur
Bilanzierung der Photochemischen Ozonproduk-
tion, Inaugural-Dissertation zur Erlangung des
Doktorgrades des Fachbereichs Naturwissen-
schaften II der Bergischen Universitat-
Gesamthochschule Wuppertal, 1992. :
Flocke, F, A. Volz-Thomas, and D. Kley, Measurements
of alkyl nitrates in rural and polluted air masses,
Atmos. Environ., 25A(9), 1951-1960, 1991.
Flocke, F., A. Volz-Thomas, and D. Kley, The use of
alkyl nitrate measurements for the characteriza-
tion of the ozone balance in southwestern
Germany, EOS 7$ (43), 180, 1993.
Follows, M.J., On the cross-tropopause exchange of air,
7. Atmos. Sci., 49, 879-82, 1992.
Gidel, L.T., and M.A. Shapiro, General circulation mod-
el estimates of the net vertical flux of ozone in the
lower stratosphere and the implications fpr the tro-
pospheric ozone budget, 7. Geophys. Res., 85,
4049-4058, 1980. :
Greenberg, J.P., et al, Diurnal variability of; methane,
nonmethane hydrocarbons, and carbon monoxide
at Mauna Loa, 7. Geophys. Res., 97, 10395, 1992.
5.24
-------
TROPOSPHERIC OZONE PROCESSES
Gregory, G.L., B.E. Anderson, L.S. Warren' E.V:
Browell, D.R. Bagwell, and C.H. Hudgins, tropo-
spheric ozone and aerosol observations: The
Alaskan Arctic, J. Geophys. Res., 97, 16451-
16471, 1992.
Gruzdev, A.N., and S.A. Sitnov, Tropospheric ozone
annual variation and possible troposphere-strato-
sphere coupling in the Arctic and Antarctic as
derived from ozone soundings at Resolute and
Amundsen-Scott stations, Tellus, 45B, 89-98,
1993.
Guicherit, R., and H. van Dop, Photochemical produc-
tion in western Europe (1971-1975) and its
relation to meteorology, Atmos. Environ., 11,
145-156, 1977.
Guicherit, R., Ozone on an urban and regional scale -
with special reference to the situation in the Neth-
erlands, in Tropospheric Ozone—Regional and
Global Scale Interactions, edited by I.S.A. Itsaks-
en, NATO AS! Series C. Vol. 227, D. Reidel Publ.,
Dordrecht, The Nederlands, 49-62, 1988.
Haagen-Smit, A.J., Chemistry and physiology of Los
Angeles smog, Indust. Eng. Chem., 44, 1342-
1346, 1952.
Hameed, S., J.P. Pinto, and R.W. Stewart, Sensitivity of
the predicted CO-OH-CH4 perturbation to tropo-
spheric NOX concentrations, /. Geophys. Resi, 84,
763-768, 1979.
Harris, G.W., D. Klemp, T. Zenker, and J.P. Burrows,
Tunable diode laser measurements of trace gases
during the 1988 Polarstern cruise and intercom-
parisons with other methods, J. Atmos. Chem., 15,
315-326, 1992.
Haynes, P.H., CJ. Marks, ME. Mclntyre, T.G. Shep-
herd, and K.P. Shine, On the "downward control"
of extratropical diabatic circulations by eddy-
induced mean zonal forces, J. Atmos. Sci., 48,651-
78, 1991.
Heikes, B., Formaldehyde and hydroperoxides at Mauna
Loa Observatory, J. Geophys. Res., 97, 18001-
18013, 1992.
Hess, G.D., F. Carnovale, M.E. Cooper, and G.M.
Johnson, The evaluation of some photochemical
smog reaction mechanisms—I. Temperature and
initial composition effects, Atmos. Environ., 26,
625-641, 1992a.
Hess, G.D;, F. Carnovale, M.E. Cooper, and G.M.
Johnson, The evaluation of some photochemical
smog reaction mechanisms—-II. Initial addition of
alkanes and alkenes, Atmos. Environ., 26, 643-
651, 1992b.
Hess, G.D., F. Carnovale, M.E. Cooper, and G.M.
Johnson, The evaluation of some photochemical
smog reaction mechanisms—HI. Dilution and
emissions effects, Atmos. Environ., 26, 653-659,
1992c. ;
Holton, J.R., On the global exchange of mass between
the stratosphere and the troposphere, Man. Wea.
Rev., 47, 392-4, 1990. i
Honrath, R.E., and D.A. Jaffe, The seasonal cycle of ni-
trogen oxides in the Arctic troposphere at Barrow,
Alaska, / Geophys. Res\ 97, 20615-20630, 1992.
Hov, 0., Changes in tropospheric ozone: A simple model
experiment, in Ozone in the Atmosphere, Proceed-
ings of the Quadrennial Ozone Symposium 1988
and Tropospheric Ozone Workshop, Gottingen,
edited by R.D. Bojkov and P. Fabian, A. Deepak
Publ., Hampton, Virginia!, 557-560, 1989.
Hiibler, G., et al.. Airborne measurements of total reac-
tive odd nitrogen (NOy), J. Geophys. Res., 97,
9833-9850, 1992a. j
Hiibler, G., et al., Total reactive oxidized nitrogen (NOy)
in the remote Pacific trojposphere and its correla-
tion with O3 and CO: Mauna Loa Observatory
Photochemistry Experiment 1988, J. Geophys.
Res., 97, 10427-10477, 1992b.
Huebert, B.J.', et ai, Measurements of the nitric acid to
NOX ratio in the troposphere, /. Geophys. Res., 95,
10193-10198, 1990. j
Jacob, D.J., S.-M. Fan, S.C. Wofsy, P.A. Spiro, P.S. Bak-
win, J. Ritter, E.V. Browell, G.L. Gregory, D.R.
Fitzjarrald, and K.E. Moore, Deposition of ozone
to tundra, J. Geophys. Res., 97, 16473-16479,
1992. |
Jacob, D., J.A. Logan, G.M. Gardener, R.M. Yevich,
CM. Spivakovsky, and S!C. Wofsy, Factors regu-
lating ozone over the United States and its export
to the global atmosphere, /. Geophys. Res., 98,
14817-14826, 1993. :
Jobson, B.T., et al., Measurements of Ca-Q hydrocar-
bons during the Polar Sunrise 92 experiment, J.
Geophys. Res., in press, 1994.
5.25
-------
TROPOSPHERIC OZONE PROCESSES
Johnson, I.E., and I.S.A. Isaksen, Tropospheric ozone
chemistry: The impact of cloud chemistry, J. At-
mos. Chem., 16, 99-122, 1993.
Kirchhoff, V.W.J.H.,A.W. Setzer, and M.C. Pereira, Bio- s
mass burning in Amazonia: Seasonal effects on
atmospheric Os and CO, Geophys. Res. Lett., 16,
469-472, 1989.
KJrchhoff, V.WJ.H., and E.V.A. Marinho, Layer en-
hancements of tropospheric ozone in regions of
biomass burning, Atmos. Environ., 28, 69-74,
1994.
Kirchhoff, V.W.J.H., Z. Nakamura, E.V.A. Marinho, and
M.M. Mariano, Excess ozone production in Ama.-
zonia from large scale burnings, /. Atmos.
Terrestrial Phys., 54,583-588, 1992.
Kley, D., H. Geiss, and V.A. Mohnen, Concentrations
and trends of tropospheric ozone and precursor
emissions in the USA and Europe, in The Chemis-
try of the Atmosphere: Its Impact on Global
Change, edited by J.G. Calvert, Blackwell Sci.
Publ., Oxford, 245-259, 1994.
Kley, D., A. Volz, and F. Miilheims, Ozone measure-
ments in historic perspective, in Tropospheric
Ozone—Regional and Global Scale Interactions,
edited by I.S.A. Isaksen, NATO ASI Series C. Vol.
227, D. Reidel Publ., Dordrecht, The Netherlands,
63-72, 1988.
Lamarque, J.-R, and P.G. Hess, Cross-tropopause mass
exchange and potential vorticity budget in a simu-
lated tropopause folding, J. Atmos. Sci., 51,
2246-2269, 1994.
Lcighton, P.A., Photochemistry of Air Pollution, Aca-
demic Press, San Diego, California, 1961.
Lelieveld, J., and PJ. Crutzen, Influences of cloud pho-
tochemical processes, on tropospheric ozone,
Nature, 343,227-233, 1990.
Lelieveld, J., and P.J. Crutzen, Role of deep cloud con-
vection in the ozone budget of the troposphere,
Science, 264, 1759-1761, 1994.
Levy, H.B., J.D. Mahlman, WJ. Moxim, and S. Liu,
Tropospheric ozone: The role of transport, /. Geo-
phys. Res., 90, 3753-3771, 1985.
Lin, X., M. Trainer, and S.C. Liu, On the nonlinearity of
the tropospheric ozone production, J. Geophys.
Res., 93, 15879-15888, 1988.
Liu, S.C., M. McFarland, D. Kley, O. Zafiriou, and B.
Huebert, Tropospheric NOX and 03 budgets in the
equatorial Pacific, J. Geophys. Res., 88, 1349-
1368,1983. ;
Liu, S.C., M. Trainer, F.C. Fehsenfeld, D.D. Parrish, E.J.
Williams, D.W. Fancy, G. Hiibler, and P.C. Mur-
phy, Ozone production in the rural troposphere
and the implications for regional and global ozone
distributions, /. Geophys. Res., 92, 4191-4207,
1987.
Liu, S.C., and M. Trainer, Response of tropospheric
ozone and odd hydrogen radicals to column ozone
change, /. Atmos. Chem., 6, 221-233, 1988.
Liu, S.C., et al., A study of the photochemistry and
ozone budget during the Mauna Loa Observatory
Photochemistry Experiment, J. Geophys. Res., 97,
10463-10471, 1992.
Logan, J.A., M.J. Prather, C.S. Wofsy, and M.B. McEl-
roy, Tropospheric chemistry: A global perspective,
J. Geophys. Res., 86, 7210-7254, 1981. \
Logan, J.A., Tropospheric ozone: Seasonal bfehaviour,
trends, and anthropogenic influence, /. Geophys.
Res., 90, 10463-10482, 1985. ;
Logan, J.A., Ozone in rural areas of the United States, /.
Geophys. Res., 94, 8511-8532, 1989. :
Low, P.S., P.S. Kelly, and T.D. Davies, Variations in sur-
face ozone trends over Europe, Geophys. Res.
Lett., 79,1117-1120, 1992. :
Marenco, A., J.C. Medale, and S. Prieur, Study of tropo-
spheric ozone in the tropical belt (Africa,
. America) from STRATOZ and TROPOZ cam-
paigns, Atmos. Environ., 24A, 2823-2834, 1990.
McConnell, J.C., G.S. Henderson, L. Barrie, ). Botten-
heim, H. Niki, C.H. Langford, ahd E.M.
Templeton, Photochemical bromine production
implicated in Arctic boundary-layer ozone deple-
tion, Nature, 355, 150-152, 1992. ]
Mihelcic, D., A. Volz-Thomas, H.W. Patz, D. ;Kley, and
M. Mihelcic, Numerical analysis of ESR spectra
from atmospheric samples, /. Atmos. Chem., II,
271-297,1990. ;
Mihelcic, D., and A. Volz-Thomas, The ratio,of hydro-
peroxy to organic peroxy radicals: Direct
measurements by matrix isolation/ESR spectros-
copy, EOS 74 (43), 167, 1993. •
5.26
-------
TROPOSPHEflIC OZONE PROCESSES
Mount, G.H., and F.L. Eisele, An intercomparison of
tropospheric OH measurements at Fritz Peak Ob-
servatory, Colorado, Science, 256, 1187-1190,1992.
Murphy, D.M., D.W. Fahey, M.H. Proffitt, S.C. Liu,
K.R. Chan, C.S. Eubank, S.R. Kawa, and K.K.
Kelly, Reactive nitrogen and its correlation with
ozone in the lower stratosphere and upper tropo-
sphere,/. Geophys. Res., 98, 8751-8774, 1993.
Murphy, D.M., and D.W. Fahey, An estimate of the flux
of stratospheric reactive nitrogen and ozone into
the troposphere, /. Geophys. Res., 99, 5325-5332,
1994.
NRC, Rethinking the Ozone Problem in Urban and Re-
.gional Air Pollution, National Research Council,
National Academy Press, Washington, D.C., 1991.
Oltmans, S.J., WE. Raatz, and W.D. Komhyr, On the
transfer of stratospheric ozone into the tropo-
sphere near the North Pole, /. Atmos. Chem., 9,
245-53, 1989.
Oltmans, S.J., and H. Levy, II, Seasonal cycle of surface
ozone over the western North Atlantic, Nature,
358, 392-394, 1992.
Oltmans, S.J:, and H. Levy, II, Surface ozone measure-
ments from a global network, Atmos. Env., 28,
9-24, 1994.
Parrish, D.D., M. Trainer, E.J. Williams, D.W. Fahey, G.
Hiibler, C.S. Eubank, S.C. Liu, P.C. Mmphy, D.L.
Albritton, and F.C. Fehsenfeld, Measurements of
the NOX-O3 photostationary state at Niwot Ridge,
Colorado,/. Geophys. Res., 91, 5361-5370, 1986.
Parrish, D.D., J.S. Holloway, M. Trainer, P.C. Murphy,
G.L. Forbes, and F.C. Fehsenfeld, Export of North
American ozone pollution to the North Atlantic
Ocean, Science, 259, 1436-1439, 1993.
Penkett, S.A., N.J. Blake, P. Lightman, A.R.W Marsh, P.
Anwyl, and G. Butcher, The seasonal variation of
nonmethane hydrocarbons in the free troposphere
over the North Atlantic Ocean: Possible evidence
for extensive reaction of hydrocarbons with the
nitrate radical, /. Geophys. Res., 98, 2865-2885,
1993.
Perner, D., et al., OH radicals in the lower troposphere,
Geophys. Res. Lett., 8, 466-468, 1976.
Perner, D., et al.. Measurements of tropospheric OH
concentrations: A comparison of field data with
model prediction, /. Atmos. Chem., 5, 185-216,
1987.
Perros, P.E., Large scale distribution of peroxyacetylni-
trate from aircraft' measurements during the
TROPOZ II experiment, /. Geophys. Res., 99,
8269-8280, 1994. !
Pickering, K.E., A.M. Thompson, J.R. Scala, W.-K. Tao,
and J. Simpson, Ozone production potential fol-
lowing convective ', redistribution of biomass
burning emissions, /. Atrnos. Chem., 14, 297-313,
1992a. :
Pickering, K.E., A.M. Thornpson, J.R. Scala, W.-K. Tao,
R.R. Dickerson, and J. Simpson Free tropospheric
ozone production following entrainment of urban
plumes into deep convection, /. Geophys. Res., 97,
17985-18000, 1992bl
Platt, U., D. Perner, J. SchrSder, C. Kessler, and A. Toe-
nissen, The diurnal variation of NOs, /. Geophys.
Res., 86, 11965-11970, 1981.
Platt, U., A.M. Winer, H.W,! Biermann, R. Atkinson, and
J.N. Pitts, Jr., Measurement of nitrate radical con-
centration in continental air, Environ. Sci.
Technol, 75,365-369, 1984.
Platt, U., M. Rateike, W. Junkermann, J. Rudolph, and
D.H. Ehhalt, New tropospheric OH measure-
ments,/. Geophys. Res., 93, 5159-5166, 1988.
Platt, U., and M. Hausmahn, Spectroscopic measure-
ment of the free radicals NO3, BrO, IO, and OH in
the troposphere, Res.'. Chem. Interned., 20, 557-
578, 1994. |
Plumb, R.A., and M.K.W! Ko, Interrelationships be-
tween mixing ratios;of long-lived stratospheric
constituents, /. Geophys. Res., 97, 10145-10156,
1992. !
Poppe, D., et al., Comparison of measured OH concen-
trations with model calculations, /. Geophys. Res.,
99, 16633-16642, 1994.
Poulida, O., et ai. Trace gas concentrations and meteo-
rology in rural Virginia 1. Ozone and carbon
monoxide, /. Geophys. Res., 96, 22461-22475,
1991. |
Poulida, O., Observations and Photochemistry of Reac-
tive Trace Gases in the Atmosphere, Dissertation
at the Faculty of the Graduate School of the Uni-
versity of Maryland, 1,993.
5.27
-------
TROPOSPHERIC OZONE PROCESSES
Prather, M, R.G. Derwent, D.H. Ehhalt, P. Fraser, E.
Sanhueza, and X. Zhou, Chapter 2: Other trace
gases and atmospheric chemistry, Radiative Forc-
ing of Climate Change 1994, Report to IPCC,
Intergovernmental Panel on Climate Change, in
preparation, 1994.
Price, J.D., and G. Vaughan, Statistical studies of cut-off
low systems, Annales Geophysicae, 10, 96-102,
1992.
Price, J.D., and G. Vaughan, On the potential for strato-
sphere-troposphere exchange in cut-off low
systems, Quart. J. Roy. Met. Soc., 119, 343-365,
1993.
Prospero, J.M., R. Schmitt, E. Cuesas,, D. Savoie, W.
Graustein, K. Turekian, A. Yak-Thomas, SL Qlt-
mans, H. Levy, and A. Diaz, Temporal: variability
of ozone and aerosols ia the free troposphere over
'the Eastern North Atlantic^ submitted to Nature,
1993.
Pszenny, A.A.P., W.C. Keene, D.J. Jacob, S. Fan, J.R.
Maben, M.P. Zetwo, M. Springer-Young, and J.N.
Galloway, Evidence of inorganic gases other than
hydrogen chloride in marine surface air, Geophys.
,Res. Lett., 20,699-702, 1993.
Ridley, B.A., S. Madronich, R.B. Chatfield, J.G. Walega,
R.E. Shetter, M.A. Carroll, and D.D. Montzka,
Measurements and model simulations of the
photostationary state during the Mauna Loa
Observatory Photochemistry Experiment: Impli-
cations for radical concentrations and ozone
production and loss rates, J. Geophys. Res., 97,
10375-10388, 1992.
Rosenlof, K.H., and J.R. Holton, Estimates of the strato-
spheric residual circulation using the downward
control principle, J. Geophys. Res., 98, 10465-
10480,1993.
Rudolph, J., B. Vierkom-Rudolphr and F.X. Meixner,
Large-scale distribution of peroxyacetylnitrate.
Results from-the STRATOZ III flights, /. Geo-
phys. Res., 92, 6653-6661, 1987.
Sandholm, S.T., et aL, Summertime tropospheric obser-
vations related to NxOy distributions and
partitioning over Alaska: Arctic Boundary Layer
Expedition 3A, J. Geophys. Res., 97, 16481-
16509, 1992.
Sandroni, S., D. Anfossi, and S. Viarengo, Surface ozone
levels at the end of the nineteenth century in South
America, /. Geophys. Res., 97, 2535-2540, 1992.
Savoie, D.L., J.M. Prospero, S.J. Oltmans,; W.C.
Graustein, K.K. Turekian, J.T. Merrill, and H.
Levy II, Sources of nitrate and ozone in the marine
boundary layer of the tropical North Atlantic, /.
Geophys. Res., 97, 11575-11589, 1992. ;
Scheel, H.E., E.G. Brunke, and W. Seller, Trace gas mea-
surements at the monitoring station Cape Point,
South Africa, between 1978 and 1988, /. Atmos.
Chem., 11, 197-210, 1990. |
Scheel, H.E., R. Sladkovic, and W. Seller, Ground-based
measurements of ozone and related precursors at
47°N, 11°E, in Proceedings EUROTRAC Sympo^
slum. '92, edited by P.M. Borreli' et: all,. SPB
Academic Publ., The Hague-,. The Netherlands,,
104-108, 1993. ;
Schmitt, R., B. Schreffier,. and I. Levin, Effects: of long-
range transport on atmospheric trace constituents
at the Baseline Station Tenerife (Canary Islands),
J. Atmos. Chem., 7, 335-351, 1988. ;
Schmitt, R., and L. Hanson, Ozone in the free tropo-
sphere over the North Atlantic: Production and
long-range transport, EUROTRAC Annual Report
Pan 9 (TOR), 112-118, 1993. !
Schnell, R.C., S.C. Liu, S.J. Oltmans, R.S. Stone, D.J.
Hofmann, E.G. Dutton, T. Deshler, W.T. Sturges,
J.W. Harder, S.D. Sewell, M. Trainer, ancl J.M.
Harris, Decrease of summer tropospheric; ozone
concentrations in Antarctica, Nature, 35,726-729,
1991. :
Seila, R.L./W.A. Lonneman, M.E.R. Santoyo, and J.T.
Ruiz, VOCs in Mexico City ambient air, in Pro-
ceedings of the International Symposium on
Measurement of Toxic and Related Air Pollutants,
Air and Waste Management Association,' Pitts-
burgh, Pennsylvania, 616-621,1993.
Seiler, W., and J. Fishman, The distribution of carbon
monoxide and ozone in the free troposphere, J.
Geophys. Res., 86, 7255-7265, 1981.
Shapiro, M.A., T. Hampel, and A.J. Krueger, The Arctic
tropopause fold, Man. Wea. Rev., 115, 444-454,
1987. i
5.28
-------
TROPOSPHERIC OZONE PROCESSES
Sillman, S., J. A,Logan, and S.C. Wofsy, A regional scale
model for ozone in the United States with subgrid
representation of urban and power plant plumes, /.
Geophys. Res., 95, 5731-5748, 1990.
Simmonds, P.O., Tropospheric ozone research and glo-
bal atmospheric gases experiment, Mace Head,
Ireland, EUROTRAC Annual Report Part 9 (TOR),
234-242, 1993.
Singh, H.B., et al., Atmospheric measurements of PAN
and other organic nitrates at high latitudes: Possi-
ble sources and sinks, /. Geophys. Res., 97,
16511-16522,1992.
Singh, H.B., etal.. Summertime distribution of PAN and
other reactive nitrogen species in the northern high
latitude atmosphere of eastern Canada, J. Geo-
phys. Res., 99, 1821-1836, 1994.
Smit, H.G.J., D. Kley, S. McKeen, A. Volz, and S. Gilge,
The latitudinal and vertical distribution of tropo-
spheric ozone over the Atlantic Ocean in the
Southern and Northern Hemispheres, in Ozone in
the Atmosphere, Proceedings of the Quadrennial
Ozone Symposium 1988 and Tropospheric Ozone
Workshop, Gottingen, edited by R.D. Bojkov and
P. Fabian, A. Deepak PubL, Hampton, Virginia,
419-422, 1989.
Smit, H.G.J., D. Kley, H. Loup, and W. Strater, Distribu-
tion of ozone and water vapor obtained from
soundings over Jiilich: Transport versus chemistry,
in Proceedings EUROTRAC Symposium '92, edit-
ed by P. Borrell et al., SBP Academic PubL, The
Hague, The Netherlands, 145-148, 1993.
Smit, H.G.J., and D. Kley, Vertical distribution! of tropo-
spheric ozone and its correlation with water
vapour over the equatorial Pacific Ocean between
160E and 160W, EOS 74 (43), 115, 1993.
Staehelin, J., J. Thudium, R. Buehler, A. Volz-Thomas,
and W. Graber, Trends in surface ozone concentra-
tions at Arosa (Switzerland), Atmos. Environ., 28,
75-87, 1994.
Tang, X.Y., Air quality measurements in the People's
Republic of China, MILOX (Mid-Latitude Eco-
systems and Photochemical Oxidants) Committee
Meeting, Atlanta, Georgia, September, 1993.
Thompson, A.M., and RJ. Cicerone, Possible perturba-
tions to atmospheric CO, CH4, and OH, J.
Geophys. Res., 91, 10853-10864, 1986.
Thompson, A.M., New ozone hole phenomenon, Nature
352, 282-283, 1991.
Thompson, A. M., et al., Ozone observations and a mod-
el of the marine boundary layer photochemistry
during SAGA 3, J. Geophys. Res., 98 16955-
16968,1993. !
Thompson, A.M., K.E. Pickering, R.R. Dickerson, W.G.
Ellis, Jr., D.J. Jacob, J.R. Scala, W.-K. Tao, D.P.
McNamara, and J. Simpson, Convective transport
over the central United States and its role in re-
gional CO and ozone budgets, /. Geophys. Res.,
99, 18703-18711, 1994.
Thorncroft, C.D., B.J. Hosikins, andM.E. Mclntyre, Two
paradigms of baroclinic-wave life-cycle behav-
iour, Quart. J. Roy. Met. Soc., 119, 17-55, 1993.
Trainer, M., E.T. Williams, D.D. Parrish, M.P. Buhr, E.J.
Allwine, H.H. Westberg, EC. Fehsenfeld, and S.C.
Liu, Models and observations of the impact of nat-
ural hydrocarbons in rural ozone, Nature, 329,
705-707, 1987. i
Trainer, M., M.P. Buhr, CJ.M. Curran, EC. Fehsenfeld,
E.Y. Hsie, S.C. Liu^ R.B. Norton, D.D. Parrish,
EJ: Williams, B.WJ Gandrud, B.A. Ridley, J.D.
Shelter, E.J. Allwine, and H.H. Westberg, Obser-
vations and modeling of the reactive nitrogen
photochemistry at a1 rural site, /. Geophys. Res.,
96, 3045-3063, 1991}.
Trainer, M., et al.. Correlation of ozone with NOy in
photochemically aged air, /. Geophys. Res., 98,
2917-2925, 1993. ;
Volz; A., and D. Kley, Evaluation of the Montsouris se-
ries of ozone measurements made in the
nineteenth century, Nature, 332, 240, 1988.
Volz, A., et at., Ozone projduction in the Black Forest:
Direct measurements of ROi, NOX, and other
relevant parameters,! in Tropospheric Ozone-
Regional and GlobaV'Scate Interactions, edited by
I.S.A. Isaksen, NATO ASI Series C. Vol. 227, D.
Reidel PubL, Dordrecht, The Netherlands, 293-
302, 1988.
Volz-Thomas, A., Trends in photo-oxidant concentra-
tions, in Proceedings EUROTRAC Symposium
'92, edited by P.M. Borrell et al, SPB Academic
PubL, The Hague, The Netherlands, 59-64, 1993.
5.29
-------
TROPOSPHERIC OZONE PROCESSES
Vote-Thomas, A., et al., Photo-oxidants and precursors
at Schauinsland, Black Forest, in Proc. EU-
ROTRAC Symposium '92, edited by P.M. Borrell,
et al., SPB Academic Publishing, The Hague, The
Netherlands, 98-103,1993a.
Volz-Thomas, A., et al., Photo-oxidants and precursors
at Schauinsland, Black Forest: Results from con-
tinuous measurements of VOC and organic ni-
trates, EUROTRAC Annual Report Part 9,
International Scientific Secretariate, Garmisch-
Partenkirchen, 147-156,1993b.
Volz-Thomas, A., et al., Concurrent measurements of
carbon monoxide, ozone, NOy, PAN, and aerosols
at Izana, Tenerife, EOS 74 (43), 145, 1993c.
Vukovich, F.M., et al. On the relationship between high
ozone in the rural boundary layer and high pres-
.sure systems, Atmos. Environ., 11,967-983, 1977.
Vukovich, F.M., J. Fishman, and E.V. Browell, The res-
ervoir of ozone in the boundary layer of the eastern
United States and rts potential impact on the global
tropospheric ozone budget, J. Geophys. Res., 90,
5687-5698, 1985.
Wahner, A., F. Rohrer, D.H. Ehhalt, B. Ridley, and E.
Atlas, Global measurements of photocfajemically
active compounds, in Proceedings of 1st IGAC
Scientific Conference: Global Atmospheric-Bio-
Spheric Chemistry, Eilat, Israel, 205-222, 1994.
Wang, W.C., Y.L. Yung, A.A. Lacis, T. Moand, and J,E.
Hansen, Greenhouse effects due to man-made per-
turbations of trace gases, Science, 194, 685-690,
1976.
Wakamatsu, S., I. Uno, H. Ueda, K. Uehara, and H.
Tateishi, Observational study of stratospheric
ozone intrusions into the lower troposphere,
Atmos. Environ., 23, 1815-1826, 1989.
Wayne, R.P., I. Barnes, P. Biggs, J.P. Burrows, C.E.
Canosa-Mas, J. Hjorth, G. LeBras, G.K. Moortgat,
D. Perner, G. Poulet, G. Restelli, and H.
Sidebot-
tom, The nitrate radical: Physics, chemistry, and
the atmosphere 1990, Atmos. Environ., 25A, l-
203, 1991.
Weinheimer, A.J., et al., Meridional distributions of
NOx, NOy, and other species in the lower strato-
sphere and upper troposphere during AASE II,
accepted for publication in Geophys. Res. Lett.,
1994.
WMO, Atmospheric Ozone 1985, World Meteorological
Organization Global Ozone Research and Moni-
toring Project - Report No. 16, Geneva, 1986.
WMO, Scientific Assessment of Ozone Depletion: 1991,
World Meteorological Organization Global Ozone
Research and Monitoring Project - Report No. 25,
Geneva, 1992.
Wofsy, S.C., G.W. Sachse, G.L. Gregory, D.R. Blake,
J.D. Bradshaw, S.T. Sandholm, H.B. Singh, J.A.
Barrick, R.C. Harriss,. R.W. Talbot, M.A.
Shipham, E.V. Browell, D.J. Jacob, and J.A. Lo-
gan, Atmospheric chemistry in the Arctic and
Subarctic: Influence of natural fires, industrial
emissions, and stratospheric inputs, /.; Geophys.
Res.r 97, 16731-16746, 1992. ;
Xiuli, Z.r L. Chao, D. Guoan, L. Qijun, T. lie; T. Wang,
and M.O. Rodgers, Studies of ozone and its pre-
cursor variations at Linan, China during the
PEM-WESTA experiment, accepted for publica-
tion inJ. Geophys. Res., 1994.
Zetsch, C., and W. Behnke, Photocatalysis of tropo-
spheric chemistry by sea spray, EUROTRAC
Annual Report Part 6 (GCEMALIPP)^ 121-126,
1992. |
5.30
-------
PARTS
MODEL SIMULATIONS OF GLOBAL bzoNE
Chapter 6
Model Simulations of Stratospheric Ozone
Chapter 7
Model Simulations of Global Tropospheric Ozone
-------
-------
CHAPTER 6
Model Simulations of Stratospheric Ozone
Author:
M.K.W. Ko
Co-authors:
A. Ibrahim
I. Isaksen
C. Jackman
F. Lefevre
M. Prather
P. Rasch
R. Toumi
G. Visconti
Contributors:
S. Bekki
G. Brasseur
C. Briihl
P. Connell
D. Considine
P.J. Crutzen
E. Fleming
J. Gross
L. Hunt
D. Kinnison
S. Palermi
Th. Peter
G. Pitari
K. Sage
T. Sasaki
X.Tie
D. Weisenstein
D.J. Wuebbles
-------
-------
CHAPTER 6
MODEL SIMULATIONS OF STRATOSPHERIC OZONE
Contents !
SUMMARY i
o.l
6.1 INTRODUCTION '. , ,
u.j
6.2 COMPONENTS IN A MODEL SIMULATION 6 4
6.2.1 Source Gases and Radical Species fi
-------
-------
STRATOSPHERIC MODELS
SUMMARY
Model Simulations of Stratospheric Ozone
Multi-dimensional models are designed to provide simulations of the large-scale transport in the stratosphere.
This transport rate is combined with the local chemical production and removal
distribution of ozone as a function of longitude, latitude, height, and season.
i
There is strong observational evidence that heterogeneous chemistry (hydrolysis! of N2O5 and C1ONO2) is oper-
ating on surfaces of the aerosol particles in the stratospheric sulfate layer. There is a general agreement on how
this should be represented in the models. Models that include these reactions produce calculated ozone decreases
(between 1980 and 1990) that are larger and in better agreement with the observed trend than those produced by
models that include only gas-phase reactions. All model simulations reported here include these two reactions.
Both three-dimensional and two-dimensional models have been used in simulating polar stratospheric cloud
(PSC) chemistry in the vortex and how the equatorward transport of chemically perturbed polar air may affect
ozone at midlatitudes. Our lack of understanding of the detailed mechanisms for denitrification, dehydration, and
transport processes reduces our confidence in these model predictions. i
No multi-year simulation has been performed to date using three-dimensional models. Two-dimensional (lati-
tude-altitude) models remain the primary tools for extensive diagnostic studies and multi-year simulations.
How well do models simulate the distributions and trends of ozone in the; stratosphere?
UPPER STRATOSPHERE j
I
The model-simulated ozone concentration in the upper stratosphere is typically 20% smaller than the observed
values, a problem that has been identified previously. This suggests that there is a problem with our understand-
ing of the photochemistry in that region. I
The model-calculated ozone trends above 25 km due to emission of halocarbons between 1980 and 1990 are in
reasonable agreement with the trends (both in the altitude profile and latitudinal variation) derived from the
satellite measurements. Most of the model results did not consider radiative feedback and temperature trends that
are likely to reduce the predicted ozone decreases by about a factor of 0.8. i
LOWER STRATOSPHERE J
The models underestimate the amount of ozone in the lower stratosphere at high latitudes during winter and
spring. This, coupled with the model-calculated behaviors of other trace gases, indicates that the models do not
have a good representation of the transport processes in those seasons. !
i
I
The partitioning of the radical species in the lower stratosphere is influenced to a| large extent by the hydrolysis
rates of N2O5 and C1ONO2. This, in turn, affects the calculated ozone response in the lower stratosphere to
increases in chlorine and bromine. The trend in the polar region is also affected by
-------
STRATOSPHERIC MODELS
COLUMN ABUNDANCE ;
• The model-simulated ozone columns in the tropics are within 10% of the observed values. However, some
models underestimate the spring maximum in the Northern Hemisphere by as much as 30%.
• The models calculate a trend in the tropics of about -1% per decade in the column abundance of ozone due to
emissions of halocarbons between 1980 and 1990. This is consistent with the trend derived from the Dobson
stations, the Solar Backscatter Ultraviolet (SBUV) instrument, and the Total Ozone Mapping Spectrometer
(TOMS). The model-calculated trend in the tropics is largely a result of the calculated ozone decrease above 25
km. . ;
• The decreases in column ozone at high latitudes calculated by models that include hydrolysis of NyOs and
ClONOa as the only heterogeneous reactions are between 2% to 3% per decade. This is smaller than the observed
negative trends of 4%-8% per decade at the northern high latitudes, and 8%-14% at southern high latitudes out-
side the vortex.
• Models with PSC chemistry calculate a trend at high latitudes comparable to observation. However, the trend at
midlatitudes is still small compared to the observed decrease of 4%-6% during winter and spring (Northern
.Hemisphere), and winter, and summer (Southern Hemisphere).
• A larger trend can. be obtained at midlatitudes by including the effects from export of chemically perturbed air
from the polar region, by adjusting the transport, or by invoking additional chemical ozone removal'cycles. The
importance of the processes has not been resolved because of the lack of laboratory and- field, data:.
f
• The increase in aerosol loading' due to the eruption of Mt. Pinatubo was predicted to perturb the lower strato-
sphere. An idealized simulation was designed to isolate the effect of the photochemical response to a; uniform
thirty-fold increase in aerosol loading starting in June that decays with a time constant of 1 year. The model-
calculated decreases range from 2% to 8% around 50°N in the spring after the prescribed increase, ;with the
calculated decrease diminishing to zero over a five-year period.
Mode! Predictions of Future Trends
• Using an emission scenario that is designed to represent global compliance with the international agreements, the
calculated chlorine loading in the stratosphere reaches its maximum value about 3-5 years after the prescribed
tropospheric organic chlorine concentration achieves its maximum value. The maximum calculated chlorine and
bromine concentrations and the lowest ozone values occur within 2 years of each other in this scenario.
• ' 'Comparison of the model results indicates that although there are significant differences among the model-calcu-
lated local photochemical rates and transport rates, the rates from each individual model combine to produce
reasonable present-day ozone distributions and the 1980 to 1990 ozone trend. However, as the atmosphere is
perturbed farther away from its present state (e.g., large increase in aerosol loading, changes due to long-term
trends of N2O, CKi, and halocarbons), the model-predicted responses differ by larger amounts. Current efforts
aimed at direct validation of the transport process and photochemical process will help to resolve the differences
and bolster our confidence in the model predictions.
6.2
-------
SI
6.1 INTRODUCTION
Ozone concentrations in the atmosphere are main-
tained by the balance between photochemical production
(mainly the photolysis of O2) and removal by photo-
chemical reactions associated with the hydrogen (HOX),
nitrogen (NOX), chlorine (C1OX), and bromine (BrOx)
radicals. However, this balance is not always local be-
cause an ozone molecule created at one location can be
transported to another location before it is photochemi-
cally destroyed. Ozone concentration in the lower
stratosphere is maintained by a balance among the fol-
lowing processes: production in the tropics, transport to
mid- and high latitudes, photochemical removal in the
mid- and high latitudes, and removal from the strato-
sphere by stratosphere/troposphere (strat/trop) exchange.
The magnitude of each term changes with seasons and
their combined value determines the seasonal behavior
of ozone. In the tropical upper stratosphere (above 30
km between 30°N and 30°S), the photochemical reac-
tions are sufficiently fast that local balance hold's and the
local ozone concentration is determined by the local pro-
duction and removal rates. However, transport still
affects ozone indirectly by modulating the concentra-
tions of the radical species.
The role of the radical species in the removal of
ozone has been confirmed by process studies using in
situ observations. The concentrations of the radical spe-
cies are maintained by photodegradation of the
corresponding source gases: H2O and methane (CKt)
for HOX, nitrous oxide (N2O) for NOX, and halogen
source gases for C1OX and BrOx. The large-scale circu-
lation that transports ozone is also responsible for the
redistribution of source gases, radicals, and other trace
gases that can affect the partitioning of the radical spe-
cies. Increases in radical concentrations (e.g., increases
in C1OX due to chlorofluorocarbons (CFCs) emitted at
the Earth's surface, and increases in NOX due to N2O
emitted at the ground and stratospheric injection of NOX
by aircraft) lead to changes in ozone. ]
In this chapter, we discuss modeling of the season-
al behavior of ozone in the stratosphere using
multi-dimensional models. The amount of ozone in the
atmosphere may be separated into three layers according
to the processes controlling the concentrations: 1000 mb
(ground) to 100 mb (16 km) in the tropics and 200 mb
(11 km) in the extra tropics; from the first layer to 10 mb
RATOSPHERIC MODELS
(30 km); and 10 mb to I nib (45 km) (see, e.g., Jackman
et ai, 1989). Ninety percent of the ozone resides in the
upper two layers, with more than two-thirds in the mid-
dle layer. Although the models include a simple version
of the troposphere, representation of many of the pro-
cesses is incomplete (see Chapter 5 and Chapter 7 in this
report for discussions of ozone in the troposphere). In
die upper layer, where ozone is controlled by local pro-
duction and removal, the1 ozone concentration can be
simulated by box models if the concentrations of the rad-
ical species and overlying ozone column are known.
The middle layer has received the most attention for sev-
eral reasons. It is where the aerosol layer and the polar
stratospheric clouds reside. The observations from the
various aircraft campaigns and satellite observations
(see Chapters 3 and 4, this report) have provided a
wealth of data for studying this middle layer.
Because of limitations in computer resources, it is
not practical to use three-dimensional models to perform
multi-year simulations to study the response of strato-
spheric ozone to perturbations of the source gases and
the radical species. These calculations have been done
using two-dimensional (latitude-altitude) zonal-mean
models. They incorporate processes that have been
proven to be important. The same models are used to
compute the atmospheric lifetimes of various trace gases
(see Kaye et al., 1994) and; the ozone depletion potential
indices for the halocarbons (see Chapter 13). While
questions can be raised regarding some aspects of the
formulation and representation of the processes in two-
dimensional (2-D) models, model results from
individual models that appeared in the literature are
found to be in reasonable! agreement with the present-
day atmosphere (within 20% of the observed ozone
column away from the pol;ir region).
This chapter reviews the recent improvements in
model formulation and discusses the strengths and
weaknesses of these models. An open letter was sent to
modeling groups to solicit; results for a number of pre-
scribed calculations. Different models have reported the
results of their calculations in the scientific literature.
More often than not, the results are not in agreement
with each other. The purpose of the prescribed calcula-
tions is to ask each model to do the same calculations
with the same input so that the model results can be com-
pared. For this reason, the criterion for choosing the
prescribed conditions is that they can be easily imple-
6.3
-------
STRATOSPHERIC MODELS
Table 6-1. Models providing results in this chapter.
Model Name
AER
CAMBRIDGE
GSFC
ITALY
T.T.NT.
MPIC
MRI
NCAR
OSLO
Institution
Atmospheric and Environmental Research
Inc.. USA
University of Cambridge, United Kingdom
NASA Goddaid Space Flight Center, USA
Universita deeli Studi L'Aquila, Italy
Lawrence Livermore Laboratory, USA
Max Planck Institute for Chemistry,
Germany
Meteorological Research Institute, Japan
National Center for Atmospheric
Research, USA
University of Oslo, Norway
Investigators
M. Ko and D. Weisenstein
i
S. Bekki •
C. Jackman. D. Considine, E. Fleming
G. Pitari. S. Palermi, G. Visconti ;
D. Kinnison. P. Connell
C. Briihl, J. Gross, P.J. Crutzen, Th. Peter
T. Sasaki
G. Brasseur, X. Tie ]
I. Isaksen i
mented, rather than being faithful to what actually oc-
curs in the atmosphere. For these calculations, it is more
meaningful for the model results to be compared with
each other rather than with observations. Clearly, com-
parison with observation still remains as the only real
test on the reliability of model results.
Modeling groups that submitted results are listed
in Table 6-1. They are all 2-D models. Most of these
models (with the exception of the CAMBRIDGE model)
have participated in one or more of the intercomparison
exercises, the latest of which took place in 1991 and
1992 (see'Prattler and Remsberg, 1993). This intercom-
parison involved 14 different groups from 6 countries.
The intercomparison was comprehensive and included:
1) source, radical, and reservoir gases important in
ozone photochemistry; 2) radioactive tracers 14C and
90Sr and the Mt. Ruiz volcanic cloud, which tested the
models* transport; and 3) a detailed model intercompari-
son of photodissociation rates, transport fluxes, and
idealized tracers that highlighted some of the models'
similarities and differences. One result of these exer-
cises was to help eliminate simple coding errors in the .
models and give more confidence that the range of pre-
dictions is due to differences in formulations and
approaches. The remaining differences will ultimately
have to be resolved by comparison with observations.
The results from these calculations will show that there
are substantial differences among the model predictions,
particularly when perturbations are large. Unfortunate-
ly, the schedule of this report does not allow enough time
to resolve all the issues. It is hoped that this will be done
soon. !
6.2 COMPONENTS IN A MODEL SIMULATION
This section discusses how the models simulate
the distributions of the source gases, the partitioning of
the radical species, and the distribution of ozOne in the
stratosphere. To simulate the distribution of ozone, the
models calculate the local production and removal rates
for ozone, and combine them with the effect of transport
to determine the ozone concentration as a function of
longitude, latitude, altitude, and season. The local pro-
duction and removal rates depend on the model-Jcomputed
distributions of the source gases and the radical species,
and the partitioning of the radicals (which in turn de-
pends on the local temperature and solar insolation).
One thing to note is that the photochemical removal rate
for ozone in most of the lower stratosphere is about 10%
per month in summer and 1 % per month in winter. Thus,
it is always necessary to consider the effect of transport
and the combined cumulative effect over several years to
assess the ozone response. This is to be contrasted with
situations where activation of the chlorine radicals in the
polar vortex leads to a rapid ozone removal rate of 1%
per day.
6.4
-------
STRATOSPHERIC MODELS
6.2.1 Source Gases and Radical Species
The models simulate the following processes in
the life cycle of a source gas released in the troposphere:
the cycling of the source gas between the troposphere
and the stratosphere via strat/trop exchange; the photo-
chemical reactions that release the radical species; the
subsequent redistribution of the radical species by the
large-scale transport; and the partitioning of the radical
species into the active and reservoir species. A molecule
in the stratosphere can either be photochemically re-
moved, or it will spend, on average, about three years
before it is transported back to the troposphere (see e.g.,
Holton, 1990). The three-year residence time corre-
sponds to the average for all material in the stratosphere.
Clearly, material introduced to the stratosphere near the
tropopause will have a much shorter residence time. The
exchange between the troposphere and stratosphere is
simulated in the models in terms of the large-scale ad-
vection and eddy transport. This is probably adequate
for source gases such as N2O and the CFCs, and the rad-
ical families Cly, Bry, and NOy, whose distributions are
relatively uniform. A more sophisticated treatment is
needed for cases involving direct injection of radical
species, such as injection of chlorine radicals by the
space shuttle solid rocket engine and injection of NOX by
high-flying aircraft. ',
6.2.1.1 HALOGEN SPECIES
The odd chlorine (Cly) and bromine (Bry) species
in the stratosphere come from degradation of the source
gases. Among the source gases that have been measured
in the atmosphere, the atmospheric burdens of methyl
chloride (CH3C1), methyl bromide (CH3Br), and other
bromomethanes are thought to be maintained, in part, by
natural' sources. Other man-made sources include the
chlorofluorocarbons (CFCs), the hydrochlorofluorocar-
bons (HCFCs), the bromomethanes (mainly methyl
bromide), and the halons in the stratosphere. Photo-
degradation of the CFCs takes place almost exclusively
in the stratosphere. The hydrogenated halogen species
can be broken down by photochemical reactions in both
the troposphere and stratosphere. The Cly and Bry spe-
cies released in the troposphere will be washed out
relatively quickly and will not be transported to the
stratosphere. Thus, source gases that react in the tropo-
sphere will deliver less of their chlorine or bromine to
the stratosphere. The radical species released in the
stratosphere are redistributed in the stratosphere and
eventually removed from the stratosphere by the large-
scale transport that parameterizes strat/trop exchange in
the models. While in the stratosphere, they will be parti-
tioned into the active species (Cl, CIO, C12O2, BrO) and
the reservoir species (HC1, C1ONO2, HOC1; HBr,
BrONO2, HOBr). The active species participate directly
in the ozone removal cycles! Observed concentrations of
the reservoir species provide an important check for the
model results. •,
Model calculations have been used to simulate the
distribution of the chlorine Radicals released by specific
source gases in the present-day stratosphere (see
Weisenstein et al., 1992). 'this can be used to estimate
the individual contribution! of a specific source gas to
chlorine loading and ozone depletion. A similar break-
down can also be obtained using observed concentrations
of the source gases in the lower stratosphere (Kawa et
al, 1992; Woodbridge etaL, 1994).
Two other sources for chlorine radicals were dis-
cussed in Chapter 2. These are deposition of chlorine by
solid-fuel rockets and injection of HC1 into the strato-
sphere by violent volcanic eruptions. These sources are
not included in the model simulations. The estimated
input of 0.7 kiloton (Cl)/yr from solid-fuel rockets is
small compared to the annual input of 300 kiloton (Cl)/
yr from the current inventory of organic halocarbons in
the atmosphere (Prather et al., 1990a). Theoretical cal-
culations discussed in Chapter 3 show that HC1 will be
scavenged in the volcanic plume (Tabazadeh and Turco,
1993). This, together with the lack of observed increase
in HC1 after eruptions (Wallace and Livingston, 1992;
Mankin etaL, 1992), supports the conclusion that volca-
nic eruptions contribute little to stratospheric chlorine.
6.2.1.2 THE ODD NITROGEN SPECIES
The odd nitrogen species are introduced into the
stratosphere by several sources. The major natural
source of NOy is from the reaction of N2O with O('D),
producing two NO molecules (Crutzen, 1970; McElroy
and McConnell, 1971). This is why changes in concen-
tration of N2O will affect • the concentration of NOy
radicals and ozone. Reaction of N2O with excited O2
molecules has been suggested as a possible source
(Toumi, 1993) but cannot be quantified because of lack
of rate data. Other suggested continuous natural sources
6.5
-------
STRATOSPHERIC MODELS
Table 6-2. Comparison of sources and sinks for odd nitrogen species in the stratosphere.
SOURCES
Nitmns oxide oxidation N?O + OOP) -»2NO
Transport of NOT produced by lightning in the troposphere
Galactic cosmic rays (solar minimum)
(solar maximum)
Solar proton events (1972, solar maximum)
(1975, solar minimum)
Input from mesosphere and thermosphere, relativistic electron
precipitations, meteors
Nuclear explosions (1961 & 1962 nuclear tests)
Stratospheric-flying aircraft
Rocket launches
SINKS
Reforming of molecular nitrogen N + NO -» N? + O
Rainout of HNO 3 transported to the troposphere
Magnitude ;
fkiloton(N)/vr) '
600 ;
250
86
63 '
35 •
0.01 !
7
550 !
depends on emission index, fleet
size, and flight paths ;
7
i
195
750
of stratospheric NOy that have a regional impact are
galactic cosmic rays for. the polar lower stratosphere
(Wameck, 1972; Nicolet, 1975; Legrand et al, 1989),
lightning for the lower equatorial stratosphere (Noxon,
1976; Tuck, 1976; Liu et al., 1983; Ko et al, 1986; Kot-
amarthi et al., 1994), and the downward flux of odd
nitrogen from the thermosphere (Strobel, 1971; McCon-
nell and McElroy, 1973) especially in the polar region
during winter (Solomon et al., 1982; Garcia et al., 1984;
Russell et al., 1984). Sporadic natural sources of strato-
spheric NOX include meteors (Park and Menees, 1978),
solar proton events (Zadorozhny et al., 1992; Jackman,
1993), and precipitation by relativistic electrons (Callis
etal, 1991). The frequency and magnitude of these spo-
radic sources are not well quantified. Most models
include the lightning source in addition to N2O oxida-
tion, but ignore other sources. Mankind also influences '
stratospheric NOX production through atmospheric nu-
clear explosions (Johnston et al., 1973; Foley and
Ruderman, 1973), rocket launches (Karol et al., 1992;
Chapter 10 in WMO, 1992), and high-flying aircraft
(CIAP, 1975; Albritton et al, 1993).
The odd nitrogen species introduced into the
stratosphere are redistributed by the large-scale trans-
port. They are partitioned into N, NO, NO2, NO3, N2O5,
HNO3, HNO4, C1ONO2, and BrONO2. The;active spe-
cies (NOX = NO + NO2) are important in ozone control.
Besides reacting with ozone, the NOX constituents are
also important in interference reactions with Other fami-
lies (HOX, Clx, Brx) involved in ozone regulation
through reactions widi OH (forming HNOs), with CIO
(forming C1ONO2), and with BrO (forming BrONO2).
Photochemical removal occurs in the upper part of
• the stratosphere via the reaction of N with NO forming
N2. The rest of the production is balanced by transport
removal. Table 6-2 (from Jackman et al, 1980, 1990;
Prather et al, 1992) shows a comparison of the magni-
tude of some of these suggested sources and sinks of odd
nitrogen. Nitrous oxide oxidation is believed to be the
largest source, with lightning also contributing substan-
tially in the lower equatorial stratosphere. The transport
to the troposphere is thought to be the largest sink, with
the reforming of N2 also contributing significantly.
[
6.2.1.3 THE HOX SPECIES
The HOX species are produced from the reaction of
OOD) with H2O and CFLj. Reaction of excited O2 mol-
ecules with H2 has been suggested as a source (Toumi,
1993) but cannot be quantified because of'lack of rate
6.6
-------
STRATOSPHERIC MODELS
data. The reaction N2O5 + H2O -> 2HNO3 occurs on the
surfaces of the aerosol particles. After the HNO3 mole-
cules are released to the atmosphere, they can either
react with OH or be photolyzed to produce OH and NO2.
Thus, depending on the fate of the HNO3 molecules, the
reaction can be a source of OH. Model calculations
show that there is a net increase in OH when the reaction
is included (see discussion in Rodriguez et al., 1991).
Removal of the HOX species in the stratosphere is domi-
nated by reaction of OH with HO2, HNO3, HNO4, and
HC1. The species OH and HO2 participate in the ozone
removal reactions and modulate the partitioning of the
NOy, Cly, and Bry species.
The H2O concentration in the stratosphere is
maintained by oxidation of CH4 and import of H2O from
the troposphere. The exchange of H2O across the tropo-
pause is not well understood. Some models (ITALY,
LLNL, MPIC, MRI, NCAR) parameterized this by im-
posing a boundary condition along the tropopause.
Other models (AER and GSFC) keep the stratospheric
H2O concentration fixed at observed values (see section
B in Prather and Remsberg, 1993) and make adjustments
for future changes from CHt increase and from engine
emissions of stratospheric aircraft.
6.2.2 Heterogeneous Reactions and
Partitioning of the Radical Species
Studies of the Antarctic ozone hole pointed to the
importance of heterogeneous reactions in affecting
ozone in the lower stratosphere. These early modeling
studies, laboratory experiments, and field measurements
were summarized in a review paper by Solomon (1990).
Subsequent studies were reviewed in WMO (1992).
Chapters 3 and 4 presented more recent evidence that
shows that heterogeneous reactions do occur on particles
in the atmosphere at rates that are consistent with rate
constants determined in the laboratory. These reactions
are
N2O5 + H2O(on particles) -» 2HNO3 '. (6-1)
C1ONO2 + H2O(on particles) -» HOCI + HNO3 (6-2)
C1ONO2 + HCl(on particles) -> C12 + HNO3 (6-3)
N2O5 + HCl(on particles) -» HNO3 + C1NO2 (6-4)
HOCI -(- HCl(on particles) -> C12 + H2O (6-5)
In each reaction, a gas molecule (e.g., N2Os) is as-
sumed to collide with a particle and proceed to react with
another molecule (H2O or HC1) already on the particle.
As discussed in Chapter 3, these reactions occur on liq-
uid or frozen sulfate particles and on polar stratospheric
clouds (PSCs) at different rates. The effectiveness of
each reaction in altering the partitioning of the radical
species depends on how last the heterogeneous conver-
sion rate is compared to' the gas-phase reactions that
control the partitioning in specific regions of the atmo-
sphere. Because HCI is much more soluble on PSCs,
reactions (6-3) through (6-5) are more effective on PSCs
than on liquid sulfate particles. A common effect of the
first four reactions is to decrease the NOx/NOy ratio,
with the net effect of reducing the ozone destruction due
to the NOX loss cycle. At the same time, the reduction in
NOX also inhibits the foimation of C1ONO2, leaving
more of the active chlorine in the form of CIO, and in-
creases the C1OX removal of ozone. The additional
HNO3 produced in the reaction also increases OH and
the removal of ozone due to the HOX cycle.. The last
three reactions involve direct activation of chlorine spe-
cies by converting HCI to active, chlorine. The inclusion
of these reactions in the models has brought the model
results in closer agreement with observations (see Chap-
ters 3 and 4). |
The information on the reaction rate constants in-
dicates that reaction (6-1) has the dominant effect at
normal stratospheric temperatures at midlatitudes (see
discussion in Hanson et al., 1994). Reaction (6-1) re-
duces the efficiency of the NOX cycle, while both the
HOX and C1OX cycles are enhanced. As a result, the HOX
cycle is the dominant ozone removal cycle in the lower
stratosphere. This has beep confirmed using direct ob-
servations of OH and HO2 in the lower stratosphere
(Wennberg et al., 1994). The net effect on the local re-
moval rate of ozone is small for normal aerosol loading,
so that the present-day ozone abundances calculated
with and without heterogeneous chemistry are within
10% of each other (Rodriguez et al., 1991; Weisensiein
etai, 1991, 1993; McElroy;e/a/. 1992). However, these
same reactions make the model-calculated ozone more
sensitive to increases in chlorine and less sensitive to
added nitrogen-containing radicals.
Reaction (6-2) has a more noticeable impact on the
partitioning of the radical species for temperatures less
than 200 K and/or unde^ enhanced aerosol loading
6.7
-------
STRATOSPHERIC MODELS
(Hanson et al, 1994). Indirect evidence for this reaction
was reported in Solomon et al. (1993) and Sanders et al.
(1993), who detected enhanced OC1O in the Antarctic,
consistent with C1ONO2 hydrolysis on ML Pinatubo
aerosols before the onset of PSCs. The effect of reac-
tions (6-3) through (6-5) appears to be limited to ice
surfaces, but could be important on sulfate particles at
high latitudes under very cold temperatures (Hanson et
al, 1994). We discuss below how models attempt to
simulate the effects of these reactions.
6.2.2.1 HETEROGENEOUS CHEMISTRY ON THE SULFATE
LAYER
Reactions (6-1) and (6-2) proceed on liquid sulfate
aerosol particles that are present in the global sulfate lay-
er throughout the lower stratosphere. Molina et al.
(1993) reported experimental results that show that if the
temperature is below 200 K, the activation may also take
place on solid H2SO4 hydrates. The rates of the reac-
tions depend on the surface area density and the water
content of the aerosol, and possibly the phase of the
aerosol particle. In the models, a first-order reaction rate
constant is defined for each reaction as the product of the
collision frequency of the gas-phase reactant with the
aerosol particles in the sulfate layer and the sticking co-
efficient (Y). which is the reaction probability per
collision. The collision frequency depends on the sur-
face area density of the sulfate particles. The effect of
the varying water content and phase of the aerosol is pa-
rameterized in the models by defining an effective Y in
terms of the local temperature and concentration of wa-
ter vapor. However, there is observational evidence that
indicates that the phase of the aerosol particles may also
depend on the history of the particles, and not just on
local conditions. A final assumption made in the models
is that the products of the reaction are released to the
atmosphere. Thus, there is no sequestering of the reac-
tion products.
Most model studies have assumed a Y value of Q. 1
for reaction (6-1). Recent results reported by Fried et al.
(1994) indicate that Y for (6-1) may vary between 0.077
to 0.15 at 230 K for H2SO4 weight percent between 64%
to 81%. The extrapolated rate in the atmosphere based
on their semi-empirical model ranges from 0.03 to 0.15.
Other studies (Tolbert et al, 1993; Fried et al, 1994)
discussed whether uptake of formaldehyde may change
the composition of the aerosol and affect the rvalues.
The effect of such variation for reaction (6-1) has not
been explored.
Hanson et al. (1994) recommended the following
expression for Y for reaction (6-2) :
where W is the weight percent of acid, defined as
T(0.6246Z- 14.458) + 3565
W= T(-0.19988) + 1.3204Z + 44.777
with Z = In (partial pressure H2O (mb)), T is the temper-
ature in K. Hanson et al. (1994) also provided parameters
for reactions (6-3) and (6-5). They concluded from their
model calculation that the reactions should be included
in simulating the ozone behavior at high latitude winter
under enhanced aerosol conditions. The calculations in
this chapter include reactions (6-1) and (6-2) as the only
heterogeneous reactions on sulfate particles.
There are additional problems specific to simulat-
ing the effects of these reactions in a 2-D zonal-mean
model. The model results presented in this chapter use a
prescribed zonal-mean aerosol surface density specified
as a function of altitude and latitude (Chapter 8, WMO,
1992) derived from the Stratospheric Aerosol and Gas
Experiment (SAGE) observations. If we assume that the
surface area density is constant in the zonal direction, a
constant value for Y in (6- 1 ) would mean that the conver-
sion rate can be represented reasonably well as a
zonal-mean rate. On the other hand, the parameteriza-
tion for reaction (6-2) depends on local temperature and
partial pressure of H2O. As the dependence on these
zonally varying quantities become more nonlinear, sim-
ulating the effect of the conversion as a zpnal-mean
quantity becomes more problematic (Murphy and Ravi-
shankara, 1994; Considine et al, 1994). The effects of
longitudinal temperature fluctuation on the zbnal-mean
rate of reaction (6-2) has been studied by Pitari (1993a).
The conversion rate experienced by an air parpel follow-
ing the actual trajectory in the polar vortex was found to
be as much as a factor of 10 larger than the rate calculat-
ed using the zonal-mean temperature.
6.2.2.2 HETEROGENEOUS CHEMISTRY ON PSCs
Modeling the effects of polar stratospheric clouds
(PSCs) involves two steps, the model must simulate the
removal of H2O and HNO3 vapor when the particles are
6.8
-------
STRATOSPHER/C MODELS
formed, and the effects of heterogeneous conversions
that occur on the surfaces. Inside the polar .vortex,, the
conversion rates due to PSC chemistry are so fastthat the
amount converted is limited by the availability of the re-
actants (N2O5, C1ONO2, HOC1, HC1) once the particles
are formed. As a result, the calculated repartitioning de-
pends less on the details of how the reactions are
parameterized.
Early 2-D model studies of the effects of polar het-
erogeneous processes parameterized the heterogeneous
reactions as first-order conversion rates for the gas-phase
reactants triggered by location and season (Chipperfield
and Pyle, 1988; Isaksen et al, 1990) or by the zonal-
mean temperature falling below a threshold value
(Granier and Brasseur, 1992). In the latter case, the
threshold zonal mean temperature was picked to give a
reasonable PSC frequency of occurrence. Denitrifica-
tion was included in Isaksen et al (1990) by ad hoc
removal of 50% of the HNO3 in the PSC regions. Grani-
er and Brasseur (1992) included denitrification and
dehydration for Type II PSCs by introducing a first-order
removal rate for H2O and HNO3 with a time constant of
5 days when the zonal-mean temperature falls below the
threshold value. Denitrification was included for Type I
PSCs using a first-order removal rate for HNO3 with a
time constant of 30 days. To obtain the surface area den-
sity, a log-normal size distribution was assumed. In the
3-D model studies of Chipperfield et al (1993) and
Lefevre et al (1994), the amounts of H2O, and HNO3
condensed to form Type I and Type II PSCs were calcu-
lated assuming thermodynamic equilibrium using the
local model temperature, H2O, and HNO3 concentra-
tions. The surface area densities were calculated
assuming that the particles have radii of 1 urn and 10 |j.m
for Type I and Type II PSCs, respectively. Sedimenta-
tion was included for Type II particles in the transport of
the condensed material. Pitari et al. (1993) developed a
code in their 2-D model in which PSC occurrence and
surface area were calculated rather than prescribed.
They used a tracer continuity equation for condensed
material with a production term that included terms pa-
rameterizing condensation, coagulation, sedimentation,
and rainout. Different treatments for the uptake of HC1
were used in the models. Pitari et al. (1993) ignored the
uptake of HC1. Chipperfield et al. (1993) and Lefevre et
al. (1994) assumed that HC1 is incorporated in the PSCs
using the mole fractions given by Hanson and Mauers-
berger (1988). In all cases, it is assumed that the reaction
rate can be represented by ah effective Y.
Modeling such processes on PSCs in 2-D models
presents special challenges. First, the motions of air-par-
cels are typically not zonally symmetric. The effectiveness
of reactions (6-3) through (6-5) depends on the availabil-
ity of sunlight to photolyze C12 and C1ONO2 to form Cl
and CIO. It is not clear whether a full air-trajectory cal-
culation is needed to take into account the solar
insolation experienced by the air parcel, or whether the
situation can be approximated by an average exposure to
PSCs over several trips around the globe. The problems
will likely be most severe at the beginning and end of the
polar winters, especially in j the Arctic, which experi-
ences large temperature fluctuations and azonal motions
throughout the winter. The Southern Hemisphere vortex
in the depth of the winter is more uniformly cold and
zonally symmetric. Secondly, it is not clear that using
the zonal-mean temperature alone can capture the com-
plexity of the different temperatures experienced by an
air parcel. Peter et al. (1991) developed a way to use
climatological temperature statistics to derive probabili-
ties for PSC formation as & function of latitude and
altitude for both Type I and Type II PSCs. This formed
the basis of methods that other studies used to predict
surface area densities without relying solely on zonal
mean temperatures (Pitari et al., 1993; Grooss et al,
1994;Considinee?a/., 1994).|
I
6.2.3 Transport and Ozone
If ozone is calculated assuming local photochemi-
cal equilibrium (i.e., local production balanced by local
removal), the calculated column abundance will have its
maximum value of 700 Dobson units (DU) in the tropics,
decreasing to about 200 DU |in the summer high lati-
tudes. The observed behavior'of the column abundance
of ozone (minimum at the tropics and maximum at high
latitude) is a good indication that transport plays an im-
portant role in redistributing ozone from the production
region in the tropics to high latitudes.
Transport of trace gases in three-dimensional
chemistry-transport models (CTMs) is based on either
three-dimensional winds from| general circulation mod-
els (GCMs) or data-assimilated winds derived from
observations. Because of the limitation in computation-
al resources, it is not yet practical for 3-D CTMs to
6.9
-------
STRATOSPHERIC MODELS
predict evolution of chemical species over time periods
much longer than a few years. The same limitation also
precludes incorporating full chemistry into a GCM to
calculate ozone and winds interactively. Thus, the GCM
winds are calculated using prescribed ozone based on
observation.
In 2-D models, transport is represented by advec-
tion from the zonal-mean velocities and eddy mixing
coefficients. Most models used prescribed velocity and
eddy coefficients with seasonal variations. The same
circulation and temperature are used year after year to
simulate the climatological mean state. However, previ-
ous studies (Tung and Yang, 1988; Schneider et al,
1991; Jackman et al, 1991; Yang et al., 1991) have
shown that the observed interannual variations in tem-
perature would induce corresponding variations in the
transport circulation leading to changes in ozone of
about 3% to 4%. Variations in the circulation caji also
come from the quasi-biennial oscillation (QBO) in the
equatorial winds. Gray and Pyle (1989) and Gray and
Dunkerton (1990) produced a QBO in ozone in their 2-D
model with interactive dynamics by parameterizing the
QBO in the equatorial winds through specification of
damping of waves. In Gray and Ruth (1993), a QBO in
the equatorial winds was introduced into the model by
relaxing the model winds toward the monthly mean ob-
served winds. The calculated ozone QBO showed
anomalies of ±6 Dobson units (±3%) at the tropics and
±12 DU (±2%) at midlatitudes. The broad patterns were
shown to be in agreement with the anomalies derived
from TOMS (Lait et al., 1989), although the amplitude
was larger in the model.
6.2.3.1 RELATION TO OBSERVATION
Most applications of 3-D CTMs are formulated as
initial value problems where the concentrations of the
trace gases are first initialized from observations, and the
models are then used to simulate the evolution of the
trace gases (typically for a season) for comparison with
observations. Granier and Brasseur (1991) used a. mech-
anistic 3-D model with rather detailed chemistry to
investigate the mechanisms responsible for ozone deple-
tion over the Antarctic and the Arctic. Kaye et al.
(1991), Douglass et al. (1991), and Rood et al. (1991)
used a simple parameterized chemistry to assess the im-
portance of chemical processing in polar regions during
the winters of 1979 and 1989. The transport of chemical
tracers in those studies was driven by winds from the
STRATAN assimilated system (Rood et al, 1989).
Chipperfieldera/. (1993) and Lefevre ef a*. (1994) sim-
ulated the behavior of chemical constituents in the Arctic
lower stratosphere during the winters of 19,89-1990, and
1991-1992, respectively. These models used analyzed
winds and temperature from the European Centre for
Medium-Range Weather Forecasts (ECMWF). The sim-
ulations reproduce successfully the activation of
atmospheric chlorine in polar regions and predict the de-
pletion of ozone in PSC-processed air. While the
simulations cannot be used to predict the long-term be-
havior of the trace gases, they provide the opportunity to
diagnose observations and to quantify the different pro-
cesses that have led to the observed ozone depletion.
Chipperfield et al. (1993), for example, quantified the
respective contribution of the different catalytic cycles
responsible for the destruction of ozone!in the Arctic
lower stratosphere during the 1989-1990 winter.
In 2-D models, the relation to observation is less
straightforward. In models that use the residual mean
formulation, the velocity and eddy mixing coefficient
can be related to observed quantities as follows. The
vertical velocity is related to the ratio of the local diabat-
ic heating rate and the lapse rate. Comparison of the
vertical velocity in the model with the diabatic heating
rate calculated from observed ozone and temperature
(Rosenfield et al, 1987) and the lapse rate provides a
reference point (see Prather and Remsberg, 1993). The
values of the eddy diffusion coefficient Kyy can be com-
pared . with values derived using mixing rates for
potential vorticity (Newman et al, 1988); The interac-
tion between the vertical velocity and the eddy mixing
determines the shapes of the surfaces of constant mixing
ratios in the lower stratosphere. Measured concentra-
tions of different trace gases indicate that'they share the
same mixing surfaces in the lower stratosphere when the
local photochemical time constant is longer than the
transport time constant. This sharing of the mixing ratio
surfaces is evident in that an x-y plot of the mixing ratios
of two long-lived trace gases shows a compact curve
(Plumb and Ko, 1992). This feature is present in both
observations and model results (see section H in Prather
and Remsberg, 1993). The mixing ratio surfaces in a
model defined by the advection velocity and the eddy
diffusion coefficient help to determine 'the latitudinal
6.10
-------
STRATOSPHERIC MODELS
gradient of the model-calculated column abundance of
ozone.
Finally, the simulated distributions of the long-
lived trace gases from 2-D models can be compared to
observations. Simulations of source gases N2O and CRU
were reasonable when compared to SAMS (Stratospher-
ic and Mesospheric Sounder), ATMOS (Atmospheric
Trace Molecule Spectroscopy), and balloon measure-
ments at mid- to high latitudes between 20 and 30 km;
however, the variabilities near the winter poles were
more difficult to simulate (see Prather and Remsberg,
1993). It was noted in Prather and Remsberg (1993) that
direct comparison of model results for the sources gases
or transient tracers (such as the radioisotopes '^C and
^Sr from nuclear weapons tests) with observation is dif-
ficult because the transport can vary significantly from
year to year, with the quasi-biennial oscillation leading
to two distinctly separate modes of stratospheric circula-
tion. The transport as formulated in the 2-D models can,
at best, represent the averaged transport on a seasonal
time scale and does not provide any specific information
on the transport of the trace gases on shorter time scales.
Such information has to come from 3-D CTMs using
three-dimensional winds from a data assimilation proce-
dure or similar analysis using observations. Analysis of
such results should provide the information necessary to '
assess the appropriateness of the transport parameteriza-
tion in the 2-D models.
6.23.2 TRANSPORT BETWEEN THE POLAR VORTICES AND
MlDLATITUDES
The representation of either a closed or a leaky
vortex is a major challenge for models. This is particu-
larly problematic for 2-D models, where the inherent
dependence on diffusion coefficients does not allow fora
completely satisfactory representation of either process.
Previous attempts by 2-D models (Sze et al., 1989; Chip-
perfield and Pyle, 1988) to simulate the effect of export
of ozone-poor air from the breakdown of the Antarctic
vortex suggest that the dilution process could have a
large effect on the ozone behavior year-round in the
southern midlatitudes. The results of Sze et al. (1989)
showed that for an imposed ozone hole with 50% reduc-
tion in the column, the calculated ozone column at 30°S
and in the tropics decreased year-round by 3% and 0.5%,
respectively. In contrast, the results from Chipperfield
and Pyle (1988) showed a decrease of less than 0.5%
northward of 40°S. Prather et al. (1990b) used the God-
dard Institute for Space Studies (GISS) 3-D CTM to
assess the magnitude of the dispersion of ozone-depleted
air over several months following the breakdown of the
Antarctic polar vortex and obtained a 2% decrease in to-
tal ozone year-round at 30°S.
Prather and Jaffe (1990) used a 3-D CTM to look
at the effects of the export of chemically perturbed air.
Toumi et al. (1993) suggested that polar-processed air
reaching midlatitudeii is expected to contain large
amounts of C1ONO2 amd may also play a part in affect-
ing the ozone trend. Gariolle et al. (1990) used the 3-D
general circulation model of Meteo-France (Emeraude)
to examine the evolution of the Antarctic polar vortex.
They found ozone reduction (about 2%) at midlatitudes
in September well before the vortex breakdown. More
recently, Mahlman et al. (1994) used the Geophysical
Fluid Dynamics Laboratory (GFDL) SKYHI GCM to
show that, with the 25% depletion in total ozone calcu-
lated over Antarctica during the spring season, the ozone
column abundance at die equator was reduced by 1% by
the end of a 4.5-year model experiment, and the local
ozone concentration in the lower stratosphere was re-
duced by 5%. [
The studies of Kaye et al. (1991) and Douglass et
al. (1991), in which the transport of chemical tracers was
driven by assimilated winds., concluded that the transport
of processed air in the Arctic: to midlatitudes was limited.
Lefevre et al. (1994) reported the simulation of the be-
havior of the chemical [constituents in the Arctic lower
stratosphere during the winter of 1991-1992. The model
used analyzed winds and temperature (from ECMWF)
and included a comprehensive scheme for gas-phase re-
actions, as well as a parameterization of heterogeneous
reactions occurring on the surface of nitric acid trihy-
drate (NAT) and ice particles in polar stratospheric
clouds, and heterogeneous processes on the surface of
sulfate aerosol particles. The model results showed that
the combined effects of PSC processing in the vortex,
vortex erosion, and aerosol processing at midlatitudes
led to significant ozorie reductions in the Northern
Hemisphere during January 1992. However, chemical
processes produced only a limited fraction of the ozone
deficit observed at high latitudes during a period domi-
nated by a strong blocking anticyclone over the North
Atlantic.
6.11
-------
STRATOSPHERIC MODELS
6.233 MODELS WITH INTERACTIVE DYNAMICS
The results presented in this chapter are mostly
from model simulations in which the temperature and
circulation are kept fixed. It is clear that the thermal
structure and the transport circulation will change as the
trace gas concentrations change. Changes can be due to
changes in ozone or changes in other greenhouse gases.
Decrease of ozone in the stratosphere and increases in
greenhouse gases will cause a cooling of the strato-
sphere. In addition, changes in ozone near the
tropopause and increases in greenhouse gases will cause
a warming of the troposphere. We will restrict the dis-
cussion in this section to the effect of the cooling in the
stratosphere. The effects from changes in the tropo-
sphere will be discussed in Section 6.4.3.
First-order effects on the coupling of ozone, tem-
perature, and wave feedback are relatively well
understood, and much of the relevant work is summa-
rized in earlier WMO publications. However, the
thermal structure of the atmosphere is controlled by a
delicate balance between radiative processes (which are
related to ozone) and dynamical processes. At the same
time, ozone is controlled by a delicate balance between
chemical production and destruction (which depends on
the thermal structure) and dynamical transport. Thus,
processes that appear to be of secondary importance can
act to tip the balance in perturbation studies.
Previous studies ignoring heterogeneous reactions
(Nicoli and Visconti, 1982; Schneider et al., 1993) sug-
gested that the cooling of the middle atmosphere could
be a mechanism for increasing ozone because the ozone-
removing cycles are less efficient at lower temperatures.
Thus, this temperature feedback is a negative feedback
in that the model-calculated ozone decrease will be re-
duced. However, a cooler stratosphere could lead to an
enhanced occurrence of PSCs (Peter et al, 1991; Austin
et ai, 1992) resulting in increased chlorine activation,
giving rise to the possibility of a Northern Hemisphere
ozone hole. Pitari et al. (1992) used results from a sim-
ple 3-D model to show that the ozone response to C0'2
doubling is distinctly different if PSCs are present and
heterogeneous reactions on PSCs are included. They
showed that the large stratospheric cooling caused by the
CO2 increase would induce a substantial polar ozone de-
crease despite the fact that the rates of homogeneous
catalytic cycles are reduced.
The changes in local heating will also lead to
changes in the circulation, and have an attendant effect
on,the transport of heat, momentum, and trace species.
For example, latitudinal changes in the ozone distribu-
tion (i.e., the ozone hole) can lead to substantial changes
in the persistence and strength of the polar vortex, and
thus enhance the chlorine-catalyzed ozone reduction in
polar regions. Several GCM studies examined the cou-
pling between temperature change and the ozone hole.
Kiehl et al. (1988), using the National Center for Atmo-
spheric Research (NCAR) Community Climate; Model
(CCM2), found that the introduction in the model of a
prescribed Antarctic ozone hole produced in the polar
stratosphere a cooling of approximately 5 K during the
month of October, and introduced a possible delay in the
timing of the final wanning. A similar cooling was cal-
culated by Cariolle et al. (1990) and Prather et al.
(1990b) using the Meteo-France model and the GISS
model, respectively. However, the results of Mahlman et
al. (1994) show a larger sensitivity, where a 25% reduc-
tion in ozone produces a temperature reduction of 8 K.
While attempts to implement full chemistry
schemes in GCMs are still limited by computational re-
sources, there has been important progress in including
interactive dynamics in 2-D models. Interactive models
can be separated into groups according to the treatment
of the forcing term for the zonal-momentum equation.
The first group uses externally specified momentum
fluxes (Harwood and Pyle, 1975; Vupputuri, 1978; Gar-
cia and Solomon, 1983) or calculates the fluxes from the
gradient of the zonal mean potential vorticity from exter-
nally specified Kyy (Ko et al., 1993). Feedback in these
models is limited to changes induced by changes in local
heating rates. A second group of models calculates the
forcing term explicitly from the zonal waves computed
in the model. This latter approach can, in principle, ac-
count for the effect of the interaction'between the waves
and mean circulations. Examples in these groups are the
models of Brasseur et al. (1990), Garcia et al. (1992),
Garcia and Solomon (1994), and Kinnersly and Har-
wood (1993). ;
6.3 COMPARISON OF MODEL RESULTS WITH
OBSERVATION
i
If the models are designed to simulate the behavior
of ozone, an obvious question concerns how well they
6.12
L
-------
STRATOSPHERIC MODELS
simulate the ozone behavior in the present-day atmo-
sphere and the observed changes in the past decades.
Another question is to what extent we can trust the mod-
el predictions. Ironically, one cannot answer those
questions by simply comparing the model-simulated
ozone directly with the observations. The reasons are as
follows. The winds and temperature in the global mod-
els represent climatological averaged states. It would
not be appropriate to compare the model simulations
with the observed behavior in any one particular year.
The behavior of ozone is the net effect from many com-
peting mechanisms. Thus, it is difficult to come to any
definitive conclusion about the role of any specific
mechanism by simply looking at whether the model-
simulated ozone values agree with observations. The
balance among these mechanisms in the future atmo-
sphere could be very different from that in the
present-day atmosphere. The important thing is not only
whether we have the proper balance in the present-day
atmosphere, but whether the correct physics has been in-
cluded so that we can predict with confidence how
changes in these terms will affect ozone.
Comparison of model results with observations
has to be done indirectly after further processing of the
observations and/or model-simulated results. One ex-
ample is the process study that prescribes values for
winds, temperature, and concentrations of some of the
trace gases based on observations. A more restricted
simulation is performed to calculate the remaining trace
gases. A comparison is then made for the restricted set
to test the few mechanisms that control the behavior of
those species. Examples of these are the studies that use
data-assimilated winds to isolate the short-term trans-
port, and modeling studies associated with aircraft
campaigns that test the mechanisms for the photochemi-
cal partitioning. Another example is the intercomparison
exercise in Prather and Remsberg (1993) that calculates
the relative abundance of the odd-nitrogen and chlorine
species in the altitude range of 20 to 40 km, constrained
by observed concentrations from ATMOS. Other meth-
ods have been developed specifically for ozone. In
previous studies to obtain trends in the column abun-
dance of ozone, analyses were performed to take out the
quasi-biennial oscillations and the 11-year solar cycle
effects to obtain an ozone trend that can be ascribed to
changes in trace gases (see, e.g., Bojkov 1987; Reinsel et
al., 1994; Stolarski et ai, 1992). The derived trend is
then compared to a model simulation that examines the
effect of changes in trace gases on ozone. We will dis-
cuss some of the model results in Section 6.3.2.
6.3.1 Present-Day Atmosphere
,!
In the comparisons shown below, the University of
Oslo (OSLO), NCAR, zind Max Planck Institute for
Chemistry (MPIC) modeling groups submitted results
from calculations that include chemical reactions on
PSC surfaces. The Goddard Space Flight Center
(GSFC) group submitted! two sets of results, one with
and one without polar heterogeneous processes.
6.3.1.1 OZONE m THE UPPER STRATOSPHERE
Several problems identified previously in the up-
per stratosphere have not been resolved. The Model &
Measurement Intercomparison Workshop (Prather and
Remsberg, 1993) confirmed previous findings that mod-
el-calculated O3 around 40 krn is 20% to 40% smaller
than the values derived from the Solar Backscatter Ultra-
violet (SBUV) measurement (see Figure 6-1). Recent
analysis by Eluszkiewicz land Allen (1993) indicates a
deficit of 8% to 20% even when observations are used to
constrain the concentrations of the radical species.
Previous suggestions that vibrationally excited ox-
ygen molecules may produce ozone in the upper
stratosphere (Slanger et al, 1988; Toumi et al, 1991;
Toumi, 1992) are found to be ineffective because of rap-
id quenching (Patten et al, 1994). The values for the
C1O/HC1 ratio derived from measurements (Stachnik et
al, 1992) are found to be smaller than model-calculated
values. Recent model simulations show that the effects
of assuming a branching that produces HC1 from the re-
action of CIO with OH (McElroy and Salawitch, 1989;
Natarajan and Callis, 1991) are to increase the calculated
ozone concentration at 2 mb (Chandra et al, 1993) and
to decrease the calculated decadal ozone trend at the
same altitude (Toumi and Bekki, 1993). However, the
results from Chandra et all (1993) show that even with
the branching, the calculated ozone concentration is still
20% too small in the summer months.
Although the amount, of ozone in the upper layer is
relatively small and the error may not affect the model-
calculated ozone column, jthe discrepancy may be an
indication that there is missing chemistry in the models.
There is a need to obtain simultaneous measurements of
6.13
-------
STRATOSPHERIC MODELS
O3 Annual Average Scenario 1,1990 at 44 km
i-
e
'x
-90
Q.
-2=4
CO
I3
'x
* NOAA
X NIMBUS
A OSLO
NCAR
D MRI
r MPIC
o ITALY
v LLNL
A GSFC
• CAMBRIDGE
• AER
j <—i—i—i—i—•-
-60
-30
0
Latitude
30
60
90
O3 Scenario I, 1990 at 44 km, 40 Peg. S
X NOAA
X NIMBUS
A OSLO
* NCAR
D MRI
T MPIC
o ITALY
V LLNL
A GSFC
• CAMBRIDGE
• AER
67
Month
10 11 12 13
Figure 6-1. Comparison of the model-calculated ozone concentrations,ai 44 km (|J"b) for 1990 ^h °bser
^SSSSZ^
IfunctfonTffl? The lower panel shows the calculated concentrations at 40°S for four seaspns.
6.14
-------
STRATOSPHERIC MODELS
ozone, temperature,, and radical species such as OH,
HO2, CIO, and NO2 in the upper stratosphere to help re-
solve this.
63.1.2 OZONE COLUMN
Figure 6-2a shows the calculated column abun-
dance of ozone for the 1990 condition. The model
results are within 20% of the observations away from the
polar region. The zonal-mean total ozone derived from
the Total Ozone Mapping Spectrometer (TOMS) obser-
vation indicates that the spring maximum in the
Northern Hemisphere extends all the way to the pole,
while the Southern Hemisphere shows a sub-polar max-
imum, with the largest value occurring at about 60°S.
This has been attributed to the different surface topogra-
phies in the two hemispheres inducing different
circulations, resulting in a more stable vortex that encir-
cles the pole in the Southern Hemisphere. By adjusting
the circulation and the eddy diffusion coefficients, most
models succeeded in producing these features. Hou et
al. (1991) discussed the relative roles of the circulation
and eddy diffusion coefficients in determining the result
in the Atmospheric and Environmental Research, Inc.
(AER) model. However, none of the models simulates
the isolation of the air in the vortex. Thus, it is question-
able whether the models produce the observed ozone
behavior by simulating the actual mechanisms occurring
in the atmosphere.
In Prathef and Remsberg (1993), the model-calcu-
lated ozone distributions were compared with the
average of the 1979 and 1980 observed distribution.
This was done to minimize the ozone QBO in the obser-
vation. The difference (in Dobson units) between the
model-calculated total ozone for 1980 and the averaged
observed abundance is plotted in Figure 6-2b. The cal-
culated total ozone values in most models are within 20
Dobson units (10%) of the observed value in the tropics.
The models also calculate smaller column ozone than
the observed values during the spring maxima in polar
regions, up to 100 DU (30%) smaller in some cases.
6.3.2 Ozone Trends Between 1980 and 1990
63.2.1 MECHANISMS THAT CAN AFFECT THE OZONE
TREND
The distribution of ozone can be modified in many
ways. The concentrations of the radical species can be
increased by the introduction of additional source gases.
or direct introduction of ra'dicat species, such as injection
of chlorine radicals by the space shuttle solid rocket en-
gine (WMO, 1992) and injection of NOX by high-flying
aircraft (WMO, 1992; this report). The partitioning of
the radical species can be affected by changes in temper-
ature, which affect the reaction rate constants and the
frequency of occurrence of the PSCs. Analyses of tem-
perature records (see, e.g., Spencer and Christy, 1993;
Oort and Liu, 1993) suggested a cooling trend of about
0.4 K/decade. This cooling may be a result of the in-
crease in CO2 and ozone depletion that occurred in this
period. The partitioning can also be affected by changes
in surface areas of the sulfate layer that affect the rate of
heterogeneous conversion.: Observations (see Chapter 3,
WMO [1992]) showed that the aerosol loading has been
. decreasing after the eruption of El Chichon in 1982.
Other works suggested that aircraft emission of SO2
from combustion of aviation fuel may have increased the
sulfate loading in the past decade (see Hofmann, 1991;
Bekki and Pyle, 1992).
Other mechanisms that cam affect the ozone trend
are the QBO in equatorial winds (which has a period of 2
years), the 11 -year solar cyclic, and the El Nino/Southern
Oscillation (ENSO) with a period of about 4 years.
Modeling of the ozone QBO was reviewed in Section
6.2.3. Previous studies using 2-D models (Brasseurand
Simon, 1981; Garcia etal., 1984;Callise/a/., 1985) pro-
vided quantitative estimates; for the sensitivity of ozone
to long-term variations in sblar flux at ultraviolet (UV)
wavelengths. Results from -four 2-D models containing
gas-phase chemistry only that were reported in WMO
(1990) indicate that the global ozone content is 2% larg-
er at solar maximum than at solar minimum. Results
from models with heterogeneous chemistry are available
from several recent studies. lUnfortunately, it is difficult
to compare the results because each work used different
assumptions on the variation of the solar flux. Huang
and Brasseur (1993) reported that total ozone at solar
maximum is 0.5% smaller at winter high latitudes and
0.5% larger at the tropics compared to the values at solar
minimum. Brasseur (1993) jreported that total ozone is
1% larger at the tropics and 1.5% larger at high latitudes
at solar maximum compared to solar minimum when a
3% change in solar flux between 208-265 nm is as-
sumed. Fleming et al. (1994) estimated that annual
averaged total ozone between 45°N and 45°S is about
6.15
-------
STRATOSPHERIC MODELS
AER - O3 Column 1990
Cambridge - O3 Column 199p
2 3 4 5 6 7 8 9 10 11 12 13
TIME (MONTH)
GSFC - O3 Column 1990
2 3 4 5 6 7 8 9 10 11 12 13
TIME (MONTH)
MPIC - O3 Column 1990
2 3 4 5 6 7 8 9 10 11 12 13
TIME (MONTH)
OSLO - O3 Column 1990
2 3 4 5 6 7 8 9 10 11 12 13
TIME (MONTH)
NCAR.- O3 Column 1990
2 3 4 5 6 7 8 9 10 11 12 13
TIME (MONTH).
23456789 10 11
TIME (MONTH)
ITALY - O3 Column 1990
12 13
5 -
23456789 10 11
TIME (MONTH)
MRI-O3 Column 1990 ;
12 13
< -60 -
2 3 4 5 6 7 8 9 10 11 12 13
TIME (MONTH)
LLNL- O3 Column 1990
2 3 4 5 6 7 8 9 10 11 12 13
TIME (MONTH)
Figure 6-2a. Model-simulated column abundance of ozone for 1990 conditions. The contour levels are in
steps of 20 Dobson units. ;
6.16
-------
STRATOSPHERIC MODELS
AER - TOMS
CAMBRIDGE - TOMS
90
60
30
o
-30
-60
-90
J FMAMJ JASONDJ
TIME (MONTH)
GSFC - TOMS
90
60
30
0
-30
-60
-90
90
60
30
0
-30
-60
-90
FMAMJ JASONDJ
TIME (MONTH)
MPIC - TOMS
FMAMJ JASONDJ
TIME (MONTH)
NCAR - TOMS
o
UJ
Q,
111
Q
90
60
30
0
-30
-60
-90
J F M A
MJ JASO
TIME (MONTH)
I
ITALY • TOMS
NDJ
a
at
Q,
111
Q
90
60
30
0
-30
-60
-90
J F M A
MJ JASO
TIME (MONTH)
MRI - TOMS
NDJ
O
LU
UJ
Q
90
60
30
0
-30
-60
-90
90
60
30
0
-30
-60
-90
J FMAMJ JASO
TIME (MONTH)
OSLO r TOMS
NDJ
FMAMJ JASONDJ
TIME (MONTH)
JFMAMJJASO
TIME (MONTH)
NDJ
Figure 6-2b. The differences (in Dobson units) between the model-calculated column abundance of ozone
for 1980 and the average of the 1979 and 1980 observed column from TOMS.! The contour levels are in
steps of 10 Dobson units. ;
6.77
-------
STRATOSPHERIC MODELS
6.18
-------
3.5% larger in 1985 and 1979 (solar maxima) than in
1985 (solar minimum). These values are to be compared
with the value of 1.2% derived from the statistical analy-
sis of the Dobson data on ozone and the FIQ 7 solar flux
through 1984 (Reinsel et al, 1987), and the 1-2% value
cited in Chapter 7 of WMO (1990). A review of the ef-
fects of ENSO on ozone can be found in Zerefos el al.
(1992). Analyses of the observations indicate that there
was a 2% ozone decrease in the tropics after the large
ENSO event in 1982-1983. No modeling work has been
done to simulate the suggested mechanisms to produce
the ozone response.
6.3.2.2 MODEL RESULTS
As discussed in the beginning of Section 6.3, the
effects of the 11-year solar cycle and QBO are subtract-
ed from the ozone trend using statistical techniques.
Here, we compare this remaining trend to the model-cal-
culated trend due to changes in other trace gases. The
trends in the surface concentrations of the halocarbons,
CH4, and N2O are discussed in Chapter 2 of this report.
The modelers were asked to perform a calculation in
which the changes in the surface concentrations of the
source gases are as given in Table 6-3. With the excep-
tion of the CAMBRIDGE model, all models kept the
temperature, circulation, and surface area of the sulfate
particles constant in the calculation. The CAMBRIDGE
model includes dynamics feedback in its calculation.
The model-calculated changes in ozone between
1980 and 1990 are shown in Figure 6-3a. The effect of
PSC chemistry is included in the OSLO, NCAR, and
MPIC models. The GSFC model results shown corre-
spond to the case without PSC chemistry. Note that most
models show a calculated decrease of about 1-2% in the
tropics, increasing to 4% at the high latitudes. Com-
pared to the derived trend reported in Stolarski et al.
(1992), models without PSC chemistry fail to reproduce
the following features: the over 6% decrease north of
50°N during March; the 6% decrease south of 40°S
throughout the year; and the large decrease in the Ant-
arctic polar vortex. Including PSC chemistry in the
model will help to produce some of these features. The
OSLO, NCAR and MPIC results all showed decreases of
about 7% at northern high latitudes. The calculated de-
creases for the southern high latitudes range from 7% to
9%. The GSFC model with PSC chemistry shows calcu-
lated decreases.of about 3% at northern high latitudes
SI RATOSPHERIC MODELS
and up to 9% in the south, i Figure 6-3b shows the calcu-
lated trend as a function!of latitude compared to the
derived trend between 1980 and 1990. The model re-
sults agree well with the derived trend in the tropics.
Only models with PSC chemistry calculate the large
trend at high latitudes. Around 40° latitudes, the ob-
served ozone trend is between -4% to -8% per decade in
winter, and -4% to -6% per decade in spring and fall.
These are to be compared with the model-calculated val-
ues of -2% to -3% per decade year round. Thus, the
model-calculated trends are a factor of 1.3 to 3 smaller
than the observed trends, depending on season.
Figure 6-4 shows ! the calculated percentage
change in the local concentration of ozone between 1980
and 1990. The model-calculated ozone trends for the
past decade are typically 8% to 12% between 30°N and
50°N at 40 km. These values are too large compared to
the trend derived from SAGE I and SAGE II (McCor-
mick et al., 1992) and that derived from the Umkehr data
(see WMO, 1992). However, a recent study (Hood etai,
1993) of the SBUV data indicates that the trend may be
larger and somewhat closer to the model-calculated
trends. As discussed in Section 6.3.1, a smaller trend can
be obtained if a branching !for production of HC1 is as-
sumed for the reaction of OH with CIO (Toumi and
Bekki, 1993). The model-calculated trend would also be
smaller if the feedback effects from the cooling of the
stratosphere due to the ozone decrease were included in
the models. This temperature feedback is included in the
CAMBRIDGE model only. Calculations from models
(Schneider et al., 1993) indicate that the feedback will
provide a 20% compensation in the calculated ozone de-
crease, t
Results in Figure 6-4 show that none of the models
reproduced the 5% to 10% per decade decrease in ozone
in the midlatitude lower stratosphere derived from the
SAGE data (McCormick era/., 1992). There are sugges-
tions as to how a larger decrease can be calculated in the
models. One suggestion is that the transport parameter-
izations in the models fail tcj represent how ozone at high
latitudes can affect the midtatitude region. A more real-
istic representation of the transport may give a larger
ozone decrease. Another suggestion is that there may be
missing photochemistry. Solomon et al. (1994a) showed
that if IO is assumed to react with CIO and B'rO at suffi-
ciently fast rates, the calculated ozone decrease in the
lower stratosphere will be larger.
6.19
-------
STRATOSPHERIC MODELS
Cambridge - O3 Col. % Dlff. 1j9(KI980
AER - 03 Col. % Diff. 1 990-1 9BU
LATITUDE (DEG)
? g 8 o 8 g 8
— i — i — i — i i
1
LATITUDE (DEG) LATITUDE (DEG)
& & & « o> co & g § o § § S
...... -2- " '•---.-
.• '
.--•"" --. |
2 3 4 5 6 7 8 ' 9 10 11 12 1C
TIME (MONTH)
GSFC - 03 Col. % Diff. 1990-1980
. , ,...--
-2 -'"
~l
. -• '
-.-• -2-....
<-•
90 ~
^ 60-
CD
g 30-
g 0-
| -30-
-1 -60 -
-90.
» 1
on
LATITUDE (DEG)
g 8 S o 8 § 5
2 3 4 5 6 7 8 9 10 11 12 13
TIME (MONTH)
MPIC - 03 Col. % Diff. 1990-1980
' .•'.'•.• '-. :••• .-•' .--'\--''
, . -.. v^; S-:-.: :'--;..-••'
) •
> -
D- •;;;;;. ..'.:::2.--.
o-'.v;;::':-^ • . •••-'-}••:":/.'">"• ••'"•V
n -Y-.J^ — -i- — ' '"•' '•'••"• ' •••'; '-
Li -i 1 ' '
LATITUDE (DEG)
CD C7) W S S C
^ . - - -
»•
•.-••" ' "f
2 3 4 5 6 7 8 9 10 11 12 13
TIME (MONTH)
ITALY - 03 Col. % Diff- .1 990-1 98Q
...-•'" ""•• ; -
- .-2 ' •-•:••
..-••• ^ •••-.-•
2 3 4 5 6 7 8 9 10 11 12 13
TIME (MONTH)
MRI - 03 Col. % Diff. 1 990-1 980
-•"" '•--'
-2- •- "
) - :
)-
3' ...•••• ----2-.i...
° o i A q R 7 8 9 10 11 .12 1
TIME (MONTH)
90
— 60
CD
g 30
g o
p -30
3-60
-90
IMUAri - VJO v->UI. /o LSIII. 1^"^ •—
.S^;-;:"-------''""'""'""'------:--
•
90
60
o
L" 30
I °
^ 30
• 3-60
-90
0
:^1
-
_ — i "
1 ? 3 '
23456789
TIME (MONTH)
4 5 6 7 8 9 10 11 12 13
TIME (MONTH) '-
6.20
-------
STRATOSPHERIC MODELS
Dobson Total Ozone Trend - Year Round
2
CD
TJ
S
CD
TJ
C
CD
TnMo
TOMS
i SBUV
DOBSON
MPIC
D MRl
A OSLO
-10 Lad
-90 -60
-30 0 30
sin(Latitude)
60 90
Dobson Total Ozone Trend - Dec-Jan-Feb
2
TOMS
I SBUV
I DOBSON
-90 -60
sin(Latitude)
30 60 90
Dobson Total Ozone Trend - Mar-Apr-Mav
21—' '
-90 -60
sin(Latitude)
60 90
Dobson Total Ozone Trend - Jun-Jul-Aug Dobson Total Ozone Trend - Sep-Oct-Nov
73
C
Q)
-8
-10
-12
-14
AER
CAMBRIDGE
ilF~8-psc
ITALY
MPIC
MRl
NCAR
OSLO.
I TOMS
I SBUV
I DOBSON
-90 -60 -30
sin(Latitude)
30
60 90
AER !
CAMBRIDGE
GSFC-PSC
GSFC
ITALY •
MPIC
MRl
1 TOMS
I SBUV
I DOBSON
-90 -60
-30 0
sin(Latitude)
60 90
Figure 6-3b. Model-calculated changes in column ozone between 1980 and 1990 compared to derived
trends. i
6.21
-------
STRATOSPHERIC MODELS
CAMBRIDGE - O3 % Difference Iran 1980 (Mar)
so
4]
40
!
i 30
23
20
-«
45
40
1 30
'2S
20
•S
SO
49
40
I 30
2S
20
.
SO
4S
«
I*
* x
2'
2(
i:::::V3:::::"- " ;;:::::-o\ J
:'.-"'-2'". -2--.':}i':i
/- -v •-. : :' ••...,-:
r — "\ i * ' 'v '
> -60 -30 0 30 60 80
GSFC - O3 % Difference from 1980 (Mar)
S€^;|S|g
( \ i ( •• **• '•••• ;
- \v^^ \ "•••---::;:;;;;;:
K) ' -tO -30 0 30 60 9
UltoxM
MPIC - O3 % Difference from 1980 (Mar)
*» *• •• "*O -*AfS *•*."• j
5=^::::::=s|ii^,;
g--^^§|,lt-:;
NCAR - O3 % Difference from 1980 (Mar)
• I* Q *•••. ******•••••••». ...».-•••••-••-. . 1 1 * • • •".**•"•
T ^ " ^^ — "i *.
'iii^^^r
sor
4i '-
40 -
Ą
i30;
25 -
20
15
•«
45
40
135
0
I30
25
20
15
Cl -S
45
40
fss
I 30
25
2C
15
SO
'. 4.
- 4(
; Ł 3.
5 31
2
1
o . ""•-.'
-c- • - . . r.
d—J-' •' •' "
..i'. ._.-'. ixf. i N , .'•! .'•.,, i ....
> -60 -30 0 30 60 90
ITALY - O3 % Difference from 1980 (Mar)
j|i::::EE3||S
O "• \ A 1 • " '
I \\\ /// . ::H
0 -60 -30 0 30 60 90
Ulitudo
MRI - O3 % Difference from 1980 (Mar)
'••:••;•••-.. '•• ' _.."•''."-••""•. :
(?\\(1S\\CJ
90 -60 -30 0 30 60 90
OSLO - O3 % Difference from 1980 (Mar)
>- ...-•'' ''-•-. ••'.'••'. "
.90 -60 -30 0 30 60 91
<0 -30 0 30
Latitude
60 • !»
Figure 6-4 Model-calculated local change in ozone between 1980 and 1990 for March condition. PSC
chemistry is excluded except for the OSLO, MPIC, and NCAR models. Contour levels are 2%, 1 %, 0, -1 %,
-2%, and -4%, and in steps of 2% thereafter.
6.22
-------
STRATOSPHERIC MODELS
6.3.3 Effects of the Mt. Pinatubo Eruption
Volcanic eruptions introduce large amounts of
SO2 into the stratosphere that will be oxidized to form
sulfate aerosol. Model simulations (Golombek and
Prinn, 1993; Pitari et ai, 1993) have shown that the
background stratospheric aerosol layer can be explained
in terms of the present input of SO2, and OCS (carbonyl
sulfide). However, the lack of detailed knowledge on the
microphysics of particle formation precludes a detailed
prediction on how the surface area will change.
Prior to the formation of the volcanic aerosol, SO2
chemistry can affect the photochemical removal rate of
ozone in the tropics for the initial months after the erup-
tion (Bekki et al, 1993). The increase of the aerosol
surface area available for heterogeneous processes is the
most immediate effect of aerosol changes (Hofmann and
Solomon, 1989). However, modeling studies (Miche-
langeliera/., 1989;BrasseurandGranier, 1992;Kinneef
al., 1992; Pitari, 1993b; Pitari and Rizi, 1993; Schoeberl
et al., 1993) have shown that other effects may be as im-
portant. The UV flux is increased substantially above
the aerosol layer and may decrease below; thus affecting
the photolysis rates. Another effect is related to the heat-
ing of the aerosol layer due to the absorption of solar and
terrestrial radiation. The additional heat source can
modify the dynamics and affect the reaction rates of
those catalytic cycles whose reaction rates depend on
temperature.
63.3.1 RADICAL SPECIES
The effects of the volcanic aerosol on several radical
species were reviewed in Chapter 4. The observations
and the accompanying modeling studies indicate that the
behaviors of the radical species are in qualitative agree-
ment with enhanced processing on the surface of the
volcanic aerosols. These include observation of NOX/
NOy ratios from aircraft (Fahey et al., 1993; Kawa et al.,
1993), observation of N2O5 and HNO3 from ATMOS
(Rinsland et al., 1994), measurement of CIO from air-
craft (Avallone et al., 1993; Wilson et al., 1993), and
measurement of CIO, NO, and O3 (Dessler et al., 1993)
and NO2 and HNO3 (Webster et al, 1994) from bal-
loons. In addition, there are column measurements of
C1ONO2, HC1, and HNO3 from aircraft (Toon et al.,
1993) and ground-based measurements of NO2
(Johnston et al., 1992; Mills et al., 1993; Coffey and
M:ankin, 1993; Koike et la/., 1994; Solomon et al.,
1994b) and HNO3 ( Koike et al., 1994). Solomon et al.
(1993) reported that the enhanced level of OC1O ob-
served over McMurdo Station during autumn of 1992 is
consistent with expected ef fects from the enhanced con-
version of C1ONO2 via reaction (6-2).
i
63.3.2 OZONE BEHAVIOR IN THE TROPICS IN LATE 1991
Using satellite and lidar measurements, Labitzke
and McCormick (1992) concluded that the monthly av-
eraged zonal mean 30-mb (24 km) temperatures at 20°N
in September and October! 1991 are as much as 2.5 K
warmer than the 26-year average. Warming in the equa-
torial region was measured t'o be as high as 4 K. DeFoor
et al. (1992) deduced from' lidar data that there was a
total lift of 1.8 km in the tropics 100 days after the erup-
tion. There are some disagreements on how the eruption
has affected ozone because of the difficulty in isolating
the effects of the QBO and other mechanisms that cause
interannual variations of ozone. Using the Nimbus-7
TOMS and the NOAA-11 satellite Solar Backscatter Ul-
travioIet/2 (SBUV/2) spectrometer data, Chandra (1993)
suggests that the maximum change in column ozone at-
tributed to the Mt. Pinatubo 'eruption may not be greater
than a 2-4% decrease at mid- and low latitudes a few
months after the eruption after removing the effect of the
QBO. Schoeberl et al. (1993) used a different method in
analyzing the Nimbus 7 TO'MS data and derived a de-
crease of 5-6% in column ozone between 12°Nand 12°S
between June and December 1991. Grant et al. (1992,
1994) compared the electrochemical concentration cell
(ECC) sondes data and the airborne UV Differential Ab-
sorption Lidar (DIAL) data to Stratospheric Aerosol and
Gas Experiment II (SAGE JJ) climatology and deduced a
column ozone decrease in the: tropics of 9% ± 4% in Sep-
tember, 1991. ;
Bekki et al. (1993) investigated the role of gas-
phase sulfur photochemistry ;on ozone in the first month
following the eruption. Most other studies did not in-
clude this on the assumption that its effect is short-lived.
Kinne et al. (1992), Brasseur and Granier (1992), Pitari
and Rizi (1993), Kinnison et al. (1994), and Tie et al.
(1994) investigated the coupled radiative-dynamical per-
turbation on ozone following; the eruption and provided
diagnostics to estimate the contributions from dynamics,
radiation, and heterogeneous processing. All models es-
timated a net increase in heating of about 0.3 to 0.4 K/
6.23
-------
STRATOSPHERIC MODELS
day. However, different approaches were used to deter-
mine how this extra heating is to be partitioned into
wanning or enhanced vertical motion. The studies of
Brasseur and Granier (1992), Tie et al. (1994), and Pitari
and Rizi (1993) used the dynamics equations in their re-
spective models to apportion the heating. The calculated
decrease in tropical ozone in late 1991 is 9% in Pitari and
Rizi (1993), which results from a 4% decrease from
changes in photolysis rate, a 4% decrease from increased
heterogeneous processing, and 1% decrease from tem-
perature and circulation changes. The calculated
decrease in Tie et al (1994) is 2%, which results from a
2% decrease caused by changes in photolysis rates, a 2%
decrease from changes in temperature and circulation,
and a' 2% increase from changes in heterogeneous pro-
• cessing. The studies of Kinnison etal. (1994) provided
separate estimates under the assumption that all the heat-
ing is dissipated either by local warming or by enhanced
upward motion. The calculated decrease is 2% (-1.5%
from motion and -0.5% from heterogeneous processing)
if it is assumed that the extra heating goes to enhanced
upward motion, and 1% (-0.5% from temperature
change and -0.5% from heterogeneous processing) if it is
assumed that all the heating is balanced by warming.
The work of Kinne et al. (1992) estimated an uplifting of
1.7 km after accounting for the wanning using the ob-
served temperature change'. They used a simple 1-D
mechanistic model to estimate an ozone decrease of
10%.
6.33.3 OZONE BEHAVIOR IN 1992 AND 1993
Gleason et al. (1993) reported that during 1992,
TOMS on the Nimbus-7 satellite measured global aver-
age total ozone to be 1-2% lower than expected if ozone
is assumed to be decreasing at the same linear trend in
the past decade. These results are consistent with analy-
sis of the TOMS and Meteor 3 data (Herman and Larko,
1994), which showed that the 1993 ozone amount is
12.5% below the historical mean (from 1979) at high lat-
itude, 7% at midlatitude, and 4% at low latitude. Low
ozone for the winter of 1992-1993 was also reported
from the Microwave Limb Sounder (MLS) instrument
on the Upper Atmosphere Research Satellite (UAfiS)
(Froidevaux etal., 1994) and the NOAA-11 SBUV/2 in-
strument (Planet et al., 1994). Froidevaux et al. (1994)
also emphasized examining the latitude and height be-
havior of the observed ozone decrease to try to identify
the causes for the lower values.
In Pitari and Rizi (1993), the model calculated a
decrease in ozone of about 12% at 60°N in March 1992.
Diagnostic results showed that this is a combination of a
12% decrease due to heterogeneous chemistry,; a 4% de-
crease due to changes in photolysis rate, and a 4%
increase due to changes in transport. In contrast, the ad-
ditional ozone (about 4%) transported into the region
from the strengthening of the mean circulation in the
Kinnison et al. (1994) study tends to cancel me reduc-
tion of ozone due to the increase in heterogeneous
conversion rates, producing changes in ozone that do not
agree well with observed data. Tie et al. (1994) showed
that the changes in ozone at northern high latitudes are
-10% in spring of 1992 and -8% in spring of 1993. Be-
cause so many different mechanisms can change ozone
after the eruption, it is difficult to understand the ozone
response by comparison of model-simulated ozone with
observations alone. Additional diagnostics based on ob-
servations are needed to isolate the effects of the
different mechanisms.
63.3.4 ISOLATING THE EFFECTS OF HETEROGENEOUS
PROCESSING
Results from Rodriguez et al. (1994) and Kinnison
et al. (1994) showed that the effects of increased hetero-
geneous processing from the Mt. Pinatubo aerosol
caused an additional 2-5% decrease in ozone at mid- to
high latitudes in the winter of 1993. However, the results
-of Pitari and Rizi (1993) and Granier and Brasseur
(1992) indicated that the change in aerosol would lead to
a 10% decrease in ozone column due to heterogeneous
chemistry alone. It is difficult to compare the model pre-
dictions because each model used a different set of
parameters to describe the aerosol loading and its decay.
In an attempt to see if the model predictions will agree
better if the models use uniform input, we prescribed the
following set of simulations. The first simulation calcu-
lates the behavior of ozone using t$e surface
concentrations for trace gases as prescribed in Table 6-3
while keeping the aerosol surface area at the background
value. The second calculation uses the same surface
concentrations but assumes the aerosol surface area in-
creases by a factor of 30 in June 1991. The excess
surface area is assumed to decay with an exponential
6.24
-------
STRATOSPHERIC MODELS
Table 6-4. Mixing ratios for halocarbons (irv pptv) for Scenario II.
year
1992
1995
2000
2005
2010
CFC-11
281.8
290.4
284.6
278.0
264.3
CFC-12
487.6
513.7
528.6
532.4
526.9
CFC-113
79.1
87.5
85.5
82.9
79.6
CC\4
110.5
113.7
118.5
122.8
117.7
GH3CC13
178.1
159.3 .
75.8
42.5
22.0
CHsBr
14.1
14.7
15.4
16.4
17.6
time constant of 1 year. The simulation is to include only
the effect of enhanced heterogeneous processing. The
differences between the ozone in the two simulations
(second simulation minus the first) are given in Figure
6-5.
Prather (1992) investigated the potential for a non-
linear, catastrophic loss of stratospheric ozone if the
aerosol density were greatly increased following a mas-
sive eruption. None of the models indicates that such a
situation was reached in the Mt. Pinatubo case. Figure
6-5a shows the results for northern midlatitudes, indicat-
ing that the effect of enhanced processing is to decrease
the ozone. The results fall into three groups: about -3%
(AER and LLNL), about -5% (GSFC and MPIC), and
about -8% (ITALY and NCAR). The results for the trop-
ics are given in Figure 6-5b. The ozone decrease ranges
from less than 0.5% to 2.5%.
It is unclear what the causes are for the differences
in the model predictions. Possible explanations include
the different treatments used in calculating the concen-
tration of N2O5 and the different effects of reaction (6-2)
in the models caused by different temperatures being
used. The AER model and the LLNL model use an ex-
plicit diurnal variation in calculating N2O5, while other
models use various methods to estimate the N2Os con-
centration from an averaged sun condition.
6.4 RESULTS FROM SCENARIO
CALCULATIONS
For the purpose of a model intercomparison, we
have prescribed two scenarios for the source gases. The
surface concentrations of the species are specified as
functions of time as given in Tables 6-3 and 6-4. Values
prior to 1990 are based on available observations. The
growth rate for N2O is based on previous estimates of
0.25% per year. Khalil and Rasmussen (1992) showed
that the actual increase in the past decade has been very
variable, ranging from 0 5 ppbv per year to 1.2 ppbv per
year. For CH4, a linear growth rate of 13 ppbv per year is
assumed after 1992. Recent observations for CHU
(Dlugokenckyefa/., 1994; Khalil and Rasmussen, 1993)
indicate that the CH4 growth rate has slowed to as little
as 2 ppbv per year. !
The surface concentration for the CH3C1 is set at
600 pptv. Surface concentrations for the CFCs, HCFCs,
halons, and CH3Br were calculated using a box model
with assumed emissions and the reference lifetimes giv-
en in Chapter 13. In;Scenario I (Table 6-3), the
emissions for the halocarbons follow the guidelines in
the Amendments to the Montreal Protocol. For CH3Br,
it is assumed that a background of 9 pptv is maintained
by natural sources. Emission of anthropogenic CH3Br
assumes a schedule that maintains constant emission at
the 1991 level. This, when combined with the natural
sources, results in a surface concentration of 14.2 pptv
after the year 2000. The substitute HCFCs are a combi-
nation of HCFC-22, : HCFC-141b, HCFC-142b,
HCFC-123, and HCFC-124. The Ozone Depletion Po-
tential (ODP)-weighted annual production is taken to be
3.1% of the OOP-weighted emissions in 1990. In addi-
tion to the basic scenario, results are also presented for a
second scenario (Table 6;-4) where we assume partial
compliance with the Protocol for CFC-11, CFC-12,
CFC-113, CH3CC13, and CCL,. The emission for CH3Br
is also assumed to be larger, resulting in a surface con-
centration of 17.6 pptv1 in 2010. The Scenario II
calculation extends only to 2010.
6.4.1 Chlorine and Bromine Loading
Figure 6-6a shows the model-calculated chlorine
concentrations for 58 km at 50°N. The observed concen-
trations of HCI from the ATMOS instrument for 1985
(Zander et a/., 1990) and 1992 (Gunson et ai, 1994) are
6.25
-------
STRATOSPHERIC MODELS
Ozone Column Monthly % Difference
AER
..40deg. N
_ 50 deg. N
_60deg. N
90 91 92 93 94 95 96 97
Date
ITALY
..40 deg. N
_50 deg. N
-60 deg. N
90 91 92 93 94 95 96 97
MPIC
.-40 deg. N
_ 50 deg. N
-60 deg. N
GSFC
. 40 deg. N
.50 deg. N
-60 deg. N
90 91 92 93 94 95 96 97
Date
LLNL
1-3
sol with the column calculated where there is a 30-fold increase in aerosol surface area in June 1991 with the
excess aerosol decaying with a time constant of 1 year. ;
6.26
-------
STRATOSPHERIC MODELS
1.0
0.5
0.0
-0.5
-1.0
-1.5
-2.0
-2.5
-3.0 i
AER
•-•-10deg. S
- • 0 deg.
— 10 deg. N
90 91 92 93
1.0
0.5
0.0
-0.5
-1.0
-1.5
-2.0
-2.5 -
-3.0
Date
ITALY
95 96 97
••••10deg.S
- - 0 deg.
—10 deg. N
1.0
0.5
0.0
1-0.5
0>
Q -1.0
Q>
§-1-5
Q.
-2.0
-2.5
-3.0
GSFC
•••MOdeg. S
- i 0 deg.
— 10 deg.. N
1.0
0.5
0.0
i -0-5
a
a -1.0
I -1.5*.
a.
-2.0 -
-2.5 -
90-91 92 93 94 195
Date '
LLNL i
96 97
-3.0
••-•,10deg. S
- -:0deg.
— .10 deg. N
1.0
0.5
o.o
0>
§-0.5
I
5 -1.0
1
I -1.5
o.
-2.0 -
-2.5 -
-3.0
90 91 92 93 94 95 96 97
Date I
NCAR '
••••10 deg. S
- - 0 deg.
— TO deg. N
90 91 92 93 . 94 <5 96 97
Date
Figure 6-5b. Same as Figure 6-5a except for 10°S, Equator, and 10°N.
6.27
-------
STRATOSPHERIC MODELS
Cly Trend (or Scenario I at SON, 58 km
1980
1990
2000
2010 2020 2030
Year
2040 2050
Cly Trend for Scenario I at SON, 22 km
1980
1990
2000
2010 2020
Year
2040
2050
Figure 6-6a. Upper panel: Model-calculated con-
centration for chlorine for Scenario I at 58 km, 50°N
for March. The ETCL is the mixing ratio of the chlo-
rine atoms bound in the source gases at the
surface. It is calculated using the boundary values
given in Table 6-3. The measured value of HCI (I)
from the ATMOS instrument for the 1985 SL-3 mis-
sion (Zander et al., 1990) and the 1992 ATLAS-1
mission (Gunson, et al., 1994) are shown for com-
parison.
Figure 6-6b. Lower panel: Model-calculated con-
centration for chlorine for Scenario I at 22 km, 50°N
for March. The EESC curves are calculated using
the boundary values of the chlorine source gases.
It corresponds to the mixing ratio of the chlorine at-
oms bound in the sources gases and weighted by
the OOP of the source gas. The second curve is
multiplied by 0.7, which is the fraction of CF:C-11
dissociated at that altitude.
included in the figure for comparison. Also Deluded in
the graph is the curve labeled the equivalent tropospheric
chlorine loading (ETCL), which is defined as the sum of
the mixing ratios of the chlorine atoms in the source
molecules at the ground. The model-calculated Cly con-
centrations can be compared with the ETCL curve after
allowing for the time lag to transport the source gases to
the stratosphere and the redistribution of the radical spe-
cies. With the prescribed surface concentrations of the
halocarbons given in Table 6-3, the ETCL reaches a
maximum in 1994. The calculated chlorine concentra-
tion reaches the maximum around the year 1998.
The model-calculated Cly concentrations at 22 km
are shown in Figure 6-6b. Estimates for the chlorine
concentrations at 20 km between 60°N and 80°N based
on measured concentrations of the organic chlorine spe-
cies range from 1 to 2 ppbv for outside and inside the
vortex, respectively (Kawa et al., 1992). The observed
value inside the vortex should be more representative of
the concentration at 22 km because of the occurrence of
diabatic descent in the vortex. The EESC (equivalent
effective stratospheric chlorine) curve is defined by the
sum of the mixing ratios of the chlorine atoms in the
source molecules at the ground, weighted by the respec-
tive ODPs. It corresponds to the chlorine loading values
used in Chapter 13 of this report. It can be compared
with the chlorine concentration in the lower strato-
sphere. One of the curves in Figure 6-6b is obtained by
multiplying the EESC Values by 0.7, approximately the
fraction of CFC-11 dissociated at 22 km and 50°N. The
calculated Cly concentrations at 22 km among the mod-
els differ by about 1 ppbv. The models that calculate
smaller concentrations of Cly also calculate smaller con-
centrations of ozone. This is probably related to the
position of the tropopause and the strength of the circula-
tion in the models.
The time behavior of the ETCL and EESC curves
agrees well with the model-calculated curve except that
the calculated concentration in the lower stratosphere
lags the EESC curve by 3-4 years and the calculated
concentration in the upper stratosphere lags the ETCL
curve by 4-5 years.
The calculated bromine concentrations are shown
in Figure 6-7. The curve representing the equivalent tro-
pospheric bromine loading (ETBL) is included in Figure
6-7a. As in the case of the model-calculated Cly, the
6.28
-------
40
35
30
2S
'< 20
IS
10
5 -
Bry Trend tor Scenario I at SON, 58 km
1980
2010 2020
Year
40
35
30
25
20
15 r
10
5 L
STRATOSPHERIC MODELS
Bry Trend for Scenario I at SON, 22 km
2050
ot—
1980
1990
2000
2010 2020
V'ear
2030
2040
2050
Figure 6-7a.
Model-calculated bromine concentration for Scenario I at 58km 50°W for March The ETBL is
Ł, iSe IT bound in the "™ 9ases at the surfaoe- " ls •**
Figure 6-7b. Model-calculated bromine concentration for Scenario I at 22km 50°N for March.
model-calculated Bry values in the lower stratosphere
separate into the same two groups.
A comparison of the chlorine loadirig and bromine
loading between Scenarios I and II is shown in Figures
6-8 and "6-9, respectively, for the AER and GSFC mod-
els. At 2010, the calculated concentrations of chlorine
and bromine in the lower stratosphere for Scenario II are
larger than those calculated for Scenario I by 200 pptv
and 2 pptv, respectively.
6.4.2 Calculated Ozone Trend
Figure 6-10 shows the calculated trends for col-
umn ozone between 1980 and 2050. for Scenarios I and
II. Note that the calculated ozone columns in 1980 are
quite different among the models, ranging from 380
Dobson units to 470 DU at 60°N; and 300 DU to 420 DU
at 40°N. The calculated percent change relative to 1980
is plotted in Figure 6-11. The maximum decrease of
about 6% to 9% is calculated around the year, 2000.
Larger decreases are calculated by the OSLO and MPIC
models, which include PSC chemistry. The extra Cly
and Bry in Scenario II cause an extra 1.5% decrease in
ozone at high latitudes for the AER and GSFC models,
which do not include PSC chemistry. The MPIC and
OSLO models, which include PSC chemistry, calculate
an additional 3% decrease in ozone at 60°N in March.
The models also show different results in the rate
of recovery. Figure 6-12 shows the model-calculated
ozone behavior for 50°N. The CAMBRIDGE, GSFC,
and MRI models reach the 1980 ozone values before
2030, while the other models show a much slower recov-
ery. Additional calculations were performed to check
the sensitivity of the ozone response to changes in N2O,
CH4, and aerosol surface 'area. With the chlorine
concentration fixed at the 2050 level, approximately 2
ppbv, calculations were performed to determine how the
model-calculated ozone will change under the assump-
tions listed in Table 6-5. All the models agree that a
decrease in CFLt would produce a decrease in column
ozone. The GSFC model and the LLNL model are more
sensitive to changes in CPtt. All the models agree that a
decrease in N2O would produce an increase in column
ozone. A doubling of aerosol surface area in 2050 has an
effect ranging from neutral to;a slight increase in column
ozone. The models do not agree on the sign of the ozone
column change when these perturbations are combined.
6.4.3 Effects from Greenhouse Gases
i ,
In Section 6.3.2.2, we discussed dynamics feed-
back as a negative feedback in the upper stratosphere,
/.
-------
STRATOSPHERIC MODELS
Cly Trend from AER. SON, 22J
-------
STRATOSPHERIC MODELS
AER - Scenario I and II
if 460
S 440
8 420
S 400
i 380
S 360
S 340
« 320
g 300
19
F460
440
1 42°
- 400
§380
= 360
O 340
« 320
g 300
°m
19
•c 460
i 440
i 42°
S 400
i 380
= 360
S 340
J 320
O 280
260
•jtm ;
60 1990 2000 2010 2020 2030 2040 20-
Year
GSFC - Scenario I and II
60 1990 2000 2010 2020 2030 2040 20!
Year
MPIC - Scenario I and II
L d§i!j :
>-— =—
-
-
1980 1990 2000 2010 2020 2030 2040 2OSO
Year
400
c- 460
S 440
« 420
B 400
| 380
2 360
O 340
« 320
S 3oa
CAMBRIDGE - Scenario I
1980 1990 2000 2010 2020 2030 2040 2050
Year
480
•S 460
S 4*0
o 420
2-400
§ 380
= 360
S 340
• 320
g 300
ITALY - Scenario 1
1980 1990 2000 2010 2020 2030 2040 2050
Year
OSLO - Scenario I and II
f 460
3 440
I 420
S 400
J= 360
360
O 340
J 320
S 3CO
1980 1990 2000 2010 2020 2030 2O4C 2050
Year
Figure 6-10. Model-calculated ozone trend for March for Scenarios I and II.
AER - Scenario I and II
••40 deg. N
- 50 deg. N
-60deg. N
1980 1990 2000 2010 2020 2030 2O40 2050
Year
GSFC - Scenario I and II
• 40 deg. N
- 50 deg. N
-60 deg. N
1980 1990 2000 2010 2020 2030 2040 2050
Year
MPIC - Scenario I and II
1980 1990 2000 2010 2020 2030 2040 2050
Year
CAMBRIDGE - Scenario I
•• 40 deg. N
- 50 deg. N
-60 deg. N
1980 1990 2000 2010 2020 2030 2040
Year
ITALY - Scenario I
•• 40 deg. N
- 50 deg. N
-60 deg. N
1980 1990 2000 2010 2020 2030 2O40
Year
OSLO - Scenario I and II
40 deg. N
- 50 deg. N
-60 deg. N
1980 1990 2000 2010 2020 2030 2040^2050
Figure 6-11. Model-calculated percent change in ozone for March for Scenarios I and II.
6.31
-------
STRATOSPHERIC MODELS
Scenario I Percent Difference at SON
1980
1990 2000 2010 2020
Year
2030
2040
2050
tern and cooling of the middle atmosphere. Ozone is a
primary absorber of solar radiation, wanning the middle
atmosphere and reducing the solar radiation reaching the
surface. It is also a greenhouse gas, and thus acts to
warm the surface/troposphere system. Studies (Ram-
aswamyefa/., 1992; Wang era/., 1993) have shown that
the latitudinal distribution of the total greenhouse warm-
ing effect due to CO2, CH* N2O, and CFCs decreases
more sharply at mid- and high latitudes when the ob-
served changes in the ozone distribution were
, considered.
Using results from a GCM, Rind et al. (1990)
found that both the vertical and latitudinal structure of
the temperature change following a doubling of (X>2 re-
sults in increases in the propagation of planetary waves-
into the stratosphere, and increased potential energy in
the lower stratosphere. These processes generated an
increase in the eddy energy in the middle atmosphere
with an attendant increase in wave forcing and residual
circulation, tending to warm high latitude.- This change
in the transport circulation should affect the ozone distri-
bution and, consequently, the thermal forcing of the
for March at 50°N for Scenario I. The results from
e boundary conditions appropriate for the years.
atmosphere. This effect on ozone was not included in
the Rind et al. (1990) calculation but was acknowledged
to be of potential importance. Austin et al. (1992) used a
middle atmosphere GCM, which included a relatively
comprehensive ozone photochemistry that was radia-
tively interactive, in a CO2 doubling scenario. The
model was found to respond very differently to die CO2
doubling when fully interactive ozone was included, par-
ticularly with respect to the occurrence of a stratospheric
warming. They also found a much larger ozone response
to CO2 doubling, with reductions in ozone by as much as
150 Dobson units during Arctic spring. In contrast, Pi-
tari et al. (1992) found only a 10 Dobson unit reduction
in column ozone amounts during the Northern Hemi-
sphere spring minimum. ;
In addition to'these changes in the middle atmo-
sphere circulation, theoretical results (e.g., Geller and
Alpert, 1980) and modeling studies (e.g., Hansen et al.,
1983; Boville, 1984) demonstrated a troposphere re-
sponse to perturbations in the stratospheric dynamics.
Both the stationary and transient components of tropo-
spheric wave structures can be modulated by the
6.32
-------
STRATOSPHERIC MODELS
Table 6-5. Sensitivity studies illustrating the model-calculated ozone responses.
2050
baseline
B
C
D
N20
ppbv
360
7
-------
STRATOSPHERIC MODELS
Bekki, S., and J.A. Pyle, Two-dimensional assessment of
the impact of aircraft sulfur emissions on die
stratospheric sulfate aerosol layer, /. Geophys.
Res., 97, 15839-15847, 1992.
Bekki, S., R. Toumi, and J.A. Pyle, Role of sulfur photo-
chemistry in tropical ozone changes after the
eruption of. Mount Pinatubo, Nature, 362, 331-
333, 1993.
Bojkov, R.D., The 1983 and 1985 anomalies in ozone
distribution in perspective, Man. Wea. Rev., 115,
2187-2201, 1987.
Boville, B.A., The influence of the polar night jet on the
tropospheric circulation in a GCM, J. Atmos. ScL,
41, 1132-1142, 1984.
Brasseur, G., The response of the middle atmosphere to
long-term and short-term solar variability: A two-
dimensional model, J. Geophys. Res., 98,
23079-23090, 1993.
Brasseur, G., and P.C. Simon, Stratospheric chemical
and thermal response to long-term variability in
solar UV irradiance, J. Geophys. Res., 86, 7343-
7362, 1981.
Brasseur, G., M.H. Hitchman, S. Walters, M. Dymek, E.
Falise, and M. Pirre, An interactive chemical-dy-
namical radiative two-dimensional model of the
middle atmosphere, J. Geophys. Res., 95, 5639-
5655, 1990.
Brasseur, G.P., and C. Granier, Mount Pinatubo aerosols,
chlorofluorocarbons, and ozone depletion, Sci-
ence, 257, 1239-1242, 1992.
Callis, L.B., J.C. Alpert, and M.A. Geller, An assessment
of thermal, wind and planetary wave changes in
the middle and lower atmosphere due to 11-year
UV-flux variation, J. Geophys. Res., 90, 2231-
2282, 1985.
CalHs, L.B., et al. Precipitating relativistic electrons:
Their long-term effect on stratospheric odd nitro-
gen levels, /. Geophys. Res., 96,2939-2976, 1991.
Cariolle, D., A. Lasserre-Bigorry, J.-F. Royer, and J.-F.
Geleyn, A general circulation model simulation of
the springtime Antarctic ozone decrease and its
impact on mid-latitudes, J. Geophys. Res., 95,
1883-1898, 1990.
Chandra, S., Changes in stratospheric ozone and temper-
ature after the eruption of Mt. Pinatubo, Geophys.
Res. Lett., 20, 33-36, 1993.
Chandra, S., C.H. Jackman, A.R. Douglass, E.I. Flem-
ing, and D.B. Considine, Chlorine catalyzed
destruction of ozone: Implications for ozone vari-
ability in the upper stratosphere, Geophys. Res.
Lett., 20, 351-354, 1993.
Chipperfield, M.P., and J.A. Pyle, Two-dimensional
modeling of the Antarctic lower stratosphere, Geo-
phys. Res. Lett., 15, 875-878, 1988. ;
Chipperfield, M.P., D. Cariolle, P. Simon, R. Ramoroson,
and D.J. Lary, A three-dimensional modeling
study of trace species in the Arctic lower strato-
sphere during winter 1989-1990, /. Geophys. Res.,
95,7199-7218, 1993.
CIAP Monograph 3, The Stratosphere Perturbed by
Propulsion Effluents, DOT-TST-75-53/ U.S. De-
partment of Transportation, Washington, D.C.,
1975.
Coffey M.T., and W.G. Mankin, Observations pf the loss
of stratospheric NC>2 following volcanic:eruptions,
Geophys. Res. Lett., 20, 2873-2876, 1993.
Considine, D.B., A.R. Douglass, and C.H. Jackman, Ef-
fects of a PSC parameterization on ozone
depletion due to stratospheric aircraft in a 2D
model, J. Geophys. Res., 99, 18879-18894, 1994.
Crutzen, P.J., The influence of nitrogen oxides on the at-
mospheric ozone content, Quart. J. Roy. Meteorol.
Soc., 96, 320-325, 1970. '•
DeFoor, T.E., E. Robinson, and S. Ryan, Early lidar ob-
servations of the June 1991 Pinatubo eruption
plume at Mauna Loa Observatory, Hawaii, Geo-
phys. Res. Lett., 19, 187-190, 1992.
Dessler, A.E., et al., Balloonborne measurements of
CIO, NO and Os in volcanic cloud: An analysis of
heterogeneous chemistry between 20 and 30 km,
Geophys. Res. Lett., 20, 2527-2530, 1993.
Dlugokencky, E.J., K.A. Masarie, P.M. Lang, P.P. Tans,
L.P. Steele, and E.G. Nisbet, A dramatic decrease
in the growth rate of atmospheric methane in the
northern hemisphere during 1992, Geophys. Res.
Lett., 21, 43-48, 1994. !
Douglass, A.R., R.B. Rood, J.A. Kaye, R.S. Stolarski,
D.J. Allen, and E.M. Larsen, The influence of po-
lar stratospheric heterogeneous processes on
reactive chlorine at mid-latitudes: Three-dimen-
sional model implications, Geophys., Res. Lett.,
78,25-28,1991. ,
6.34
-------
STRATOSPHERIC MODELS
Eluszkiewicz J., and M. Allen, Global analysis of the
ozone deficit, /. Geophys. Res., 98, 1069-1082,
1993.
Fahey, D.W., et al. In situ measurements constraining
the role of sulphate aerosols in mid-latitude ozone
depletion, Nature, 363, 509-514, 1993.
Fleming, E.L., S. Chandra, C.H. Jackman, D.B. Consi-
dine, and A.R. Douglass, The middle atmospheric
response to short and long term solar UV varia-
tions: Analysis of observations and 2D model
results, J. Atmos. and Terrestrial Phys., in press
1994.
Foley, H.M., and M.A. Ruderman, Stratospheric NO
production from past nuclear explosions, /. Geo-
phys. Res., 78,4441-4450, 1973.
Fried, A., B.E. Henry, J.G. Calvert, and M. Mozur-
kewich, The reaction probability of N2O5 with
sulfuric acid aerosols at stratospheric temperatures
and compositions, /. Geophys. Res., 99 3517-
3532, 1994.
Froidevaux, L., J.W. Waters, W.G. Read, L.S. Elson,
D.A. Flower, and R.F. Jarnot, Global ozone obser-
vations from UARS MLS: An overview of zonal
mean results, J. Atmos. ScL, in press, 1994.
Garcia, R.R., and S. Solomon, A numerical model of the
zonally averaged dynamical and chemical struc-
ture of the middle atmosphere, J. Geophys. Res.,
88, 1379-1400, 1983.
Garcia, R.R., S. Solomon, R.G. Roble, and D.W. Rusch,
A numerical response of the middle atmosphere to
the 11-year solar cycle, Planet. Space ScL, 32,
411-423, 1984.
Garcia, R.R., F. Stordal, S. Solomon, and J.F. Kiehl, A
new numerical model of the middle atmosphere 1.
Dynamics and transport of tropospheric source
gases, J. Geophys. Res., 97, 12967-12991, 1992.
Garcia, R.R., and S. Solomon, A new numerical model
of the middle atmosphere 2. Ozone and related
species,/ Geophys. Res., 99, 12937-12951, 1994.
Geller, M.A., and J.C. Alpert, Planetary wave coupling
' between the troposphere and the middle atmo-
sphere as a possible sun-weather mechanism, J.
Atmos. ScL, 37, 1197-1215, 1980.
Gleason, J.F., et al., Record low global ozone in 1992,
Science, 260, 523-526, 1993.
Golombek, A., and R.G. Prinn, A global 3-D model of
the stratospheric acid layer, J. Atmos. Chem., 16,
179-199,1993. i
Granier, C, and G. Brasseur, Ozone and other trace gases
in the Arctic and Antarctic regions: Three-dimen-
sional model simulations, J. Geophys. Res., 96
2995-3011,1991. ;
Granier, C., and G. Brasseur, Impact of heterogeneous
chemistry on model predictions of ozone changes,
J. Geophys. Res., 97, 18015-10833, 1992.
Grant, W.B., etal. Observations of reduced ozone con-
centrations in the tropical stratosphere after the
eruption of Mt. Pinatubo, Geophys. Res. Lett 19
1109-1112,1992. I
Grant, W.B., et al.. Aerosol-associated changes in tropi-
cal stratospheric ozone following the eruption of
Mt. Pinatubo, /. Geophys. Res., 99, 8197-8211
1994.
Gray, L.J., and J.A. Pyle, A two-dimensional mean cir-
culation model of the quasi-biennial oscillation of
ozone, J. Atmos. Sci.,'46, 203-220, 1989.
Gray, L.J., and TJ. Dunkenton, The role of the seasonal
cycle in the quasi-biennial oscillation of ozone, /.
Atmos. ScL, 47, 2429--2451, 1990.
Gray, L.J., and S. Ruth, The modeled latitudinal distribu-
tion of the ozone quasi-biennial oscillation using
observed equatorial winds, J. Atmos. ScL, 50,
1033-1046, 1993.
Grooss, J.U., T. Peter, C. Briihl, and P.J. Crutzen, The
influence of high flying aircraft on polar heteroge-
neous chemistry, in Impact of Emissions from
Aircraft and Spacecraft upon the Atmosphere:
Proceedings of an International Scientific Collo-
quium, edited by U. Schumann and D. Wurzel,
DLR-Mitteilung 94-06, 229-234, 1994.
Gunson, M.R., et al., Increase in levels of stratospheric
chlorine and fluorine loading between 1985 and
1992, Geophys. Res. Lett., 21, 2223-2226, 1994.
Hansen, J.G., et al.. Efficient three-dimensional global
models for climate studies: Models I and II, Mon.
Wea. Rev., Ill, 609-662, 1983.
Hanson, D., and K. Mauersberger, Solubility and equi-
librium vapor pressures of HC1 dissolved in polar
stratospheric cloud material: Ice and the trihydrate
of nitric acid, Geophys. Res. Lett., 15, 1507-1510
1988. I
6.35
-------
STRATOSPHERIC MODELS
Hanson, D.R., A.R. Ravishankara, and S. Solomon, Het-
erogeneous reactions in sulfuric acid aerosol: A
framework for model calculations, J. Geophys.
Res., 99,3615-3629, 1994.
Harwood, R.S., and J.A. Pyle, A two-dimensional mean
circulation model for the atmosphere below 80
km, Quart. J. Roy. Meteorol. Soc., 101, 723-748,
1975.
Herman, J.R., and D. Larko, Low ozone amounts during
1992-1993 from Nimbus 7 and Meteor 3 total
ozone mapping spectrometers, J. Geophys. Res.,
99,3483-3496, 1994.
Hofmann, D.J., Aircraft sulphur emissions, Nature, 349,
659, 1991.
Hofmann, D.J., and S. Solomon, Ozone destruction
through heterogeneous chemistry following the
eruption of El Chichdn, /. Geophys. Res., 94,
5029-5041, 1989.
Holton, J.R., On the global exchange of mass betv/een
the stratosphere and troposphere, J. Atmos. ScL,
47,392-395, 1990.
Hood, L.L., R.D. McPeters, J.P. McCormick, L.E.
Flynn, S.M. Hollandsworth, and J.F. Gleason, Al-
titude dependence of stratospheric ozone trends
based on Nimbus 7 SBUV data, Geophys. Res.
Lett., 20, 2667-2670, 1993.
Hou. A.Y., H.R. Schneider, and M.K.W. Ko, A dynami-
cal explanation for the asymmetry in
zonally-averaged column abundances of ozone be-
tween northern and southern springs, /. Atmos.
Sci., 48,547-556, 1991.
Huang, T. Y.W., and G. Brasseur, The effect of long-term
solar variability in a two-dimensional interactive
model of the middle atmosphere, J. Geophys. Res.,
98, 20413-20427, 1993.
Isaksen, I.S.A., B. Rognerud, F. Stordal, M.T. Coffey,
and W.G. Mankin, Studies of Arctic stratospheric
ozone in a 2-D model including some effects of
zonal asymmetries, Geophys. Res. Lett., 17, 557-
560, 1990.
Jackman, C.H., Energetic particle influences on NOy
and ozone in the middle atmosphere, Geophysical
Monograph 75, IUGG vol. 15, 131-139, 1993.
Jackman, C.H., J.E. Frederick, and R.S. Stolarski, Pro-
duction of odd nitrogen in the stratosphere and
mesosphere: An intercomparison of source
strengths, J. Geophys. Res., 85, 7495-7505, 1980.
Jackman, C.H., A.R. Douglass, P.O. Guthrie, and R.S.
Stolarski, The sensitivity of total ozone and ozone
perturbation scenarios in a two-dimensional mod-
el due to dynamical inputs, J. Geophys. Res., 94,
9873-9887, 1989. ;
Jackman, C.H., A.R. Douglass, R.B. Rood,' R.D. Mc-
Peters, and P.E. Meade, Effect of solar proton
events on the middle atmosphere during the past
two solar cycles as computed using a two-dimen-
sional model, J. Geophys. Res., 96, 7417-7428,
1990.
Jackman, C.H., A.R. Douglass, S. Chandra, R.S. Stolar-
ski, J.E. Rosenfield, and J.A. Kaye,' Impact of
interannual variability (1979-1986) of transport
and temperature on ozone as computed using a
two-dimensional photochemical model, J. Geo-
phys. Res., 96, 5073-5079, 1991.
Johnston, H., G. Whitten, and J. Birks, Effect of nuclear
explosions on stratospheric nitric oxide and ozone,
J. Geophys. Res., 78, 6107-6135, 1973.
Johnston, P.V., R.L. McKenzie, J.G. Keys, and W.A.
Matthews, Observations of depleted stratospheric
NOa following the Pinatubo volcanic eruption,
Geophys. Res. Lett., 19, 211-213, 1992.
Karoi, I.L., Y.E. Ozolin, and E.V. Rozanov, Effect of
space rocket launches on ozone, Annales Geo-
physicae, 10, 1810-1814, 1992. :
Kawa, S.R., et al.. Photochemical partitioning of the re-
active nitrogen and chlorine reservpirs in the
high-latitude stratosphere, J. Geophys. Res., 97,
7905-7923, 1992. . '.
Kawa, S.R., et al. Interpretation of NOx/NOy observa-
tions from AASE-II using a model of chemistry
along trajectories, Geophys. Res. Lett., 20, 2507-
2510, 1993. :
Kaye, J.A., et al., Spatial and temporal variability of the
extent of chemically processed stratospheric air,
Geophys. Res. Lett., 18, 29-32, 1991.'
Kaye, J.A., S.A. Penkett, and P.M. Ormond, ;eds., Report
on Concentrations, Lifetimes, andl Trends of
CFCs, Halons, and Related Species, NASA Refer-
ence Publication 1339, 1994. :
Khalil, M.A.K., and R.A. Rasmussen, The global
sources of nitrous oxide, J. Geophys. Res., 97,
14651-14660, 1992. !
6.36
-------
STRATOSPHERIC MODELS
Khalil, M.A.K., and R.A. Rasmussen, Decreasing trend
of methane: Unpredictability of future concentra-
tions, Chemosphere, 26, 803-814, 1993.
Kiehl, J.T., B.A. Boville, and B.P. Briegleb, Response of
a general circulation model to a prescribed Antarc-
tic ozone hole, Nature, 332, 501-504, 1.988.
Kinne, S., O.B. Toon, and M.J. Prather, Buffering of
stratospheric circulation by changing amounts of
tropical ozone: A Pinatubo case study, Geophys.
Res. Lett., 19, 1927-1930, 1992.
Kinnersley, J.S., and R.S. Harwood, An isentropic
2-dimensional model with an interactive parameter-
ization of dynamical and chemical planetary-wave
fluxes, Quart. J. Roy. Meteorol. Soc., 119, 1167-
1193, 1993.
Kinnison, D.E., K.E. Grant, P.S. Connell, D.J. Wueb-
bles, and D.A. Rotman, The chemical radiative
effects of the Mt. Pinatubo eruption, J. Geophys.
Res., in press, 1994.
Ko, M.K.W., M.B. McElroy, O.K. Weisenstein, and
N.-D. Sze, Lightning: A possible source of strato-
spheric odd nitrogen, J. Geophys. Res., 91,
5395-5404, 1986.
Ko, M.K.W., H.R. Schneider, R.-L. Shia, O.K. Weisen-
stein, and N.-D. Sze, A two-dimensional model
with coupled dynamics, radiation, and photo-
chemistry: 1. Simulation of the middle
atmosphere, J. Geophys. Res., 98, 20429-20440
1993.
Kodera, K., Influence of the stratospheric circulation
change on the troposphere in the northern hemi-
sphere winter, in The Role of the Stratosphere in
Global Change, edited by M.-L. Chanin, Springer-
Verlag, 227-243, 1993.
Kodera, K., M. Chiba, K. Yamazaki, and K. Shiba, A
possible influence of the polar night stratospheric
jet on the subtropical tropospheric jet, J. Meteorol.
Soc. Japan, 69, 715-721, 1991.
Koike, M., et al., Impact of Pinatubo aerosols on the par-
titioning between NO2 and HNO3, Geophys. Res.
Lett., 21, 597-600, 1994.
Kotamarthi, V.R., M.K.W. Ko, O.K. Weisenstein, and
N.-D. Sze, The effect of lightning on the concen-
tration of odd nitrogen species in the lower
stratosphere: An update, J. Geophys. Res., 99
8167-8173, 1994.
Labitzke, K., and M.P. McCormick, Stratospheric tem-
perature increases due to Pinatubo aerosols,
Geophys. Res. Lett., 19, 207-210, 1992.
Lait, L.R., M.R. Schoeberl, and P.A. Newman, Quasi-
biennial modulation of the Antarctic ozone
depletion, /. Geophys. Res., 94 11559-11571
1989. I '
Lefevre, R, G.P. Brasseur, I. Folkins, A.K. Smith, and P.
Simon, Chemistry of the 1991 -1992 stratospheric
winter: Three-dimensional model simulations, /.
Geophys. Res., 99, 8183-8195, 1994.
Legrand, M.R., F. Stordzil, I.S.A. Isaksen, and B. Rog-
nerud, A model study of the stratospheric budget
of odd nitrogen, including effects of solar cycle
variations, Tellus, 4IB, 413-426, 1989.
Liu, S.C., M. McFarland; D. Kley, O. Zafirious, and B.
Huebert, Tropospheric NOX and O3 budgets in the
equatorial Pacific, J. Geophys. Res., 88 1360-
1368, 1983. :
Mahlman, J.D., J.P. Pinto,,and L.J. Umscheid, Transport,
radiative, and dynamical effects of the Antarctic
ozone hole: A GFDL "SKYHI" model experi-
ment, J. Atmos. Sci., 51, 489-508, 1994.
Mankin, M.G., MT. Coffey, and A. Goldman, Airborne
observations of SO;,, HC1, and O3 in the strato-
spheric plume of the Pinatubo volcano in July
1991, Geophys. Res.\Letf., 19, 179-182, 1992.
McConnell, J.C., and M.B J McElroy, Odd nitrogen in the
atmosphere, /. Atmos. Sci., 30, 1464-1480, 1973.
McCormick, M.P., R.E. Veiga, and W.P. Chu,'Strato-
spheric ozone profile and total ozone trends
derived from the SAGE I and SAGE II data, Geo-
phys. Res. Lett., 19, 269-272, 1992.
McElroy, M.B., and J.C. filcConnell, Nitrous oxide: A
natural source of stratospheric NO, /. Atmos. Sci
28, 1095-1098, 1971.
McElroy, M.B., and R.J. Salawitch, Changing composi-
tion of the global stratosphere, Science 243
763-770, 1989. 1
McElroy, M.B., R.J. Salawitch, and K. Minschwaner,
The changing stratosphere, Planet. Space Sci., 40,
373-401,1992. ,.
Michelangeli, D.V., M. Allen, and Y.L. Yung, El Chichon
volcanic aerosols: Impact of radiative, thermal,
and chemical perturbations, J. Geophys. Res., 94,
18429-18443, 1989.
6.37
-------
STRATOSPHERIC MODELS
Mills, M.J., et al.. On the relationship between strato-
spheric aerosols and nitrogen dioxide, Geophys.
Res. Lett., 20, 1187-1190, 1993.
Molina M J., et al., Physical chemistry of the H2SC>4/
HNO3/H2O system: Implications for polar strato-
spheric clouds. Science, 261, 1418-1423, 1993.
Murphy, D.M., and A.R. Ravishankara, Temperature av-
erages and rates of stratospheric reactions,
Geophys. Res. Lett., in press, 1994.
Natarajan, M., and L.B. Callis, Stratospheric photo-
chemical studies with Atmospheric Trace
Molecule Spectroscopy (ATMOS) measurements,
/. Geophys. Res., 96,9361-9370, 1991.
Newman, P.A., M.R. Schoeberl, R.A. Plumb, and J.E.
Rosenfield, Mixing rates calculated from potential
vorticity,/ Geophys. Res., 93, 5221-5240, 1988.
Nicolet, M., On the production of nitric oxide in the
stratosphere by cosmic rays in the mesosphere and
stratosphere. Planet. Space ScL, 23, 637-649,
1975.
Nicoli. M.P., and G. Visconti, Impact of coupled pertur-
bations of atmospheric trace gases on earth's
climate and ozone, Pure Appl. Geophys., 120,626-
641, 1982.
Noxon, J.F., Atmospheric nitrogen fixation by lightning,
Geophys. Res. Lett., 3,463-465, 1976.
Oort, A.H., and H. Liu, Upper-air temperature trends
over the globe, 1958-1989, J. Climate, 6, 292-307,
1993.
Park, C.^and G.P. Menees, Odd nitrogen production by
meteoroids, J. Geophys. Res., 83, 4029-4035,
1978.
Patten, K.O., Jr., P.S. Connell, D.E. Kinnison, D.J.
Wuebbles.T.G. Slanger, and L. Froidevaux, Effect
of vibrationally excited oxygen on ozone produc-
tion in the stratosphere, J. Geophys. Res., 99,
1211-1224,1994.
Peter, T., C. Briihl, and P.J. Crutzen, Increase in the PSC
formation probability caused by high-flying air-
craft, Geophys. Res. Lett., 18, 1465-1468, 1991.
Pitari, G., Contribution to the ozone trend of heteroge-
neous reactions of C1ONO2 on the sulfate aerosol
layer, Geophys. Res. Lett., 20,2663-2666, 1993a.
Pitari, G., A numerical study of the possible perturbation
of stratospheric dynamics due to Pinatubo aero-
sols: Implications for tracer transport, J. Atmos.
Sci., 50, 2443-2461, 1993b.
Pitari, G., S. Palermi, G. Visconti, and R. Prinn, Ozone
response to a CO2 doubling: Results from a strato-
spheric circulation model with heterogeneous
chemistry, J. Geophys. Res., 97,5953-5962, 1992.
Pitari, G., and V. Rizi, An estimate of the chemical and
radiative perturbation of stratospheric ozone fol-
lowing the eruption of Mt. Pinatubo, J. Atmos.
ScL, 50, 3260-3276, 1993. !
Pitari, G., V. Rizi, L. Ricciardulli, and G. Visc0nti, High
speed civil transport impact: Role of sulfate, nitric
acid trihydrate and ice aerosols studied with a two-
dimensional model including aerosol physics, J.
Geophys. Res., 98, 23141-23164, 1993.
Planet, W.G., et al.. Northern hemisphere total ozone
values from 1989-1993 determined with the
NOAA-11 Solar Backscatter Ultraviolet (SBUV/
2) instrument, Geophys. Res. Lett., 21, 205-208,
1994.
Plumb, R.A., and M.K.W. Ko, Interrelationships be-
tween mixing ratios of long-lived stratospheric
constituents, J. Geophys. Res., 97, 10145-10156,
1992.
Prather, M.J., Catastrophic loss of stratospheric ozone in
dense volcanic clouds. J. Geophys. Res.< 97,10187-
10191, 1992.
Prather, M., and A.M. Jaffe, Global impact of the Antarc-
tic ozone hole: Chemical propagation, J. Geophys.
Res., 95, 3473-3492, 1990. !
Prather, M.J., M.M. Garcia, A.R. Douglass,! C.H. Jack-
man, M.K.W. Ko, and N.-D. Sze, The space
shuttle's impact on the stratosphere, /. Geophys.
Res., 95, 18583-18590, 1990a.
Prather, M., M.M. Garcia, R. Suozzo, and D, Rind, Glo-
bal impact on the Antarctic ozone hole: Dynamical
dilution with a three-dimensional chemical trans-
port model, J. Geophys. Res., 95, 3449-3471,
1990b.
Prather, M.J., et al., The Atmospheric Effects of Strato-
spheric Aircraft: A First Program Report, NASA
Reference Publication 1272,46-50, January 1992.
Prather, M.J., and E.E. Remsburg, eds., The Atmospheric
Effects of Stratospheric Aircraft: Report of the
1992 Models and Measurements, Workshop,
NASA Reference Publication 1292,; Vol. 1, Na-
tional Aeronautics and Space Administration,
Washington, D.C., 1993.
6.38
-------
STRATOSPHERIC MODELS
Ramaswamy, V.M., D. Schwarzkopf, and K.P.,Shine,
Radiative forcing of climate from halocarbon-
induced global stratospheric ozone loss, Nature,
355,810-812, 1992.
Reinsel, G.C., et al, Statistical analysis of total ozone
and stratospheric umkehr data for trends and solar
cycle relationship, /. Geophys. Res., 92, 2201-
2209, 1987.
Reinsel, G.C., et al., Seasonal trend analysis of pub-
lished ground-based and TOMS total ozone data
through 1991, /. Geophys. Res., 99, 5449-5464,
1994.
Rind, D., R. Suozzo, N.K. Balachandran, and M.J.
Prather, Climate change and the middle atmo-
sphere. Part I: The doubled CO2 climate, /. Atmos.
ScL, 47, 475-494, 1990.
Rinsland, C.P., et al., Heterogeneous conversion of
N2O5 to HNO3 in the post-Mt. Pinatubo eruption
stratosphere, /. Geophys. Res., 99, 8213-8219,
1994.
Rodriguez J.M., M.K.W. Ko, and N.-D. Sze, The role of
heterogeneous conversion of N2O5 on sulphate
aerosols in global ozone losses, Nature, 352, 134,
1991.
Rodriguez, J.M., M.K.W. Ko, N.-D. Sze, C.E. Heisey,
G.K. Yue, and M.P. McCormick, Ozone response
to enhanced heterogeneous processing after the
eruption of Mt. Pinatubo, Geophys. Res. Lett., 21,
209-212, 1994.
Rood, R., D.J. Allen, W.E. Baker, D.J. Lamich, and J.A.
Kaye, The use of assimilated stratospheric data in
constituent transport calculations, J. Atmos. ScL,
46, 687-701, 1989.
Rood, R.B., et al., Three-dimensional simulations of
wintertime ozone variability in the lower strato-
sphere, J. Geophys. Res., 96, 5055-5071, 1991.
Rosenfield, J.E., M.R. Schoeberl, and M.A. Geller, A
computation of the stratospheric diabatic circula-
tion using an accurate radiative transfer model, J.
Atmos. ScL, 44, 859-876, 1987.
Russell, J.M., III, S. Solomon, L.L. Gordley, E.E. Rems-
berg, and L.B. Callis, The variability of stratospheric
and mesospheric NO2 in the polar winter night ob-
served by LIMS, J. Geophys. Res., 89,7267-7275
1984.
Sanders, R.W., et al., Visible and near-ultraviolet spec-
troscopy at McMurdo Station, Antarctica, 9.
Observations of OC1O from April to October
1991, / Geophys. Res., 98, 7219-7228, 1993.
Schneider, H.R., M.K.W. ''Ko, and C.A. Peterson, Inter-
annual variations of ozone: Interpretation of 4
years of satellite observations of total ozone, /.
Geophys. Res., 96, 2889-2896, 1991.
Schneider, H.R., M.K.W. Ko, R.-L. Shia, and N.-D. Sze,
A two-dimensional model with coupled dynamics,
radiative transfer, and photochemistry, 2. Assess-
ment of the response of stratospheric ozone to
increased levels of CO2, N2O, CH4, and CFC, J.
Geophys. Res., 98, 20441-20449, 1993.
Schoeberl, M.R., P.K. Bhartia, and E. Hilsenrath, Tropi-
cal ozone loss following the eruption of Mt.
Pinatubo, Geophys. Res. Lett., 20, 29-32, 1993.
Slanger, T.G., L.E. Jusinski, G. Black, and G.E. Gadd, A
new laboratory source of ozone and its potential
atmospheric implications, Science, 241, 945-950
1988. ;
Solomon, S., Antarctic ozone: Progress toward a quanti-
tative understanding,; Nature, 347, 347-354, 1990.
Solomon, S., PJ. Crutzen, and R.G. Roble, Photochemi-
cal coupling between the thermosphere and the
lower atmosphere, 1. Odd nitrogen from 50 to 120
km, J. Geophys. Res., 87, 7206-7220, 1982.
Solomon, S., R.W. Sanders; R.R. Garcia, and J.G. Keys,
Increased chlorine dioxide over Antarctica caused
by volcanic aerosols; from Mt. Pinatubo, Nature,
363, 245-248, 1993.
Solomon, S., R.R. Garcia, land A.R. Ravishankara, On
the role of iodine in ozone depletion, J. Geophys.
Res., 99, 20491-20500, 1994a.
Solomon, S., et al., Visible and near-ultraviolet spectros-
copy at McMurdo1 Station, Antarctica 10.
Reductions of stratospheric NO2 due to Pinatubo
aerosols, J. Geophys. Res., in press, 1994b.
Spencer, R.W., and J.R. Christy, Precision lower strato-
spheric temperature monitoring with the MSU:
Technique, validation and results 1979-1991, /
Climate, 6, 1194-1204, 1993.
Stachnik, R.A., J.C. Hardy! J.A. Tarsala, J.W. Waters,
and N.R. Erickson, Submillimeterwave hetero-
dyne measurements of stratospheric CIO, HC1, O3
and HO2: First results, Geophys. Res. Lett., 19,
1931-1934, 1992. j
6.39
-------
STRATOSPHERIC MODELS
Stolarski, R., R. Bojkov, L. Bishop, C. Zerefos, J. Stae-
helin, and J. Zawodny, Measured trends in
stratospheric ozone, Science, 256, 342-349, 1992.
Strobel, D.F., Odd nitrogen in the mesosphere, J. Geo-
phys. Res., 76, 8384-8393, 1971.
Sze, N.-D., M.K.W. Ko, O.K. Weisenstein, J.M. Rod-
riguez, R.S. Stolarski, and M.R. Schoeberl,
Antarctic ozone hole: Possible implications for
ozone trends in the Southern Hemisphere, J. Geo-
phys. Res., 94,11521-11528, 1989.
Tabazadeh, A., and R.P. Turco, Stratospheric chlorine
injection by volcanic eruptions: HC1 scavenging
and implications for ozone, Science, 260, 1082-
1086. 1993.
Tie, X., G.P. Brasseur, B. Briegleb, and C. Granier, Two-
dimensional simulation of Pinatubo aerosol and its
effect on stratospheric ozone,/. Geophys. Res., 99,
20545-20562, 1994.
Tolbert, M.A., J. Pfaff, I. Jayaweera, and M.J. Prather,
Uptake of formaldehyde by sulfuric acid solu-
tions: Impact on stratospheric ozone, /. Geophys.
•Res., 98,2957-2962,1993.
Toon, O., et al., Heterogeneous reaction probabilities,
solubilities and physical state of cold volcanic
aerosols, Science, 261, 1136-1140, 1993.
Toumi, R., An evaluation of autocatalytic ozone produc-
tion from vibrationally excited oxygen in the
middle atmosphere, J. Atmos. Chem., 15, 69-77,
1992.
Toumi, R., A potential new source of OH and odd-nitro-
gen in the atmosphere, Geophys. Res. Lett., 20,
25-28, 1993.
Toumi, R., and S. Bekki, The importance of the reactions
between OH and CIO for stratospheric ozone,
Geophys. Res. Lett., 20, 2447-2450, 1993.
Toumi, R., B.J. Kerridge, and J.A Pyle, Highly vibra-
tionally excited oxygen as a potential source of
ozone in the upper stratosphere and mesosphere,
Nature, 351, 217-219, 1991.
Toumi, R., R.L. Jones, and J.A. Pyle, Direct stratospher-
ic ozone depletion by C1ONO2 photolysis, Nature,
365, 37-39, 1993.
Tuck, A.F., Production of nitrogen oxides by lightning
discharges. Quart. J. Roy. Meteorol. Soc., 102,
749-755, 1976.
Tung, K.K., and H. Yang, Dynamical component of sea-
sonal and year-to-year changes in Antarctic and
global ozone, J. Geophys. Res., 93, 12537-12559,
1988. I
Vupputuri, R.K.R., The structure of the natural strato-
sphere and the impact of chlorofiuoromethanes on
the ozone layer investigated in a 2-D time depen-
dent model, Pure Appl. Geophys., 117, 448-485,
1978.
Wallace, L., and W. Livingston, The effect of the Pinatu-
bo cloud on hydrogen chloride and hydrogen
fluoride, Geophys. Res. Lett., 19, 1209, 1992.
Wang, W.-C., Y.-C. Zhuang, and R.D. Bojkov, Climate
implications of observed changes in ozone vertical
distributions at middle and high latitudes of the
northern hemisphere, Geophys. Res. Lett., 20,
1567-1570, 1993. '
Warneck, P., Cosmic radiation as a source of odd nitro-
gen in the stratosphere, J. Geophys. Res., 77,
6589-6591, 1972.
Webster, C.R., R.D. May, M. Allen, L. Jaegle, and M.P.
McCormick, Balloon profiles of stratospheric
NO2 and HNO3 for testing the heterogeneous hy-
drolysis of N2Os on sulfate aerosols, Geophys.
Res. Lett., 21, 53-56, 1994. |
Weisenstein, D., M.K.W. Ko, J.M. Rodriguez, and N.-D.
Sze, Impact of heterogeneous chemistry on mod-
el-calculated ozone change due to high speed civil
transport aircraft, Geophys. Res. Lett., 18, 1991-
1994, 1991. :
Weisenstein, D., M.K.W. Ko, and N.-D. Sze, The chlo-
rine budget of the present-day atmosphere: A
modeling study, J. Geophys. Res., 97, 2547-2559,
1992. i
Weisenstein, O.K., M.K.W. Ko, J.M. Rodriguez, and
N.-D. Sze, Effects on stratospheric: ozone from
high-speed civil transport: Sensitivity to strato-
spheric aerosol loading, J. Geophys. Res., 98,
23133-23140, 1993.
Wennberg, P.O., et ai, Removal of stratospheric O3 by
free radicals: In situ measurements of OH, HO2,
NO, NO2, CIO and BrO, Science, 266, 398-404,
1994. ;
Wilson, J.C., et al., In situ observations of aerosol and
chlorine monoxide after the 1991 eruption of Mt.
Pinatubo: Effects of reactions on sulfate aerosol,
Science, 261, 1140-1143, 1993. !
6.40
-------
Woodbridge, E.L., et al, Estimates of total organic and
inorganic chlorine in the lower stratosphere from
in situ measurements during AASEII, J. Geophys.
Res., in press, 1994.
WMO, Report of the International Ozone Trends Panel:
1988, World Meteorological Organization Global
Ozone Research and Monitoring Project - Report
No. 18, Washington, D.C., 1990.
WMO, Scientific Assessment of Ozone Depletion: 1991,
World Meteorological Organization Global Ozone
Research and Monitoring Project - Report No. 25,
Geneva, 1992.
Yang, H., E. Olaguer, and K.K. Tung, Simulation of the
present-day atmospheric ozone, odd nitrogen,
chlorine and other species using a coupled 2-D
model in isentropic coordinates, /. Atmos Sci 48
442-471, 1991.
STRATOSPHERIC MODELS
I
Zadorozhny, A.M., G.A. Tuchkov, V.N. Kikhtenko, J.
Lastovicka, J. Boska, and A. Novak, Nitric oxide
and lower ionosphere quantities during solar parti-
cle events of October 1989 after rocket and
ground-based measurements, / Atmos. Terr.
Phys., 54, 183-192, 1992.
Zander, R., M.R. Gunson, J.C. Foster, C.P. Rinsland, and
J. Namkung, Stratospheric CIONO2, HC1 and HF
concentration profiles derived from Atmospheric
Trace Molecule Spsctroscopy experiment Space-
lab 3 observations: An update, /. Geophys. Res
95,20519-20525,1990.
Zerefos, C.S., A.F. Bais, I.C. Ziomas, and R.D. Bojkov,
On the relative importance of quasi-biennial oscil-
lation and El Nino/Southern Oscillation in the
revised Dobsort total ozone records, J. Geophys
Res., 97, 10135-10144, 1992.
6.41
-------
-------
CHAPTER?
Model Simulations of Global
Tropospheric Ozone
Lead Author:
F. Stordal
Co-authors:
R.G. Derwent
I.S.A. Isaksen
D.Jacob
M. Kanakidou
J.A. Logan
MJ. Prather
Contributors:
T. Berntsen
G.P. Brasseur
PJ. Crutzen
J.S. Fuglestvedt
D.A. Hauglustaine
C.E. Johnson
K.S. Law
J. Lelieveld
J. Richardson
M. Roemer
A. Strand
DJ. Wuebbles
-------
-------
CHAPTER 7 j
MODEL SIMULATIONS OF GLOBAL TROPOSPHERIC OZONE
Contents ;
I
SCIENTIFIC SUMMARY !
? 7.1
7.1 INTRODUCTION !
i :..... 1...7.3
7.2 3-D SIMULATIONS OF THE PRESENT-DAY ATMOSPHERE- '
EVALUATION WITH OBSERVATIONS '_ I
7.2.1 Atmospheric Transport
7.2.2 Nitrogen Oxides ; _'_" ' '; 7A
7.2.3 Hydroxyl Radical ZZZZZ T ?'5
7.2.4 Continental-Scale Simulations of Ozone ' " 7"5
7.3 CURRENT TROPOSPHERIC OZONE MODELING ..j
7.3.1 Global and Continental-Scale Models J
7.3.2 Limitations in Global Models ;
' 7.12
7.4 APPLICATIONS I
7.4.1 'Global Tropospheric OH 1 7'13
7.4.2 Budgets of NOy ZZZZZ r ?'13
7.4.3 Changes in Tropospheric UV ZZ 1 ?"14
7.4.4 Changes Since Pre-industrial Times '". [ 7'15
7.5 INTERCOMPARISON OF TROPOSPHERIC CHEMISTRY/TRANSPORT MODE! S 7 16
7.5.1 PhotoComp: Intercomparison of Tropospheric Photochemistry , 7'17
7.5.2 Intercomparison of Transport: A Case Study of Radon ZZ T 7'19
7.5.3 Assessing the Impact of Methane Increases \ "
r 7.24
REFERENCES i
' r : 7.29
-------
-------
TROPOSPHERIC MODELS
SCIENTIFIC SUMMARY
r
,D models cmnot tmspott
7J
-------
-------
TROPOSPHERIC MODELS
7.1 INTRODUCTION
Tropospheric models contain mathematical for-
mulations of the life cycles of the major tropospheric
source gases and the photochemistry, transport, and sur-
face exchange processes that couple them together and
to the life cycle of tropospheric ozone. They are used to
quantify the importance of the various terms in the life
cycles and budgets for ozone as well as for methane and
other ozone precursors. They allow an estimation of the
concentration distribution of the main tropospheric oxi-
dant, the hydroxyl (OH) radical in the troposphere, and
of the processes by which it is controlled. The strong
chemical tie between ozone and several other climate
gases causes tropospheric ozone to be very important in
the regulation of the Earth's climate. This indirect cli-
matic role of ozone comes in addition to the direct
climate effect of ozone due to its radiative properties.
The processes governing the tropospheric ozone
budget are described in Chapter 5 and summarized in
Figure 7-1. A substantial amount of the tropospheric
ozone is produced in the stratosphere and transported to
the troposphere at high and middle latitudes. The in situ
photochemical production is several times larger than
the import from the stratosphere, but is to a large extent
counteracted by chemical loss. The relative importance
of these processes to the ozone budget remains a topic
for future research. In the boundary layer, ozone is de-
posited at the surface and produced on urban and
regional scales that are not adequately resolved in global
models. Transport processes, especially vertical trans-
port of O3 and its shorter-lived precursors such as
NOX (=NO+NO2) and non-methane hydrocarbons
(NMHC), affect tropospheric chemistry and determine
the level of change in O3 concentration in the upper tro-
posphere, thereby strongly influencing the global budget
of tropospheric ozone.
The development of our understanding of the tro-
pospheric chemistry of ozone has been driven forward
Global Ozone Models - Components
Import from
the stratosphere
Chemical
Production
fCH4 -1
NOX,< CO >,hu
I.NMHCJ
Chemical
Destruction
O3, H,O, hu
Regional export
of ozone and
precursors
I Export to the
| stratosphere
High latitudes
Low latitudes
r°cestsef governing the global tropospheric ozone budget. The major components are import
and loss'deposition at the ground-*nd 020ne
7.3
-------
TROPOSPHERIC MODELS
by a combination of careful field observation, laboratory
investigation, and theoretical modeling. Modeling may
point to hitherto undiscovered relationships between
trace gases and processes, and observations can chal-t
lenge our theoretical understanding, leading to the
development of a more complete explanation of atmo-
spheric systems.
Recently, theoretical modeling has been given
heightened importance, particularly for ozone, through
its role in explaining the relationship between atmo-
spheric composition and the emissions of trace gases
from human activities. Theoretical modeling offers the
prospect of being able to unravel the cause of the trends
and the possible role of human activities in them. This
naturally leads to an important question as to whether
any of these observed trends will continue in the foresee-
able future. Furthermore, models offer the possibility to
estimate future changes in ozone resulting from changes
in emissions of ozone precursors.
Whether any of the conclusions derived from
models concerning trends in ozone concentrations actu-
ally describe what happens in the real atmosphere
depends on the adequacy and completeness of their for-
mulation, which is tied to our understanding of physical
and chemical processes in the troposphere, as well as on
the accuracy of their input data. Testing of models in-
volves comparison with observations, which are
inevitably limited in their accuracy and coverage in
space or time. In a complex model, it is difficult to ex-
plain the good agreement with observations often found
in some atmospheric regions for some species and the
rather poorer agreement sometimes found elsewhere.
Tropospheric models are in their infancy at present; the
global data sets required to validate them adequately are
not yet available, nor is the computer capacity to handle
all the processes that are believed to be important.
This chapter briefly surveys, in Section 7.2, the
successes and problems revealed in recent 3-D model
simulations of transport and chemistry in the tropo-
sphere. Most attention has been devoted to global
models. In order to successfully model troposphere
ozone globally, it is necessary to describe regional
ozone, since the global picture is only a conjunction of
regional parts. A few recent continental-scale model
studies are therefore also assessed in Section 7.2.
A range of global-scale models have been used for
studies of tropospheric ozone. Section 7.3 presents a
short compilation of such 2-D and 3-D models. The sec-
tion compares the zonally averaged ozone distribution
and budget terms for stratosphere/troposphere exchange
fluxes, chemical production and loss, and surface depo-
• sition in several of the models currently used for global
ozone studies. A survey of the major limitations in cur-
rent models is also included in Section 7.3, whereas
Section 7.4 presents global model integration of some
selected applications of key relevance for past, current,
and future tropospheric ozone. ;
As a part of the IPCC (1994) assessment as well as
this assessment (Section 7.5), a comparison of global
chemical models, that were used to calculate the effects
of changes in methane (CH4) on chemistry and climate
forcing, was performed. Two standard atmospheric
simulations were specified as part of the model inter-
comparison: global transport of short-lived, gases, and
photodissociation and chemical tendencies in tropo-
spheric air parcels.
A third model intercomparison on simulation of a
methane increase in today's atmosphere (also a part of
IPCC, 1994) is also included in Section 7.5. This serves
as the only example in this chapter of possible future
changes in tropospheric ozone due to changes in ozone
precursors. The previous ozone assessment (WMO,
1992) includes a thorough discussion of future changes
in ozone due to changes in several precursor gases.
7.2 3-D SIMULATIONS OF THE PRESENT-DAY
ATMOSPHERE: EVALUATION WITH
OBSERVATIONS
7.2.1 Atmospheric Transport
Transport of chemical species in global 3-D
models includes terms from both the grid-resolved circu-
lation (winds) and from parameterized subgrid processes
(convection, small-scale eddies). A number of recent
studies have used chemical tracers with'well-known
sources and sinks to test specific features of model
transport: interhemispheric exchange with chlorofluoro-
carbons (CFCs) and 85Kr (Prather et a/., I987; Jacob et
aL, 1987); convection over continents and long-range
transport of continental air to the oceans with 222Rn (Ja-
cob and Prather, 1990; Feichter and Crutzen, 1990;
Balkanski and Jacob, 1990; Balkanski et al, 1992);
transport and deposition of aerosols with 2'°Pb and 7Be
7.4
-------
1
(Brostetal., 1991;Feichterefa/., 1991;BalkanskT«a/L.
1993). These simulations show that global 3-D models
can provide a credible representation of atmospheric
transport on both global and regional scales. Some ma-
jor difficulties remain in simulating subgrid processes
involved in interhemispheric exchange, convective mass
transport, and wet deposition of aerosols. Work is also
needed to test the simulation of stratosphere-troposphere
exchange; chemical tracers such as bomb-generated
14CC>2 can be used for that purpose.
7.2.2 Nitrogen Oxides
Global 3-D simulations of NOX and nitric acid
(HNO3) including sources from combustion, lightning,
soils, and stratospheric injection have been reported by
Crutzen and Zimmerman (1991) and Penner et al.
(1991). The Geophysical Fluid Dynamics Laboratory
(GFDL) 3-D model with three transported species (NOX,
peroxyacetyl nitrate (PAN), and HNO3) has been used to
simulate the global distributions of NOy and individual
reactive nitrogen species resulting from stratospheric in-
jection (Kasibhatla et al., 1991), fossil fuel combustion
(Kasibhatla era/., 1993), and aircraft (Kasibhatla, 1993).
The same model including all sources of NOX has been
used to simulate the pre-industrial, present, and future
deposition of nitrate (Galloway et al., 1994) and the im-
pact of pollution-generated 03 on the world's crop
production (Chameides et al., 1994). The Oslo 3-D
model has been used to study the global distribution of
NOX and NQy (Berntsen and Isaksen, 1994).
The models of NOX and NOy have been, in gener-
al, fairly successful at reproducing observations in
polluted regions. Concentrations of NOX in remote re-
gions of the troposphere (e.g., the south Pacific) tend to
be underestimated, sometimes by more than an order of
magnitude. Possible explanations include an underesti-
mate of the lightning source (Penner et al., 1991), and
chemical cycling between NOX and its oxidation prod-
ucts by mechanisms that are not yet well understood
(Chatfield, 1994; Fan et al., 1994).
7.2.3 Hydroxyl Radical
Estimates of the global OH distribution have been
made in a number of 3-D model studies of long-lived
gases removed from the atmosphere by reaction with OH
(Spivakovsky etal., 1990a, b; Crutzen and Zimmerman,
ROPOSPHERIC MODELS
1991; Fung et al., 1991; tie et al., 1992; Easter et al
1993; Berntsen and Isaksen, 1994). These estimates
have generally been done by using climatological distri-
butions for the principal chemical variables involved in
OH production and loss (O3, NOX, CO, Cft,) and com-
puting OH concentrations :with a photochemical model.
Exceptions are the works I of Crutzen and Zimmerman
(1991) and Berntsen and Isaksen (1994), where O3 and
NOX concentrations were computed within the model in
a manner consistent with the computation of OH con-
centrations. The accuracy of the global mean OH
concentration obtained by the various models appears to
be within 30%, as indicated by simulations of methyl
chloroform, CH3CC13 (Spivakovsky et al., 1990a; Tie et
al., 1992). The seasonality of OH at midlatitudes ap-
pears to be well captured, as indicated by a recent
simulation of I4CO (Spivakovsky and Balkanski, 1994).
7.2.4 Continental-Scale Simulations of Ozone
t
The budget of ozone over the North American con-
tinent in summer was examined recently using the
results of a 3-D model simulation (Jacob etal., 1993a, b).
The model was evaluated by comparison with measure-
ments of ozone, NOX, carbon monoxide (CO), and
hydrocarbons. The model captures successfully the de-
velopment of regional high-ozone episodes over the
eastern U.S. on the back side of weak, warm, stagnant
anticyclones. Ozone production over the U.S. is strong-
ly NOx-lirnited, reflecting the dominance of rural areas
as sources of ozone on the regional scale. About 70% of
the net ozone production inithe U.S. boundary layer is
exported, while the rest is deposited within the region.
Only 6% of NOX emitted injthe U.S. is exported out of
the boundary layer as NOX or peroxy-acyl nitrates (e.g.,
PAN), but this export contributes disproportionately to
the U.S. influence on global tropospheric ozone because
of the high ozone production efficiency per unit NOX in
remote air. Jacob et al. (1993b) estimate that export of
U.S. pollution supplies 35 Tg ozone to the global tropo-
sphere in summer (90 days),;half of which is produced
downwind of the U.S., following export of NOX. Recent
comparison of O3-CO correlation in the model and in the
observations at sites in the United States and downwind
lends support to the model estimate for export of O3 pol-
lution from North America (Chin et al., 1994).
The ozone model of EMEP MSC-W (European
Monitoring and Evaluation Programme, Meteorological
7.5
-------
TROPOSPHERIC MODELS
Synthesizing Centre-West) has been used to study pho-
tochemistry over Europe for two extended summer
periods in 1985 and 1989 (Simpson, 1993), in combina-
tion with observations made in the EMEP program. The
model describes the boundary layer, combining trajecto-
ries in a regular geographical grid over Europe. It is
different from the models listed in Table 7-1, and is lim-
ited in the context of large-scale ozone modeling mainly
by its neglect of explicit representation of free tropo-
spheric processes. Significant differences in the
concentrations of the photo-oxidants were observed and
modeled between the two summer seasons that were
studied. The modeled ozone concentrations compare
satisfactorily with observations, particularly in 1989.
The study showed that NOX limits ozone formation in
the European boundary layer in most locations, whereas
NMHCs limit the production mostly in polluted areas.
Flat0y et al. (1994) present results from a set of
simulations with a three-dimensional mesoscale chemis-
try transport model driven by meteorological data from a
numerical weather prediction model with an extensive
treatment of cloud physics and precipitation processes.
New formulations for the vertical transport of chemical
tracers in connection with convective plumes and the
compensating sinking motion, and the calculation of
photolysis rates in clouds, are employed. The chemistry
transport model is used to calculate ozone and other
chemical species over Europe over a 10-day period in
July 1991, characterized by warm weather and frequent
cumulus episodes. When modeled vertical ozone pro-
files are compared to ozone soundings, better correlation
is found than for calculation without convection, indicat-
ing that physical processes, especially convection, can
dominate in the vertical distribution of ozone in the free
troposphere, and that sinking air that compensates for
convective updrafts is important for the tropospheric
ozone budget.
7.3 CURRENT TROPOSPHERIC OZONE
MODELING
Modeling tropospheric ozone is probably one of
the most difficult tasks in atmospheric chemistry. This is
due not only to the large number of processes that con-
trol tropospheric ozone, but even more to interactions of
processes occurring on different spatial and temporal
scales (Section 7.1, Figure 7-1). The field of tropospher-
ic ozone modeling is currently under rapid development.
To cover various spatial scales with limited com-
puter resources, different types of models have been
used. 2-D models have been widely used for several
years to study tropospheric ozone on a global scale. 3-D
models covering the global scale have only recently been
developed. An accurate representation of the 3-D trans-
port is needed in models, especially in order to describe
distributions of species with a chemical lifetime of the
order of days or weeks (like NOX and ozone) in areas
where the transport is efficient, as, e.g., in convective
cells.
7.3.1 Global and Continental-Scale Models
CATEGORIES OF MODELS
Several chemistry transport models (CTMs) have
been used to study ozone and precursor molecules in the
troposphere and in general to understand processes and
budgets of atmospheric constituents. A list of models is
given in Table 7-1, where models have been grouped in
four categories. ,
The first group is 2-D zonally averaged models.
Such models have been used for several years to study
global distributions of ozone and precursors in the cur-
rent atmosphere. To represent the various processes
explained in Figure 7-1, they contain detailed and rela-
tively similar schemes of ozone photochemistry. The
transport is described by a meridional circulation, and
relatively large diffusion is included to account for trans-
port due to wave activity. Only a few 2-D models
represent convection explicitly. Most of the models have
been used to study changes in ozone, some in the past
and most of the models in the future, due to changes in
emissions of ozone precursors (NOX, CH/v, CO, NMHC;
see Figure 7-1) and in physical variables such as temper-
ature, water vapor, and UV radiation. With currently
available computing resources, such models can, e.g., be
used to predict ozone changes over several decades for a
range of trace gas emission scenarios.
The next three categories contain 3-D models.
One group of 3-D models uses monthly averaged wind
fields to transport tracers, and therefore also need rela-
tively efficient diffusion to account for transport due to
winds that change on a day-to-day basis. However, the
7.6
-------
TROPOSPHERIC MODELS
Table 7-1. Current 2-D (global) and 3-D (global and mesoscale) Chemistry-Transport Models.
Model
2-D models
UK Met Office
Harwell
Univ Cambridge
Univ Oslo
Univ Bergen
TNO
NCAR/CNRS
MPI-tropo
LLNL
3-D monthly average
Moguntia
Images
3-D synoptic global
LLNL
GFDL/GIT
GISS/Harvard
Univ Oslo
3-D synoptic mesoscale
GISS/Harvard
Univ Bergen
References
Derwent (L994)
Johnson (1993); Johnson et al. (1992)
Law and Pyle (1993a, b) j
Fuglestvedt et al. (1994a, b) j
Strand and Hov (1993; 1994) \
Roemer and van der Hout (1992)
Hauglustaine et al. (1994) !
Singh and Kanakidou (1993); Kanakidou et al. (1991)
Wuebbles et al. (1993); Patten et dl. (1994)
Lelieveld (1994)
Mttller and Brasseur (1994)
Pennerera/. (1991; 1994)
Kasibhatla et al. (1991; 1993)
Spivakovsky et al. (1990a, b)
Berntsen and Isaksen (1994)
Jacob etal. (1993a, b)
Flat0y (1994); Flat0y et al. (1994)!
TNO = Netherlands Organization for Applied Scientific Research; NCAR = National Center for Atmospheric
Research; CNRS = Centre National de la Recherche Scientifique; MPI = Max-Planck Institute; LLNL =
Lawrence Livermore National Laboratory; GFDL = Geophysical Fluid Dynamics Laboratory; GIT = Georgia
Institute of Technology; GISS = Goddard Institute for Space Studies !
models in this category include detailed photochemical
schemes. In the last two categories, the models use daily
varying windfields and describe either the global scale or
mesoscales. Only recently, 3-D models of this category
have been developed to include detailed ozone chemis-
try. Applications and further development of such
models are expected in the near future. Some models
included in Table 7-1 have been used to study other trace
gases, e.g., NOy. Work is currently going on to include
ozone chemistry in some of these models.
MODELED OZONE DISTRIBUTIONS
Zonally averaged ozone distributions from several
of the models listed in Table 7-1 are shown in Figure 7-2.
The distributions that are shown are all for near-solstice
I
conditions, for January iknd July. Although all model re-
sults represent the current atmosphere, there are
differences between the: models in the choices of bound-
ary conditions and in the emissions of chemical ozone
precursors. j
The models agree on the general feature of the
zonally averaged ozone distribution. The vertical distri-
7.7
-------
TROPOSPHERIC MODELS
UKMO 2D Jonuory
is
•*yiu
90S 60S SOS 0 30N 60N 9071.
Oslo Univ 2D Jonuory
90S 60S SOS 0 JON 60N 90N
Bergen Univ 2D Jonuory
15
90S 60S JOS 0 SON 60K 90N
TNO 2D Jonuory
90S 60S 30S 0 JON 60N 90N
lotituda
15
I'0
UKMO 20 July
90S 60S 30S 0 JON EON SON
Oslo Univ 20 July
905 60S 305 0 JON 60N 90N
Bergen Univ 20 July
o
90S
60S 30S 0 SON 60N 90N
TNO 20 July
90S 60S 30S 0 JON 60N 90N
lotituda
Figure 7-2. Latitude by altitude contours of zonally averaged ozone mixing ratios as calculated in eight
global ozone models. The models are listed in Table 7-1. Data represent mid-January and mid-July condi-
tions for the current atmosphere. (Continued on page 7.9.) ;
7.8
-------
NCAR/CNRS 20 Jonuory
~ 10 •
15
90S 60S 30S 0 30N SON 90N
ILNL. 2D Jonuory
15
SOS 60S 30S 0 JON 60N 90N
IMAGES Jonuory
15
90S 60S 30S 0 30N 60N 90N
Oslo Univ 3D January
90S 60S 30S 0 30N 60N 90N
latitude
TROPOSPHERIC MODELS
NCAR/CNRS 2D July
15
90S 60S 30S 0 30N 60N 90S
LLNL 20 July
15
90S 60S 30S 0 SON 60N 90N
IMAGES July
15
90S 60S 30S 0 30N 60N 90N
Oslo Univ 3D July
90S 60S 30S 0 30N 60N 90N
latitude
Figure 7-2, continued.
7.9
-------
TROPOSPHERIC MODELS
bution, with maximum values in the upper troposphere
and minimum values at the surface, reflects mainly the
import of ozone from the stratosphere and deposition at
the ground. It is also clear that current global tiropo-
spheric ozone models are able to reproduce gross
features of observed ozone distributions (see Section
7.5.3 below).
The modeled mixing ratios in the tropics at the 10
km level are in the range 40-60 ppb and the boundary
layer values about 10-30 ppb. Generally the models give
higher ozone mixing ratios over the Northern Hemi-
sphere (NH) than over the Southern Hemisphere (SH)
during summertime. The modeled ozone levels in the
lowest few kilometers at northern middle latitudes are in
the range 30-50 ppb in July. In January the correspond-
ing values are 10-30 ppb in the SH. Comparison and
interpretation of the ozone levels in the region of largest
importance for radiative forcing (upper troposphere/
lower stratosphere) are difficult due to insufficient infor-
mation about the tropopause levels in the models. The
ozone levels in this region are to a high degree deter-
mined by processes in the lower stratosphere, where
ozone mixing ratios or fluxes through the tropopause are
fixed in most models. The latitudinal distribution varies
considerably between the models, reflecting clearly the
efficiency of the horizontal diffusion adopted in the
model, as discussed below in Section 7.5.2, with the
least latitudinal gradients in some of the 2-D models.
GLOBAL OZONE BUDGETS
From some of the models listed in Table 7-1, glo-
bal budget numbers are available that can be used to
explore the relative roles of the processes governing
tropospheric ozone, as explained in Figure 7-1. Strato-
sphere/troposphere exchange, photochemical reactions,
and surface deposition are identified as the three major
classes of processes governing the tropospheric ozone
budget. There are substantial differences between the
relative importance of these processes, in the way they
are represented in current models, as can be seen from
Table 7-2.
There is a factor 3 spread in the stratosphereftropo-
sphere exchange fluxes and the surface deposition values
between the models. This merely reflects the large un-
certainty in our knowledge of the efficiency of these
processes. The models usually either fix the flux
through the tropopause or fix the ozone mixing ratios in
the lower stratosphere, strongly tying the flux to obser-
vations. The most recent estimate of the ozone flux
across the tropopause is based on aircraft measurements
(Murphy etal., 1993; see discussion in Chapter 5), yield-
ing values in the range 240-820 Tg (O3)/yr, which are
comparable with or slightly less than previous estimates
(Danielsen and Mohnen, 1977; Gidel and Shapiro, 1980;
Mahlman et al., 1.980; see also Chapter 5). The spread in
values for surface deposition is presumably reflecting
differences in, e.g., vertical transport through the bound-
ary layer. Observations that can narrow the uncertainty
in its efficiency do not exist. There is currently therefore
little basis for judging which models calculate the most
realistic tropospheric ozone budget terms. ,
The even larger differences in the budgets for net
photochemical production of ozone (more than a factor
6) do not necessarily imply that the photochemical
schemes in the models are very different. The net pro-
duction is a small difference between large production
and sink terms. This is illustrated in Table'7-2, showing
also globally integrated values for the most important
individual source and loss mechanisms (see Chapter 5)
in one 2-D model (Derwent, 1994). In this model the
total production and the total loss is about 4 times larger
than the flux from the stratosphere, whereas the net pro-
duction comes out as a number that is much smaller than
the stratospheric flux.
It is obvious that differences in the import and ex-
port terms also influence the net chemical production,
since the budget balances in the models. A model that,
e.g., has a large import from the stratosphere or an ineffi-
cient deposition at the ground, estimates high ozone
concentrations in the troposphere, thereby increasing the
chemical loss, since the ozone (or excited atomic oxygen
produced from ozone) participates itself in the loss reac-
tions (see Chapter 5 and Table 7-2), and since the
photolysis of ozone initiates oxidation processes influ-
encing production as well as loss reactions for ozone.
ASPECTS OF ZONAL ASYMMETRIES ;
Two-dimensional tropospheric chemistry models
calculate zonally averaged trace gas distributions, and
therefore neglect zonal asymmetries. Yet they capture
the coarse features of the ozone distribution and they are
useful tools for sensitivity studies and analyses. Howev-
7.70
-------
TROPOSPHERIC MODELS
Table 7-2 Examples of globally integrated budget terms for tropospheric ozone for the
current and pre-mdustnal atmospheres, as calculated in various models, In TgT'
a)
b)
c)
(1)
(2)
(3)
'(4)
(5)
(6)
(7)
(8)
Model/Investigator
UOslo 3-D/Berntsen
Moguntia 3-D/Lelieveld
Cambridge 2-D/Law
UKMO 2-D/Derwent
TNO 2-D/Roemer
CNRS/NCAR 2-D/
Hauglustaine
UBergen 2-D/Strand
AER 2-D/Kotomarthi
Ref.
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
Present atmosphere
Chema
295
427
1021
343*
728
216
1404
416
Stratb
846
528
601
1077
962
408
533
610
Depc
-1178
-953
-1622
-1420
-1690
-612
-1937
-1026
! Pre-industrial
atmosphere
Chern
1
1 -87
, -195
-75
Strat
552
962
458
Dep
-465
-767
-424
Chem: The numbers represent net photochemical production, which is a small difference between large
production and loss terms (see text and the panel below). Since the budgets balance, net chemical
production must respond to stratosphere/troposphere exchange and surface deposition by changing the
ozone abundances. } B &
Strat: Net flux from the stratosphere to the troposphere, fixed or parameterized in the models
Dep: Surface deposition.
Berntsen and Isaksen (1994)
Lelieveld (1994).
Law and Pyle (I993a, b) + personal communication
Derwent (1994) + personal communication
Roemer and van der Hout (1992)
Hauglustaine et al. (1994)
Strand and Hov (1994)
Kotomarthi, personal communication. AER = Atmospheric and Environmental Research, Inc.
* Individual chemical terms are as follows:
Term
HO2 + NO
CH3O2 + NO
RO2 + NO
Total production
O('D) + H2O
O3 + OH
O3 + HO2
O3 + NMHC
O3 + NO (net loss)
O(3P) (net loss)
Total loss
Net chemical production
Strength
3117
1006
462
-1704
-410
-1719
-178
-177
-54
4585
-4242
343
7.77
-------
TROPOSPHERIC MODELS
er, quantitative assessments based on these models will
remain uncertain, due to limitations in their ability to
simulate realistically global transport of tracers (see Sec-
tion 7.5.2), and due to the lack of possibility to resolve
longitudinal variations in several species of key impor-
tance for the ozone chemistry, in particular in NOX
mixing ratios, observed between continental and oceanic
regions. The errors resulting from the assumption of
longitudinally uniform emissions have been evaluated
by use of a 3-D global monthly averaged model (Kana-
kidou and Crutzen, 1993). Longitudinally varying JNO,
CaHe, and C3H8 emissions lead to significantly lower O3
and OH concentrations, especially in the middle and low
troposphere in the tropics and at northern midlatitudes,
than when zonally averaged emissions were used. The
computed discord varies with latitude and height and
was locally as high as 80% for OH concentrations in the
tropics and 60% at midlatitudes.
On the other hand, there is no guarantee that 3-
dimensional models can simulate correctly the NOX
distributions and total nitrate observations in the tropo-
sphere and, in particular, in remote marine locations
(Pennerera/., 1991; Kasibhatla era/., 1993; Gallardo et
ai, 1994). This can be due to limitations in knowledge
of emissions, but also in the simulation of atmospheric
chemistry and transport from source areas in the model.
For example, Gallardo et al. (1994), testing various sce-
narios of distribution of NOX emissions from lightning,
found that a convection-related lightning distribution
could improve considerably the agreement with observa-
tions of NOX and total nitrate in remote oceanic areas.
THE ROLE OF CONVECTION
Representation of convection in global models re-
quires parameterization, due to the small scale of the
process. It therefore needs special consideration. One
consequence of convection is that ozone precursors
(NOX, CO, and NMHC), once they are transported to the
free troposphere, have a longer chemical lifetime, allow-
ing them to be transported over long distances and
contribute to ozone formation downwind of the convec-
tive cell. This has been confirmed in a model study of
deep tropical convection events observed during the
Amazon Boundary Layer Experiment 2A (ABLE 2A),
showing enhanced ozone formation in the middle and
upper troposphere (Pickering et al., 1992). Meteorolog-
ical and trace gas observations from convective episodes
were analyzed in the study with models of cumulus con-
vection and photochemistry. The level of formation of
free tropospheric ozone was shown to depend on the sur-
face trace gas emissions entrained in the ; cumulus
convective events.
On the other hand, boundary layer air poor in NOX
depresses the upper tropospheric ozone formation fol-
lowing convective events. This was found in a study
based on aircraft data from the Stratosphere-Tropo-
sphere Exchange Project (STEP) and the Equatorial
Mesoscale Experiment (EMEX) flights off northern
Australia, again using cumulus cloud and photochemical
models (Pickering et al., 1993). A 15-20% reduction in
the rate of O3 production between 15 and 17 km was the
largest perturbation calculated for these experiments due
to convection events. The study also showed that O3
production between 12 and 17 km would slow down by a
factor of 2 to 3 in the absence of NOX from lightning.
A secondary effect of deep cumulus convection is
associated with downward transport of ozone and NOX-
rich air from the upper troposphere in the cumulus
downdrafts. The downward transport of ozone and NOX
brings these components into regions where their life-
time is much shorter than in the upper troposphere.
Lelieveld and Crutzen (1994) have used a 3-D global
model to quantify this effect. Their calculations give a
decrease in total tropospheric ozone concentrations of
20% when deep convection is included in the calcula-
tions. The results are partly due to a corresponding
decrease in the global column of NOX of about 30%.
However, the decreased downward transport of ozone
resulted in an increase in the oxidation capacity of the
troposphere. Inclusion of convection increased methane
destruction by about 20% and CO destruction by about
10%.
7.3.2 Limitations in Global Models
Modeling of ozone production in the troposphere
is very sensitive to the assumed strengths and distribu-
tion of sources of ozone precursors. Estimates of
emissions, in particular of natural NOX (lightning and
surface sources) and hydrocarbons (isoprene and ter-
penes), are associated with large uncertainties that yield
uncertainties also in modeled ozone production. Accu-
rate emission data of O3 precursors are clearly needed to
correctly simulate tropospheric chemistry. \
7.12
-------
TROPOSPHERIC MODELS
A global CTM needs to represent the transport of
trace gases from their source to their sink regions. The
mass flux can be formulated as a function of the wind
velocity. The spatial and temporal resolution of wind
data, as provided by data assimilation or models, is limit-
ed. Advection in a CTM therefore captures only a
fraction of the total transport, and sub-grid processes
need to be parameterized, limiting the accuracy of trans-
port of ozone and other trace gases in the CTM. Such
sub-grid processes include transport within the boundary
layer, exchange between the boundary layer and the free
troposphere, convective transport, small-scale mixing
processes, and tropopause exchange. Inaccuracies in
model representation of sub-grid processes is of largest
importance for trace gases with lifetimes of the order of
days or weeks, like ozone and NOX. This is of particular
importance since NOX influences ozone chemically in a
nonlinear way.
A variety of heterogeneous chemical reactions can
affect the tropospheric ozone budget. The accuracy of
global ozone models is limited by the fact that the paths
and rates of such reactions are uncertain and by the fact
that such processes take place on spatial scales that are
not resolved in global models. Such heterogeneous reac-
tions include oxidation of N2O5 to nitrate, loss reactions
for ozone, removal of formaldehyde, and the separation
of chemical ozone precursors inside (HO2) and outside
(NO) cloud water droplets.
Gas phase chemical kinetics and photochemical
parameters are reasonably well established. However,
calculation of photodissociation rates is difficult in re-
gions with clouds and aerosols, due to difficulties in
model representation of optical properties, and the small
spatial and temporal scale of clouds.
A wide range of hydrocarbons take part in ozone
production in the troposphere in the presence of NOX.
The formulations of the degradation mechanisms of hy-
drocarbons can be important sources of uncertainty, in
tropospheric ozone models. The oxidation chains of the
dominant natural hydrocarbons, isoprene and terpenes,
are still not well known. Furthermore, the number of
emitted hydrocarbons is so large that they can only be
represented in models in groups (lumped species).
Finally, development of global ozone models is
hampered by the lack of extensive data sets for observed
species distributions. In order to test and validate mod-
els, measurements are heeded for several key species on
synoptic and even smaller scales.
i
7.4 APPLICATIONS !
i
Numerical models oifer the possibility to assess
the role of certain processes on a global scale. This sec-
tion presents some selected applications of models,
when they have been used to perform global integration
of key processes of importance for tropospheric chemis-
try and the ozone budget. ;
7.4.1 Global Tropospheric OH
The hydroxyl radical, OH,', is produced from O3
following photolysis to the Excited state O('D) and its
reaction with H2O. In turn,! the HOX family (OH, HO2,
H2O2) is involved in the production of tropospheric O3
in reactions with NOX (NO H- NQ2). The reaction of OH
with NO2 also provides the terminating step in NOx-cat-
alyzed production of O3 by converting NOX into HNO3.
Furthermore, HOX also removes ozone in NOx-poor en-
vironments. Thus the tropospheric chemistry of O3 and
OH are intertwined, and any possible calibration of mod-
eled OH adds confidence to the simulation of
tropospheric O3.
OH concentrations respond almost instantly to
variations in sunlight, H2O, O3, NO and NO2 (NOX),
CO, CH4, and NMHC; and therefore the OH field varies
by orders of magnitude in space and time. Observations
of OH can be used to test the! photochemical models un-
der specific circumstances, but are not capable of
measuring the global OH field. Therefore we must rely
on numerical models and suirogates to provide the glo-
bal and seasonal distribution1 of OH; these models need
to simulate the variations in Jsunlight caused by clouds
and time-of-day in addition to the chemical fields. These
calculations of global tropospheric OH and the conse-
quent derivations of lifetimes have not changed
significantly and are still much, the same as in the
AFEAS (Alternative Fluorocarbons Environmental Ac-
ceptability Study) Report (WMO, 1990); we cannot
expect such calculations to achie:ve an accuracy much
better than ±30%.
We can derive some properties of the global OH
distribution by observations of trace gases whose abun-
dance is controlled by reactions with OH, in conjunction
7.13
-------
TROPOSPHERIC MODELS
with some model for their emissions and atmospheric
mixing. For example, the trace gases methyl chloroform
(CH3CC13), 14CO, and HCFC-22 have been used to de-
rive empirical OH and thus test the modeled OH fields.
These gases (1) are moderately well mixed, (2) have well
calibrated and well measured atmospheric burdens, and
(3) have small or well defined other losses. However,
they primarily test only the globally, annually averaged
OH concentration, and even this average quantity is
weighted by the distribution and reaction rate coefficient
of OH with the gas. Some model studies have used
CH3CC13 (Spivakovsky et al, 1990a) and 14CO (Der-
went, 1994) to test their ab initio calculations of
tropospheric OH. Such studies have argued that the ob-
served seasonal distributions support the modeled
seasonal distribution of OH, but such seasonality also
results from transport and depends on the rate of mixing
between the aseasonal tropics and the midlatitudes.
The lifetimes for many ozone-depleting and
greenhouse gases depend on tropospheric OH, and
at this stage of model development we rely on the
empirical values. CH3CC13 fulfills all of the above re-
quirements for calibrating tropospheric OH. It has the
further advantage that its tropospheric distribution and
reaction rate are similar to many of the other gases in
which we are interested. A recent assessment (Kaye et
al., 1994) has reviewed and re-evaluated the lifetimes of
two major industrial halocarbons, methyl chloroform
(CH3CC13) and CFC-11. An optimal fit to the observed
concentrations of CHaCC^ from the five Atmospheric
Lifetime Experiment/Global Atmospheric Gases Experi-
ment (ALE/GAGE) surface sites over the period
1978-1990 was done with a pair of statistical/atmospher-
ic models (see Chapter 3 in Kaye et al., 1994). The
largest uncertainty in the empirical CH3CC13 lifetime,
5.4 ± 0.6 yr, lies currently with the absolute calibration.
The implication of a trend in this lifetime, presumably
due to a change in tropospheric OH (Prinn et al., 1992),
is sensitive to the choice of absolute calibration. Analy-
ses of the tropospheric budgets of the radio-isotope
. 14CO (Derwent, 1994) and HCFC-22 (Montzka et al.,
1993) complement th*is analysis and confirm the empiri-
cal estimate of tropospheric OH.
7.4.2 Budgets of NOy
Concentrations of NOX are critical for ozone pro-
duction. A central difficulty in modeling global ozone is
to predict the distribution of NOy components including
the large variability observed on small scales, the trans-
port out of the boundary layer, and chemical recycling of
nitrogen reservoir species. It is a problem that the rela-
tive roles of sources of tropospheric NOX (surface
emissions, lightning, transport from the stratosphere,
and aircraft emissions) in generating observed levels are
not quantitatively well known. This section describes a
few recent model studies addressing the role of emis-
sions, transport, and chemical conversion of reactive
nitrogen compounds.
A 3-D global chemistry-transport model has been
used to assess the impact of fossil fuel combustion emis-
sions on the fate and distribution of NOy components in
various regions of the troposphere (Kasibhatla et al.,
1993). It was found that wet and dry deposition of NOy
in source regions remove 30% and 40-45% of the emis-
sions, respectively, with the remainder being exported
over the adjacent ocean basins. The fossil fuel source
was found to account for a large fraction of the observed
surface concentrations and wet deposition fluxes of
HNO3 in the extratropical North Atlantic, but to have a
minor impact on NOy levels in the remote tropics and in
the Southern Hemisphere. :
Another global 3-D model study has calculated the
effect of organic nitrates, which can act as reservoirs for
NOX and therefore redistribute NOX in the troposphere
(Kanakidou et al., 1992). During their chemical forma-
tion, the organic nitrates may capture NOX that can be
released after transport and subsequent decomposition
away from source regions. The importance of hydrocar-
bons in the formation of peroxyacetyl nitrate (PAN),
which is the most abundant nitrate measured in the tro-
posphere, was demonstrated in the study, which also
included comparison with observations. According to
the model calculations, the efficiency of acetone in pro-
ducing PAN in the middle and high troposphere of the
NH ranges between 20 and 25%. This relationship be-
tween acetone and PAN concentrations has also been
observed during the Arctic Boundary Layer Expedition
(ABLE) 3B experiment. The observed concentrations of
acetone and PAN were much higher than those calculat-
ed by the model, which takes into account ethane and
propane photochemistry only. Consideration of the oxi-
dation of higher hydrocarbons and of direct emissions of
acetone is therefore needed to explain the observed con-
centrations (Singh et al., 1994).
7.14
-------
TROPOSPHERIC MODELS
An analysis of data from ABLE-3 A using a photo-
chemical model has shown that PAN and other organic
nitrates act as reservoir species at high latitudes for NOX
that is mainly of anthropogenic origin, with a minor
component from NOX of stratospheric origin (Jacob et
al., 1992). This tropospheric reservoir of nitrogen is
counteracting 03 photochemical loss over western Alas-
ka relative to a NOx-free environment The concentrations
of 03 in the Arctic and sub-arctic troposphere have been
found to be regulated mainly by input from the strato-
sphere and losses of comparable magnitude from
photochemistry and deposition (Singh et al., 1992; Ja-
cob etai, 1992).
Based on 2-D model calculations, the previous
ozone assessment (WMO, 1992) showed that injection
of NOX directly into the upper troposphere from com-
mercial aircraft is substantially more efficient in
producing ozone than surface-emitted NOX. Model tests
that have been performed show that the ozone-forming
potential of NOX emitted from airplanes depends on,
e.g., transport formulation, injection height, and the re-
moval rate. Furthermore, making reliable quantitative
estimates of the ozone production due to the aircraft
emission is also difficult, as it is nonlinear and it depends
strongly on the natural emissions and the background
concentrations of NOX, which are not well characterized
(see discussion in Chapter 11).
7.4.3 Changes in Tropospheric UV
Reductions in ozone column densities due to en-
hanced ozone loss in the stratosphere will lead to
enhanced UV penetration to the troposphere, causing
chemical changes. Such increased UV levels have been
observed in connection with reduced ozone column den-
sities during the last few years (WMO, 1992; Smith et
al., 1993; Kerr and McElroy, 1993; Gleasoo et al.,
1993). The significance for tropospheric chemistry of
enhanced UV fluxes is that they affect the lifetimes of
key atmospheric compounds like CO, CH4, NMHC,
hydrofluorocarbons (MFCs), and hydrochlorofluoro-
carbons (HCFCs) and the photochemical production and
loss of tropospheric ozone (Liu and Trainer, 1988; Briihl
and Crutzen, 1989; Madronich and Granier, 1992; Fu-
glestvedtefa/., 1994a). The main cause of this change is
that enhanced UV radiation increases O('D) production,
which in turn will lead to enhanced tropospheric OH
levels.
The atmospheric lifetimes of the above-mentioned
chemical compounds will be reduced since reactions
with OH represent the main sink. The reduced growth
rate of CHj that has been observed during the last decade
could, at least partly, be due to decreased lifetime result-
ing from enhanced UV fluxes. Fuglestvedt et al. (1994a)
have estimated that approximately 1/3 of the observed
reduction in growth rate during the 1980s is due to en-
hanced UV radiation resulting from reduced ozone
column densities over the same time period. In the same
study it was found that tropospheric ozone was reduced
in most regions. It was only at middle and high northern
latitudes during limited time periods in the spring where
NOX levels were sufficiently high, that ozone was in-
creased due to enhanced U V radiation. There might also
be significant changes in UV radiation, and thereby in
chemical activity, due to changes in the reflecting cloud
cover and due to backscatter by anthropogenic sulfate
aerosols (Liu et al., 1991). I Marked changes in the ratio
of scattered UV-B radiation to direct radiation have also
been observed in New Zealand after the Mount Pinatubo
eruption (see Chapter 9).
7.4.4 Changes Since Pre-industrial Times
Measurements of the Ichemical composition of air
samples extracted from ice eqres have been compared to
measurements of the present atmosphere, revealing that
methane volume mixing ratios have increased from
about 800 ppb to about 1700 ppb since the pre-industrial
period (see Chapter 2). The methane increase may have
reduced the OH concentration and the oxidizing effi-
ciency of the atmosphere. However, an increase in
production of ozone and thus also OH will also accom
pany growing CH* levels. As a result of industrialization,
extensive anthropogenic emiissions of the ozone precur-
sors CO, NOX, and NMHC have also occurred,
increasing ozone on a local and regional scale and, to a
lesser extent, on a global scale.
The temporal trends in tropospheric ozone in the
past are difficult ta calculate particularly because of the
critical role of surface NO* emissions. A few model
studies of impacts of anthropogenic emissions since pre-
industrial times predict large increases in ozone (Roemer
and van der Hout, 1992; Hauglustaine et al., 1994;
Lelieveld, 1994). The predicted changes in ozone during
the time of industrialization are not inconsistent with ob-
servations (see Chapter 1). Ipredictions of ozone change
7.15
-------
TROPOSPHERIC MODELS
have, however, only been made with a limited number of
models. It has been done with models describing in prin-
ciple the main processes governing the ozone budget.
However, more detailed models are needed to calculate
quantitatively reliable temporal trends in ozone.
A few models have estimated globally averaged
strengths of the various budget terms for ozone under
prc-industrial conditions. The numbers are given in Ta-
ble 7-2. Only changes in emissions of gases like CRj,
CO, NOX, and NMHC have been considered in the mod-
el experiments, whereas the meteorology and the
transport of atmospheric species have been assumed to
be unchanged. Despite the wide spread in the strength of
individual processes regulating global tropospheric
ozone (Section 7.3.1), there is good agreement between
the models in the changes in the ozone chemistry that
may have occurred during the time of industrialization.
According to the model calculations, both production
and loss were weaker in pre-industrial times, and the
chemistry has changed from being a net sink to a net
source of ozone. The global burden of tropospheric
ozone increased in these model studies by 55-70% over
the time of industrialization, supporting the assumption
that the observed marked increase in ozone over the last
approximately 100 years has at least partly been due to
anthropogenic emissions.
In a review paper on the oxidizing capacity of the
atmosphere, Thompson (1992) compiled changes in. glo-
bal OH since pre-industrial times as calculated in several
global models. There is consensus that OH has de-
creased globally since the pre-industrial times.
However, there is a substantial spread in the estimates,
which range from only a few to about 20%.
7.5 INTERCOMPAR1SON OF TROPOSPHERIC
CHEMISTRY/TRANSPORT MODELS
The observed changes in the cycles of many atmo-
spheric trace gases are expected, and often observed, to
produce a chemical response. For example, we have ac-
cumulated evidence that tropospheric ozone in the
northern midlatitudes has increased substantially, on the
order of 25 ppb, since pre-industrial times. During this
period, the global atmospheric concentration of CHLt has
increased regularly, and the emissions of NOX and
NMHC, at least over northern midlatitudes, have: also
increased greatly. An accounting of the cause of die 03
increases, particularly to any specific emissions, re-
quires a global tropospheric CTM, preferably a 3-D
model. A CTM provides the framework for coupling
different chemical perturbations that are, by definition,
indirect and thus cannot be evaluated simply with linear,
empirical analyses. We are placing an increasing re-
sponsibility on CTM simulations of the atmosphere
(e.g., the GWP calculation for CHt in IPCC 1994) and
should therefore ask how much confidence we have in
these models. Models of tropospheric chemistry and
transport have not been adequately tested in comparison
with those stratospheric models used to assess ozone de-
pletion associated with CFCs (e.g., WMO, 1990, 1992;
Prather and Remsberg, 1992). In addition; the greater
heterogeneity within the troposphere (e.g., clouds, con-
vection, continental versus marine boundary layer)
makes modeling and diagnosing the important chemical
processes more difficult. This section presents a begin-
ning, objective evaluation of the global CTMs that
simulate tropospheric ozone.
There are numerous published examples of indi-
vidual model predictions of the changes in tropospheric
03 and OH in response to a perturbation (e.g., pre-indus-
trial to present, doubling CHLj, aircraft' or surface
combustion NOX, stratospheric ©3 depletion). Since
these calculations in general used different assumptions
about the perturbants or the background atmosphere, it is
difficult to use these results to derive an \ assessment.
Further, we need to evaluate how representative those
models are with a more controlled set of simulations and
diagnostics.
Thus, two model intercomparisons and one assess-
ment are included as part of this report: (1) prescribed
tropospheric photochemical calculations that test the
modeling of 03 production and loss; (2) transport of
short-lived radon-222 that highlights differences in
transport description between 2-D and 3-D global chem-
ical tracer models in the troposphere; and (3) assessing
the impact of a 20% increase in CH4 on tropospheric O3
and OH. All of these studies were initiated as blind inter-
comparisons, wirn model groups submitting results
before seeing those of others. The call for participation
in (1) and (3) and preliminary specifications went out in
June 1993; the first collation of results was reported to
the participants in January 1994; and the final deadline
for submissions to this report was June 1994. In the
transport study, no obvious mistakes in performing the
7.16
-------
Table 7-3. Initial values used in PhotoComp.
TROPOSPHERIC MODELS
ALTITUDE
(km)
T(K)
p (mbar)
N (#/cm3)
H2O (% v/v)
03 (ppb)
NOX (ppt)
HN03 (ppt)
CO (ppb)
CH4 (ppb)
NMHC
MARINE
0
288.15
1013.25
2.55E19
1.0
30
10
100
100
1700
none
LAND+BIO
0
288.15
1013.25
2.55E19
1.0
30
200
100
100
1700
none
FREE
8
236.21
356.50
1.09E19
0.05
100
100
100
100
1700
none
; PLUME/X +
PLUME/HC
4
i
; 262.17
i 616.60
; 1.70E19
i 0.25
> 50
; 10000
! 100
600
1700
H2 = 0.5 ppm; H2C>2 = 2 ppb for all cases. BIO case equals LAND but with 1 ppb isioprene.
PLUME without NMHCs (/X) and with NMHCs (/HC). Initial values of NMHC (ppb): C2H6 = 25, C2H4 = 40,
C2H2 = 15, C3H8 = 15, C3H6 = 12.5, C4H10 = 5, toluene = 2, isoprene = 0.5.
(Note: Integrations were performed for 5 days starting July 1, with solar zenith angle 22 degrees.)
case studies were found, and detailed results will be pub-
lished as a WCRP (World Climate Research
Programme) workshop report. In the photochemical
study and methane assessment, about half of the results
contained some obvious errors in the setup, diagnosis, or
model formulation that were found in January 1994.
Most participants identified these errors and chose to re-
submit new results. Removal or correction of obvious
errors did not eliminate discrepancies among the mod-
els, and significant differences still remain and are
presented here. The current list of contributions is iden-
tical to the parallel IPCC Assessment. The combination
of these intercomparisons provides an objective, first
look at the consistency across current global tropospher-
ic chemical models.
7.5.1 PhotoComp: Intercomparison of
Tropospheric Photochemistry
An evaluation of the chemistry in the global CTMs
is not easy. There are no clear observational tests of the
rapid photochemistry of the troposphere that include the
net chemical tendency of O3 and are independent of
transport. Furthermore, uncertainties in the kinetic pa-
rameters would probably encompass a wide range of ob-
servations. Thus, we chose an engineering test
(PhotoComp) in which all chemical mechanisms and
data, along with initial conditions, were specified exact-
ly as in Table 7-3. Atmospheric conditions were
prescribed (July I, U.S. standard atmosphere with only
molecular scattering and O2 + O3 absorption) and the air
parcels with specified initial conditions were allowed to
evolve in isolation for five days with diurnally varying
photolysis rates (J's). PhotoComp becomes, then, a test
of the photochemical solvers used by the different
groups in which there is only one correct answer. For
most of these results, many models give similar answers,
resulting in a "band" of consensus, which we assume
here to be the correct numerical solution. The 23 differ-
ent model results submitted to PhotoComp are listed in
Table 7-4. '•
The PhotoComp cases were selected as examples
of different chemical environments in the troposphere.
The wet boundary layer is the most extensive, chemical-
ly active region of the troposphere. Representative
conditions for the low-NOx oceans (case: MARINE) and
the high-NOx continents (case: LAND) were picked. In
MARINE, ozone is lost rapidly (-1.4 ppb/ day), but in
7.77
-------
TROPOSPHERIC MODELS
Table 7-4. Models participating in the PhotoComp and delta-CH4 intercomparisons.
Code
A
B
B&
C
D
E
F
G
H
I
J
K
L
M&
N
O&
P
P&
Q
R&
S
T
T&
U
Y#
Z#
Affiliation
U. Mich.
UKMetO/UEAnglia
UEA-Harwell/2-D
U. Iowa
UC Irvine
NASA Langley
AER (box)
Harvard
NASA Ames
NYU-Albany
JQlich
GFDL
Ga. Tech.
U. Camb/2-D
U. Camb (box)
LLNU2-D
LLNL/3-D
H
NASA Goddard
AER/2-D
Cen. Faible Rad.
U. Oslo/3-D
M
NILU
U. Wash.
Ind. Inst. Tech.
Contributor*
Sandy Sillman
Dick Derwent
Claire Reeves
Gregory Carmichael
Michael Prather
Jennifer Richardson
Rao Kotamarthi
Larry Horowitz
Bob Chatfield
Shengxin Jin
Michael Kuhn
Lori Perliski
Prasad Kasibhatla
Kathy Law
Oliver Wild
Doug Kinnison
Joyce Penner
Cynthia Atherton
Anne Thompson
Rao Kotomarthi
Maria Kanakidou
Terje Berntsen
Ivar Isaksen
Frode Stordal
Hu Yang
Murari Lai
(e-mail)
(sillman@madlab.sprl.umich.edu)
(rgderwent@email.meto.govt.uk) :
(c.reeves @ uea.ac.uk)
(gcarmich@icaen.uiowa.edu) ;
(prather@halo.ps.uci.edu)
(richard@sparkle.larc.nasa.gov) •
(rao@aer.com) :
(lwh@hera.harvard.edu) i
(chatfield@clio.arc.nasa.gov) ;
0"in@mayfly.asrc.albany.edu) :
(ICH304@zam001.zam.kfa-juelich.de)
(lmp@gfdl.gov)
(psk@gfdl.gov) '
(kathy@atm.ch.cam.ac.uk)
(oliver@atm.ch.cam.ac.uk)
(dkin@cal-bears.llnl.gov [
(pennerl@llnl.gov)
(cyndi@tropos.llnl.gov) ;
(thompson@gatorl.gsfc.nasa.gov)
(rao@aer.com) ••
(mariak@asterix.saclay.cea.fr)
(terje.berntsen@geofysikk.uio.no)
(ivar.isaksen@geofysikk.uio.no))
(frode@nilu.no) '
(yang@amath.washington.edu) :
(mlal@netearth.iitd.ernet.in)
Notes:
* Only a single point-of-contact is given here; for other collaborators see appropriate references.
# Results for photolysis rates only.
& Also did delta-CH4 experiment in a 2-D.or 3-D model.
NYU = New York University; NILU = Norwegian Institute for Air Research
LAND, the initial NOX boosts ©3 levels. Over the conti-
nental boundary layer, NOX loss is rapid and the high
NOX levels must be maintained by local emissions. In
addition, these high-NOx, ozone-producing regions have
the capability of exporting significant amounts of 03
(and its precursors) to the free troposphere (Pickering et
a/., 1992; Jacob et al, 1993a, b). Rapid O3 formation
has been observed to occur in biomass burning plumes,
and the rate is predicted to depend critically on whether
hydrocarbons are present (PLUME/HC) or not (PLUME/
X). In the dry upper troposphere (FREE), 63 evolves
very slowly, less than 1%/day, even at NOX levels of 100
ppt.
The photolysis of 03 yielding O('D) is the first
step in generating OH, and it controls the net production
of 03. Tropospheric values peak at about 4-8 km be-
cause of molecular scattering. Model predictions for this
J at noon, shown in Figure 7-3a, fall within a band, ±20%
7.75
-------
TROPOSPHERIC MODELS
of the mean value, if a few outliers are not considered.
These differences are still large considering that all mod-
els purport to be making the same calculation. Another
key photolysis rate, that of NO2 in Figure 7-3b, shows a
similar range of results but with a more distinct pattern: a
majority clusters within 5% of one another, and the re-
maining results are systematically greater or smaller by
about 15%. It appears likely that this discrepancy may
be caused by the different treatments of scattering, be-
cause NO2 photolysis peaks at about 380 nm, where the
only significant cause of extinction is Rayleigh scatter-
ing. It is likely that such model differences could be
reduced to the 5% level with some modest effort
The photolysis of O3 and subsequent reaction with
H2O (reaction 2) drives the major loss of O3 in MARINE
(0 km, 10 ppt NOX), as shown in Figure 7-3c. The spread
in results after 5 days, 21 to 23 ppb, or ±12% in O3 loss,
does not seem to correlate with the O3 photolysis rates in
Figure 7-3a. Also shown in Figure 7-3c is the evolution
of 03 in LAND (0 km, 200 ppt NOX). The additional
NOX boosts O3 for a day or two and doubles the discrep-
ancy in the modeled ozone. The disagreement here is
important since most tropospheric O3 is destroyed under
these conditions in the wet, lower troposphere.
In the cold, dry upper troposphere, the net tenden-
cy of O3 is for slow loss, even with initially 100 ppt of
NOX, as shown for FREE in Figure 7-3d. The divergence
of results is disturbing but limited to a few models (i.e.,
losses after 5 days range from 2 to 4 ppb). These differ-
ences are not likely to affect the ozone budget for the
majority of models. In contrast, the production of O3 in
a N0x-rich PLUME (4 km, 10 ppb NOX) without non-
methane hydrocarbons is rapid and continues over 5
days, as shown in Figure 7-3e. Model agreement is ex-
cellent on the initial increases from 30 to 60 ppb O3 in 48
hours, but starts to' diverge as NOX levels fall. When
large amounts of NMHC are included in PLUME+HC
(also Figure 7-3e), ozone is produced and NOX depleted
rapidly, in less than one day. Differences among models
become much greater, in part because different chemical
mechanisms for NMHC oxidation were used. (The reac-
tion pathways and rate coefficients for chemistry with
CH4 as the only hydrocarbon have become standardized,
but different approaches are used for non-methane hy-
drocarbons.) Some of these differences become even
more obvious in the NOX predicted for PLUME+HC as
shown in Figure 7-3f. By day 3, NOX levels in individual
models are nearly constant but with a large range, from 0
to 50 ppt. ;
The 24-hour averaged OH concentrations are
shown for LAND and MARINE in Figure 7-3g. Values
are high for LAND during the first day and demonstrate
the dependence of OH on NOX, which begins at 200 ppt
and decays rapidly to about 10 ppt by day 4. The diver-
gence among LAND results is greater than MARINE, in
part due to the larger differences in the residual NOX left
from the initial value. Modeled OH values generally fall
within a ±20% band. Tlfiis variation in OH between
models, however, does notcorrelate obviously as expect-
ed with any other model differences such as the
photolysis rate of O3 or the abundance of NOX.
While O3 and OH may seem only moderately sen-
sitive to numerical treatment of the photochemistry,
some minor species appear to be less constrained. The
MARINE results for noontime formaldehyde (CH2O),
shown in Figure 7-3h, reach an approximate steady state
by day 5, but the range in model results is large, about a
factor of 2. !
These results show that basic model-to-model dif-
ferences of 30% or more exist In the calculations of O3
change and OH concentrations. This spread is not a true
scientific uncertainty; but presumably a result, of differ-
ent numerical methods that could be resolved given
some effort, although no single fix, such as O3 photoly-
sis rates, would appear to reduce the spread. A more
significant uncertainty in the current calculations of O3
tendencies is highlighted by the parallel experiments
with and without NMHC: tlje sources, transport and ox-
idation; and in particular the correlation of NOX
emissions and NMHC emissions on a fine scale, may
control the rate at which NOX produces O3.
7.5.2 Intercomparison of Transport:
A Case Study of Fladon
1
A critical element in calculating tropospheric
ozone is the transport of short-lived tracers such as NOX
and O3. The model comparison that tested atmospheric
transport was carried out primarily as a WCRP Work-
shop on short-range transport of greenhouse gases as a
follow up to a similar workshop on long-range transport
of CFC-11 (December 1991). A detailed description is
being prepared as a WCRP P'.eport. The basic intercom-
parison examined the global J distribution and variability
predicted for 222Rn emitted ubiquitously by decay of ra-
7.19
-------
TROPOSPHERIC MODELS
12
T fl)
^
1BOS 2E-05 3E-OS 4E-05
J(O3->O(1D)+O2) noon (/sec)
5E-05
12-
•c 8
-ffl 4'
TE
! TEC
...,R..
T feC
UlTFJ
i>
Ł
;D
"§
0.002 0.004 0.006 0.008 0.01 0.012 0.014
J(NO2->NO+O) noon (/sec)
100-
98-
w 96-
LU
111
Ł 94^
92
90
•X3GOO("
OH
CH
3
days
Figure 7-3. Results from the PhotoComp model intercomparison of 23 models (2 with only J-values); see
Table 7-4 for the key letters and Table 7-3 for the initial conditions. Photolysis (J) rates for 03 to O(1 D) (a) and
for NOa (b) are for local noon, July 1,45°N, U.S. Standard Atmosphere. Results are reported for altitudes of
0, 4, 8, and 12 km. For clarity, the letter codes have been offset in altitude here, and in time-of-day in
subsequent panels. Ozone mixing ratios are shown for noon in the boundary layer LAND (c, upper case
codes) and MARINE (c, lower case) cases, for the FREE troposphere (d) case, and finally for the biomass
(continued on page 7.21)
7.20
-------
TROPOSPHERIC MODELS
160
~. 160
i
120
i2 1°°
Si
2 60
20
.xxxxx
-JF-ft
B
L
......E.
$Ł
H
c
O
c
•fQ-N0-
—*—A—
days
o
Q.
inn
PO
80
70
60'
50'
in-
30'
20'
10
n-
i
1
Q
G"*""
U
.I..LL
*n r
I
I
r
i
i
Lc
, N
'G
u o
i T
"fT
c c
*•
i da
N
G
"u "o-
TV
n n
4
ys
N
c
^0
u
TT
L, LŁ
................
C
~""<^"o"
u
..Ł.... T..
* 'J
"• EH
6
g
0.12
0.00
035
Q.
&
00.25
0 0*
LU
^ 0.15
oc
2 ov
0 05-
o-
XXXXX i
_*_
— P"— ----- -
AB
T
:
! ..
f "
•i
i- 1
-c
j
D
_.
"fujO"
TT
1
j :
; da
"""
c-.-j..
-^
"^g"
f
: t
ys
i
C..J...
B
A 0
"•KR
u ^r
"W"
5
1
.-C...-L.
&5 D
-l^t-
6
Figure 7-3, continued. '
burning PLUME, without (e, lower case codes) and with NMHC (e, upper case). O3 was initialized at the 'X1
Noontime NOX mix.ng ratios are shown for the PLUME case with NMHC (f); whereas 24-hour average val
ues
7.27
-------
TROPOSPHERIC MODELS
Table 7*5. Models participating in the Rn/Pb transport Intel-comparison.
Model
Code
Contributor
CTMs established: 3-D synoptic
' CCM2
ECHAM3
GFDL
GISS/H/I
KNMI
LLNL/Lagrange
LLNL/Euler
LMD
TM2/Z
CTMs under development: 3-D synoptic
CCC
LaRC
LLNL/Impact
MRI
TOMCAT
UGAMP
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
CTMs used in assessments: 3-D/2-D monthly average
Moguntia/3-D
AER/2-D
UCamb/2-D
HarweIl/2-D
UWash/2-D
16
17
18
19
20
Rasch
Feichter/Koehler
Kasibhatla
Jacob/Prather
Verver
Penner/Dignon
Bergman
Genthon/Balkanski
Ramone/Balkanski/Monfray
Beagley
Grose
Rotman
Chiba
Chipperfield
P. Brown
Zimmerraann/Feichter
Shia
Law
Reeves
M. Brown
KNMI = Koninklijk Nederiands Meteorologisch Instituut; LaRC = NASA Langley Research Center
dium in soils. The radon is treated as an ideal gas with
constant residence time of 5.5 days. Although NOX
would seem a more relevant choice for these model com-
parisons, the large variations in the residence time for
NOX (e.g., <1 day in the boundary layer and 10 days in
the upper troposphere) make it difficult to prescribe: a
meaningful experiment without running realistic chem-
istry, a task beyond the capability of most of the
participating models. Furthermore, the nonlinearity of
the NOX-OH chemistry would require that all major
sources be included (see Chapter 5), which again is too
difficult for this model comparison.
Twenty atmospheric models-(both 3-D and 2-D)
participated in the radon/lead intercomparison for CTMs
(see Table 7-5). Most of the participants were using es-
tablished (i.e., published), synoptically varying (».&,
with daily weather) 3-D CTMs; several presented results
from new models under development. Among these syn-
optic CTMs, the circulation patterns represented the
entire range: grid-point and spectral, first generation
climate models (e.g., GFDL and GISS), newly devel-
oped climate models (e.g., CCM2 and ECHAM3), and
analyzed wind fields from ECMWF (European Centre
for Medium-Range Weather Forecasts) (e.g., TM2Z and
KNMI). One monthly averaged 3-D CTM and four lon-
gitudinally and monthly averaged 2-D models also
participated.
We have a limited record of measurements of
222Rn with which to test the model simulations; Some of
these data are for the surface above the continental
sources (e.g., Cincinnati, Ohio), and some are from is-
lands far from land sources (e.g., Crozet I.). The former
7.22
-------
TROPOSPHERIC MODELS
RADON-222 STATISTICS FOR JUN-AUG; MODELS (CASE A) AND OBSERVATIONS
300
200
100
0
i Cincinnati, mixed layer 2 pm
!
[ -ifiiiiisja-j
T
;
!'
,
hoss
I
• ™
1 1 -----
OBS 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
CM
01
2f
0
CC
LU
U
O
O
Rf\
OU
40
20
0
25
20
15
10
5
0
_ Crozet, surface T [396
- ~ T T I T' ,. T 1 I I T
^.. AAlJ.-,J,3Tl6 X JL i 1 _
tjrf
OBS 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Hawaii, 300 mb
I
i • l
T
B'T I T I T T ~, _, I :'
. -SSilBsBiii^^iiJ,!,,,^ .-
OBS . 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
2 n ml 7f^;S?T;Rl2^nCenlati? St!fiStiCS f°r Jun-Ju|-Au9 at Cincinnati, Ohio (40N 84W, mixed layer at
2 p m.), Crozet I (46S 51 E, surface), and over Hawaii (20N 155W, 300 mbar). Modeled time series show
S (S^d6d b°X)' and medians (white band)" IdentificaSon coSel Ire Jven in
Wa" falkanski er a/- 1 992> show ^e same statistics; but for Cincinnati (Gold
t H K -°X 9'!eS thf 'nterannual ran9e °^ June-August means; and for Crozet (Polian etal.,
the shaded box g,ves typical background concentrations with the top of the vertical bar, a typical
sites show a diurnal cycle, with large values at the sur-
face at night when vertical mixing is suppressed. The
latter sites show a very low-level background, with large
events lasting as long as a few days. An even more lim-
ited set of observations from aircraft over the Pacific
(e.g., 300 mbar over Hawaii) shows large variations with
small layers containing very high levels of radon, obvi-
ously of recent continental origin. A set of box plots in
Figure 7-4 summarizes the observations of radon at each
of these three sites and compares with model predictions
(see Table 7-5 for model codes). At Cincinnati, the syn-
optic 3-D CTMs generally reproduce the mean
afternoon concentrations in the boundary layer, although
some have clear problems! with excessive variability,
possibly with sampling the boundary layer in the after-
noon. At Crozet, most of the synoptic models can
reproduce the low background with occasional radon
"storms." In the upper troposphere over Hawaii, the one
set of aircraft observations shows occasional, extremely
high values, unmatched by any model; but the median
7.23
-------
TROPOSPHERIC MODELS
value is successfully simulated by several of the synoptic
3-D CTMs. The monfhlv averaged moH^s could not, of
course, simulate any of the time-varying observations.
The remarkable similarity of results from the syn-
optic CTMs for the free-tropospheric concentrations of
Rn in all three experiments was a surprise to most partic-
ipants. All of the established CTMs produced patterns
and amplitudes that agreed within a factor of two over a
dynamic range of more than 100. As an example, the
zonal mean Rn from case (i) for Dec-Jan-Feb is shown
for the CCM2 and ECHAM3 models in Figure 7-5a-b.
The two toothlike structures result from major tropical
convergence and convective uplift south of the equator
and the uplift over the Sahara in the north. This basic
pattern is reproduced by all the other synoptic CTMs. In
Jun-Jul-Aug (not shown) the 5-contour shifts north of
the equator, and again, the models produce similar pat-
terns. In contrast, the 2-D model results, shown for the
AER model in Figure 7-5c, have much smoother latitu-
dinal structures, do not show the same seasonally, and,
of course, cannot predict the large longitudinal gradients
expected for Rn (similar arguments hold for NOX; see
Kanakidou and Crutzen, 1993). Results from the Mo-
guntia CTM (monthly average 3-D winds) fell in
between these two extremes and could not represent
much of the structures and variations predicted by the
synoptic CTMs.
Such differences in transport are critical to this as-
sessment. Both NOX and O3 in the upper troposphere
have chemical time scales comparable to the rate of ver-
tical mixing, and the stratified layering seen in. the
monthly averaged models is likely to distort the impor-
tance of the relatively slow chemistry near the
tropopause. Compared with the synoptic models, it is
also obvious that the monthly averaged models would
transport surface-emitted NOX into the free troposphere
very differently, which may lead to inaccurate simula-
tion of total NOX concentrations. The 2-D models
appear to have a clear systematic bias favoring high-alti-
tude sources (e.g., stratosphere and aircraft) over surface
sources (e.g., combustion) and may also calculate a very
- different ozone response to the same NOX perturbations.
The participating synoptic CTMs are derived from
such a diverse range of circulation patterns and tracer
models that the universal agreement is not likely to be
fortuitous. It is unfortunate that we lack the observations
to test these predictions. Nevertheless, it is clear that the
currently tested 2-D models, and to a much lesser extent
the monthly averaged 3-D modeK have a fundamental
tlaw in transporting tracers predominantly by diffusion,
and they cannot simulate the global distribution of short-
lived species accurately. The currently tested synoptic
3-D CTMs are the only models that have the capability
of simulating the global-scale ransport of NOX and 03;
however, this capability will not be realized1 until these
models include better simulations of the boundary layer,
clouds, and chemical processes.
7.5.3 Assessing the Impact 01 Methane
Increases
The impacts of methane perturbations are felt
throughout all of atmospheric chemistry from the sur-
face to the exosphere, and most of these mechanisms are
Well understood. Quantification of these effects, how-
ever, is one of the classic problems in modeling
atmospheric chemistry. Similar to the ozone;studies not-
ed above, the published meth?""-change studies have
examined scenarios that range from 700 ppb. (pre-indus-
trial) to 1700 ppb (current) to a doubling by the year
2050 (e.g., WMO, 1992), but these scenarios:are not con-
sistent across models. This delta-rH4 study was designed
to provide a common framewo. evaluating the mul-
titude of indirect effects, especially changes in 03 and
OH, that are associated with an increase in CK*. The
study centers on today's atmosphere: use eiach model's
best simulation of the current atmosphere and then in-
crease the CH4 concentration (not fluxes) in the
troposphere by 20%, from 1715 ppb to 2058 ppb (ex-
pected in about 30 years based on observed 1980-1990
trend). This increase is small enough so that perturba-
tions to current atmospheric chemistry: should be
approximately linear. The history and protocol of the
delta-CH4 assessment is the same as that of PhotoComp
described above, and the six participating research
groups are also denoted in Table 7-4.
THE CURRENT ATMOSPHERE
Important diagnostics from delta-CH4 include 03
and NOX profiles for the current atmosphere, providing a
test of the realism of each model's simulation. Typical
profiles observed for O^ in the tropics and in northern
midlatitudes over America and Europe are shown in
Figure 7-6. The corresponding calculated 63 profiles.
7.24
-------
TROPOSPHERIC MODELS
~
u
a.
I
i i i i i | i i i i i i i i i i i i i i i i i i i i i i i i i
1000
90S 60S 30S 0 30N 60N 9QN
LATITUDE (DEGREES) !
LATITUDE (DEGREES)
200
400
600
800
1000
90°S 60°S 30°S EQ 30"N 60°N 90°N
LATITUDE (DEGREES) ••
Figure 7-5. Latitude by altitude contours of chemical transport model simulations of a continental source of
radon-222. Units are 1E-21 v/v. Zonal means for the period Dec-Jan-Feb are shovyn for two 3-D models, (a)
COM2 and (b) ECHAM3, and for (c) the AER 2-D model. These results are examples form a WCRP work-
shop on tracer transport. :
7.25
-------
TROPOSPHERIC MODELS
b
16
14-
12-
-§
"S 6-
Natal
Mar-Apr-May
20 40 60 80
Ozone (ppb)
100 120
10
9-
8-
7-
6-
©
-o
NH - Jul
0/0 i
0 20 40 60 80 100 120 140 160
Ozone (ppb)
Figure 7-6. Observed mean profiles of 0$ in the tropics (Natal, panel a) and at northern midlatitudes in July
(G = Goose Bay and H = Hohenpeissenberg, panel b). Data from northern stations were averaged over
1980-1991. Tropical station shows seasons of minimum (Mar-Apr-May) and maximum (Sep-Oct-Nov)
ozone. Source: Logan, 1994; Kirchhoff et at, 1990.
shown in Figure 7-7a-b, differ by almost a factor of two,
but encompass the observations. Clear divergence of re-
sults above 10 km altitude illustrates difficulties in
determining the transition between troposphere and
stratosphere. This exercise is only the beginning of an
objective evaluation of tropospheric ozone models
through comparison with measurements.
The modeled zonal-mean NOX profiles, shown in
Figure 7-7c, differ by up to almost a factor of 10. Com-
parisons in the lowest 2 km altitude are not meaningful
since the CTMs average regions of high urban pollution
with clean marine boundary layer. The range of mod-
eled NOX values in the free troposphere often falls
outside the range of typical observations, about 20 to 100
ppt (see Chapter 5).
03 PERTURBATIONS
The predicted changes in tropospheric 63 for Jun-
Jul-Aug in northern midlatitudes and the tropics are
shown in Figure 7-8a and 7-8b for the delta-CH4 study.
Ozone increases everywhere in the troposphere, by val-
ues ranging from about 0.5 ppb to more than 5 ppb. (The
extremely high values for model P in the upper tropo-
sphere must be considered cautiously since this recent
submission has not yet been scrutinized as much as the
other results.) In general the increase is larger at midlat-
itudes, but not for all models. Results for the southern
midlatitudes in summer (Dec-Jan-Feb) (not shown) are
similar to the northern.
The large spread in these results shows that our
ability to predict changes in tropospheric 63 induced by
CH4 perturbations is not very good. This conclusion is
not unexpected given the large range in modeled NOX
(Figure 7-7c), but.the differences in 03 perturbations do
not seem to correlate with the NOX distribution in the
models. Nevertheless, a consistent pattern of increases
in tropospheric 63, ranging from 0.5 to 2.5 ppb, occurs
throughout most of the troposphere. Our best estimate is
that a 20% increase in CH4 would lead to an increase in
ozone of about 1.5 ppb throughout most of the tropo-
7.26
-------
TROPOSPHERIC MODELS
1f>
g-
n-
c
^
R
-R
P
1
P
3 '
J !
5 6
delta-O3 (ppb) @ 12S-12N /JJA
12-
10-
8-
&
4-
2-
T""
T
-T
T"
T-r
fr
B h 6 f
f «-
B f! R°
r : t^j '
?....da.^..;
r ^6 !
..r. R..;
r fPp"
St4 r
RBOR :
0 i i
0 1 2
: 'P
fi
1--
B: Hawell/2D
R: AER/2D
M: UCambCD
O: LLNL/2D
P: LLNU3D
T: UOsloOD
P -*-
P — »•
3
3456
delta-03 (ppb) @ 35N-55N /JJA
7.27
-------
TROPOSPHER1C MODELS
Table 7-6. Inferred CH4 response time from delta-CH4 simulations.
Model code
B
M
O
P
R
(R)
T
FF*
-0.20%
-0.17%
-0.35%
-0:22%
-0.26%
-0.18%
-0.34%
RT/LT** !
1.29 ;
i.23# :
1.62
1.32
1.39
1.26# ;
1.61
Notes:
* FF = feedback factor, relative change (%) in the globally averaged CR* loss frequency (i.e., (OH)) for a
+1% increase in CRt concentrations
** RT = residence time and LT = lifetime
# Uses fixed CO concentrations, underestimates this ratio.
sphere in both tropics and summertime midlatitudes.
This indirect impact on the radiative forcing is about
25% ± 15% of that due to the 343 ppb increase in CR*
alone.
RESIDENCE TIME OF CH4 EMISSIONS
Methane is the only long-lived gas that has a clearly
identified, important chemical feedback: increases in at-
mospheric CH4 reduce tropospheric OH, increase the CHt
lifetime, and hence amplify the climatic and chemical im-
pacts of a CH4 perturbation (Isaksen and Hov, 1987;
Bemtsenefai, 1992). The delta-CH4 simulations from six
different 2-D and 3-D models show that these chemical
feedbacks change the relative loss rate for CH4 by -0.17%
to -0.35% for each 1% increase in CRt concentration, as
shown in Table 7-6. This range can reflect differences in
the modeled roles of Cfy, CO, and NMHC as sinks for OH
(Prather, 1994). For example, model M, with the smallest
feedback factor, 'has fixed the concentrations of CO; and
model R has shown that calculating CO instead with a flux
boundary condition (as most of the other models have
done) results in a larger feedback. These differences can-
not be resolved with this intercomparison, and this range
underestimates our uncertainty in this factor.
Recent theoretical analysis has shown that the
feedback factor (FF) defined in Table 7-6 can be used to
derive a residence time that accurately describes the
time scale for decay of a pulse of CH4 added to the
atmosphere. Effectively, a pulse of QHU, no matter how
small, reduces the global OH levels by a similar amount
(i.e., -0.3% per +1 %). This leads to the buildup of a cor-
responding increase in the already-existing atmospheric
reservoir of Cttt, that, in net, cannot be distinguished
from a longer residence time for the initial pulse. Thus
the residence time (RT) is longer than the lifetime (LT)
derived from the budget (i.e., total abundance divided by
total losses). Prather (1994) has shown that the ratio,
RT/LT, is equal to 1/(1 + FF) and that this residence time
applies to all CH4 perturbations, positive or "negative,"
no matter how small or large, as long as the change in
CH4 concentration is not so large as to change the feed-
back factor. Based on model results, this assumption
should apply at least over a ±30% change in current CH4
concentrations. Two of the models with results in Table
7-6 have shown that small CH4 perturbations decay with
the predicted residence time. ;
Based on these limited results, we choose 1.45 as
the best guess for the ratio RT/LT, with an uncertainty
bracket of 1.20 to 1.70. The budget lifetime of CH4 is
calculated to be about 9.4 yr, using the CH3CCl3 lifetime
as a standard for OH and including stratospheric and soil
losses. Thus, the residence time for any additional emis-
sions of CH4 is 13.6 yr (11.3-16.0 yr). This enhanced
time scale describes the effective duration for all current
7.28
-------
TRQPOSPHERIC MODELS
emissions of CH*; it is independent of other emissions as
long as current concentrations of CH*, within ±30%, are
maintained. Some of this effect was included in the pre-
vious assessment as an "indirect OH" enhancement to
the size of the CH4 perturbations. Here we recognize
that the OH chemical feedback gives a residence time for
CH4 emissions that is substantially longer than the life-
time used to derive the global budgets. This effective
lengthening of a CHLt pulse applies also to all induced
chemical perturbations such as tropospheric 03 and
stratospheric H2O.
REFERENCES
Balkanski, Y.J., and D.J. Jacob, Transport of continental
air to the subantarctic Indian Ocean, Tellus, 42B,
62-75, 1990.
Balkanski, Y.J., D.J. Jacob, R. Arimoto, and M.A. Kritz,
Long-range transport of radon-222 over the North
Pacific Ocean: Implications for continental influ-
ences, J. Atmos. Chem., 14, 353-374, 1992.
Balkanski, Y.J., D.J. Jacob, G.M. Gardner, W.M.
Graustein, and K.K. Turekian, Transport and resi-
dence times of continental aerosols inferred from a
global 3-dimensional simulation of 210Pb, J. Geo-
phys. Res., 98, 20573-20586, 1993.
Berntsen,T., J.S. Fuglestvedt, and I.S.A. Isaksen, Chem-
ical-dynamical modelling of the atmosphere with
emphasis on the methane oxidation, Ber. Bunseng-
es. Phys. Chem., 96, 3, 241-251, 1992.
Berntsen, T.K., and I.S.A. Isaksen, A 3-D photochemis-
try/transport model of the global troposphere,
Report No. 89, Institute for Geophysics, Universi-
ty of Oslo, Norway, 1994.
Brost, R.A., J. Feichter, and M. Heimann, Three-dimen-
sional simulation of 7Be in a global climate model,
J. Geophys. Res., 96, 22423-22445, 1991.
Briihl, C, and P.J. Crutzen, On the disproportionate role
of tropospheric ozone as a filter against solar UV-
B radiation, Geophys. Res. Lett., 17, 703-706,
1989.
Chameides, W.L., P.S. Kasibhatla, JJ. Yienger, H. Levy
II, and WJ. Moxim, The growth of continental-
scale metro-agro-plexes, regional ozone pollution,
and world food production, Science, 264, 74-77,
1994.
Chatfield, R.B., The anomalous HNOyNOx ratio of re-
mote tropospheric air: Is there conversion of nitric
acid to formic acid and NOX?, submitted to Geo-
phys. Res. Lett., 1994.
Chin, M., D.J. Jacob, J.W. Munger, D.D. Parrish, and
B.G. Doddridge, Relationship of ozone and car-
bon monoxide over North America, /. Geophys.
Res., 99, 14565-14573, 1994.
Crutzen, P.J., and P.H. Zimmermaoin, The changing pho-
tochemistry of the troposphere, Tellus, 43AB,
136-151,1991. \
Danielsen, E.F., and V. Mohnen, Project Duststorm:
Ozone transport, in situ measurements, and meteo-
rological analysis of tropopause folding, J.
Geophys. Res., 82, 5867-5877,1977.
Derwent, R.G., The influence of human activities on the
distribution of hydroxyl radicals in the tropo-
sphere, submitted to Philosophical Transactions
of the Royal Society, 1S>94.
Easter, R.C., R.D. Saylor, and E.G. Chapman, Analysis
of mid-tropospheric carbon monoxide data using a
three-dimensional global atmospheric chemistry
numerical model, in Proceedings 20th Inter. Tech.
Meeting on Air Pollution Modeling and Its Appli-
cation, Valencia, Spain.! 1993.
Fan, S.-M, D.J. Jacob, D.L. Mauzerall, J.D. Bradshaw,
ST. Sandholm, D.R. Blake, H.B. Singh, R.W. Tal-
bot, G.L. Gregory, and G.W. Sachse, Origin of
tropospheric NOX over subarctic eastern Canada in
summer,/. Geophys. Res., 99, 16867-16877, 1994.
Feichter, J., and P.J. Crutzen, Parameterization of vertical
tracer transport due to deep cumulus convection in a
global transport model and its evaluation with
222Radon measurements, Tellus, 42B, 100-117,
1990. !
Feichter, J., R.A. Brest, and M. Heimann, Three-dimen-
sional modeling of the concentration and deposition
of 210Pb aerosols, J. Geophys. Res., 96, 22447-
22460, 1991.
Flat0y, F., Modelling of Coupled Physical and Chemical
Processes in the Troposphere over Europe, Ph.D.
Thesis, Geophysical Institute, University of Ber-
gen, Norway, 1994. i
Flat0y, R, 0. Hov, and H. Sjnit, 3-D model studies of
vertical exchange processes in the troposphere
over Europe, submitted to /. Geophys. Res., 1994.
7.29
-------
TROPOSPHERIC MODELS
Fuglestvedt, J.S., J.E. Jonson, and I.S.A. Isaksen, Effects
of reductions in stratospheric ozone on tropo-
spheric chemistry through changes in photolysis
rates, Tellus, 46B, 172-192,1994a.
Fuglestvedt, J.S., I.E. Jonson, W.-C. Wang, and I.S.A.
Isaksen, Responses in tropospheric chemistry to
changes in UV fluxes, temperatures and water va-
pour densities, to appear in Atmospheric Ozone as
a Climate Gas, edited by W.-C. Wang and I.S.A.
Isaksen, NASA ASI Series, in press, 1994b.
Fung, L, J. John, J. Leraer, E. Matthews, M. Prather, L.P.
Steele, and P.J. Fraser, Three-dimensional model
synthesis of the global methane cycle, J. Geophys.
Res., 96, 13033-13066, 1991.
Gallardo, L., H. Rodhe, and P.J. Crutzen, Evaluation of a
global 3-D model of the tropospheric reactive ni-
trogen cycle, submitted to /. Geophys. Res., 1994.
Galloway, J.N., H. Levy II, and P.S. Kasibhatla, Year
2020: Consequences of population growth and
development on deposition of oxidized nitrogen,
Ambio. 23, 120-123,1994.
Gidel, L.T., and M.A. Shapiro, General circulation esti-
mates of the net vertical flux of ozone in the lower
stratosphere and the implications for the tropo-
spheric ozone budget, /. Geophys. Res., 85,
4049-4058, 1980.
Gleason, J.F., P.K. Bhartia, J.R. Herman, R. McPeters, P.
Newman, R.S. Stolarski, L. Flynn, G. Labow, D.
Larko, C. Seftor, C. Wellemeyer, W.D. Kornhyr,
A.J. Miller, and W. Planet, Record low global
ozone in 1992, Science, 260, 523-526, 1993.
Gold, S., H. Barkhau, W. Shleien, and B. Kahn, Mea-
surement of naturally occurring radionuclides in
air, in The Natural Radiation Environment, edited
by J.A.S. Adams and W.M. Lowder, University of
Chicago Press, 369-382, 1964.
Hauglustaine, D.A., C. Granier, G.P Brasseur, and G.
Mdgie, The importance of atmospheric chemistry
in the calculation of radiative forcing on the cli-
mate system, J. Geophys. Res., 99, 1173-1186,
1994.
IPCC, Intergovernmental Panel on Climate Change, Ra-
diative Forcing of Climate Change, The 1994
Report of the Scientific Assessment Working
Group of IPCC, 1994.
Isaksen, I.S.A., and 0. Hov, Calculation of trends in the
tropospheric concentration of 03, OH, CPLt, and
NOX, Tellus, 39B, 271-285, 1987.
Jacob, D.J., M.J. Prather, S.C. Wofsy, and M.B. McEl-
roy, Atmospheric distribution of 85Kr simulated
with a general circulation model, J. Geophys. Res.,
92, 6614-6626,1987. ;
Jacob, D.J., and M.J. Prather, Radon-222 as a test of con-
vection in a general circulation model, Tellus, 42,
118-134,1990.
Jacob, D.J., S.C. Wofsy, P.S. Bakwin, S.-M. Fan, R.C.
Harriss, R.W. Talbot, J.D. Bradshaw, S.T. Sand-
holm, H.B. Singh, E.V. Browell, G,L. Gregory,
G.W. Sachse, M.C. Shipham, D.RJ Blake, and
D.R. Fitzjarrald, Summertime photochemistry of
the troposphere at high northern latitudes, J. Geo-
phys. Res., 97, 16421-16431, 1992. :
Jacob, D.J., J.A. Logan, R.M. Yevich, G.M. Gardner,
C.M. Spivakovsky, S.C. Wofsy, J.W. Munger, S.
Sillman, M.J. Prather, M.O. Rodgers, H. West-
berg, and P.R. Zimmerman, Simulation of
summertime ozone over North America, /. Geo-
phys. Res., 98, 14797-14816, 1993a.
Jacob, D.J., J.A. Logan, G.M. Gardner, R.M. Yevich,
C.M. Spivakovsky, S.C. Wofsy, S. Sillman, and
M.J. Prather, Factors regulating ozone over the
United States and its export to the global atmo-
sphere, /. Geophys. Res., 98, 14817-14826,
1993b.
Johnson, C.E., The sensitivity of a two-dimensional
model of global tropospheric chemistry to emis-
sions at different latitudes, AEA Environment &
Energy, United Kingdom Atomic Energy Authori-
ty, AEA-EE-0407, 1993.
Johnson, C., J. Henshaw, and G. Mclnnes, Impact of air-
craft and surface emissions of nitrogen oxides on
tropospheric ozone and global warming, Nature,
355, 69-71, 1992.
Kanakidou, M., and P.J. Crutzen, Scale problems in glo-
bal tropospheric chemistry modelling: Comparison
of results obtained with a three-dimensional mod-
el, adopting longitudinally uniform and varying
emissions of NOX and NMHC, Chemosphere, 26
(1-4), 787-801, 1993.
7.30
-------
TROPOSPHEFHC MODELS
Kanakidou, M., H.B. Singh, K.M.-Valentin, and PJ.
Crutzen, A two-dimensional mode! study of •
ethane and propane oxidation in the troposphere,
J. Geophys. Res., 96, 15395-15413, 1991.
Kanakidou, M., PJ. Crutzen, P.H. Zimmermann, and B.
Bonsang, A 3-dimensional global study of the
photochemistry of ethane and propane in the tro-
posphere: Production and transport of organic
nitrogen compounds, in Air Pollution Modelling
and Its Application IX, edited by H. van Dop and
G. Kallos, Plenum Press, New York, 415-426,
1992.
Kasibhatla, P.S., H. Levy II, WJ. Moxim, and W.L. Cha-
meides, The relative impact of stratospheric
photochemical production on tropopsheric NOy
levels: A model study, J. Geophys. Res., 96,
18631-18646, 1991.
Kasibhatla, P.S., NOy from sub-sonic aircraft emissions:
A global three-dimensional model study, Geo-
phys. Res. Lea., 20, 1707-1710, 1993.
Kasibhatla, P.S., H. Levy II, and WJ. Moxim, Global
NOX, HNO3, PAN, and NOy distributions from
fossil fuel combustion emissions: A model study,
J. Geophys. Res., 98, 7165-7180, 1993.
Kaye, J., S. Penkett, and P.M. Ormond (eds.), Concen-
trations, Lifetimes and Trends of CFCs, Halons,
and Related Species, National Aeronautics and
Space Administration Reference Publication No.
1339, 1994.
Kerr, J.B., and C.T. McElroy, Evidence for large upward
trends of ultraviolet-B radiation linked to ozone
depletion, Science, 262, 1032-1034, 1993.
Kirchhoff, V.W.J.H., I.M.O. da Silva, and E.V. Browell,
Ozone measurements in Amazonia: Dry season
versus wet season, J. Geophys. Res., 95, 16913-
16926, 1990.
Law, K.S., and J.A. Pyle, Modelling trace gas budgets in
the troposphere 1. Ozone and odd nitrogen, J.
Geophys. Res., 98, 18377-18400, 1993a.
Law, K.S., and J.A. Pyle, Modelling trace gas budgets in
the troposphere 2. CH* and CO, J. Geophys. Res.,
98, 18401-18412, 1993b.
Lelieveld, J., Model simulations of the 1850 and 1990
atmosphere chemical compositions, accepted for
publication in /. Geophys. Res., 1994.
Lelieveld, J., and P. Crutzen, Role of deep convection in
the ozone budget of the troposphere, Science, 264,
1759-1761, 1994.
Liu, S.C., and M. Trainer, Responses of tropospheric
ozone and odd hydrogen radicals to column ozone
change, /. Atmos. Chem., 6, 221-233, 1988.
Liu, S.C., S.A. McKeen, and S. Madronich, Effect of
anthropogenic aerosols on biologically active ultra-
violet radiation, Geophys. Res. Lett., 18,2265-2268,
1991.
Logan, J.A., Trends in the vertical distribution of ozone:
An analysis of ozonesonde data, submitted to /.
Geophys. Res., 1994.
Madronich, S., and C. Granier, Impact of recent total
ozone changes on tropospheric ozone photodisso-
ciation, hydroxyl radicals, and methane trends,
Geophys. Res. Lett., 19, 465-467, 1992.
Mahlman, J.D., H. Levy II, and WJ. Moxim, Three-di-
mensional tracer structure and behavior as
simulated in two ozone precursor experiments, J.
Atmos. Sci., 37, 655-685, 1980.
Montzka, S.A., R.C. Myers, J.H. Butler, and J.W. Elkins,
Global tropospheric distribution and calibration
scale of HCFC-22, Geophys. Res. Lett., 20, 703-
706, 1993. ;
Murphy, D.M., D.W. Fahey, M.H. Proffitt, S.C. Liu,
K.R. Chan, C.S. Eubank, S.R. Kawa, and K.K.
Kelly, Reactive nitrogen and its correlation with
ozone in the lower stratosphere and the upper tro-
posphere,/. Geophys. Res.,.98, 8751-8773-, 1993.
Miiller, J.-F., and G. Brasseur, IMAGES: A three-dimen-
sional chemical transport model of the
troposphere, accepted for publication in /. Geo-
phys. Res., 1994.
Patten, K.O., PS. Connell, D.E. Kinnison, D.J. Wueb-
bles, L. Froidevaux, and T.G. Slanger, Effect of
vibrationally excited oxygen on ozone production
in the stratosphere,!/ Geophys. Res., 99, 1211-
1223, 1994.
Penner, I.E., C.S. Atherton, J. Dignon, S J. Ghan, JJ._
Walton, and S. Hameed, Tropospheric nitrogen: A
three-dimensional study of sources, distributions,
and deposition, /. Geophys. Res., 96, 959-990,
1991. I
7.31
-------
TROPOSPHERIC MODELS
Penner, I.E., C.S. Atherton, and T.E. Graedel, Global
emissions and models of photochemically active
compounds, in Global Atmospheric-Biospheric
Chemistry, edited by R. Prinn, Plenum Publ., New
York, 223-248, 1994.
Pickering, K.E., A.M. Thompson, J.R. Scala, W.-K. Tao,
' and J. Simpson, Ozone production potential fol-
lowing convective redistribution of biomass
•burning emissions, / Atmos. Chern., 14, 29T-313,
1992.
Pickering, K.E., A.M. Thompson, W.-K. Tao, and TJL.
Kucsera, Upper tropospheric ozone production
following mesoscale convection during STEP/
EMEX.7. Geophys. Res., 98, 8737-8749, 1993.
Polian, G., G. Lambert, B. Ardouin, and A. Jegou, Long-
range transport of radon in sub-Antarctic and
Antarctic areas, Tellns, 38B, 178-189, 1986.
Prather, M J., Lifetimes and eigenstates in atmospheric
chemistry, Geophys. Res. Lett., 21, 801-804,
1994.
Prather, M J., M. McEIroy, S. Wofsy, G. Russell, and D.
Rind, Chemistry of the global troposphere: Fluo-
rocarbons as tracers of air motion, J. Geophys.
Res.. 92, 6579-6613, 1987.
Prather, M 3., and E.E. Remsberg (eds.). The Atmospher-
ic Effects of Stratospheric Aircraft: Report of the
1992 Models and Measurements Workshop, Na-
tional Aeronautics and Space Administration
Reference Publication No. 1292, 1992.
Prinn, R., D. Cunnold, P. Simmonds, F. Alyea, R. Boldi,
A. Crawford, P. Fraser, D. Gutzler, D. Hartley, R.
Rosen, and R. Rasmussen, Global average concen-
tration and trend for hydroxyl radicals deduced
from ALE/GAGE trichloroethane (methyl chloro-
form) data for 1978-1990, J. Geophys. Res., 97,
2445-2461, 1992.
Roomer, M.G.M., and K.D. van der Hout, Emissions of
NMHCs and NOX and global ozone production, in
Proceedings of the 19thNATO/CCMS International
Technical Meeting on Air Pollution Modelling and
Its Application, North Atlantic Treaty Organiza-
, lion/Committee on the Challenges of Modem
Society, September 29-October 4, 1991, Crete,
Greece, Plenum Press, New York, 1992.
Simpson, D., Photochemical model calculations over
Europe for two extended summer periods: 1985
and 1989. Model results and comparison with ob-
servations, Atmos. Environ., 27A, 921-943, 1993.
Singh, H.B., D. O'Hara, D. Herlth, J.D. Bradshaw, S.T.
Sandholm, G.L. Gregory, G.W.. Sachse, D.R..
Blake, P.J. Crutzen, and M. Kanakidou, Atmo-
spheric measurements of peroxyacetyl nitrate and
other organic nitrates, at high-latitudes: Possible
sources and; smfcs, Ł Geophys. Res., 97, 16511-
16522, 1992.
Singh, H.B., and M. Kanakidou, An investigation of the
atmospheric sources and sinks of methyl bromide,
Geophys. Res. Lett., 20, 133-136, 1993.:
Singh, H.B., D. O'Hara, D. Herlth, W. Sachse, D.R.
Blake, J.D. Bradshaw, M. Kanakidou, and P.J.
Crutzen, Acetone in the atmosphere: Distribution,
sources and sinks, /. Geoghys* Res^f 99-; 1805-
1820, 1994. '
Smith, R..C.,. B.B. Prezelin, K.S. Baker, RJJ. :Bidigare,
N P. Boucher, T. Coley, D. Karstenz, S. Maclntyre,
H. Matlick, D. Menzies, M. OndrusekJ Z. Wan,
and KJ. Waters, Ozone depletion: Ultraviolet ra-
diation and phytoplankton biology in Antarctic
waters, Science, 255, 952-959, 1992.
Spivakovsky, C.M., R. Yevich, J.A. Logan, S.C. Wofsy,
M.B. McEIroy, and M.J. Prather, Tropospheric OH
in a three-dimensional chemical tracer model: An
assessment based on observations of CHjCCl^, J.
Geophys. Res., 95, 18441-18472, 1990a •
Spivakovsky, CM., S.C. Wofsy, and M.J. Prather, A nu-
merical method for parameterization of atmospheric
chemistry: Computation of tropospheric OH, J.
Geophys. Res., 95, 18433-18440, 1990b.
Spivakovsky, C.M., and YJ. Balkanski, Tropospheric
OH: Constraints imposed by observations of
14CO and methyl chloroform, in Report of the
WMO-Sponsored Meeting of Carbon Monoxide
Experts, edited by P.C. Novelli and R.M. Rosson,
Global Atmospheric Watch, World Meteorological
Organization, Geneva, in press, 1994.
Strand, A., and 0. Hov, A two-dimensional zonally aver-
aged transport model including convective
motions and a new strategy for the numerical solu- •
tion,/. Geophys. Res., 98, 9023-9037, 1993.
7.32
-------
Strand, A., and 0. Hov, Two-dimensional global study of
the tropospheric ozone production, accepted for
publication in J. Geophys. Res., 1994.
Tie, X., C.-Y. Kao, EJ. Mroz, R.J. Cicerone, EN. Alyea,
and D.M. Cunnold, Three-dimensional simula-
tions of atmospheric methyl chloroform: Effect of
an ocean sink, J. Geophys. Res., 97, 20751-20769,
1992.
Thompson, A.M., The oxidizing capacity of the Earth's
atmosphere: Probable past and future changes,
Science, 256, 1157-1165, 1992.
TROPOSPHERIC MODELS
WMO, Scientific Assessment of Ozone Depletion: 1989,
World Meteorological Organization Global Ozone
Research and Monitoring Project - Report No. 20,
Geneva, 1990.
WMO, Scientific Assessment of Ozone Depletion: 1991,
World Meteorological Organization Global Ozone
Research and Monitoring Project - Report No. 25,
Geneva, 1992.
Wuebbles, D.J., J.S. Tamaresis, and K.O. Patten, Quanti-
fied Estimates of Total GWPs for Greenhouse
Gases Taking Into Account Tropospheric Chemis-
try, Lawrence Livermore National Laboratory
UCRL-115850, 1993.
7.33
-------
-------
PART 4
CONSEQUENCES OF OZONE CHANGE
Chapter 8
Radiative Forcing and Temperature Trends
Chapter 9
Surface Ultraviolet Radiation
-------
-------
CHAPTERS
Radiative Forcing and Temperature Trends
Lead Author:
K.P. Shine
Co-authors:
K. Labitzke
V. Ramaswamy
P.C. Simon
S. Solomon
W.-C. Wang
Contributors:
C. Bruhl
J. Christy
C. Granier
A.S. Grossman
J.E. Hansen
D. Hauglustaine
H.Mao
A.J. Miller
S. Pinnock
M.D. Schwarzkopf
R. Van Dorland
-------
-------
CHAPTER 8
RADIATIVE FORCING AND TEMPERATURE TRENDS
i
Contents !
SCIENTIFIC SUMMARY '
: 8.1
8.1 INTRODUCTION....
• ; 8.3
8.2 RADIATIVE FORCING DUE TO OZONE CHANGE...
8.2.1 Recent Calculations .' .' ; "" 3
8.2.1.1 Stratospheric Ozone Change
8.2.1.2 Tropospheric Ozone Change !
8.2.1.3 Net Effect of Ozone Change ...'..."...". '•
8.2.2 Intel-comparison of Models Used to Calculate Radiative Forcing ......] o'c
8.2.3 General Circulation Model Calculations i
8.2.4 Attribution of Ozone Radiative Forcing to Particular Halocarbons "' a i n
8.2.5 Outstanding Issues 8'1U
8.11
8.3 OBSERVED TEMPERATURE CHANGES !
8.3.1 Effects of the Volcanic Eruptions, Especially Mt. Pinatubo ' c"]0
8.3.2 Long-Term Trends... «' 8'12
G — 817
8.3.3 Interpretation of Trends ""'
• 8.17
8.4 HALOCARBON RADIATIVE FORCING
8.4.1 Comparison of IR Absorption Cross Sections ' "„'
8.4.2 Comparison of Radiative Forcing Calculations j 0 or,
o.20
REFERENCES.. '
!- 8.23
-------
-------
] RADIATIVE FORCING
j
SCIENTIFIC SUMMARY !
I. Radiative Forcing and Climatic Impact of Ozone Change
• Recent one-dimensional studies support earlier conclusions that between 1980 and li 990 the observed decrease in
stratospheric ozone has caused a negative global-mean radiative forcing (i.e., it acts to cool the climate) that is
about -0.1 Wnr2; this can be compared to the direct radiative forcing due to changes in CO2; CHt, N2O, and the
chlorofluorocarbons (the "well-mixed" gases) of about 0.45 Wnr2 over the same period. An accurate assessment
of the potential climatic effect of changes in stratospheric ozone is limited by the lack of detailed information on
the ozone change, especially in the vicinity of the tropopause. A limited sensitivity study using different assump-
tions about the vertical profile of ozone loss indicates that, for the same change in total column ozone, the ozone
forcing could conceivably be up to a factor of two less negative. j
Model simulations and deductions from limited observations of the increase in tropospheric ozone since pre-
industrial times suggest a positive global-mean forcing that may be around 0.5 Wrn^; this can be compared to the
direct radiative forcing due to changes in well-mixed gases of about 2.4 Wnr2 over the same period. Particular
difficulties are the verification of the geographical extent of the tropospheric ozone increases and the problems in
accurately specifying the vertical profile of the change. ;
On the basis of these estimates, the net global-mean radiative forcing due to ozone changes is likely to have been
positive since pre-industrial times. However, Chapter 1 indicates that tropospheric ozone has changed little since
1980 and so, since then, the stratospheric ozone change is likely to have dominated, giving a negative global-mean
forcing due to ozone changes. '
t
An intercomparison of radiation codes used in assessments of the radiative forcing due to ozone changes has
shown that, in general, differences between these codes can be explained; the intercomparison highlights the
detail with which solar and thermal infrared processes must be represented.
The only general circulation model (GCM) experiment to investigate the climatic impact of observed lower
stratospheric ozone loss between 1970 and 1990 indicates the expected surface coolinig in response to the negative
radiative forcing. :
Other GCM simulations raise important, but as yet unresolved, issues about the way in which the global mean of
a spatially very inhomogeneous radiative forcing, such as that due to ozone change, Can be directly compared to
the global-mean forcing due to, for example, changes in well-mixed greenhouse gases.
The previous assessment noted that the negative radiative forcing due to stratospheric ozone loss in the 1980s is of
a similar size as, but the opposite sign to, the positive direct radiative forcing due to halocarbons over the same
period. Attempts have been made to partition the indirect negative forcing due to ozone loss amongst individual
halocarbons. They have emphasized that bromocarbons are, on a per molecule basis, much more efficient at
destroying ozone than chlorofluorocarbons (CFCs). One evaluation attributes about 50% of the 1980-1990 neg-
ative forcing to the CFCs; since the CFCs dominate the direct forcing, their net effect is reduced by about 50% due
to the ozone loss. Carbon tetrachloride and methyl chloroform each contribute about 20%, while bromocarbons
contribute about 10% to the indirect forcing; these species contribute relatively little to the direct forcing so their
net effect is likely to be negative.
8.1
-------
RADIATIVE FORCING
II. Lower Stratospheric Temperature Trends! *
• The temperature of the lower stratosphere showed a marked rise, of about 1 deg C-in the global mean, due to the
radiative effects of increases in stratospheric aerosol loading following the eruption of Mt. Pinatubo in June 1991.
The maximum wanning occurred in the six months following the eruption; temperatures have now returned to
near pre-emption levels. This warming has been successfully simulated by GGMs.
• Data from radiosondes over the past three decades and satellite observations since the late 1970s continue to
support the existence of a long-term global-mean cooling of the lower stratosphere of about 0.25 to 0.4 deg C/
decade; there is some indication of an acceleration in this cooling during the 1980s, but the presence of large
temperature perturbations induced by volcanic aerosols makes trend analysis difficult.
• Model calculations indicate that ozone depletion is likely to have been the dominant contributor to the tempera-
ture trend in the lower stratosphere since 1980 and is much more important than the contribution of well-mixed
greenhouse gases. In addition, observed temperature trends since 1979 are found to be significantly negative at
the same latitudes and times of year as significant decreases in column ozone, with the exception of the southern
midlatitudes in midwinter. However, there are other potential causes of lower stratospheric temperature change
(such as changes in stratospheric water vapor or cirrus clouds) whose contributions are difficult to quantify.
III. Radiative Properties of Halocarbon Substitutes
• More laboratory measurements, of the infrared absorption cross sections of actual and proposed substitutes to
chlorofluorocarbons have become available, including molecules not hitherto reported in assessments. These
cross sections have been included in radiative transfer models to provide estimates of the radiative forcing per
molecule. The radiative forcing estimates are subjectively estimated to be accurate to within 25%, but not all of
the studies have yet been reported in adequate detail in the open literature; this hinders a detailed understanding of
differences in existing estimates.
8.2
-------
RADIATIVE FORCING
8.1 INTRODUCTION
In recent ozone assessments, changes in the radia-
tion balance have been an issue because (i) the
chlorofluorocarbons (CFCs) and their hydrohalocarbon
replacements are powerful absorbers of infrared radia-
tion and (ii) changes in stratospheric ozone have been
shown to cause a significant radiative forcing of the sur-
face-troposphere system (WMO, 1992). Hence both
factors are of potential importance in understanding cli-
mate change.
The aim of this Chapter is to provide an update of
research in these two areas. In addition, significant
changes in stratospheric temperature have been reported
in recent assessments (WMO, 1988); it is believed that
changes in the radiative properties of the stratosphere are
an important part of the cause of the temperature trends.
Changes in ozone appear to be of particular significance;
in turn, ozone may itself be affected by the temperature
changes. This Chapter will update both the observations
of temperature changes and our understanding of their
causes; we will concentrate on the lower stratosphere
because ozone losses are believed to be greatest in this
region and also because it is changes in this region that
are believed to be of most climatic significance.
It is not our aim to provide an overall assessment
of our understanding of the radiative forcing of climate
change. Such an assessment will be part of IPCC
(1994). Instead we concentrate on the radiative forcing
due to halocarbons and ozone change, building on the
discussion in IPCC (1994). In common with the rest of
this assessment, we will also consider the role of tropo-
spheric ozone changes.
One important discussion in IPCC (1994) con-
cerns the utility of the entire concept of radiative forcing.
Radiative forcing is defined as the global-mean change
in the net irradiance at the tropopause following a
change in the radiative properties of the atmosphere or in
the solar energy received from the Sun. As discussed in
IPCC (1994), the preferred definition includes the
concept of stratospheric adjustment, in which the strato-
spheric temperatures are allowed to alter such as to
re-establish a state of global-mean radiative equilibrium;
this process is of particular relevance when considering
the forcing due to stratospheric ozone change.
The utility of radiative forcing has been based on
model results indicating that the climate response is es-
sentially independent of the forcing mechanism.
Thus, a radiative forcing of x Wm'2 due to a change
in greenhouse gas concentration would give essen-
tially the same climate response as x Wnr2 at the
tropopause due to, say, a change in solar output; this'
is despite the fact that the two mechanisms involve
rather different partitioning of the irradiance change
between the surface and troposphere as well as be-
tween different latitudes and seasons.
Recent, and rather preliminary, results from
general circulation models (GCMs) indicate that this
relationship may not be so well-behaved for radiative
forcings due to changes in ozone and tropospheric
aerosols, where there are strong vertical, horizontal,
and temporal variations in both the concentrations of
the species and their changes with time.
If this work were! to be confirmed, it would
make it more difficult to intercompare the forcings in
Wm" between different climate change mecha-
nisms; in particular it might be difficult to add the
forcings from different mechanisms to achieve a
meaningful total forcing.';
Hence radiative forcing must be used with
some caution, although niuch more work is needed to
investigate whether the concept lacks general validi-
ty. Radiative forcing remains a useful measure for
intercomparing different Icalculations of ozone forc-
ing and for intercomparing the strength of different
halocarbons. j
8.2 RADIATIVE FORCING DUE TO OZONE
CHANGE ;
8.2.1 Recent Calculations
8.2.1.1 STRATOSPHERIC OZONE CHANGE
The principal features outlined in IPCC (1992)
and WMO (1992) concerning the net radiative forc-
ing of the surface-troposphere system due to ozone
depletion in the lower stratosphere are:
a) the distinction between the solar component
that acts to heat the surface-troposphere system
and the longwave component that tends to cool
it;
b) the difference between the instantaneous forc-
ing (referred to as "Mode A" in WMO, 1992)
8.3
-------
RADIATIVE FORCING
and the forcing calculated using adjusted strato-
spheric temperatures (referred to as "Mode B" in
WMO, 1992); the consequent cooling of the lower
stratosphere enhances the longwave radiative ef-
fects to give a net negative forcing.
These features have been supported by several
model studies (Ramaswamy et al., 1992; Wang et al.,
1993; Shine, 1993; Karol and Frolkis, 1994), as well as
by an intercomparison exercise to be discussed in Sec-
tion 8.2.2.
An accurate knowledge of the magnitude of the
ozone loss in the lower stratosphere in different regions
is crucial in evaluating the global-mean radiative forc-
ing. Chapter 1 indicates the possibility of a small ozone
depletion in the tropics. The radiative forcing conse-
quences of such a loss would depend on the vertical
profile of the loss profile; it could be significant if it were
to be concentrated in the lower stratosphere (Schwarz-
kopf and Ramaswamy, 1993).
The radiative forcing is strongly governed by the
shape of the vertical profile of the ozone loss, particular-
ly in the vicinity of the tropopause (Wang et al., 1980;
Lacis et al., 1990). While it is unambiguously clear that
a loss of ozone in the lower stratosphere will lead to a
negative radiative forcing of the surface-troposphere
system, the precise value is dependent on the assumed
altitude shape of the ozone change.
Schwarzkopf and Ramaswamy (1993) examine
this problem by using 1978-1990 ozone changes at alti-
tudes above 17 km derived from Stratospheric Aerosol
and Gas Experiment (SAGE) observations; they make a
range of assumptions about how the ozone changes be-
tween the tropopause and 17 km. The results can be
quoted in terms of the radiative forcing per Dobson unit
(DU) change in stratospheric ozone. Their results de-
pend on latitude, mainiy because a smaller fraction of
the ozone depletion is located near the tropopause at
higher latitudes.. In the tropics, the estimates range from
0.007 to 0.01 WnT2/DU; at midlatitudes the relative un-
certainty is greater with a range of 0.003 to 0.008 Wm'2/
DU. Such values are obviously dependent on the SAGE
ozone change profiles used in the calculations.
Wang et al. (1993) report instances where the
change in stratospheric ozone results in a wanning rather
than a cooling of the surface-troposphere system; this
can be explained by the fact that the position of the
tropopause in these calculations is such that some of the
ozonesonde-observed increases in tropospheric ozone
are attributed to the lower stratosphere. Hauglustaine et
al. (1994) also find that the decrease in stratospheric
ozone causes a warming of the surface-troposphere sys-
tem. They used a 2-D chemical-dynamical-radiative
model to simulate changes in concentration of a number
of gases, including ozone, since pre-industrial times; the
sign of the stratospheric ozone effect in their model ap-
pears to be because, in the Northern Hemisphere at least,
their model simulates less ozone depletion in the lower
stratosphere than is indicated by recent observations -
their fractional ozone loss is found to be highest in the
mid- to upper stratosphere, where a decrease in ozone
leads to a positive radiative forcing (Lacis et al., 1990).
The model of Hauglustaine et al. (1994) does include an
interactive dynamical response, so they are not depen-
dent on assumptions such as those required when
applying fixed dynamical heating.
These recent model studies highlight the need for a
detailed consideration to be given to the vertical profile
of the depletion and for consistency between the tropo-
pause level chosen to estimate the surface-troposphere
forcing and the altitude profile of the ozone change.
Model-dependent factors are also significant for the
accuracy of the computed forcing, as will be discussed in
Section 8.2.2. :
The overall effect of observed stratospheric ozone
depletion on radiative forcing has not been significantly
updated since WMO (1992), which reported a forcing of
about -0.1 WnV2 between 1980 and 1990. Hansen et al.
(1993) compute a global mean change of -0:2 ±0.1
Wm"2 between 1970 and 1990. Such values represent a
small but not negligible offset to the greenhouse forcing
from changes in CO2, CH*, N2O, and CFCs (the so-
called "well-mixed" gases) that result in a forcing of
about 0.45 Wm'2 between 1980 and 1990. The results of
Schwarzkopf and Ramaswamy (1993) show that differ-
ent assumptions about the vertical profile of ozone
change, for the same change in total column ozone,
could conceivably result in an ozone forcing lip to a fac-
tor of two less negative.
8.2.1.2 TROPOSPHERIC OZONE CHANGE
Estimates for the global effect of tropospheric
ozone increases are scarcer, mainly because of the diffi-
culties in making global estimates of ozone change from
8.4
-------
RADIATIVE FORCING
the limited observations that are available. The model
study of Hauglustaine et al. (1994) found that their sim-
ulated tropospheric ozone increases contributed about
0.5 Wm'2 to the radiative forcing; this can be compared
to the forcing of about 2.4 Wm'2 due to the changes in
well-mixed gases since pre-industrial times (IPCC,
1990). (See Chapter 7 for an assessment of the ability of
current models to represent tropospheric ozone chang-
es.) Marenco et al. (1994) have used observations from
France in the late nineteenth century, together with re-
cent observations of the meridional distribution of
tropospheric ozone, to make a simple radiative forc-
ing estimate; they derive a global-mean radiative forcing
since pre-industrial times of 0.6 Wnr2. Fishman (1991),
using observed trends, estimates that between 1965 and
1985, a 1%/year trend in tropospheric ozone applied
over the entire Northern Hemisphere implies an approx-
imate global-mean forcing of 0.15 WnT2, or about 20%
of the effect of well-mixed gases over this period. Par-
ticular difficulties in all studies of the radiative forcing
due to tropospheric ozone change are the verification of
the geographical extent of the ozone increases and the
problems in accurately specifying the vertical profile of
ozone change.
8.2.1.3 NET EFFECT OF OZONE CHANGE
While it is clear that a tropospheric ozone increase
would lead to a positive radiative forcing, and that this
would be opposite to the effect due to the lower strato-
spheric losses, the sign of the net effect is uncertain
(WMO, 1986; Lacis et al., 1990; Karol and Frolkis,
1994; Schwarzkopf and Ramaswamy, 1993; Wang et al,
1993). Wang et al point out that the net forcing due to
the total ozone change in the atmosphere at the locations
of the sonde measurements could be positive or negative;
at Hohenpeissenberg the net forcing due to ozone
changes between the 1970s and 1980s was calculated to
be positive and comparable to that due to the increases in
the well-mixed greenhouse gases over the same period.
Lacis et al. (1990), on the other hand, found that for the
period 1970-1982, the forcing due to the Hohenpeissen-
berg changes was negative; however the uncertainty, due
to uncertainties in the trend estimate, was large. The ef-
fects due to the total atmospheric ozone change are
extremely sensitive to the vertical profile of the changes
(both in the troposphere and the stratosphere), the tropo-
pause level assumed, and, the latitude and season.
On the basis of thjese estimates, the net global-
mean radiative forcing due to ozone changes is likely to
have been positive since jpre-industrial times. However,
Chapter 1 indicates that tropospheric ozone has changed
little since 1980 and so; since then, the stratospheric
ozone change is likely to have dominated, giving a nega-
tive global-mean forcing due to ozone changes.
8.2.2 Intel-comparison of Models Used to
Calculate Radiative Forcing
The recent work described above highlights the
need to understand the reported differences in radiative
forcing due to ozone change. Whilst the overall features
of these differences were attributable to different as-
sumptions about the vertical profile of ozone change, it
60.0
10.0
Figure 8-1. Idealized change in ozone as a func-
tion of height used solely for the purposes of the
intercomparison of radiative codes. The strato-
spheric change is based on the midlatitude S2
profile of Schwarzkopf and Ramaswamy (1993)
which was derived from SAGE/SAGE II measure-
ments during the 1980s above 17 km, then
decreasing linearly with altitude to zero at the tropo-
pause at 13 km. The tropospheric increase is an
idealized one of 10% up to 8 km, then decreasing
linearly with altitude to zero at the tropopause." The
stratospheric decrease is 15.5 Dobson units and
the tropospheric increase is 3.5 Dobson units. The
results shown in Figure 8-2 to 8-4 are calculated
using the change in the stratosphere only.
5.5
-------
RADIATIVE FORCING
Table 8-1. Participants in Ozone Radiative Forcing Intercomparison and brief description of
model type. It is emphasized that the spectral resolution of a radiative transfer scheme is not always
a good indication of its accuracy.
Group
GFDL
(Geophysical Fluid
Dynamics Laboratory)
GFDL (2)
KNMI
(Koninklijk Nederlands
Meteorologisch Instituut)
KNMI (2)
LLNL
(Lawrence Livermore
National Laboratory)
NCAR/CNRS
(National Center for
Atmospheric Research/
Centre National de la
Recherche Scientifique)
Reading
(University of Reading)
Reading (2)
SUNY
(State University of New
York)
Thermal IR scheme
10 cnr1 narrow-band code —
Ramaswamy et al. (1992)
Line-by-line code - Schwarzkopf
and Pels (1991)
Wide-band scheme amended from
Morcrette (1991) to include more
trace gases
As KNMI(l) but including 14 p.m
band of ozone
25 cm-' narrow-band scheme —
Grossman and Grant (1994)
Longwave Band Model (LWBM)
100 cm-' resolution - Briegleb (1992)
10 cnr1 narrow-band scheme -
Shine (1991)
As Reading but excluding 1'4 u,m
and microwave bands of ozone
Wide-band scheme — Wang et al.
(1991)
Solar Scheme
Wide-band code based on Lacis
and Hansen (1974) - two bands in
UV and visible
Wide-band delta-Eddington
scheme from Morcrette (1991) -
one band in UV and visible -
As KNMI
Narrow-band code with 126
bands between 175-725 nm using
adding method for scattering -
Grossman et al. (1993)
Wide-band scheme based on
Lacis and Hansen (1974) - Kiehl et
al. (1987)
Delta-Eddington scheme from '_
Slingo and Schrecker (1982) with
10 bands in UV and visible
As Reading
Wide-band scheme based on •
Lacis and Hansen (1974) - Kiehl et
al. (1987) '
is important to isolate to what extent differences are due
to the radiative transfer methods employed in the stud-
ies. An intercomparison of results from different
radiative codes was initiated to study this issue. The re-
sults are reported in more detail in Shine et al. (1994);
the main conclusions of the study will be described here.
The intercomparison used tightly specified input
parameters to ensure that all groups were using the same
conditions. A midlatitude summer clear-sky atmosphere
was used with a spectrally constant surface albedo of
0.1. The solar forcing was calculated using an effective
daylength and mean solar zenith angle appropriate to 15
April at 45°N. The vertical profile of ozone and ozone
change was specified; the ozone change is shown in Fig-
ure 8-1 and is described in the caption. Groups were
asked to provide the change in solar and thermal infrared
irradiances at the tropopause for both the instantaneous
and adjusted forcings (calculated using the fixed dynam-
ical heating assumption [e.g., WMO, 1992]). Three
different cases were considered: (i) stratospheric deple-
tion only, (ii) tropospheric increase only, and (iii) both
stratospheric depletion and tropospheric increase. The
results from the case with stratospheric depletion only
will be concentrated on here. \
Six groups participated in at least part of the com-
parison. They are listed in Table 8-1 along with an
8.6
-------
RADIATIVE FORCING
0.40
0.30
Solar (W/sq.m) Q2Q
0.10
0.00
0.00
-0.02
Infrared (W/sq.m) "°-04
-0.06
Instantaneous Forcing
Stratosphere Only
ULLllLJJ
Rdg Rdg(2) ' KNMI KNMI(2) CNRS SUNY GFDL GFDL{2)
Rdg Rdg{2) KNMI KNMI(2) CNRS SUNY GFDL GFDL(2) LLNL
Net (W/sq.m)
0.20
0.10
0.00
Illllll I
Rdg Rdg(2) KNMI KNMI(2) CNRS SUNY GFDL GfDL(2) LLNL
MODEL I
indication of the nature of the radiative transfer codes.
Some groups contributed results from more than one
model configuration.
Figure 8-2 shows the change in solar, thermal in-
frared, and net irradiance at the tropopause for the
instantaneous, stratospheric-depletion case. The spread
in solar results is quite marked, ranging from 0.23 to
0.31 Wnr2; these differences will be discussed later. In
all cases the net change is positive, indicating a tendency
to warm the surface-troposphere system.
The most important conclusions from the instanta-
neous case concern the thermal infrared. First, the
narrow-band calculations are in very good agreement
with the Geophysical Fluid Dynamics Laboratory
(GFDL) line-by-line calculations. Second, it became
apparent that the results were splitting into two classes -
those calculations that included the 14 Jim band of ozone
(which is spectroscopically very weak compared to the
9.6 u.m band) obtained a forcing of about -0.07 Wnr2,
whilst those without this band reported a value of about
-0.05 Wm-2; i.e., the 14 urri band contributes about 30%
of the total forcing. The University of Reading calcula-
tions were repeated with and without the 14 urn band,
and this band was indeed shown to explain the differ-
ence. It should be noted that this band is not included in
many general circulation model calculations. Further,
the 14 [im band contributes only 2% of the change in
irradiance for the change in tropospheric ozone only, be-
cause in the troposphere jthis band is more heavily
overlapped by pressure-broadened lines of carbon diox-
ide; this indicates that models that neglect the 14 ^m
band will give greater relative1 weight to tropospheric
ozone changes compared to stratospheric ozone changes.
S.7
-------
RADIATIVE FORCING
Adjusted Forcing
Stratosphere Only
Infrared (W/sq.m)
Net(W/sq.m)
-0.05
-0.15
-0.25
-0.35
-0.45
0.00
-0.05
Rdg Rdg(2) KNMI KNMI(2) CNRS SUNY GFDL GFDL(2) LLNL
Rdg Rdg(2) KNMI KNMI(2) CNRS SUNY GFDL GFDL(2) LLNL
Rdg Rdg(2) KNMI KNMI(2) CNRS SUNY GFDL GFDL(2) LLNL
MODEL
Figure 8-3. As in Figure 8-2. but after allowing for stratospheric temperature changes ("adjusted;forcing"
using Fixed Dynamical Heating) shown in Figure 8-4.
Figure 8-3 shows the changes in solar, thermal
infrared, and net irradiances after allowing for strato-
spheric adjustment The solar changes are only very
slightly affected by the adjustment process, but the
stratospheric cooling decreases the thermal infrared (and
hence net) irradiance by typically 0.3 Wnr2 compared to
the instantaneous case. The relative effect of the 14 UJTI
band is less than in the instantaneous case, but it is more
important in an absolute sense and contributes about
-0.03 Wm'2. All models now indicate that a decrease in
lower stratospheric ozone leads to a cooling tendency for
the surface-troposphere system, but the spread in the re-
• suits is greater than for the instantaneous forcing (Figure
8-2).
Figure 8-4 shows the temperature changes calcu-
lated by each of the models; again there is a substantive
spread. The effect of the adjustment process can be as-
certained by computing the change in the net irradiance
between the instantaneous and adjusted calculations; the
results agree to within 10%. This agreement is better
than might be anticipated from Figure 8-4. However,
calculations with the Reading model indicate that most
of the change in net irradiance at the tropopause when
adjustment was included came from the temperature
changes within about 3 km of the tropopause; at these
levels the temperature changes predicted by the models
are in much better agreement. ;
The results were interpreted by Shine et al. (1994)
as follows. About 50% of the modeled temperature
change is due to the change in solar heating rates. Alti-
tudes nearest the tropopause are most influenced by the
longer wavelength (Chappuis) absorption bands of
ozone; the models are in much better agreement about
the change in these bands than the changes at shorter
8.8 .
-------
RADIATIVE FORCING
Ozone Change in Stratosphere Only
75.0 -
55.0
to
T3
13
35.0
15.0
-3.0
SUNY
Q----O KNMI-lo
* ° KNM!-hi
NCAR/CNRS
GFDL
-2.0 -1.0
Temperature Change (deg C)
81 1 T
Solutions
? 9o '"strato,?Pneric temperature (deg C) as a function of altitude computed by the partici-
a'atlFOrCing lntercomParison for the change in stratospheric ozon?shown in HgTre
approx.mat.on is used. The KNMI results are presented at two vertical
wavelengths, so that the effect of adjustment is more
similar in the models. The results can be brought into
good agreement for the adjusted case by (i) adding the
effect of the 14 urn band to those calculations that do not
include it and (ii) using a single value of the solar irradi-
ance change, rather than using the solar irradiance
change calculated by each model independently. The
major conclusion from this is that the main reason for
inter-model differences is the way the solar forcing is
calculated; it is this aspect of the calculations most in
need of scrutiny in each model. As reported by Shine et
al. (1994), high-resolution calculations of the solar irra-
diance change appear to be in good agreement; hence,
the spread in shortwave results is not believed to repre-
sent the actual uncertainty in modeling irradiances, but is
more a reflection of the simplifications used in existing
parameterizations.
8.2.3 General Circulation Model Calculations
Hansen etal. (1992, 1994) have used a general cir-
culation model (GCM) to evaluate the relative effects of
changes in the well-mixed greenhouse gases and ozone
upon the surface temperature.! A sequence of model runs
with the 1970-1990 increases; in the well-mixed green-
house gases only is comparediwith a sequence including
observed stratospheric ozone changes; the members of
each sequence differ only in their initial conditions. It is
estimated that the 1970-1990 modeled surface warming
(0.35 deg C) due to greenhouse gases is reduced by 15%
due to the ozone changes (see Figure 8-5). There is a
considerable spread among the different GCM realiza-
tions in the sequence of i experiments performed;
however, the results indicate that the cooling induced by
ozone loss has the potential to reduce the warming effect
due to the halocarbon increases over the time period con-
sidered. The results from i this study are broadly
5.9
-------
RADIATIVE FORCING
AHMGG (C02+CH4 +CFCs +N20)'-5 expls
AHMGG5 meon
AHMGG+ A03: 5 experiments
AHMGG+AOy'meon
i i i 1 I I—I—I—I—1
-0.
1970
I960
1990
Figure 8-5. Transient global surface temperature
change due to changes in greenhouse gases, as
simulated by the GISS GCM (Hansen etal., 1992,
1994). Five experiments were run with homoge-
neously mixed greenhouse gases (HMGG) (CO2,
ChU, NaO, and CFCs). Five additional experiments
were run with an ozone change inferred from the
Total Ozone Mapping Spectrometer (TOMS) and
placed entirely in the 70-250 mb layer in addition to
the changes in HMGG.
consistent with the expected temperature changes antici-
pated from the radiative forcing calculations.
Hansen et al. (1994) also investigate the climate
sensitivity to changes in the vertical profile of ozone us-
ing a simplified 3-D model; all previous studies have
used 1-D models. Such an investigation of parameter
space would be difficult with a full GCM because of the
computational cost. Instead, Hansen et al. use a sector
version of the 9-level Goddard Institute for Space Stud-
ies (GISS) GCM they call the "Wonderland" Climate
Model (see also Hansen et al., 1993). The surface tem-
perature response is a strong function of the height of the
ozone change for two reasons. First, as is well-estab-
lished (see Section 8.2.1), the radiative forcing is a
strong function of the height of the ozone change; in ad-
dition, the climate sensitivity (i.e., the surface warming
per unit radiative forcing) is found to be a function of the
height of the ozone change. This sensitivity is most
marked in experiments that allow cloud feedbacks; in
experiments with large-and idealized perturbations in
ozone, the climate sensitivity to changes in tropospheric
ozone is substantially modified when cloud feedbacks
are included.
The results of Hansen et al. (1994) have not yet
been reported in detail and must be regarded as prelimi-
nary. In addition, the sensitivity to cloud feedbacks is
likely to vary considerably amongst different GCMs be-
cause of the recognized difficulties in modeling clouds
in GCMs (e.g., IPCC, 1990, 1992). However, if con-
firmed by other studies, the new results could have very
significant implications for the way the possible climatic
impacts of ozone changes are assessed.
As discussed in IPCC (1994) available GCM sim-
ulations raise important, but as yet unresolved, issues
about the way in which the global mean of a spatially
very inhomogeneous radiative forcing, such as that due
to ozone change, can be directly compared to the global-
mean forcing due to, for example, changes in well-mixed
greenhouse gases.
8.2.4 Attribution of Ozone Radiative Forcing to
Particular Halocarbons
As discussed in earlier chapters, the weight of evi-
dence suggests that heterogeneous chemical reactions
involving halocarbons are the cause of the observed low-
er stratospheric ozone depletion. Since several of these
compounds, particularly the CFCs, exert a (direct) posi- '
tive radiative forcing, the (indirect) negative radiative
forcing due to the chemically induced ozone loss has the
potential to substantially reduce the overall contribution
of the halocarbons to the global-mean greenhouse forc-
ing, particularly over the past decade.
Daniel et al. (1994) have employed simplified
chemical considerations and partitioned the total direct
and the total indirect forcing among the various halocar-
bons (see also the discussion in Chapter 13 and Figure
13-9). The indirect effect is strongly dependent upon the
effectiveness of each halocarbon for ozone destruction.
On a per-molecule basis, bromine-containing com-
pounds are estimated to contribute more to the indirect
effect because they are more effective ozone depletors
than chlorine-containing compounds, whilst the chlori-
nated compounds have a much bigger direct effect
because they are stronger absorbers in the infrared. Thus
for the total halocarbon forcing up to 1990, Daniel et al.
estimate that the indirect effect of the CFCs is about 20%
of the direct effect but has the opposite sign. For the
halons, though, the (indirect) cooling effect is about 3
times larger than the warming due to the direct effect.
Nevertheless, because of their greater concentrations.
8.10
-------
RADIATIVE FORCING
the CFCs are estimated to have contributed about 50% of
the total indirect forcing; the precise value depends on
the value used for the effectiveness of bromine, relative
to chlorine, at destroying ozone.
For the period 1980 to 1990, Daniel et al. attribute
about 50% of the negative forcing to the CFCs, about
20% each to carbon tetrachloride and methyl chloro-
form, and about 10% to the bromocarbons. The net
(direct plus indirect) forcing for the CFCs is about 50%
of their direct effect, while the net forcing for carbon tet-
rachloride, methyl chloroform, and the bromocarbons is
likely to have been negative.
The analysis suggests that the net forcing by halo-
carbons was probably quite strong in the 1960s and
1970s (see Figure 8-6); then, when the ozone decrease
became more marked, the growth in the net forcing de-
creased substantially. Using projections for the change
in halocarbons over the next century (based on the
Copenhagen Amendment to the Montreal Protocol and
projections of possible hydrofluorocarbon (HFC) use),
Daniel et al. estimate that the cooling effect due to ozone
depletion will soon begin to decrease;- by the latter half
of next century, the positive forcing due to HFCs will be
the dominant contributor to the radiative forcing due to
halocarbons.
8.2.5 Outstanding Issues
Radiative forcing due to a specified ozone change,
as a function of the ozone altitude, is qualitatively well
understood. However, as revealed by the intercompari-
son exercise, approximate radiative methods appear to
differ in their estimates; it is important that such differ-
ences be understood. The principal limitation inhibiting
an accurate estimate of the global ozone forcing is the
lack of knowledge of the precise vertical profile of
change with latitude and season.
In contrast to the well-mixed greenhouse gases,
ozone change causes a more complicated radiative forc-
ing - neither the ozone profile nor the change are
uniform in the horizontal and vertical domains. The spa-
tial patterns of the radiative forcing differ for changes in
ozone and the well-mixed gases. Even the apportion-
ment of the radiative forcing between the surface and the
troposphere is different for ozone compared to the well-
mixed greenhouse gases (WMO, 1986, 1992). Ozone
forcings have not been used for systematic climate stud-
ies analogous to those carried out for the well-mixed
Hatocarbon Instantaneous Radiative Forcing
CM
-0.2
I95O
2000
2O5O
2IOO
Year
Figure 8-6. Radiative forcing (Wnr2) due to chang-
es in halocarbons (labeled "heating") and an
estimate of the associated stratospheric ozone loss
(labeled "cooling") and net change using observed
halocarbon changes and an "optimistic" scenario
based on the Copenhagen Amendment to the Mon-
treal Protocol and assuming bromine is 40 times
more effective at destroying ozone than chlorine
From Daniel et al. (1994).
greenhouse gases. A remaining question is the degree to
which the irradiance change at the tropopause is a rea-
sonable indicator of the surface temperature response in
the case of ozone changes; it is for the well-mixed green-
house gases, but preliminary work indicates that it may
not be for ozone.
A further complication that needs to be explored is
that changes in ozone in the vicinity of the tropopause
have the potential to alter tropopause height. Thus, the
energy received by the surface-troposphere system may
be different in a model that allows changes in the tropo-
pause height than in a model with a fixed tropopause.
The vertical resolution of a model in the upper tropo-
sphere could then be an important consideration.
As discussed in earlier chapters, there is evidence
that heterogeneous chemistiy on sulfate aerosol leads to
enhanced ozone loss. The observations of unusually low
ozone in the northern midlatitudes during the winter of
8.11
-------
RADIATIVE FORCING
1992-93 and spring of 1993 suggest a possible link with
the Mt. Pinatubo aerosols. Such a volcano-ozone link
would imply an enhancement of the transient negative
radiative forcing owing to the presence of unusually
large volcanic sulfate aerosol concentrations.
Additional complications in the determination of
the ozone forcing are uncertainties in the feedbacks re-
lated to chemical processes. One example of this is the
connection between stratospheric ozone loss, OH, and
methane lifetimes. A depletion of stratospheric ozone
would lead to an enhancement in tropospheric UV radia-
tion, which in turn increases the rate of production of OH
and destruction of methane (see, e.g., Chapter 7 and
Madronich and Granier [1992, 1994]). However, it is
important to note that photochemical oxidation of meth-
ane and other species that react with OH takes place
largely in the tropics, where ozone losses are small or not
statistically significant. Thus a quantitative assessment
of this effect requires consideration not only of tropo-
spheric chemistry but also the latitudinal distribution of
ozone depletion (particularly the tropical trends and the
sensitivity to them). A cooling of the lower stratosphere
due to the ozone loss can affect the water vapor mixing
ratios there, with the potential to alter heterogeneous
chemical reactions. Also, changes in methane in the
stratosphere as a consequence of the altered tropospheric
processes could be accompanied by changes in strato-
spheric water vapor that, in turn, would affect the
radiation balance.
8.3 OBSERVED TEMPERATURE CHANGES
A large number of factors can influence strato-
spheric temperatures (see, e.g., Randel and Cobb, 1994).
Natural phenomena can result in a change in the radia-
tive fluxes in the stratosphere, such as changes in solar
output or in aerosols resulting from volcanic eruptions.
Internal variability of the climate system, such as the
quasi-biennial o'scillation and the El Nino-Southern Os-
cillation, can induce dynamical effects that result: in
temperature change by advection and, if ozone changes
as a result, also by radiative processes: Additionally, hu-
man activity is resulting in changes in a number of
radiatively active constituents, such as ozone and carbon
dioxide, and these can perturb the radiation balance and
hence the temperature; attempts to detect trends due to
human activity require consideration of the natural pro-
cesses. Most recent work on temperature trends has con-
centrated on the lower stratosphere, so we will
concentrate on this region in this section; the particular
emphasis will be on (i) the impact of Mt. Pinatubo and
(ii) the detection of long-term trends.
8.3.1 Effects of the Volcanic Eruptions,
Especially Mt. Pinatubo
A number of studies have reported the lower
stratospheric warming following the eruption of Mt. Pi-
natubo (Labitzke and McCormick, 1992; Angell, 1993;
Spencer and Christy, 1993; Christy and Drouilhet,
1994); this warming is associated with the increased ab-
sorption of upwelling thermal infrared radiation and
solar radiation by the stratospheric aerosol layer (see,
e.g., WMO, 1988).
Angell (1993), from a selection of radiosonde sta-
tions, finds that the warming of the lower stratosphere
following both Agung and El Chichon was igreatest in
the equatorial zone and least in the polar zones. The
warming following El Chichon was slightly greater than
following Agung everywhere except the south polar
zone. Preliminary analysis for Mt. Pinatubo indicated
that, in the north extratropics and the tropics,'the warm-
ing following this eruption was comparable to the
warming following Agung and El Chich6n. However, in
south temperate and south polar zones, the wanning fol-
lowing Mt. Pinatubo is considerably greater, perhaps due
to a contribution from the eruption of Volcan: Hudson in
Chile. Globally, the warming of the lower stratosphere
following Mt. Pinatubo is greater.than that following El
Chich6n and Agung.
Figure 8-7 (updated from Labitzke and;Van Loon,
1994) shows the Northern Hemisphere annual area-
weighted temperature series from an analysis of
radiosonde data. The times of the Agung, El Chichon,
and Mt. Pinatubo eruptions are marked, although it
should be noted that other, less intense volcanic eruptions
during this period probably led to some enhancement of
the stratospheric aerosol load (e.g., Robock, 1991; Sato
et ai, 1993). Whilst the Northern Hemisphere post-Pi-
natubo warming is clear, particularly at 50 mbar, it is not
obviously larger than that due to El Chichon.
More recently, satellite observations from Channel
4 of the Microwave Sounder Unit (MSU) on the NOAA
polar-orbiting satellites have been used to monitor lower
stratospheric temperatures (e.g.. Spencer and Christy,
8.12
-------
RADIATIVE FORCING
Annual Area Weighted Means. 1O-90°N
-sac
-59.5
-65.0
-655
1985
1990
Figure 8-7. Annual mean area-weighted (10°-
90°N) temperatures (°C) at 30, 50, and 100 mb
The heavy lines are three-year running means.
Based on daily radiosonde analysis by the Free
University of Berlin (updated from Labitzke and Van
Loon, 1994). A, Ch, and P denote the times of the
Agung, El Chichon, and Mt. Pinatubo eruptions re-
spectively.
1993; Christy and Drouilhet, 1994; Randel and Cobb,
1994). The weighting function for Channel 4 peaks at
about 75 mbar with half-power values at 120 and 40
mbar (Christy and Drouilhet, 1994). Figure 8-8 shows
the global and hemispheric monthly-mean anomalies
from MSU between January 1979 and July 1994 (J.
Christy and R.' Spencer, personal communication).
From the global data set, Mt. Pinatubo gives a slightly
greater warming (about 1.1 deg C in 1991/1992) than El
Chichon (about 0.7 deg C in 1982/1983) compared to the
immediate pre-emption temperatures.
The Northern Hemisphere MSU Channel 4 data in
Figure 8-8 can be compared with the radiosonde data
analysis in Figure 8-7, although differences in the verti-
cal resolution of the data sets need to be recognized.
Both data sets are in general agreement that the immedi-
ate post-eruption warming is similar for both El Chichon
and Mt. Pinatubo. The greater wanning in the Southern
Hemisphere following Mt. Pinatubo is consistent in both
the MSU analysis (see! Figure 8-8 and Christy and
Drouilhet [1994]) and radiosonde analysis (Aneell
1993). '
Hansen et al. (1993) show that the tropical wann-
ing in the lower stratosphere associated with Mt.
Pinatubo is very well simulated by the GISS GCM with
an imposed idealized volcanic aerosol cloud.
One interesting, development in the identification
of volcanic signals is the'use of temperature and ozone
data together (Randel and Cobb, 1994; A.J. Miller, per-
sonal communication). Normally, temperatures in the
lower stratosphere and total ozone are positively corre-
lated; Randel and Cobb show that this correlation
changes sign when the lo\ver stratospheric aerosol layer
is enhanced as a result of volcanic eruptions.
8.3.2 Long-Term Trends
Both Figures 8-7 and 8-8 emphasize that the detec-
tion of long-term trends in temperatures in the lower
stratosphere will be difficult because of the episodic and
frequent volcanic eruptions that cause a major perturba-
tion to those temperatures. An additional problem
concerns the quality of available radiosonde data (see,
e.g., IPCC, 1992;Gaffen, 1994; Parker and Cox, 1994).'
Changes in instrumentation ascent times, and reporting
practices introduce a number of time-varying biases that
have not yet been property characterized; they indicate
the need for some caution when using data primarily in-
tended for weather forecasting for climate trend
analysis. Nevertheless, since 1979, comparison of inde-
pendent MSU Channel 4 data with radiosonde analyses
in the lower stratosphere has shown good agreement
(Oort and Liu, 1993; Christy and Drouilhet, 1994).
Considering all available radiosonde reports for
the period December 1963 through November 1988,
Oort and Liu (1993) infer ja trend in the global lower
stratospheric (100-50 mbar) temperature of -0.4 ±0.12
deg C/decade; the cooling trend is apparent during all
seasons and in both hemispheres. These results were
8.13
-------
RADIATIVE FORCING
O
o>
CD
•o
^^ 4
111
cc
a:
LU
a.
LU
79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94
79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94
Rgure 8-8. Global and hemispheric monthly-mean lower stratospheric temperature anomalies (from 1982-
1991 means) from the MSU Channel 4 from January 1989 to July 1994. The solid line indicates the
12-month running mean. (Data from J.R. Christy and R. Spencer. See text for a description of the weighting
function of MSU Channel 4.) i
compared with earlier estimates by Angell (1988), who
used a subset of 63 sonde stations, for the same time pe-
riod; Oort and Liu find that their global and hemispheric
trends agree with AngelFs within the error bars, although
Angell's larger Southern Hemisphere trends are believed
to be associated with undersampling. These trend analy-
ses are also consistent with the findings in Miller et al.
(1992). IPCC (1992) combines the analysis of Oort and
Liu with more recent data from Angell to deduce a glo-
bal trend of-0.45 deg C/decade between 1964 and 1991
for the 100-50 mbar layer.
A concern expressed in IPCC (1992) was that
trend analyses starting in 1964 may be biased by the
warming associated with the eruption of Agung in 1963;
however, Oort and Liu (1993) extend their own Northern
Hemisphere analysis and Angell's global analysis back
to December 1958 and find the decadal trends to differ
little from those calculated for the period December
1963 to November 1988. :
Latitudinal profiles of the estimated trends from
Oort and Liu (1993) (see Figure 8-9) show that the cool-
ing of the lower stratosphere has occurred everywhere,
8.14
-------
RADIATIVE FORCING
0.0
O
90-60S 60-30S 30-10S 10-10N10-30N 30-60N 60-90N | NH SH WORLD
nmth Sc^f ?• PL °f the estimated trends in ^e annual mean temperatures (in deg C/decade)
from the GFDL radiosonde analysis (after Oort and Liu, 1993) for the 100-50 mb layer during the
December 1963 - November 1989. The 95% confidence limits are also shown. The hemisphSnd
mean changes are shown on the right of the figure. '
but that the strongest temperature decreases (-1 deg C/
decade) have occurred in the Southern Hemisphere ex-
tratropics, strongly suggesting an association with the
Antarctic ozone hole.
Labitzke's (personal communication) analysis of
Northern Hemisphere sonde data indicates an annual
mean trend of -0.2 to -0.4 deg C/decade between 1965
and 1992 between 30 and 80 mbar at most latitudes;
however, the' trend varies greatly from month to month
both in size and in sign, and is most difficult to detect in
the extratropics during the Northern Hemisphere winter
when interannual variability is substantial. This sug-
gests that the winter months are best avoided for
long-term trend detection. There is an indication that the
trend during springtime is more negative over the period
1979-1993 than over the period 1965-1993. Figure 8-10
shows the analyses for May for these two periods; for
substantial regions the trend is almost double for the lat-
er period, although it must be noted that the significance
level is much lower, as it contains fewer data.
For the shorter period available from MSU Chan-
nel 4 observations, trends are clearly sensitive to the
period of analysis (Figure 8-8); Christy and Drouilhet
(1994) report a trend of -0.26 deg C/decade for the peri-
od January 1979 to November 1992, but comment that,
because of the effects of the volcanoes, its significance is
hard to assess. Downward trends are most marked for
the temperatures in the lower stratosphere of the polar
cap regions (defined as being 67.5° to 83.5°), being -0.78
deg C/decade for the north polar cap and -0.90 deg C/
decade for the south polar caps for the period January
1979 to January 1994 (J.R. Christy, personal communi-
cation).
For the period 1979-1991, Randel and Cobb
(1994), using MSU data, linfer a significant cooling of
the lower stratosphere over the Northern Hemisphere
midlatitudes in winter-spring (with a peak exceeding
-1.5 deg C/decade) and over Antarctica in the Southern
Hemisphere spring (peak exceeding -2.5 deg C/decade)
(Figure 8-11); the overall space-time patterns are similar
to those determined for ozone trends. The Northern
Hemisphere trends derived, from MSU data are in good
agreement with the sonde analysis from the Free Univer-
sity of Berlin (McCormack and Hood, 1994; K.
Labitzke, personal communication).
In summary, the available analyses continue to
support the conclusions of WMO (1992) that the lower
stratosphere has, on a global-mean basis, cooled by
about 0.25-0.4 deg C/decade in recent decades, although
more work on the quality of the archived data sets is
clearly warranted.
In the upper stratosphere and mesosphere there is
linle new to report beyond the discussion in WMO
(1992). Upper stratospheric temperature trends based on
8.15
-------
RADIATIVE FORCING
TREND (deg C/decade) May (1965-1993)
PROB May (1965-1993)
«« SON 60N 70N,HN 9CX
TREND (deg C/decade) May (1979-1993)
--0 !/////.<
/ / ^ / / /
TOM JON 40N SON
70H MN 9CN
PROB May (1979-1993)
MN
-------
satellite, rocket, and lidar data do not lead to a clear con-
clusion concerning trends; mesospheric coolings of
several deg C per decade in the past decade have been
deduced (see also Chanin, 1993; Kokin and Lysenko,
1994).
8.3.3 Interpretation of Trends
Models indicate that the loss of ozone in the lower
stratosphere leads to a decrease in the temperature there
(WMO, 1992 and Section 8.2.1.1). One-dimensional
models, such as the Fixed Dynamical Heating (FDH)
and the Radiative-Convective Models, compute signifi-
cant temperature changes of several tenths of a deg C/
decade in the lower stratosphere due to the ozone
changes of the past decade (WMO, 1992; Miller et ai,
1992; Shine, 1993; 'Karol and Frolkis, 1994; Ra-
maswamy and Bowen, 1994). It is this temperature
change in the FDH models that determines, to a substan-
tial extent, the negative forcing due to the ozone losses
(see Section 8.2.1.1).
The cooling trends in the lower stratosphere, either
from the long-term records or those over the past decade,
are too negative to be attributable to increases in the
well-mixed greenhouse gases (mainly CO2) alone
(Miller et al., 1992; Hansen et al., 1993; Shine, 1993;
Ramaswamy and Bowen, 1994). In contrast, models
employing the observed ozone losses yield a global tem-
perature decrease that is broadly consistent with
observations. This strongly suggests that, among the
trace gases, stratospheric ozone change is the dominant
contributor to the observed cooling trends. However, the
potential competing effects due to unknown changes in
other radiative constituents (e.g., ice clouds, water vapor,
tropospheric aerosols, and tropospheric ozone: Hansen
etai, 1993; Ramaswamy and Bowen, 1994) make it dif-
ficult to rigorously quantify the precise contribution by
ozone to the temperature trends.
McCormack and Hood (1994) calculate the tem-
perature decreases using an FDH model employing the
ozone changes deduced from Solar Backscatter Ultra-
violet (SBUV) observations for the period 1979-1991;
the temperature changes are comparable to or slightly
less than the decadal change inferred from satellite and
radiosonde data in regions where the observed trends are
statistically significant. Importantly, the modeled latitu-
dinal and seasonal dependences are in reasonable
agreement with the observations.
RADIATIVE FORCING
Temperature trend (deg C/year)
MONTH
Figure 8-11. Latitude-time sections of zonal-mean
lower stratospheric temperature trends in deg C/
year calculated from MSU Channel 4 data (Randel
and Cobb, 1994) for the period 1979-1991. Stip-
pling denotes regions where the statistical fits are
not different from zero at the 2c level. (See text for
description of the weighting function of MSU Chan-
nel 4.)
General circulation model (GCM) studies with
imposed ozone losses in the lower stratosphere also ob-
tain a temperature decrease in this region. Hansen et al.
(1993) obtain a cooling in the lower stratosphere that is
qualitatively consistent with and, in the global mean,
agrees well with the decadal trend (-0.4 deg C) inferred
from radiosonde observations. Another GCM study (V.
Ramaswamy, personal communication) finds a similar
cooling of the lower stratosphere and shows that the
FDH temperature changes exhibit a qualitatively similar
zonal pattern to the GCM re'sults.
Mahlman etal. (1994) present a three-dimensional
chemical-radiative-dynamical investigation of the cli-
matic effects due to the Antarctic ozone losses. The
transport of ozone and the ozone losses are handled ex-
plicitly, although the modeled Antarctic ozone loss is
somewhat less than observed. There is a decrease of the
lower stratospheric temperatures in the Southern Hemi-
sphere that is consistent with the observed trends. An
important aspect of the GCM calculations is that they,
simulate a slight cooling in the lower stratosphere at low-
er latitudes as a dynamical consequence of extratropical
ozone depletion; this is in contrast to FDH models which
8.17
-------
RADIATIVE FORCING
calculate temperature changes only at latitudes of ozone
change. Thus the presence of equatorial cooling in ob-
servations (see, e.g.. Figure 8-9) cannot be used as a
simple discriminator of whether ozone depletion has oc-
curred in the equatorial lower stratosphere (see Chapter
1). In addition, the simulation of Mahlman et al. (1994)
shows a dynamically induced heating in the Antarctic
mid-stratosphere as a consequence of the loss of ozone
in the lower stratosphere; such dynamical effects need to
be taken into account when attempting to detect temper-
ature trends from other causes, such as the increased
concentrations of other greenhouse gases.
It is encouraging that both the FDH models and
the GCMs yield a cooling in the lower stratosphere that
is consistent with the magnitude inferred from observa-
tions. Precise agreement might not be expected as, in all
the model studies, the temperature changes in the lower
stratosphere are subject to uncertainties related to the as-
sumed vertical and horizontal distribution of the ozone
change and there are uncertainties in the observed
trends.
8.4 HALOCARBON RADIATIVE FORCING
8.4.1 Comparison of IR Absorption Cross
Sections
Since the review in the AFEAS (1989) report (see
also Fisher et al., 1990), further work on the absorption
cross sections of halocarbons has been reported; this is
particularly important for some of the HCFCs (hyclro-
chlorofluorocarbons) and MFCs (hydrofluorocarbons),
as some of the data used in earlier assessments were
from a single source. Recent comparisons of strengths
of many CFCs (chlorofluorocarbons) are presented in
McDaniel et al. (1991), Cappellani and Restelli (1992),
' and Clerbaux et al. (1993) and are not repeated here. For
newer HCFCs and HFCs, measurements are more limit-
ed and the available measurements are reviewed.
Molecules that are created by the destruction of halocar-
bons have the potential to cause a radiative forcing, but
their lifetimes are believed to be too short for them to be
of importance (see Chapter 12); they are therefore not
considered here.
Measurements of infrared (IR) cross sections are
normally made using Fourier transform IR spectrome-
ters and, sometimes, grating spectrometers; spectral res-
olutions range from around 0.01 cm"1 to 0.1 cm"1.
Clerbaux et al. (1993) present a detailed error estimate
with errors ranging from 1-2% for strong absorption and
3-4% for weak absorption. Cappellani and Restelli
(1992) estimate an uncertainty of 2.5% and other work-
ers estimate uncertainties of between 5 and 10%.
Table 8-2 lists the integrated absorptionCToss sec-
tions of measurements of HFCs and HCFCs known to
the authors. Measurements for a number of these mole-
cules, as w6ll as a number of halogenated ethers used as
anesthetics, are also reported by Brown et al. (1990);
however, the absorption cross sections are reported for
only a limited spectral region (800 - 1200 cm"1) that ne-
glects some important absorption features. Garland et
al. (1993) report measurements in the region 770-1430
cm"1 of the absorption cross sections of HFC-236cb,
HFC-236ea, and HFC-236fa, as well as the fluorinated
ether E-134. Because the results are presented as rela-
tive cross sections, they are not included in Table 8-2;
their integrated strengths are reported to be between 1.5
and 2.3 times stronger than CFC-11.
As examples of the degree of agreement in the
near-room temperature measurements, for HCFC-123,
HCFC-141b, and HCFC-142b the spread of results is
more than 25% of the mean cross section; however, the
spread between the results from the two published stud-
ies (Cappellani and Restelli, 1992; Clerbaux etai, 1993)
is generally smaller. For HFC-134a the spread is about
10%. Detailed descriptions, including temperatures and
pressures of the measurements, are not available for all
the data sets, so it is difficult to comment on the discrep-
ancies. Except for HCFC-22, only Cappellani and
Restelli (1992), Clerbaux et al. (1993), and Clerbaux
and Colin (1994) have published the details of their mea-
surements of HFC/HCFC cross sections and presented
measurements for a range of temperatures. In general,
the change in integrated cross section over the range of
temperatures is less than 10%, although the two groups
do not always agree on the sign of the temperature effect.
The spread of results puts a limit on our knowledge of
the accuracy with which the radiative forcing due to
these gases can currently be modeled.
8.18
-------
I RADIATIVE FORCING
abSOrpU°n cross sections of HpC and HCFCs in units of xlCH?
Gas
HCFC-22
HFC-23
HFC-32
HFC-41
HCFC-123
HCFC-124
HFC- 125
HFC-134a
HCFC-141b
HCFC-142b
HFC- 143
HFC-143a
HFC-152a
HCFC-225a
HCFC-225b
HFC-227ea
Clerbaux 1
10.0-10.3
12.2-12.9
14.4
16.1
12.7-12.6
6.8-7.8
10.8-11.1
7.0-6.9
7.1-6.9
17.5-17.7
16.5-15.6
BriihP
13.6
14.8
15.7
12.5
9.0
12.1
11.4
5.9
21.2
_Gehring3
8.9
9.5
15.1
14.5
11.8
6.5
9.2
12.7
Majid3
9.5
10.6
12.2
7.1
9.6
6.1
Hurley*
127
63
1 7
12.5
13.0
76
10.3
7.3
i
10.9-10.3 8.3-9.0
f
-•I
13.1-12.8
i
14.1-13.2
113-10/7
1
7.5-6.9
|
2.
3.
4.
5.
6.
Quoted Tt^or 92387TgMCOrreSPOnd * ^ * ** * "* ^ * ^^'^ "*« only one value
s quoted, it is for 287 K. Measurements m spectral interval 600-1500 cm-i at 0.03 cm-i resolution HFC-
143 is from Clerbaux and Colin (1994). «»uiuuon. nr^
C. Briihl, personal communication of room temperature measurements at Max-Planck Institute-Mainz-
cTvTnT5 T\ mtTal 50(M400 Cm" aPPr°X- ValUCS SUPPliŁd in -'^^ «- at 296 K) "-
converted by multiplying by 296 -s- (273 x 2.687x1019)
rRatm a** l^pfr eraL ^"^ meaSUrementS made at ro™ temperature. Values reported as cm-«
(atm cm at STpy-i Converted by multiplying by 1 + 2.687x10-9. Some authors (Clerbaux et al. 1993
and Cappellam and Restelli, 1992) convert assuming the gas amounts are atm cm at 296 K; we have
been unable to resolve this with D. Fisher. If original units are indeed at 296 K instead of STP the
values m the above table should be multiplied by 1.08.
M. Hurley and T J Wallington, personal communication of measurements by Ford; integrated cross
ecaons denved by S. Pinnock (University of Reading). Measurements in spectral interval 700-3800
^.described *
to values at 233K and
- M—
8:19
-------
RADIATIVE FORCING
8.4.2 Comparison of Radiative Forcing
Calculations
In IPCC (1990, 1992, 1994) a specific definition
of radiative forcing was adopted such that:
The radiative forcing of the surface-troposphere
system (due to a change, for example, in green-
house gas concentration) is the change in net
irradiance (in Wnr2) at the tropopause after al-
lowing for stratospheric temperatures to
re-adjust to radiative equilibrium.
The tropopause is chosen because, in simple mod-
els at least, it is considered that in a global and annual-
mean sense, the surface and troposphere are so closely
coupled that they behave as a single thermodynamic sys-
tem (see, e.g., Rind and Lacis, 1993; IPCC, 1994) This
follows earlier work (e.g., Ramanathan et al, 1985,
Hanson etal, 1981, and references therein). One advan-
tage of allowing for the stratospheric adjustment is that
the change in the net irradiance at the top of the atmo-
sphere is then the same as the change at the tropopause;
this is not the case when stratospheric temperatures are
not adjusted (see Hansen et al., 1981).
In preparing this review some difficulty has been
experienced in intercomparing work performed by dif-
ferent authors, because some have applied the term
"radiative forcing" to the instantaneous change in tropo-
pause irradiance, not allowing for any change in
stratospheric temperature. In other works, it is not clear
which definition of radiative forcing has been adopted.
It is also emphasized here that the forcing should be cal-
culated as a global mean using appropriate vertical
profiles of temperature, trace gas concentrations, and
cloud conditions - again, it is not always clear, in pub-
lished estimates, what conditions are being used for
calculations. An added problem is that if perturbations
used to calculate the forcings are too small, the results
can be affected by computer precision, an effect that will
vary between models, depending on their construction.
Finally, when results are presented as ratios of forcings
to other gases (e.g., CO2 or CFC-11) rather than as abso-
lute forcings (e.g., as Win'2 ppbv1), it is important to
know the absolute forcing of the reference gas to rigor-
ously intercompare different works. Again, such
information is not always presented.
The first calculations of the radiative forcing due
to a large range of HFCs and HCFCs were reported in
Fisher et al. (1990). More recent calculations include
those of Shi (1992), Briihl (personal communication),
Clerbaux et al. (1993) and Clerbaux and Colin (1994);
C. Granier (personal communication) has updated the
Clerbaux et al. (1993) calculations to account for the ef-
fect of clouds, and these new values are used here. The
results from these sources differ in general, because of
the use of different radiation schemes, different spectro-
scopic data, different assumptions about vertical profiles
of temperatures, clouds, etc., and whether stratospheric
adjustment was included.
The Fisher et al (1990) values for this class of
gases were instantaneous forcings. The effect of adjustment
can be estimated from the 1-dimensional radiative-con-
vective model values in Fisher et al. (1990) (see footnote
to Table 8-3). The adjustment leads to the adjusted forc-
ing being up to 10% greater than the instantaneous
forcing because an increase in the concentrations of
these gases generally leads to a warming of the lower
stratosphere, increasing the downwelling thermal infra-
red irradiance at the tropopause. The values most
affected are those for the more heavily fluorinated gases
(such as HFC-125 and HFC-134a).
Table 8-3 lists recent estimates of the strengths of
the HFC and HCFCs, on a per-molecule basis, relative to
CFC-11. The variations of the relative forcings from dif-
ferent studies show little consistency. The same spectral
data in different radiation schemes do not always give
the same relative forcings amongst the HFCs and
HCFCs; and schemes using different spectral data do not
always show differences that would be anticipated from
the cross sections used. For the majority of gases, Shi
(1992) computes a radiative forcing weaker than those
given in IPCC (1990), by as much as 30% for HFC-125.
The results from Briihl and Clerbaux et al. generally
show no systematic difference compared with the Atmo-
spheric and Environmental Research, Inc. (AER) values.
For only two gases is there a consistent and large devia-
tion from AER values: HFC-125 and HFC-152a.
More systematic work needs to be done to estab-
lish the effect of factors such as overlap with other
species, stratospheric adjustment, the vertical profile of
the absorber, and the dependence of the calculation on
the spectral resolution, if the range in the current esti-
mates is to be understood better.
Table 8-4 presents our recommended forcings for
a wide range of gases, all relative to CFC-11. We have
8.20
-------
t-3. Radiative forcing due to HFC and HCFCs on a per-m
for gases for which more than one assessment is available.
RADIATIVE FORCING
molecule basis relative to
Gas
AERl
HCFC-22
HCFC-123
HCFC-124
HFC-125
HFC-134a
HCFC-141b
HCFC-142b
HFC-143a
HFC-152a
HFC-227ea
AER adj2
pont3
Shi-*
0.86
0.80
0.87
1.08
0.77
0.62
0.82
0.63
0.53
Bi
0.92
0.82
0.93
1.19
0.84
0.64
0.89
0.68
0.56
GranierS
0.80
0.69
0.81
0.89/0.93
0.71
0.57
0.76
0.65
0.44/0.46
-71.24
0.75
0.67
0.88
0.74
0.65
0.64
0.63
0.50
0.44
0.79
0.80
0.95
0:91
0.66
0.68
0.93
0.58
0.48
1.09
0.87
0.89
0.93
0.90
0.78
0.65
0.81
0.49
1.
2.
4.
5.
6.
AER results from Fisher et al. (1990). They are instantaneous forcings - '
AER-adjusted forcings deduced from results in Fisher et al (1990), Tables 3 and 4. These authors wrote
Ae chmate sensitivity A. in terms of instantaneous forcing so that the surface temperature change equals
the Td ?VT r Tng- ^ dimate SenSidvity " rcaS°nably Dependent of gas when using
T1 (e'8\ ' 3nd LadS' 1993>- ^ 3djUSted f°rcin*' rel'^e * CFC-11, can bf
of a P ^ thVnStantane°US forci"S relative <° CFC-ll by the ratio of the instantaneous
of a given gas to the instantaneous sensitivity for CFC-11
beuctlVrA^T etal (1!90)- ThCy ^ instantaneous f°™&. The adjusted forcings could
be deduced as for the AER forcings above. Second values, where quoted, are more recent values from D
risner (personal communication). i
n99m(™2) T !nClUdeS °VerlaP Whh methanC "nd nitr°US °Xide USin* sP^ral data from Fisher et
(1990). The calculations are not believed to include stratospheric adjustment
overTarfwith^1^50^ C°mmUniCati°n) ^^ MPI-Mainz absorption cross sections and including
overlap with methane and nitrous oxide. j .
From C. Granier (personal communication) using measurements from Clerbaux et al (1993) The results
include clouds and do not include stratospheric adjustment; the original C.erbaux etal. vies were for
clear stcies.
chosen to retain values used in IPCC (1990) where there
was neither a large nor a consistent deviation from more
recent calculations. We replace the earlier values for
HFC-125 and HFC-152a by those from C. Granier (per-
sonal communication). In cases where details of
calculations have not been provided, we simply take the
means of available estimates. It is subjectively estimated
that these values are accurate to within about 25%, but it
is anticipated that further revision will be necessary in
the future. Another feature of Table 8-4 is that an at-
tempt is made to classify HFC/HCFC species on the
basis of likely emissions (using information from A.
McCulloch [personal communication]).
Estimates for the forcing due to increased concen-
trations of HFCs and HCFCs between 1990 and 2100
include 0.15 Wnr2 by Daniel:er al (1994) and between
0.2 and 0.4 Wm'2 by Wigley (1994). For the sets of as-
sumptions used by .these authors, the forcing is a small
fraction of the estimates of forcing due to increases in
the well-mixed greenhouse ^ases between 1990 and
2100; the various scenarios used in IPCC (1992) give
that forcing to lie between 3.4 and 8.5 Wm'2. The actual
radiative forcing due to futurb emissions of the HFCs
and HCFCs depends critically on factors such as growth
rates in emissions and the precise mix of species used.
5.27
-------
RADIATIVE FORCING
Tames-,.
dX is the perturbation to the volume mix,ng rat,o
of CFC-11 in ppbv.
1.00
1.45
0.93
1.18
1.04
0.41
023
CFCs «nd other controlled chlorinated species
CFC-11 CFC13
CFC-12 CF2C12
CFC-113 CF2C1CFC12
CFC-114 CF2C1CF2C1
CFC-115 CF3CF2C1
•Carbon tetrachloride CC14
Methyl chloroform CH3CC13
HFC/HCFCs in production now and likely to be widely used
HCFC-22 CHF2C1 137
HCFC-141b CH3CFC12 0.3
HCFC-142b CH3CF2C1 -12
HFC-134a CF3CH2F -04
•HFC-32 CH2F2 1-06
HFC/HCFCs in production now for specialized end use
HCFC-123 CF3CHC12 0.72
HCFC-124 CF3CHFC1 0.88
•HFC-125 CF3CHF2 1-03
HFC-143a CH3CF3 L03
•HFC-152a CH3CHF2 1-02
•HCFC-225C3 CF3CF2CHC12 0.72
«HCFC-225cb CC1F2CF2CHC1F 0.87
HFC/HCFCs under consideration for specialized end use
•HFC-23W CHF3 ^
•HFC-134 ' CHF2CHF2 1-08
•HFC-143 CH2FCHF2 0.85
•HFC-227 CF3CHFCF3 0.95
•HFC-236
•HFC-245
«HFC-43-10mee
Fully fluorinated substances
*CF4
•C2F6
•C3F8
•perfluorocyclobutane
CF3CH2CF3
CHF2CF2CFH2
Other species
CFC-13
CHCI3
CH2C12
halon 1301 .
«CF3I
CC1F3
CF3Br
1.06
0.95
0.86
0.69
136
0.77
1.00
0.75
2.75
137
0.09
0.23
1.19
1.20
1.00
1.27
1 .27
1.47
1.17
0.46
022
0.86
0.62
0.82
0.77
0.40
0.80
0.87
0.90
0'.63
049
1.07
129
0.8i
O.go
0.52
1.17
1 17
093
0.44
137
1 05
1 45
j g4
2.92
1.04
0.078
0 14
129
1 71
IPCC (1990)
IPCC (1990)
IPCC (1990)
IPCC (1990) :
IPCC (1990)
IPCC (1990) ;
IPCC (1990)
IPCC (1990)
IPCC (1990)
IPCC (1990)
IPCC (1990)
Fisher (personal communication)
IPCC (1990)
IPCC (1990)
Granier (personal communication) '
IPCC (1990)
Granier (personal communication)
Granier (personal communication)
Granier (personal communication) |
Fisher (personal communication)
Fisher (personal communication)
Clerbaux and Colin (1994)
Mean of Briihl and Fisher (pers. comms.)
Fisher (personal communication)
Fisher (personal communication)
Fisher (personal communication)
Isaksen et al. (1992)
Isaksen el al. (1992)
Briihl (personal communication)
Fisher (personal communication)
Mean of Briihl and Ko (pers. comms.)
Mean of Ko era/. (1993)/Stordal etat. (1993)
Mean of Briihl and Fisher (pers. comms.)
Fisher (personal communication)
Fisher (personal communication)
IPCC (1990) [
Pinnock (personal communication)
mass factor for CC,4 has altered due to a typographical
error in IPCC 1990.
8.22
-------
RADIATIVE FORCING
REFERENCES
AFEAS, Alternative Fluorocarbons Environmental Ac-
ceptability Study, AFEAS Report - Appendix
(Volume 2) of Scientific Assessment of Strato-
spheric Ozone: 1989, World Meteorological
Organization Global Ozone Research and Moni-
toring Project - Report No. 20, Geneva, 1989.
Angell, J.K., Variations and trends in tropospheric and
stratospheric global temperatures, 1958-1987, J.
Climate, 1, 1296-1313, 1988.
Angell, J.K., Comparison of stratospheric warming fol-
lowing Agung, El Chich6n and Pinatubo volcanic
eruptions, Geophys. Res. Lett., 20,. 115-718, 1993.
Briegleb, B.P., Longwave band model for thermal radia-
tion in climate studies, /. Geophys. Res., 97,
11475-11485, 1992.
Brown, A.C., C.-E. Canosa-Mas, A.D. Parr, and R.P.
Wayne, Laboratory studies of some halogenated
ethanes and ethers: Measurements of rates of reac-
tion with OH and of infrared absorption cross
sections, Atmos. Environ., 24A, 2499-2511, 1990.
Cappellani, R, and G. Restelli, Infrared band strengths
and their temperature dependence of the hydroha-
locarbons HFC-134a, HFC-152a, HCFC-22,
HCFC-123 and HCFC-142b, Spectrochimica
Acta, 48A, 1127-1131, 1992.
Chanin, M.-L., Long-term trend in the middle atmo-
sphere temperature, in The Role of the
Stratosphere in Global Change, edited by M.-L.
Chanin, NATO-ASI Series, Springer-Verlag,
Berlin, 301-317, 1993.
Christy, J.R., and S.J. Drouilhet, Variability in daily,
zonal mean lower stratospheric temperatures, J.
Climate, 7, 106-120, 1994.
Clerbaux, C., R. Colin, P.C. Simon, and C. Granier, In-
frared cross sections and global warming
potentials of 10 alternative hydrohalocarbons, J.
Geophys. Res., 98, 10491-10497, 1993.
Clerbaux* C., and R. Colin, Determination of the infra-
red cross section and global warming potential of
1,1,1-trifluoroethane (HFC-143), Geophys. Res.
Lett., 21, 2377-2380, 1994.
Daniel, J.S., S. Solomon, and D.L. Albritton, On the
evaluation of halocarbon radiative forcing and glo-
bal warming potentials, /. Geophys. Res., in press,
1994.
Fisher, D.A., C.H. Hales, W.-C. Wang, M.K.W. Ko, and
N.-D. Sze, Model calculations of the relative ef-
fects of CFCs and their replacements on global
warming, Nature, 344, 513-516, 1990.
Fishman, J., The global consequences of increasing tro-
pospheric ozone concentrations, Chemosphere,
22, 685-695, 1991. <
Gaffen, D.J., Temporal irihomogeneities in radiosonde
temperature records, /. Geophys. Res., 99, 3667-
3776,1994.. i
Garland, N.L., LJ. Medhiirst, and H.H. Nelson, Poten-
tial chlorofluorocarbon replacements: OH
reaction rate constants between 250 and 315 K and
infrared absorption spectra, / Geophys. Res., 98,
23107-23111, 1993J
Grossman, A.S., K.E.-Graiit, and DJ. Wuebbles, Radia-
tive flux calculations at UV and visible
wavelengths, LLNL Report UCRL-ID-115336,
Lawrence Livermore. National Laboratory, Liver-
more, California, 1993.
Grossman, A.S., and K.E. Grant, A correlated k-distribu-
tion model of the heating rates for atmospheric
mixtures of H2O, C0|2,03, CH4 and N2O in the 0-
2500 cm'1 wavenumber region at altitudes
between 0-60 km, in' Proceedings of the 8th Con-
ference on Atmospheric Radiation, American
Meteorological Society, Boston, Massachusetts,
97-99, 1994. !
Hansen, J., D. Johnson, A. Lacis, S. Lebedeff, P. Lee, D.
Rind, and G. Russel, Climate impacts of increas-
ing carbon dioxide, Science, 213, 957-966, 1981.
Hansen, J., W. Rossow, and!I. Fung, Long-term monitor-
ing of global climate Iforcing and feedbacks, NASA
Conference Publications 3234, National Aeronau-
tics and Space Administration, Washington, D.C.,
1992. ;
Hansen, J., A. Lacis, R. Ruedy, M. Sato, and H. Wilson,
How sensitive is the world's climate?, National
Geographic Research and Exploration, 9, 142-
158, 1993. i
8.23
-------
RADIATIVE FORCING
Hansen, J., M. Sato, A. Lads, and R. Ruedy, The impact
on climate of ozone change and aerosols, in Pro-
ceedings of IPCC Hamburg Meeting, a joint
workshop of IPCC Working Group I and the Inter-
national Ozone Assessment Panel, Hamburg,
available from IPCC WGI Technical Support Unit,
Hadley Center, Meteorological Office, Bracknell,
UK, 1994.
Hauglustaine, D.A., C. Granier, G.P. Brasseur, and, G.
Megie, The importance of atmospheric chemistry
in the calculation of radiative forcing on the cli-
mate system, /. Geophys. Res., 99, 1173-1186,
1994.
IPCC, Climate Change. The IPCC Scientific Assessment,
Intergovernmental Panel on Climate Change,
Cambridge University Press, 1990.
IPCC, Climate Change 1992. The Supplementary Report
to the IPCC Scientific Assessment, Intergovern-
mental Panel on Climate Change, Cambridge
University Press, 1992.
IPCC, Radiative Forcing of Climate Change. An Interim
Scientific Assessment, Intergovernmental Panel on
Climate Change, to be published by Cambridge
University Press, 1994.
Isaksen, I.S.A., C. Briihl, M. Molina, H. Schiff, K.
Shine, and F. Stordal, An assessment of the role of
CF4 and C2p$ as greenhouse gases, Center for In-
ternational Climate and Research, University of
Oslo Policy Note 1992:6, 1992.
Karol. I., and V.A. Frolkis, Model evaluation of the radi-
ative and temperature effects of the ozone content
changes in the global atmosphere of the 1980s, in
Proceedings of the Quadrennial Ozone Sympo-
sium 1992, Charlottesville, Virginia, 13 June 1992,
NASA Conference Publication CP-3266, 409-
412, 1994.
, Kiehl, J.T., R.J. Wolski, B.P. Briegleb, and V. Ra-
manathair, Documentation of radiation and cloud
routines in the NCAR Community Climate Model,
NCAR Technical Note TN-288+IA, National
Center for Atmospheric Research, Boulder, Colo-
rado, 1987.
Ko, M.K.W., N.-D. Sze, W.-C. Wang, G. Shia, A. Gold-
man, F.J. Murcray, D.G. Murcray, and C.P.
Rinsland, Atmospheric sulfur hexafluoride:
Sources, sinks and greenhouse warming, J. Geo-
phys. Res., 98, 10499-10507, 1993.
Kokin, G.A., and E.V. Lysenko, On temperature trends
of the atmosphere from rocket and radiosonde
data, /. Atmos. Terr. Phys., 56, 1035-1044, 1994.
Labitzke, K., and M.P. McCormick, Stratospheric tem-
perature increases due to Pinatubo aerosols,
Geophys. Res. Lett., 19, 207-210, 1992.
Labitzke, K., and H. Van Loon, A note on trends in the
stratosphere: 1958-1992, COSPAR Colloquia Se-
ries, 5, Pergamon Press, 537-546,1994.
Lacis, A.A., and J.E. Hansen, A parameterization for the
absorption of solar radiation in the earth's atmo-
sphere, /. Atmos. ScL, 31, 118-133, 1974.
Lacis, A.A., D.J. Wuebbles, and J.A. Logan, Radiative
forcing of climate by changes in the vertical distri-
bution of ozone, /. Geophys. Res., 95, 9971-9981,
1990.
McCormack, J.P., and L.L. Hood, Relationship between
ozone and temperature trends in the lower strato-
sphere: Latitudinal and seasonal dependences,
Geophys. Res. Lett., 21, 1615-1618, 1994.
McDaniel, A.H., C.A. Cantrell, J.A. Davidson, R.E.
Shelter, and J.G. Calvert, The temperature depen-
dent infrared absorption cross sections for
chlorofluorocarbons: CFC-11, CFC-12, CFC-13,
CFC-14, CFC-22, CFC-113, CFC-114 and CFC-
115, J. Atmos. Chem, 12, 211-227, 1991.
Madronich, S., and C. Granier, Impact of recent total
ozone changes on tropospheric ozone photodisso-
ciation, hydroxyl radicals and methane trends,
Geophys. Res. Lett., 19, 465-467, 1992.
Madronich, S., and C. Granier, Tropospheric chemistry
changes due to increased UV-B radiation, in
Stratospheric Ozone Depletion—UV-B Radiation
in the Biosphere, edited by R.H. Biggs and M.E.B.
Joyner, Springer-Verlag, Berlin, 3-10, 1994.
Mahlman, J.D., J.P. Pinto, and LJ. Umscheid, Transport,
radiative and dynamical effects of the Antarctic
ozone hole: A GFDL "SKYHI" model experi-
ment, J. Atmos. Sci., 51, 489-509, 1994.
Marenco, A., H. Gouget, P. Nedelec, J-P. Pages, and F.
Karcher, Evidence of a long-term increase in tro-
pospheric ozone from Pic du Midi data series -
consequences: Positive radiative forcing, /. Geo-
phys. Res., 99, 16617-16632, 1994.
8.24
-------
Miller, A.J., R.M. Nagatani, G.C. Tiao, X.F. Niu, G.C.
Reinsel, D. Wuebbles, and K. Grant, Comparisons
of observed ozone and temperature trends in the
lower stratosphere, Geophys. Res. Lett., 19, 929-
932, 1992.
Morcrette, JJ., Radiation and cloud radiative properties
in the ECMWF forecasting system, /. Geophys.
Res., 96, 9121-9132, 1991.
Oort, A.H., and H. Liu, Upper air temperature trends
over the globe, 1958-1989,7. Climate, 6, 292-307
1993.
Parker, D.E., and D.I. Cox, Towards a consistent global
climatological rawinsonde data base, Int. J. Clima-
tology, in press, 1994.
Ramanathan, V., R.J. Cicerone, H.B. Singh, and J.T.
Kiehl, Trace gas trends and their potential role in
climate change, J. Geophys. Res., 90, 5547-5566
1985.
Ramaswamy, V., M.D. Schwarzkopf, and K.P. Shine,
Radiative forcing of climate from halocarbon-
induced global stratospheric ozone loss, Nature
355, 810-812, 1992.
Ramaswamy, V., and M.M. Bowen, Effect of changes in
radiatively active species upon the lower strato-
spheric temperatures, /. Geophys. Res., 18909-18921
1994.
Randel, W.J., and J.B. Cobb, Coherent variations of
monthly mean total ozone and lower stratospheric
temperature, J. Geophys. Res., 99 5433-5447
1994.
Rind, D., and A. Lacis, The role of the stratosphere in
climate change. Surveys in Geophysics 14 133-
165, 1993. ' -
Robock, A., The volcanic contribution to climate change
of the past 100 years, in Greenhouse Gas Induced
Climatic Change: A Critical Appraisal of Simula-
tions and Observations, edited by M.E.
Schlesinger, Elsevier, Amsterdam, 429-443, 1991.
Sato, M., J.E. Hansen, M.P. McCormick, and J.B. Pol-
lack, Stratospheric aerosol optical depths,
1850-1990, J. Geophys. Res., 98, 22987-22994
1993.
Schwarzkopf, M.D., and S.B. Pels, The simplified ex-
change method revisited: An accurate, rapid
method for computation of infrared cooling rates
and fluxes, / Geophys. Res., 96, 9075-9096, 1991.
RADIATIVE FORCING
Schwarzkopf, M.D., and V. Ramaswamy, Radiative forc-
ing due to ozone in the 1980s: Dependence on
altitude of ozone change, Geophys. Res. Lett 20
205-208, 1993.
Shi, G., Global Warming Potential due to chlorofluoro-
carbons and their substitutes, Chinese J. Atmos
Sci., 16, 240-248, 1992.
Shine, K.P., On the cause of the relative greenhouse
strength of gases such as the halocarbons, /. At-
mos. Sci., 48, 1513-1518, 1991.
Shine, K.P., The greenhouse effect and stratospheric
change, in The Role of the Stratosphere in Global
Change, edited by M.-L. Chanin, NATO ASI Se-
ries, Springer-Verlagi Berlin, 285-300, 1993
Shine, K.P., B.P. Briegleb, A.S. Grossman, D. Haug-
lustaine, H. -Mao, V. Ramaswamy, M.D.
Schwarzkopf, R. Van Dorland, and W.-C. Wang,
Radiative forcing due to changes in ozone - A
comparison of different codes, in Atmospheric
Ozone as a Climate Gas, edited by W.-C. Wang
and I.S.A. Isaksen, NATO ASI Series, Springer-
Verlag, Berlin, in press, 1994.
Slingo A., and H.M. Schrecker, On the shortwave prop-
erties of stratiform water clouds. Quart. J. Roy.
Meteorol. Soc., 108, 407-426, 1982.
Spencer, R.W., and J.R. Chiristy, Precision lower strato-
spheric temperature monitoring with the MSU:
Validation and results 1979-1991, J. Climate 6
1194-1204,1993.
Stordal, R, B. Innset, A.S. Grossman, and G. Myhre, SF6
as a greenhouse gas, Nonvegian Institute for Air
Research (NILU) Report 0-92102, 1993.
Wallington, T.J., C.A. Gierczak, J.C. Ball, S.M. Japar,
Fourier transform infrared study of the self reac-
tion of C2H5O2 radicals in air at 295 K, Int. J.
Chem. Kinet., 21, 1077-1089, 1989.
Wang, W.-C, J.P. Pinto, andY.L. Yung, Climatic effects
due to halogenated compounds in the Earth's at-
mosphere, / Atmos. Sci., 37, 333-338, 1980.
Wang W.-C., G.Y..Shi, and J.T. Kiehl, Incorporation of
the thermal radiative effect of CH4, N2O, CF2C12
and CFC13 into the NCAR Community Climate
Model,/ Geophys. Res., 96, 9097-9103, 1991.
8.25
-------
RADIATIVE FORCING
Wang, W.-C., Y.-C. Zhuang, and R.D. Bojkov, Climate
implications of observed changes in ozone vertical
distributions at middle and high latitudes of the
Northern Hemisphere, Geophys. Res. Lett., 20,
1567-1570, 1993.
Wigley, T.M.L., The contribution from emissions of dif-
ferent gases to the enhanced greenhouse effect, In
Climate Change and the Agenda for Research,
edited by T. Hanisch, Westview Press, Boulder,
Colorado, 1994.
WMO, Atmospheric Ozone, 1985: Assessment of Our
Understanding of the Processes Controlling Its
Present Distribution and Change, World Meteoro-
logical Organization Global Ozone Research and
Monitoring Project - Report No. 16, Geneva,
1986.
WMO, Report of the International Ozone Trends Panel,
World Meteorological Organization Global Ozone
Research and Monitoring Project - Report No. 18,
Geneva, 1988.
WMO, Scientific Assessment of Ozone Depletion: 1991.
World Meteorological Organization Global Ozone
Research and Monitoring Project - Report No. 25,
Geneva, 1992.
8.26
-------
CHAPTER 9
Surface Ultraviolet Radiation
Lead Author:
R.L. McKenzie
Co-authors:
M. Blumthaler
C.R. Booth
S.B. Diaz
J.E. Frederick
T. Ito
S. Madronich
G. Seckmeyer
Contributors:
S.Cabrera
M. Ilyas
J.B. Kerr
C.E. Roy
P.C.Simon
D.I. Wardle
-------
-------
CHAPTER 9
SURFACE ULTRAVIOLET RADIATION
Contents
SCIENTIFIC SUMMARY
: 9.1
9.1 INTRODUCTION....
i ; 9.3
9.2 UPDATE ON TREND OBSERVATIONS
9.2.1 Results Derived from Broad-Band Meters ' 9'3
9.2.2 Multi-Wavelength Measurements ..J ' ' "9-3
9.2.3 Status of Trend Observations r ' 9'4
: 9.4
9.3 SPECTRO-RADIOMETER RESULTS
9.3.1 Intercomparisons 9-4
9.3.2 Geographic Differences ! 9'5
9.3.3 High Latitude (North and South) '" 9'6
9.3.4 Northern Hemisphere Midlatitude 9'7
' , 9.10
9.4 IMPLICATIONS OF RECENT CHANGES j
9.4.1 Stratospheric Aerosols from the Mt. Pinatubo Eruption ' 9"12
9.4.2 Tropospheric Pollution ; ' 9'12
9.4.3 Magnitude of Changes . < 9-12
• i 9.14
9.5 UPDATE ON PREDICTIONS j
9.5.1 Semi-Empirical Method .' i 9-14
9.5.2 Calculated Changes in Clear-Sky UVUsinic^Oz^ r ?'"
9.5.3 Cloud and Albedo Effects i 9'14
9.5.4 UV Forecasts '• 9-16
! • 9.18
9.6 GAPS IN KNOWLEDGE
• ; • 9.18
REFERENCES
• i 9.18
-------
-------
I
SURFACE UV RADIATION
SCIENTIFIC SUMMARY
There is overwhelming experimental evidence that, all other things being equal, decreases in atmospheric ozone
result in UV-B increases at the Earth's surface, in quantitative agreement with predictions by radiative transfer
models. t , - .
Large UV-B increases have been observed in association with the ozone "hole" at high southern latitudes. Biolog-
ically damaging radiation at the South Pole exceeded that in the Arctic by more than a factor of two, for the same
solar zenith angle. At Palmer Station, Antarctica (64.5°S), erythemal and DNA-damaging radiation sometimes
exceeded summer maxima at San Diego (32°N). These measured differences agree well with model calculations.
Large increases in UV-B were measured, despite the natural variability in cloudiness, at northern middle and high
latitudes in 1992/93 compared with previous years. These are the first reported examples of persistent increases
associated with anomalous ozone reductions over densely populated regions.
Clear-sky UV measurements at midlatitude locations in the Southern Hemisphere are significantly larger than at
a midlatitude site in the Northern Hemisphere, in agreement with the expected differences due to ozone column
and Sun-Earth separation. !
The increases in UV resulting from ozone reductions measured by satellite from 1979 to early 1994 have been
calculated, assuming other factors such as pollution and cloudiness did not change ^systematically over this period.
The calculated increases are largest at short wavelengths and at high latitudes. Poleward of 45°, the increases are
significantly greater in the Southern Hemisphere. At 45° (N and S), the calculated increase at 310 nm was approx-
imately 8 to 10 percent over this 15-year period, but there was considerable year-to-year variability.
Tropospheric ozone and aerosols can reduce global UV-B irradiances appreciably. At some locations, tropospher-
ic pollution may have increased since pre-industrial times, leading to some decreases in surface UV radiation.
However, recent trends in tropospheric pollution probably had only minor effects on UV trends relative to the
effect of stratospheric ozone reductions.
Only a few studies have monitored UV-B over time scales of decades, and these have yielded conflicting results
on the magnitude and even sign of the trends. Some studies may have been affected by problems with instrument
stability and calibration, and local pollution trends. Recently published data from unpolluted locations appear to
show the expected increases due to ozone depletion. The baseline UV irradiances present at mid and high latitudes
before ozone depletion began are not known. >
;i
Significant improvements have been made in UV instrumentation and its calibration. Intercomparisons between
spectro-radiometers show, however, that it is still difficult to achieve absolute calibration accuracies better than ±5
percent in the UV-B region. Therefore, the detection of future trends will require careful measurements at short
wavelengths that are more sensitive to changes in ozone. ;
,i
Cloud variability causes large temporal changes in UV. Although recent advances have been made, our ability to
realistically model cloud effects is still limited. i
Scattering by stratospheric aerosols from the Mt. Pinatubo volcanic eruption did not alter total UV irradiances
appreciably, but did increase the ratio of diffuse to direct radiation. '
9.1
-------
-------
SURFACE UV RADIATION
9.1 INTRODUCTION
Although the ultraviolet (UV) region represents
only a small component of the total solar spectrum, these
wavelengths are important because the photon energies
are comparable with molecular bond energies in the bio-
sphere. The UV radiation that reaches the Earth surface
can be arbitrarily divided into 2 sub-regions: UV-B (280-
315 nm), which is strongly absorbed by ozone; and
UV-A (315-400 nm), which is only weakly absorbed by
ozone. Less than 2 percent of the extra-terrestrial solar
energy falls within the UV-B range, and only a small
fraction of this reaches the surface.
Here we review progress in our understanding of
UV at the surface since the last assessment (WMO;
1992) and attempt to identify remaining gaps in our
knowledge. Impacts of UV increases (e.g., effects on the
biosphere, including human health and materials) are
outside the scope of this report and are discussed in the
UNEP "Effects Panel" reports (1991, 1994). Impacts on
tropospheric chemistry that may result from changes in
UV radiation fields are also discussed in Chapter 5 of
this report. These may lead to either positive or negative
feedbacks to stratospheric ozone depletion (UNEP, 1991
and 1994; Madronich and Granier, 1994).
Detailed reviews of our understanding of UV at
the surface can also be found in Tevini (1993) and Young
etal. (1993).
9.2 UPDATE ON TREND OBSERVATIONS
9.2.1 Results Derived from Broad-Band Meters
Analyses of broad-band data have focused on vari-
ability in the radiation received in specific geographic
regions over time scales of months to years. The much-
discussed work of Scotto et al. (1988) showed a decline
in annually integrated irradiance measured by eight Rob-
ertson-Berger (RB) meters in the continental United
States between 1974 and 1985. The average trend based
on all stations was -0.7 percent per year, while the statis-
tically significant values for individual stations varied
from -0.5 to -1.0 percent per year. A careful.analysis of
the RB meter's operating characteristics was carried out
shortly after the publication of Scotto et al. (1988).
These studies showed that the spectral response func-
tions of selected meters were remarkably stable over
time, although small differences between instruments
existed (DeLuisi et al., 1992). As part of this evaluation,
Kennedy and Sharp (1992) found no obvious problems
in the RB meter system apart from a well-documented
temperature sensitivity. This does not appear to be a like-
ly explanation for the downward trends found by Scotto
etal. (1988). However, some of the detailed information
required to assess the stability of the RB meter network
is no longer in existence. More recent work (DeLuisi,
1993; DeLuisi et al., 1994) has uncovered a potential
shift in calibration of the RB meter network in 1980 that
could remove the downward trend found by Scotto et al.
(1988). This issue merits further attention before defini- "
tive conclusions are reached.
Frederick and Weatherhead (1992) studied the
time series of RB data-from tv/o specific sites, Bismarck
(46.8°N) and Tallahassee (30.4°N), where Dobson col-
umn ozone data were available over the period from
1974 to 1985. They found that the derived trend in clear-
sky RB data during the summer months was consistent
with that expected from the Dobson data. However, dur-
ing winter, when the measured broad-band irradiances
were very small, a pronounced downward trend near -2
percent per year exists in the RB data. This differs in sign
from spectrally weighted irradiance calculations for
clear skies based on the Dobson ozone. The winter be-
havior in the RB data sets at Bismarck and Tallahassee is
not readily explained by any known change in the atmo-
sphere above these sites. Although the influences of
cloudiness and ozone in the boundary layer can be de-
tected in the output of the RB meter (Frederick et al.,
1993a), these influences are not likely to be causes of the
winter trends in broad-band irradiance.
Blumthaler and Ambach (1990) reported an up-
ward trend in RB readings made from an unpolluted site
in the Swiss Alps at latitude 47lbN during the period 1981
through 1989. Readings were Expressed as ratios to the
total solar irradiance measured; by a pyranometer so as to
remove the effects of aerosols. These measurements
have continued, and the upward trend in the ratios was
0.7 ± 0.3 percent per year to the end of 1991, but results
from 1992 were similar to those at start of the period.
The analysis did not examine the trend by month of the
year.
Recently, Zheng and Basher (1993) reported an
upward trend in clear-sky RE! data from Invercargill,
New Zealand, at 46°S. The observation site is in an un-
9.3
-------
SURFACE UV RADIATION
polluted region where changes in aerosols were small
over the observation period. The deduced trend is anti-
correlated in the expected way with column ozone data
from the same location.
Temperature coefficients of order 1%/K have been
reported for RB meters and their derivatives (Johnsen
and Moan, 1991; Btumthaler, 1993; Dichter et ai,
1994). Of the trend analyses above, only that by
Blumthaler and Ambach (1990) applied corrections for
instrument temperature changes. New generation tem-
perature-stabilized instruments are now available and
are being tested against spectro-radiometers (Grainger et
aL, 1993; McKenzie, 1994a).
9.2.2 Multi-Wavelength Measurements
The longest time series of UV irradiance at the
ground has been published by Correll et al. (1992). A
multi-filter instrument was used in Maryland (39°N,
77°W), over the period September 1975 to December
1990. The data show a large increase in UV-B, especially
at shorter wavelengths over the period 1980 to 1987. The
authors deduce from their measurements that the "RB-
weighted" UV (over the interval 295-320 nm, however)
would have increased by 35 percent over the period
1977-78 to 1985. This increase is much larger tharfex-
pected from stratospheric ozone losses. The integral
used would, however, show greater sensitivity to ozone
loss than a real RB meter, which is more responsive at
wavelengths longer than 320 nm in the UV-A region that
are unaffected by ozone changes. A decrease in the irra-
diances after 1987 may be a consequence of changes to
the instrument at that time, though the authors speculate
that changes in aerosols and cloud conditions may have
influenced the results.
9.2.3 Status of Trend Observations
The measurement of trends in UV is challenging
from an instrumental point of view, and the availability
and deployment of instruments to monitor trends in UV
have been far from ideal. Instrument development over
the past few years has continued to address the issues of
stability, spectral response, spectral resolution, cost, and
ease of maintenance in an attempt to meet the varied
needs of the community. Short-term process studies have
revealed strong anticorrelations between ozone and UV,
in agreement with those expected from model calcula-
tions (WMO, 1992). Thus there is no doubt that, in the
absence of other changes, reductions in stratospheric
ozone will result in UV increases. However, the results
of long-term studies have been conflicting. The network
of RB meters was never designed to measure long-term
trends, and questions still remain over the ability of
broad-band meters to achieve this i aim. Evidence now
suggests that changing aerosol (and cloud) conditions
can lead to increases or decreases in UV (Justus and
Murphey, 1994). Further comparisons between RB mea-
surements and pyranometer data at other sites are
warranted. It is significant that at unpolluted sites, the
observed increases in UV are comparable with those ex-
pected from ozone changes. Even at more polluted sites
where UV has apparently not increased, it is reasonable
to assert that current UV levels are greater than they
would otherwise have been without ozone depletion.
Better instruments are now available to monitor changes.
These include improved broad-band monitors and so-
phisticated spectro-radiometers that can distinguish
between changes caused by ozone and other effects such
as aerosols and clouds. However, if current predictions
are correct (see Chapter 13), much of the expected ozone
depletion has already occurred. It will therefore be im-
portant to maintain careful calibration of these
instruments over decadal time scales if trends in U V are
to be discerned from natural variability. Although mea-
surements from polluted sites will be of interest to
epidemiologists and for process studies, instruments de-
signed to monitor trends due to ozone depletion should
generally be located at remote sites where tropospheric
changes are minimized.
9.3 SPECTRO-RADIOMETER RESULTS
The observation period from spectro-radiometers
is too short to detect trends. However, multi-year data are
now available from a network of instruments operated
by the National Science Foundation (NSF) (Booth et al,
1994) and from several other groups (Gardiner et al.,
1993; McKenzie et al., 1993; Kerr and McElroy, 1993;
Ito et al., 1994). Process studies using these data have
already provided experimental corroboraition of the
modeled relationship between ozone and UV (WMO,
1992).
9.4
-------
SURFACE UV RADIATION
100
Ł
,c
CN
I
10
UJ
o
Q
on
0.1 -
0.01
Louder, New Zealand (Feb 21,
Neuherberg, Germany (Jul 29,
,1991)
1991)
• • i i
290 295 300 305 310 315 320
WAVELENGTH (nm) j
325
330
Figure 9-1. Measured clear-sky spectral irradiances in New Zealand and Germany "for solar zenith anqle
34.3 .The ozone column was 266 Dobson units (DU) in New Zealand and 352 DU in Germany Note the
logarithmic scale on the y-axis (adapted from Seckmeyer and McKenzie, 1992).
9.3.1 Intercomparisons
The measurement of solar U V spectral irradiances
is demanding. The steep slope of the solar spectrum in
the UV-B region (Figure 9-1) poses specific instrumental
problems that must be overcome to cope with the wide
dynamic range, the need to reject stray light adequately,
and the need to align the wavelength accurately (Mc-
Kenzie et al., 1992). An additional problem concerns
tracing the absolute calibration to a common standard.
National standards laboratories themselves disagree by
more than ±2 percent in the UV-B region (Walker et al.,
1991).
Excellent radiometric stability is required to mea-
sure UV trends or geographic differences. However,
recent intercomparisons have revealed large calibration
differences between some spectra-radiometers. Major
sources of uncertainty are instability of sensitivity and
cosine errors. Agreement at the ±5 percent level (Figure
9-2) is as good as can be expected al: present (Gardiner et
al., 1993; McKenzie et al., 1993; Seckmeyer et al,
1994b). Further field and laboratory intercalibrations be-
tween instruments are required!
Given these measurement uncertainties, it will
probably be necessary to use very short wavelengths in
the UV-B that have a high sensitivity to ozone change to
detect trends in UV due to the ozone, depletions expected
over the next decade. As one moves to shorter wave-
lengths, the sensitivity to ozone reductions increases
dramatically. For example, a; 1 percent reduction in
9.5
-------
SURFACE UV RADIATION
.if 150
CALCULATED, GREEN
MEASURED. IFU (BENTHAM)
aaana MEASURED, NIWA (J-Y)
00000 MEASURED. NIWA (BENTHAM)
MEASURED. ARL (SPEX)
IO 12 14 16
Hour(NZST=GMT+l2)
Figure 9-2. Comparison between measurements made with 4 spectro-radiometers at Lauder, New Zealand,
on Feb 23, 1993. Instruments included were from National Institute of Water and Atmospheric Research,
New Zealand (2), Australian Radiation Laboratory, Australia, and Fraunhofer Institute for Atmospheric Envi-
•tonment, Germany. Clear-sky model results are shown for comparison, although the observation day was
not perfectly clear (adapted from McKenzie et a/., 1993).
ozone-.causes an increase of approximately 1 percent in
UV at 310 run, whereas the increase at 300 nm is 3 to 4
times as large (see Figure 9-12).
9.3.2 Geographic Differences
Although large geographical differences in UV-B
are expected from theoretical considerations, there have
been few published studies demonstrating measured
geographic differences in UV-B radiation. A climatology
obtained from a network of RB meters in the 1970s
(Berger and Urbach, 1982) may be biased by the strong
temperature coefficient of these instruments. Although
the UV data base is improving, it still remains largely
uncoordinated. Large latitudinal gradients have, howev-
er, been observed from the NSF network of
spectro-radiometers, as discussed in Section 9.3.3
(Booth et al, 1994).
Geographic intercomparisons based on measure-
ments from the same instrument (Seckmeyer and
McKenzie, 1992) have shown that for clear-sky observ-
ing conditions and similar solar zenith angles, UV
irradiances measured in Europe are much less than in
New Zealand (Figure 9-1). The differences are larger
than expected from calculations using an earlier ozone
climatology, though their spectral characteristics indi-
9.6
-------
SURFACE UV RADIATION
2.0
M
~Z.
I 1.5
Q
LoJ
.
^
o
o
g
^0.5
cr ' '
0.0
Measured Ratio
Calculated Ratio
MELBOURNE AUSTRALIA / LAUDER NZ
DoY=29, SZA=19.8, Ozone=259 Du
NEUHERBERG GERMANY / LAUDER NZ
DoY=194, SZA=26.3, Ozone=310 Du
290 300 310 320
330 340 350 360 370
WAVELENGTH (nm)
380 390 400
Figure 9-3. Geographic comparison between maximum clear-sky spectra measured in three countries The
ratios are with respect to a spectrum measured at Lauder on Dec. 27, 1992 (Day-of-Year [DoY] =362
sza=21.8°, ozone=2781 DU). The smooth curves show calculated ratios assuming similar albedos and aero-
sol properties (adapted from McKenzie et al., 1993)
cate that they are primarily due to ozone. This illustrates
the importance of tropospheric ozone, which has in-
creased in Europe (Staehelin and Schmid, 1991).
Data from cross-calibrated instruments have been
used to compare the maximum clear-sky irradiances
measured over several summers at three sites (McKenzie
etal., 1993). Ratios of these maximum clear-sky spectra
obtained are shown in Figure 9-3. The maximum DNA-
weighted UV (Setlow, 1974) measured in New Zealand
(45°S) was 50 percent greater than at a similar latitude in
Germany (48°N). UV irradiances in Australia (38°S)
were significantly higher than in New Zealand. Figure
9-3 also shows ratios calculated with a simple model, as-
suming no differences in aerosol loading. The calculated
differences in UV are due to differences in ozone, sun
angle, and Earth-Sun separation. Measured and calculat-
ed ratios are in agreement within experimental
uncertainties.
9.3.3 High Latitude (North and South)
Year-to-year variability in cloudiness is among the
largest sources of variance in monthly integrated UV ir-
radiance measured at the ground (Frederick et al.,
1993b; Diaz et al., 1994), althpugh this can vary from
one location to the next, depending on the timing and
severity of ozone depletions. At the NSF site in Ushuaia,
9.7
-------
SURFACE UV RADIATION
500 -r
400 -:
8 300 .
o
200 ..
11-Nov
-r 0-04
100
270 280 290 300 310 320 330 340 350 360
Day of Year, 1991
0)
-------
SURFACE UV RADIATION
Argentina (54.6°S), the lowest ozone column amounts to
date (1988-1992) were in 1992. However, the highest
UV irradiances occurred in 1990, when the ozone hole
persisted and was displaced towards South America In
December 1991, the erythemal UV (McKinlay and Dif-
fey, 1987) was 45 percent larger than the zonal mean
climatology, which is equivalent to moving 20° closer to
the equator. Because radiation at 305 nm is sensitive to
both ozone and cloud changes, whereas 340 nm radia-
tion is insensitive to ozone, Frederick et al (1993C) have
investigated the irradiance ratio I305fl340 to remove
cloud effects from UV measurements in Ushuaia Over
the summers of 1989-90 to 1992-93, these ratios were
significantly larger than those deduced from a climatolo-
gy of ozone measurements obtained over the period
1980 to 1986 (Frederick et al., 1993c).
The unique geometry at South Pole Station (90°S)
means that there are no diurnal cycles in solar zenith an-
gle. This simplifies investigation of the relationship
between UV, ozone, and other parameters. The strong
anticorrelation between UV and ozone is demonstrated
by Figure 9-4, which shows a UV maximum occurring
on 11 November 1991 (day 315); a day when ozone was
a local minimum. As is normal for Antarctica, the high-
est instantaneous UV irradiances (erythema, or UV-B)
do not occur at the time of the greatest ozone depletion,
but at a time closer to the summer solstice, combining
the effects of higher solar zenith angles with relatively
low ozone. In contrast, visible radiation increases steadi-
ly as the solar zenith angle decreases over the
observation period (Figure 9-4). Perturbations by clouds
are relatively small at this site, probably due to the high
surface albedo and to extremely cold temperatures,
winch keep clouds from becoming optically thick. Al-
though the relative increases are large, the absolute UV
irradiances at this site are still small compared with those
at mid or low latitudes:
Huge year-to-year variations in UV have been
measured at the South Pole. These correlate with the lo-
cation of the polar vortex and the persistence of
springtime ozone depletion to times when higher solar
elevations occur. Figure 9-5 shows that there are distinct
differences between the timing of the seasonal maxima
of UV-B and visible radiation. The UV has a maximum
m the spring, whereas longer-wavelength radiation
peaks near the summer solstice. UV in the range 298-
303 nm was elevated by a factor of 4 in 1992 compared
Jan-94
™«r ooul"f;wHoui]y sPectral irradiance integrated
over 298-303 nm (upper panel) and over 338-342
nm (lower panel) at the South Pole between 1991
and mid-1994. Dotted vertical lines mark the sum-
mariJq?CeS' Adapted and uPda'ted from Booth et
with 1991 (Booth etal., 1994), and the 1992 maximum
(November 29) occurred 18 days later than in 1991.
Year-to-year variations were much smaller at longer
wavelengths (338-342 nm) where ozone absorptions are
small. In the 1993 austral spring, the lowest ever total
column ozone amounts were recorded over Antarctica
bringing the highest UV irradiances for October at the
South Pole.
The effects of Antarctic ozone depletions on UV
irradiances have been clearly observed by comparisons
with Arctic data. Figure 9-6'compares noontime irradi-
ances (DNA-weighted and visible) from South Pole with
those from an Arctic site at Barrow, Alaska (71.2°N,
156.5°W) as a function of solar zenith angle. DNA-
9.9
-------
SURFACE UV RADIATION
[7 South Pole 1991
o— Barrow 1991
65
70
75 80 85 90
Solar Zenith Angle: Degree
100
South Pole 1991
Barrow 1991
65
70
75 80 85 90
Solar Zenith Angle: Degree
95
100
Figure 9-6. Comparison of time series of noontime
radiation measurements at the South Pole in 1991
(solid squares) and Barrow, Alaska, (open squares)
in the spring of 1991, plotted as a function of solar
zenith angle. The upper panel shows DNA-weight-
ed UV-B radiation, and the lower panel shows total
visible radiation, 400-600 nm (Booth etal., 1993).
weighted UV is several fold larger at South Pole during
the period of the ozone "hole," while the visible irradi-
ances are generally similar at both sites for similar solar
zenith angles. In summer, the solar elevations are larger
at Barrow than at South Pole, and the UV irradiances are
larger. The lower panel of Figure 9-6 shows that cloud
effects are relatively small at these sites. The NSF net-
work installation at Barrow, Alaska, showed
significantly elevated springtime UV-B in 1993 com-
pared with previous years (Booth et al., 1993).
At Palmer Station (64.5°S, 64°W), the highest bio-
logically weighted UV doses of the six-year NSF
network monitoring period were observed in late Octo-
ber of 1993, surpassing the previous records of early
" December, 1990. During this period, the noon readings
of biologically damaging UV even exceeded the summer
maximum measured at San Diego (32°N), as shown in
Figure 9-7. Additionally, daily integrals of biologically
weighted UV measured during the spring at Palmer Sta-
tion sometimes exceeded those measured in summer in
San Diego. Unlike the changes at the South Pole, the
large UV doses at Palmer Station may have important
biological consequences given the diversity of the ma-
rine ecosystem at these latitudes.
Large increases in spectral UV irradiance were
observed in the Southern Ocean during the spring of
1990 as ozone-depleted air in the Antarctic vortex
moved across the sampling site. These enhancements,
which were apparent at the surface and beneath the
ocean surface to depths of 35 m, were shown to have
adverse effects on marine primary production (Smith et
al, 1992; Prezelin et ai, 1994; UNEP, 1994). Calcula-
tions with a coupled atmosphere-ocean radiative transfer
model show that the effect of ozone depletion on UV-B
penetration into the water depends on solar zenith angle
and is more pronounced in spring than in summer (Zeng
etal, 1993).
9.3.4 Northern Hemisphere Midlatitude
Large ozone depletions have been measured at
mid-Northern latitudes in 1992 and 1993. In the late
winter and spring of 1993, the ozone was 7 percent be-
low the climatological envelope. Decreases were larger
at high northern latitudes, but smaller at equatorial and
southern latitudes (Herman and Larko, 1994; also see
Chapter 1). Depletions continued into the summer, when
the UV irradiances are greatest.
The study by Kerr and McElroy (1993) was the
first to show the effects of ozone depletions on integrated
daily UV-B at midlatitudes, including cloudy conditions.
Over the 4-year period to August 1993 there were large
changes in ozone measured over Toronto, Canada (44°N,
29°W). Although there are gaps in the data and the obser-
vation period is rather short, a statistical analysis was
performed. The ozone change over this period was re-
ported as -4 percent per year in the winter months, and -2
percent per year for the summer, as measured by the
Brewer spectrometer. The corresponding temporal
changes in U V were small at wavelengths above 320 nm
and increased toward shorter wavelengths. The increase
at 300 nm was 35 ± 20 percent per year in winter (when
UV flux is in any case rather weak) and 6 ± 10 percent
per year in summer. The statistical significance of these
results has been disputed (Michaels et al., 1994) because
some of the results were influenced by a few days in
9.10
-------
SURFACE UV RADIATION
Palmer
San Diego
0
0 10 20 30 40 50 ,60
Solar Zenith Angle (degrees)
70
80
.Figure 9-7. DNA-weighted noon irradiances measured in 1993 versus solar zenith angle: Palmer Station
Antarctica, compared with San Diego, USA. i
March 1993 when particularly low ozone values were
observed and because statistically significant increases
at the shorter wavelengths occurred only in the last year.
A later month-by-month analysis (Kerr and McElroy,
1994) showed that the increases at 300 nm persisted for
several months.
By 1994, UV-B measurements had reverted to lev-
els similar to those seen prior to 1992 (unpublished
data), showing that the enhancements in 1992/1993 are
better described as a perturbation, rather than a trend. In
Toronto in the summer of 1993, ozone was 7.4 percent
less than in 1989, and in the winter of 1992-93, ozone
was 10.9 percent less than in 1989-90. As can be seen
from Figure 9-8, changes in UV were small (statistically
insignificant) at wavelengths greater than 320 nm, but
were very large at 300 nm. The UV at 300 nm increased
by factors of 1.3 and 1.9 in the summer and winter re-
spectively. The increases were significant at the 95
percent confidence level. The resulting spectral 'differ-
ences in mean daily UV fluxes for the high-to-low year
comparison show clearly that they are caused by ozone.
Biologically weighted UV increases were clearly signif-
icant.
UV-B increases due to the lower ozone amounts in
1993 have also been reported in Europe. Large increases
in UV-B. in 1993 compared with 1992 were measured in
Germany by Seckmeyer et al. (1994a), despite lower
UV-A due to increased cloudiness in 1993. The UV re-
covery was incomplete at this site in mid-1994. The high
variability of cloud cover masked the detection of possi-
ble increases in- UV-B measured with broad-band
detectors, and no significant UV increases due to ozone
depletion were measured with the RB meter in Inns-
bruck (Austria) during the winter/spring of 1993
compared with the 1981-1988 period (Blumthaler et al,
1994a). !
9.11
-------
SURFACE UV RADIATION
WINTER RATIO
ooooo SUMMER RATIO
235
300 305 310 315 320 325
WAVELENGTH (nm)
Figure 9-8. Impact of low ozone over Toronto,
Canada, in 1992/1993 compared with earlier years.
The top panel .shows the mean daily UV flux as a
function of wavelength for the summers of 1989
and 1993, and the winters of 1989-90 and 1992-93.
The middle panel shows flux ratios for summer
(1993 divided by 1989) and for winter (1992-93 di-
vided by 1989-90). The bottom panel compares the
observed changes as a function of wavelength with
the ozone absorption spectrum. The log of the win-
ter ratio is used because the intensity of UV-B
radiation depends on the exponent of the absorp-
tion coefficient of ozone (adapted from Kerr and
McElroy, 1993).
Spectral UV-B measurements made during the low
ozone event of 1992/93 indicate that ozone decreases of
5-10 percent result in detectable increases of UV-B un-
der all types of weather conditions. These decreases in
ozone are similar in magnitude to long-term accumulat-
ed ozone losses at midlatitudes, as noted in Chapter 1.
The confidence with which past and future trends can be
determined will improve as the records of spectral UV-B
measurements become longer. *
9.4 IMPLICATIONS OF RECENT CHANGES
9.4.1 Stratospheric Aerosols from the Mt.
Pinatubo Eruption
Although the Mt. Pinatubo eruption reduced glo-
bal (i.e., diffuse + direct) solar irradiance at the surface,
any reductions were small in the UV region (Blumthaler
and Ambach, 1994). However, there was a marked in-
crease in the clear-sky diffuse/direct ratio throughout the
UV region (Figure 9-9), so that shaded areas received
substantially more UV in the summer following the
eruption (McKenzie, 1994b; Blumthaler and Ambach,
1994). Some decreases in global UV have been reported
(Smith et al., 1993), but these decreases may be due to
imperfect cosine responses of those instruments that un-
derestimate the diffuse component from large zenith
angles.
Model calculations suggest that aerosols from vol-
canic eruptions reduce the direct beam component, but
increase scattered skylight, so that any decreases in glo-
bal irradiance are small. Calculations show that under
some conditions, volcanic aerosols can lead to increases
at short wavelengths within the UV-B region (Michelan-
geli et al., 1992), particularly at large solar zenith angles
and for high surface albedos (Davies, 1993; Tsay and
Stamnes, 1992).
The volcanic aerosol provides sites for heteroge-
neous chemistry to occur, leading to potential losses of
ozone as discussed in Chapter 1 and Chapter 4. This
would lead to additional enhancements of UV-B.
9.4.2 Tropospheric Pollution
Although tropospheric aerosols attenuate the, di-
rect beam (Blumthaler et al., 1993), there is a lack of
consensus regarding their effect on global irradiances.
Some measurements suggest that there is only a small
9.12
-------
SURFACE UV RADIATION
1.2
1.0
O
0.8
O
Ld
^0.6
Q
Ld
00
0.4
u_
Q
0.2
0.0
YEAR , DAY SZA OZONE
Pre-Pinatubo, 1990 i 336 23.2 301
Post-Pinotubo.1991 j 338 22.9 307
290, 310 330 350 370 390 410
WAVELENGTH (nm) \
430
450
Figure 9-9. Comparison of clear-sky diffuse/direct ratios measured over Lauder, New Zealand as a function
1994a) S'm S°'ar Zen'th an°leS bef°re and after the emption °f Ml Pinatubo (from McKenzie,
effect on global irradiances (Seckmeyer and McKenzie,
1992; McKenzie et al., 1993). Other measurements
show that there are situations where they reduce UV irra-
diances considerably (Seckmeyer et al., 1994a). Some
model results suggest that aerosol effects can be large
(Liuetai, 1991).
Some regions, particularly in the Northern Hemi-
sphere, have experienced increased tropospheric
pollution (mostly sulfate aerosols and ozone) during the
last century. It has been estimated that the corresponding
UV (DNA-weighted) could have been reduced by 6-18
percent from the sulfate aerosol increases (Liu et al.,
1991) and by 3-10 percent from the tropospheric ozone
increases (UNEP, 1991) in some industrialized regions.
However, no direct information exists on pre-industrial
stratospheric ozone, precluding accurate estimates of the
net UV changes. ;
More recent tropospheric ozone trends in industri-
alized regions are estimated to contribute at most -2
percent per decade to the DNA-weighted UV, compared
to +5 to +11 percent per decade from midlatitude ozone
reductions (UNEP, 1991).; Sulfur emissions have recent-
ly decreased in some regions while increasing in others
(NRC, 1986), and the corresponding UV changes are
expected to reflect such local variations. .
Large increases in UV have been measured at high
altitudes in Europe and South,America. These altitude.
effects become more pronounced at shorter wavelengths
9.13
-------
SURFACE UV RADIATION
(Cabrera et ai, 1994; Blumthaler et al, 1994b).,At 300
nm, increases of 24 ± 4%/km have been measured in
Europe for snow-free conditions (Blumthaler et al,
1994b). U V-B increases of 18%/km have also been mea-
sured, although this included the effect of snow cover at
the high elevation site (Ambach et al., 1993). Larger gra-
dients in UV-B have been observed during the winter
near Santiago, Chile (33°S), though the same study re-
ported gradients of only 4-5%/km in less polluted
regions (Cabrera et al., 1994). The calculated gradients
for clear conditions are typically 5-8%/km (Madronich,
1993). Larger gradients result from increased tropo-
spheric ozone or aerosols.
High concentrations of tropospheric pollutant
gases (e.g., S02, NO2, ©3) can also have a significant
influence on surface UV irradiances (Bais et al., 1993).
9.4.3 Magnitude of Changes
Recent ozone losses in the Northern Hemisphere
have been much larger than expected (Herman and Lar-
ko, 1994; Chapter 4), so that UV increases are much
larger. For the first time, greatly enhanced UV was seen
for extended periods of time in heavily populated lati-
tude bands, and there may be future implications for
human health (UNEP, 1994). However, the UV irradi-
ances in 1993 were still less than for comparable
southern latitudes where ozone and aerosol concentra-
tions are lower, and where the minimum Sun-Earth
separation occurs in summer.
Previously, the Radiation Amplification Factor
(RAF) for changes in ozone was defined in terms of a
linear relationship between incremental changes in
ozone (AO3) and UV (AE):
RAF=-(AE/E)/(A03/03)
(9-1)
If this definition is (incorrectly) applied to the
large depletions in ozone that have occurred recently, the
magnitude of the deduced increase in UV is underesti-
mated. To avoid this problem, the radiative change due to
ozone depletion has been reformulated in terms of a
power law (Madronich, 1993) so that:
= ln(E*/E)/ln(03/03*),
(9-2)
where Ł* and Ł are two UV irradiances, and 03* and Oj
are corresponding ozone amounts. With this definition,
previously calculated RAF values, which agree well with
measurements (e.g., UNEP, 1991), can still be used to
•deduce the increases in UV caused by the large reduc-
tions in ozone that have occurred in Antarctica and more
recently at midlatitudes. For example, Booth and Mad-
ronich (1994) have used measurements from Antarctica
to show that the power relationship works well, even for
ozone variations of a factor of two (Figure 9-10).
9.5 UPDATE ON PREDICTIONS
9.5.1 Semi-Empirical Method
No suitable data base exists to directly measure
changes in UV that may have already occurred as a result
of ozone depletion. Unfortunately, the potential to calcu-
late temporal changes in U V at the surface is also limited
by inadequacies in our capability to model the effects of
- clouds. A semi-empirical technique has been imple-
mented to overcome these difficulties, so that UV-B can
be inferred using solar pyranometer data to estimate
cloud effects, and ozone data (Ito et al., 1994). Satellite
ozone data suitable for these studies are available from
the year 1978, when ozone depletions were small.
The relationship between pyranometer data and
ozone data to derive UV-B was verified using ground-
based measurements of UV spectra at four sites in Japan,
and the technique has been applied to infer historical
records of UV over an eleven-year period at these sites.
Over this period, the long-term changes were found to be
small compared with the year-to-year variability. The
geographical distribution of UV over Japan has also
been deduced (Ito et al., 1994).
Although the technique is imperfect, the historical
record and geographical differences derived may pro-
vide useful information for users such as
epidemiologists. The method will be more useful 'if it
can be successfully applied to biologically weighted UV
irradiances (e.g., erythemal irradiance) rather than an un-
weighted integral (290-315 nm) which is relatively
insensitive to ozone changes. ;
9.5.2 Calculated Changes in Clear-Sky UV
Using Global Ozone Measurements
A multi-layer radiative transfer model (Madron-
ich, 1993) was used to calculate UV irradiances (i.e., the
flux passing through a horizontal surface) and their
9.14
-------
200%
150%
^100%
0)
2! 50% -
o
c
0% -
-50% -
SURFACE UV RADIATION
Power RAF =1.1
linear RAF =1.1
Measured
-60%
-50%
-40% -30% -20%
Decrease in Ozone
-10%
0%
pt^^
changes over time as a function of latitude using ozone
fields from the Solar Backscatter Ultraviolet spectrome-
ter (SBUV) and.SBUV2 satellite instruments (see
Chapter 1) over the period late 1978 through early 1994.
The calculations presented are for clear-sky aerosol-free
conditions, with a constant surface albedo of 0.05. The
sensitivity of this model to changes in ozone has been
assessed previously and agrees well with measurements
(McKenzie et al, 1991; UNEP, 1991). Here, we report
calculated irradiances at selected wavelengths in the UV
region. Corresponding biologically-weighted irradi-
ances are discussed in the UNEP "Effects Panel" report
(1994).
The calculated latitudinal variation in clear-sky
UV for selected wavelengths using satellite ozone data
over the period 1979 to 1992 is shown in Figure 9-11.
The irradiances increase strongly with wavelength (note
the logarithmic scale) and have maxima near the equator.
Latitudinal gradients and hemispheric asymmetries in-
crease at shorter wavelengths, where ozone absorptions
are greatest. The hemispheric differences are most pro-
nounced at latitudes poleward of 45°. At the shortest
wavelength shown (300 nm), tlje daily spectral irradi-
ance at the South Pole is an order of magnitude greater
than at the North Pole.
The changes in these quantities over the period
1978 to 1994 (relative to the rrieari of the period) are
shown in Figure 9-12. Changes'are largest at latitudes
where ozone depletions have been most severe, so that
percentage trends increase towards the poles, with larg-
est increases in the Southern Hemisphere. The effects of
ozone reduction are much more important at shorter
wavelengths.
The calculated time dependence of changes in 310
nm UV at latitudes 45° and 55° (N and S) for the period
1979 to 1994 is illustrated in Figure 9-13. The rate of
increase in UV is not constant, but is anticorrelated with
ozone changes which include perturbations due to the
11-year solar cycle. Hemispheric jdifferences in the tim-
ing of the increases are also j apparent. Percentage
changes generally lead the absolute changes by a few
months, as expected from the timings of greatest ozone
9.15
-------
SURFACE UV RADIATION
Dally spectral Irradiance. 1979-92 mean
10s fe . . i . • i • • i ' ' ' ' ' ' '
T 10*
E
c
CM 10'
E
• 300 nm
• 310 nm
A 320 nm
O 340 nm
-90 -60 -30 0 30
Latitude
90
Figure 9-11. Calculated daily spectral irradiance,
averaged over all months of 1979-1992, at different
wavelengths. Sea level, cloudless and aerosol-'free
skies.
depletion (winter, spring) compared with the greatest
natural UV levels (summer). The absolute changes ap-
proach zero in winter, when the UV has a minimum.
At latitude 45° the trend is approximately +0.5 per-
cent per year in both hemispheres. At latitude 55° the
trends are significantly larger, particularly in the South-
em Hemisphere. Gradients are larger at shorter
wavelengths and continue to increase at higher latitudes,
where hemispheric differences become more pro-
nounced.
9.5.3 Cloud and Albedo Effects
The analysis in Section 9.5.2 assumes cloud-free
conditions. In practice, cloud variability causes large
year-to-year changes in UV. The theory of radiative
transfer through clouds is well developed, and algo-
rithms for its numerical implementation are available
(e.S., Stamnes et aL. 1988). However, the practical appli-
cation of the theory to the atmosphere is still limited
because of incomplete cloud characterization.
Cloud cover at most surface observation sites is
specified only as the fraction of sky covered by cloud,
with little or no information about the optical depth or
layering. Further, although cloud optical depth is not a
strong function of wavelength, there is a nonlinear rela-
tionship between observed cloud cover and its effect in
the UV-B region where a much larger fraction of the en-
ergy is diffuse (Seckmeyer et al., 1994a). Measured
reductions in UV-B are relatively small even for large
fractional cloud covers (Ito et al, 1994; Bais et al,
1993).
Satellite measurements of clouds are more quanti-
tative, but stratification of clouds is difficult to measure,
and the cloud cover viewed from space is not generally
the same as that viewed from the ground (Henderson-
Sellers and McGuffie, 1990). Other complications arise
from the nonlinear relationship between UV transmis-
sion and cloud optical depth, and the fact'. that cloud
effects are modulated by surface albedo (Liibin et al.,
1994). Generally, with high surface albedo, the effective
optical depth of clouds is reduced by multiple scattering
effects between the surface and the cloud base, which
Slope of annual spectral Irradiance,
relative .to 1979-93 mean
-2
-90
-60 -30 0 30
Latitude
60
Figure 9-12. Calculated rate of increase of the an-
nual spectral irradiances from 1978 to 1993. For
comparison, the negative quadrants give thp
changes in annually averaged ozone column. Val-
ues are least-squares slopes expressed as percent
of the 1979-1993 mean. Error bars are 2o.
9.76
-------
SURFACE UV RADIATION
310nm 4SN
400
310nm 55N
0 %
79 81 83 85 87 89 91 93 95
400
0 %
79 81 83 85 87 89 91 93 95
310nm 45S
400
310nin 55S
0 %
/9 81 83 85 87 89 '91
: -20
-40
400
0 %
79 81 83 85 87 89 91 93
: -20
-40
9.17
-------
SURFACE UV RADIATION
enhance the flux. Methods have, however, recently been
developed and successfully implemented to map surface
UV-B using multi-spectral satellite imagery (Lubin et
al, 1994).
9.5.4 UV Forecasts
In recent years, efforts have been made in several
countries to educate the public concerning ambient UV-
B levels. This information is often reported in the form
of a daily UV index delivered with local weather fore-
casts (c.g., Burrows et al., 1994). Most of the indices in
current use are based on erythemally weighted UV and
are reported in a variety of forms, including arbitrary
scales, weighted energy dose units, "burn times," and
others. The information would be more useful if a single
index could be agreed upon. The values forecast for
these indices can be based on measurements, or models,
or a combination of both. To be useful, such forecasts
must be capable of assimilating ozone measurements in
near real time and predicting changes in ozone fields
within a few hours. No operational forecasts currently
make realistic allowances for changes due to clouds.
Ground-truthing and verifying predictive algorithms
will be important in the development of UV indices.
9.6 GAPS IN OUR KNOWLEDGE
High quality extraterrestrial irradiances are re-
quired to test models against measurements and to
deduce accurately the spectral consequences of changes
in aerosol optical depth. New irradiance measurements
from instruments on board the Upper Atmosphere Re-
search Satellite (UARS) may fill this need (Lean et al.,
1992; Brueckner etai, 1993; Rottman etal, 1993).
Despite the importance of clouds in modulating
UV transfer through the atmosphere, our ability to model
' their effects is poor. The role of aerosols has not been
fully determined.
Detailed intercomparisons between measured and
modeled UV are now being attempted (Wang and Leno-
ble, 1994; Zeng et al, 1994). These require a wide range
of measured input parameters (e.g., aerosol and ozone
profiles, cloud cover) to constrain the models. These
measurements are often not available or inadequate. The
validity of parameterizations of these quantities is also
untested.
The achievable accuracy of UV measurements is
limited by the lack of suitable irradiance standards. Ro-
bust protocols to maintain secondary standards and to
transfer them accurately to field instruments are also
lacking.
Detailed instrument intercomparisons and instru-
ment-model comparisons are limited by our
understanding of the effects of instrument errors due to
imperfections in the cosine response. One approach
would be to develop improved detectors for which these
errors are small. In addition to cosine-weighted mea-
surements that are already available, measurements of
the angular dependence of sky radiances, altitude-depen-
dences, and direct-sun observations may be useful for
model validation (Seckmeyer and Bernhard, 1993).
Historical and geographic changes in U V radiation
are not adequately understood. The data set produced by
a network of broad-band meters would be a yaluable
source of information for the photobiology and epidemi-
ology communities. All instruments must be
characterized and calibrated in the same way. In the past
there has been a lack of international coordination. The
data from numerous, uncoordinated meters, while not
necessarily incorrect, could provide questionable, and
sometimes conflicting, information on long-term chang-
es in broad-band solar UV radiation at the ground.
There is a lack of high quality spectral measure-
ments of UV and ancillary measurements from the same
site from which photobiological effects can be evaluat-
ed, and our understanding of the reasons for changes in
UV can be improved. Useful ancillary measurements ii>
clude ozone, total solar irradiance, aerosols (turbidity),
and cloud cover.
REFERENCES
Ambach, W., M. Blumthaler, and T. Schopf, Increase of
biologically effective ultraviolet radiation with al-
titude, J. Wilderness Medicine, 4, 189-197, 1993.
Bais B.A., C.S. Zerefos, C. Meleti, I.C. Ziomas, and K.
Tourpali, Spectral measurements of solar UVB ra-
diation and its relations to total ozone, SO2, and
clouds, J. Geophys. Res., 98, D3, 5199-5204,
1993.
Berger, D.S., and F. Urbach, A climatology of sunburn-
ing ultraviolet radiation, Photochem. Photobiol.,
35, 187-192, 1982.
9.18
-------
SURFACE UV RADIATION
Blumthaler, M., Solar UV measurements, in UV-B Radi-
ation and Ozone Depletion: Effects on Humans,
Animals, Plants, Microorganisms, and Materials,
edited by M. Tevini, Lewis Publishers, Boca Ra-
ton, Florida, USA, 71-94, 1993.
Blumthaler, M., and W. Ambach, Indication of increas-
ing solar ultraviolet-B radiation in alpine regions,
Science, 248, 206-208, 1990.
Blumthaler, M., and W. Ambach, Changes in solar radia-
tion fluxes after the Pinatubo eruption, Tellus, 46B
76-78, 1994.
Blumthaler, M., W. Ambach, R. Silbernagl, and J.
Staehelin, Erythemal UV-B irradiance (Robert-
son-Berger sunburn meter data) under ozone
deficiencies in winter/spring 1993, Photochem.
Photobiol., 59, 6, 657-659, 1994a.
Blumthaler, M., M. Huber, and W. Ambach, Measure-
ments of direct and global UV spectra under
varying turbidity, in Proceedings International
Symposium on High Latitude Optics, Tromso,
Norway, 28 June-2 July, 1993.
Blumthaler, M., A.R. Webb, G. Seckmeyer, and A.F.
Bais, Simultaneous spectral UV measurements
applied to the determination of the altitude effect,
presented at WMO Meeting of Experts on UV-B
Measurements, Les Diablerets, Switzerland, 18-
21 July, 1994b.
Booth, C.R., and S. Madronich, Radiation Amplification
Factors: Improved formulation accounts for large
increases in ultraviolet radiation associated with
Antarctic ozone depletion, in Ultraviolet Radia-
tion in Antarctica: Measurements and Biological
Research, edited by C. S. Weiler and P. A. Penhale,
AGU Antarctic Research Series, volume 62,
American Geophysical Union, Washington, D.C
39-42, 1994.
Booth, C.R., T.B. Lucas, J.H. Morrow, C.S. Weiler, and
P.A. Penhale, The United States National Science
Foundation/Antarctic Program's Network for
Monitoring Ultraviolet Radiation, in Ultraviolet
Radiation in Antarctica: Measurements and Bio-
logical Research, edited by C. S. Weiler and P. A.
Penhale, AGU Antarctic Research Series, volume
62, American Geophysical Union, Washington,
D.C., 17-37, 1994.
Booth, C.R., T.M. Mestechlcina, T. Lucas, J. Tusson, and
""J. Morrow, Contrasts; in Southern and Northern
Hemisphere high latitude ultraviolet irradiance, in
Proceedings International Symposium on High
Latitude Optics, European Optical Society and In-
ternational Society for Optical Engineering,
Tromso, Norway, 28 J'une-2 July, 1993.
Brueckner, G.E., K.L. Edlow, L.E. Floyd, J.L. Lean, and
M.E. VanHoosier, The Solar Ultraviolet Spectral
Irradiance Monitor (SUSIM) experiment on board
the Upper Atmosphere Research Satellite
(UARS),/. Geophys. Res., 98, D6, 10695-10711
1993. ,
Burrows, W.R., M. Vallee, D.I. Wardle, J.B. Kerr, L.J.
Wilson, and D.W. Tarasick, The Canadian opera-
tional procedure for forecasting total ozone and
UV radiation, accept
-------
SURFACE UV RADIATION
Dichter, B.K., DJ. Beaubien, and A.R Beaubien, Effects
of ambient temperature changes on UV-B mea-
surements using thermally regulated and
unregulated fluorescent phosphor ultraviolet pyra-
nometers, in Proceedings Quadrennial Ozone
Symposium, Charlottesville, Virginia, June 1992,
NASA Conf. Public. 3266,766-769, 1994.
Frederick, J.E., and E.C. Weatherhead, Temporal chang-
es in surface ultraviolet radiation: A study of the
Robertson-Berger meter and Dobson data records,
Photochem. Photobioi, 56, 123-131, 1992.
Frederick, I.E., A.E. Koob, A.D. Alberts, and E.C.
Weatherhead, Empirical studies of tropospheric
transmission in the ultraviolet: Broad band mea-
surements, J. Appl. Meteorol, 32, 1883-1892,
1993a.
Frederick, J.E., P.F. Soulen, S.B. Diaz, I. Smolskaia,
C.R. Booth, T. Lucas, and D. Neuschuler, Solar ul-
traviolet irradiance observed from Southern
Argentina: September 1990 to March 1991, J.
Geophys. Res., 98, D5, 8891-8897,1993b.
Frederick, J.E., S.B. Diaz, I. Smolskaia, W. Esposito, T.
Lucas, and C.R. Booth, Ultraviolet solar radiation
in the high latitudes of South America, submitted
to Photochem. Photobioi., October 1993c.
Gardiner, B.C., A.R. Webb, A.F. Bais, M. Blumthaler, I.
Dirmhim, P. Foster, D. Gillotay, K. Henriksen, M.
Huber, P. J. Kirsch, P.C. Simon, T. Svenoe, P.
Weihs, and C.S. Zerefos, European intercompari-
son of ultraviolet spectroradiometers, Environ.
Techn., 14, 25-43, 1993.
Grainger, R.G., R.E. Basher, and R.L. McKenzie, UV-B
Robertson-Berger characterization and field cali-
bration, Appl. Optics, 32, 3,343-349, 1993.
Henderson-Sellers, A., and K. McGuffie, Are cloud
amounts estimated from satellite sensor and con-
ventional surface-based observations related?, Int.
J. Remote Sensing, 11, 3,543-550, 1990.
Herman, J.R., and D. Larko, Low ozone amounts during
1992-1993 from NIMBUS-7 and METEOR-3/
TOMS, /. Geophys. Res., 99, D2, 3483-3496,
1994.
Ito, T., Y. Sakoda, T. Uekubo, H. Naganuma, M. Fukuda,
and M. Hayashi, Scientific application of UV-B
observations from JMA network, in Proceedings
13th UOEH International Symposium and the 2nd
Pan Pacific Cooperative Symposium on Human
Health and Ecosystems, Kitakyushu, Japan, 13-15
October 1993, in press, 1994.
Johnsen, B., and J. Moan, The temperature sensitivity of
the Robertson-Berger sunburn meter, model 500,
/. Photochem. Photobioi. B.: Biol, 11, 277-284,
1991.
Justus, C.G., and B.B. Murphey, Temporal trends in sur-
face irradiance at ultraviolet wavelengths, J.
Geophys. Res., 99, Dl, 1389-1394, 1994.
Kennedy, B.C., and W.E. Sharp, A validation study of
the Robertson-Berger meter, Photochem. Photobi-
oi., 56, 133-141,1992.
Kerr, J.B., and C.T. McElroy, Evidence for large upward
trends of ultraviolet-B radiation linked to ozone
depletion, Science, 262, 1032-1034, 1993.
Kerr, J.B., and C.T. McElroy, Response (to comment by
Michaels et al., 1994), Science, 264, 1342-1343,
1994.
Lean, J., M. VanHoosier, G. Brueckner, D. Prinz, L.
Floyd, and K. Edlow, SUSIM/UARS observations
of the 120 to 300 nm flux variations during the
maximum of the solar cycle: Inferences for the
11-year cycle, Geophys. Res. Lett., 19+ 22, 2203-
2206,1992.
Liu, S.C., S.A. McKeen, and S. Madronich, Effect
of anthropogenic aerosols on biologically active
ultraviolet radiation, Geophys. Res. Lett.,-18, 12,
2265-2268, 1991.
Lubin, D., P. Ricchiazzi, C. Gautier, and R.H. Whritner,
A method for mapping Antarctic surface ultravio-
let radiation using multi-spectral satellite imagery,
in Ultraviolet Radiation in Antarctica: Measure-
ments and Biological Research, edited by C. S.
Weiler and P. A. Penhale, AGU Antarctic Research
Series, volume 62, American Geophysical Union,
Washington, D.C., 53-81, 1994.
Madronich, S., The atmosphere and UV-B radiation at
ground level, in Environmental UV Photobiology,
edited by A. R. Young et al.. Plenum Press, New
York, 1-39, 1993.
9.20
-------
SURFACE UV RADIATION
Madronich, S., and C. Granier, Tropospheric chemistry
changes due to increased UV-B radiation, in
Stratospheric Ozone Depletion/UV-B Radiation in
the Biosphere, edited by R. H. Biggs and M. E. B.
Joyner, NATO ASI Series I, Vol 18, Springer-Ver-
lag, 3-10, 1994.
McKenzie, R.L., UV radiation monitoring in New
Zealand, in Stratospheric Ozone Depletion/UV-B
Radiation in the Biosphere, edited by R.H. Biggs
and M.E.B. Joyner, NATO ASI Series I, Vol 18,
Springer-Verlag, 239-246, 1994a.
McKenzie, R.L., UV spectral irradiarice in New
Zealand: Effects of Pinatubo volcanic aerosol, in
Proceedings Quadrennial Ozone Symposiunii
Charlottesville, Virginia, June 1992, NASA Conf.
Public. 3266, 627-630, 1994b.
McKenzie, R.L., P.V. Johnston, M. Kotkamp, A. Bittar,
and J.D. Hamlin, Solar ultraviolet spectroradiome-
try in New Zealand: Instrumentation and sample
results from 1990, Appl. Optics, 31, 30, 6501-
6509, 1992.
McKenzie, R.L., W.A. Matthews, and P.V. Johnston, The
relationship between erythemal UV and ozone,
derived from spectral irradiance measurements,
Geophys. Res. Lett., 18, 12, 2269-2272, 1991.
McKenzie, R.L., M. Kotkamp, G. Seckmeyer, R. Erb,
C.R. Roy, H.P. Gies, and S.J. Toomey, First South-
ern Hemisphere intercomparison of solar UV
spectra, Geophys. Res. Lett., 20, 20, 2223-2226,
1993.
McKinlay, A.F., and B. L. Diffey, A reference spectrum
. for ultraviolet induced erythema in human skin, in
Human Exposure to Ultraviolet Radiation: Risks
and Regulations, edited by W. R. Passchler and B.
F. M. Bosnajokovic, Elsevier, Amsterdam, 1987.
Michaels, P.J., S.F. Singer, and P.C. Knappenberger,
Analysing Ultraviolet-B radiation: Is there a
trend? Science, 264, 1341-1342, 1994.
Michelangeli D.V., M. Allen, YX. Young, R.-L. Shia, D.
Crisp, and J. Eluszkiewicz, Enhancement of atmo-
spheric radiation by an atmospheric aerosol layer,
J. Geophys. Res., 97, Dl, 865-874, 1992.
NRC, Acid Deposition Long-Term Trends, National Re-
search Council, National Academy Press,
Washington, D.C., 1986.
Prezelin, B.B., N.P. Boucher, and R.C. Smith, Marine
primary production under the influence of the Ant-
arctic ozone hole: Icecolors '90, in Ultraviolet
Radiation in Antarcticii: Measurements and Bio-
logical Research, edited by C. S. Weiler and P. A.
Penhale, AGU Antarctic Research Series, volume
62, American Geophysical Union, Washington,
D.C., 159-186, 1994.
Rottman, G.J., T.N. Woods, amd T.P. Sparn, Solar-Stellar
Irradiance Comparison Experiment 1:1. Instru-
ment design and operation, J. Geophys. Res., 98,
D6, 10667-10677, 1993.
Scotto, J., G. Cotton, F. Urbach, D. Berger, and T. Fears,
Biologically effective ultraviolet radiation: mea-
surements in the United! States 1974-1985,
Science, 239, 762-764, 1988.
Seckmeyer, G., and R.L. McKenzie, Increased ultravio-
let radiation in New Z'-ealand (45°S) relative to
Germany (48°N), Nature, 359, 135-137, 1992.
Seckmeyer, G., and G. Bernhard, Cosine error of spec-
tral UV-irradiances, in Proceedings International
Symposium on High latitude Optics, Tromso,
Norway, 28 June-2 July, 1993.
Seckmeyer, G., B. Mayer, R. Erb, and G. Bemhard,
UVB in Germany higher in 1993 than in 1992,
Geophys. Res. Lett., 21, 7, 577-580, 1994a.
Seckmeyer, G., S. Thiel, M. Blurnthaler, P. Fabian, S.
Gerber, A. Gugg-Helmiiiger, D.-P. Hader, M. Hu-
ber, C. Kettner, U. Kohler, P. Kopke, H. Maier, J.
Schafer, P. Suppan, E. tamm, and E. Thomalia,
First intercomparison cif spectral UV radiation
measurement systems, Appl. Optics, in press,
1994b. ;
Setlow, R.B., The wavelengths in sunlight effective in
producing skin cancer: A theoretical analysis,
Proc. Nat. Acad. Sci. USA., 71(9), 3363-3366,
1974. l
Smith, G.J., M.G. White, and K.G. Ryan, Seasonal
trends in erythemal and carcinogenic ultraviolet
radiation at mid-Southern latitudes 1989-1991,
Photochem. and Photobiol., 57, 3, 513-517, 1993.
Smith, R.C., B.B. Prezelin, K.$. Baker, R.R. Bidigare,
N.P. Boucher, T. Coley, t). Karentz, S. Maclntyre,
H.A. Matlick, D. Menzies, M. Ondrusek, Z. Wan,
and K.J. Waters, Ozone depletion: Ultraviolet ra-
diation and phytoplankton • biology in Antarctic
waters. Science, 255, 952-959, 1992.
9.21
-------
SURFACE UV RADIATION
Staehelin, J., and W. Schmid, Trend analysis of tropo-
spheric ozone concentrations utilizing the 20-year
data set of ozone balloon soundings over Payerne
(Switzerland), Atmos. Environ., 25A, 9, 1739-
1749,1991.
Stamnes, K., S.-C. Tsay, W. Wiscombe, and K. Jaya-
weera, Numerically stable algorithm for
discrete-ordinate-method radiative transfer in
multiple scattering and emitting layered media,
Appl. Opt., 27,2502-2509, 1988.
Tevini, M., (ed.), UV-B Radiation and Ozone Depletion:
Effects on Humans, Animals, Plants, Microorgan-
isms, and Materials, Lewis Publishers, Boca
Raton, Florida, USA, 1993.
Tsay, S.-C., and K. Stamnes, Ultraviolet radiation in the
Arctic: The impact of potential ozone depletions
and cloud effects, J. Geophys. Res., 97, 7829-
7840,1992.
UNEP, Environmental Effects of Ozone Depletion: 1991
Update, United Nations Environment Programme,
panel report pursuant to Article 6 of the Montreal
Protocol on Substances that Deplete the Ozone
Layer, November 1991.
UNEP, Environmental Effects of Ozone Depletion: 1994
Update, United Nations Environment Programme,
panel report pursuant to Article 6 of the Montreal
Protocol on Substances that Deplete the Ozone
Layer, in preparation, 1994.
Walker, J.H., R.D. Saunders, J.K. Jackson, and K.D.
Mielenz, Results of a CCPR intercomparison of
spectral irradiance measurements by National
Laboratories, J. Res. National Institute of Stan-
dards and Technology (NIST), 96, 647-668, 1991.
Wang, P., and J. Lenoble, Comparison between measure-
ments and modeling of UV-B irradiance for clear
sky: A case study, Appl. Optics, 33, 18, 3964-
3971, 1994. :
WMO, Scientific Assessment of Ozone Depletion: 1991,
World Meteorological Organization Global Ozone
Research and Monitoring Project—Report No. 25,
Geneva, 1992.
Young, A.R., L.O. Bjom, J. Moan, and W. Nultsch (eds.),
Environmental UV Photobiology, Plenum Press,
New York, 1993.
Zeng, J., Z. Jin, and K. Stamnes, Impact of stratospheric
ozone depletion on UV penetration into the ocean
at high latitudes, in Underwater Light Measure-
ments, edited by H. Chr. Eilertsen, Proc. SPIE
2048, 1993.
Zeng, J., R.L. McKenzie, K. Stamnes, M. Winleland,
and J. Rosen, Measured UV spectra compared
with discrete ordinate method simulations, J. Geo-
phys. Res., in press, 1994.
Zheng, X., and R.E. Basher, Homogenisation and trend
detection analysis of broken series of solar UV-B
data, Theor. Appl. Climatol., 47,4,189-303, 1993.
9.22
-------
PART 5
SCIENTIFIC INFORMATION FOR FUTURE DECISIONS
Chapter 10
Methyl Bromide
Chapter 11
Subsonic and Supersonic Aircraft
Emissions
Chapter 12
Atmospheric Degradation of Halocarbon Substitutes
Chapter 13 i
Ozone Depletion Potentials, Global Warming Potentials,
and Future Chlorine/Bromine Loading;
-------
-------
CHAPTER 10
Methyl Bromide
Lead Author:
S.A. Penket
Co-authors:
J.H. Butler
M.J. Kurylo
C.E. Reeves
J.M. Rodriguez
H. Singh
D. Toohey
R. Weiss
Contributors:
M.O. Andreae
N.J. Blake
R.J. Cicerone
T. Duafala
A. Golombek
M.A.K. Khalil
J.S. Levine
M.J. Molina
S.M. Schauffler
-------
-------
CHAPTER 10
METHYL BROMIDE
Contents
SCIENTIFIC SUMMARY
r 10.1
10.1 INTRODUCTION.. !
| 10.3
10.2 MEASUREMENTS, INCLUDING INTERHEMISPHERIC RATIOS
10.2.1 Vertical Profiles '"
10.2.2 Trends IZZZ ' 10'6
10.2.3 Calibration Issues... ! 10'6
• 10.7
10.3 SOURCES OF METHYL BROMIDE j
10.3.1 The Oceanic Source ZZZZZ • "'
10.3.2 Agricultural Usage and Emission of CH3Br ""; n'
10.3.3 Biomass Burning... T
c? "*""*"""*"""*"*"**""*"""**""*••"*'•"••"**••••"•• •r»""»"-»"«-.....».....'L............ 10 0
10.3.4 Industrial Sources, including Gasoline Engine Exhaust T.IZ 10 9
10.3.5 Summary of Methyl Bromide Emissions from Individual Sources ZZZZZZ..Z.......... 10 10
10.4 SINK MECHANISMS j
10.4.1 Atmospheric Removal Processes ] '!!
10.4.2 Oceanic Removal Processes I.ZZZ ' ! !»
10.4.3 Surface Removal Processes !
1 10-13
10.5 THE ROLE OF THE OCEANS |
10.5.1 A Simple Ocean-Atmosphere Model Z.I i
.10.5.2 Oceanic Uptake and the Atmospheric Lifetime .! ' 1Q'15
10.6 MODELED ESTIMATES OF GLOBAL BUDGET !
10.6.1 Introduction ." | 15
10.6.2 Budget and the Anthropogenic Contribution .1 10'J5
10.7 STRATOSPHERIC CHEMISTRY: MEASUREMENTS AND MODELS I , n 18
10.7.1 Observations ! 1U'18
10.7.2 Laboratory Studies '' 10'18
10.7.3 Ozone Loss Rates ZZZZZZZZZZZZZZ t JQ 19
10.8 THE OZONE DEPLETION POTENTIAL OF METHYL BROMIDE 1 10 20
10.8.1 General Considerations "T" '
10.8.2 Steady-State OOP: Uncertainties j " : , ' ,
10.8.3 Time-Dependent ODPs ZZZZZI I" Q 22
10.9 CONCLUSIONS... !
: 10.23
REFERENCES !
-. -t 10.23
-------
-------
j METHYL BROMIDE
i
SCIENTIFIC SUMMARY
•
Four potentially major sources for atmospheric methyl bromide (CH3Br) have been identified: the ocean, which is
a natural source, and three others that are almost entirely anthropogenic; these are agricultural usage, which has
been reaffirmed, biomass burning, which is newly recognized, and the exhaust of automobiles using leaded gas-
oline.
The estimated uncertainty range for these sources is large, with oceans ranging from 60 to 160 ktonnes/yr, agri-
culture from 20 to 60 ktonnes/yr, biomass burning from 10 to 50 ktonnes/yr, and automobile exhaust from 0.5 to
22 ktonnes/yr. In the latter case, the range results from two conflicting assessments, which yield 0.5 to 1.5
ktonnes/yr and 9 to 22 ktonnes/yr, respectively.
I .
There are also two minor anthropogenic sources, structural fumigation (4 ktonnes/yr) and industrial emissions (2
ktonnes/yr), each of which are well quantified.
Measurements of CH3Br yield a global average ground-level atmospheric mixing ratio of approximately 11 pptv.
These measurements also have confirmed that the concentration in the Northern •Hemisphere is higher by about
30% than the concentration in the Southern Hemisphere (interhemispheric ratio of 1.3). Such a ratio requires that
the value of sources minus sinks in the Northern Hemisphere exceeds the same term in the Southern Hemisphere.
There is no clear long-term change in the concentration of CH3Br during the time period of the systematic contin-
ued measurements (1978-1992). One possible explanation is that CH3Br from automobiles may have declined
while,, at the same time, emissions from agricultural use may have increased, leading to relatively constant anthro-
pogenic emissions over the last decade.
The magnitude of the atmospheric sink of CH3Br due to gas phase chemistry is well known and leads to a lifetime
of 2 ± 0.5 yr. The recently postulated oceanic sink leads to a calculated, atmospheric lifetime due to oceanic
hydrolysis of 3.7 yr, but there are large uncertainties (1.3 to 14 yr). Thus the overall atmospheric lifetime due to
both of these processes is 1.3 yr with a range of 0.8 to 1.7 yr.
Recognizing the quoted uncertainties in the size of the individual sources of CH3Br,' the most likely estimate is
that about 40% of the source is anthropogenic. The major uncertainty in this number is the size of the ocean
source. Based on the present atmospheric mixing ratio and the current source estimate, a lifetime of less than 0.6
yr would require identification of new major sources and sinks.
The chemistry of ozone destruction by bromine in the stratosphere is now better understood. A high rate coeffi-
cient for the HC>2 + BrO reaction is confirmed and there is no evidence that it produces HBr. A conservative upper
limit of 2% can be placed on the reaction channel yielding HBr. Stratospheric measurements confirm that the
concentration of HBr is very low (less than 1 pptv) and that it is not a significant bromine reservoir.
i
The combined efficiency of the bromine removal cycles for ozone (HO2 + BrO and CIO + BrO) is likely to be
about 50 times greater than the efficiency of known chlorine removal cycles on an atom-for-atom basis.
The calculated Ozone Depletion Potential (OOP) for CH3Br is currently estimated to be 0.6 based on an atmo-
spheric lifetime of 1.3 years. The range of uncertainties in the parameters associated with the OOP calculation
places a lower limit on the OOP of 0.3.
10.1
-------
-------
METHYL BROMIDE
10.1 INTRODUCTION
Bromine atoms are highly effective in removing
ozone in the stratosphere through catalytic cycles involv-
ing free radicals such as BrO and CIO. In fact the
bromine atoms remove ozone more effectively than
chlorine atoms on an atom-for-atom basis, because the
large majority of the inorganic bromine is in a more la-
bile form capable of taking part in the ozone removal
cycles. This is discussed in more detail in Section 10.7.
The role of bromine in the distribution of strato-
spheric ozone has assumed greater prominence in the
past few years due to the re-evaluation upwards in the
efficiency of the reaction HO2 + BrO, which cycles BrO
radicals back to bromine atoms, and due to the probabil-
ity that a sizeable fraction of the main bromine source
gas to the stratosphere, methyl bromide (CH3Br), is of
anthropogenic origin. Overall, the impact on ozone of
approximately 20pptv of inorganic bromine in the
stratosphere could be equivalent to about 1000 pptv of
inorganic chlorine. This compares with a present total of
inorganic chlorine in the stratosphere in the range of
3500 pptv.
Bromine is carried into the stratosphere in various
forms such as halons and substituted hydrocarbons, of
which CH3Br is the predominant form. The halons are
rather stable in the troposphere, and their production for
consumption in developed countries ceased on 31 De-
cember 1993, under the latest Amendments to the
Montreal Protocol. Methyl bromide, on the other hand,
is much less stable in the troposphere and limitations to
its emission could have a rapid impact on the amount of
bromine carried into the stratosphere in this form. At
present, CH3Br production for consumption in devel-
oped countries is capped at 1991 levels beginning in
1995 under the terms of the Montreal Protocol. The U.S.
Environmental Protection Agency has recently an-
nounced a phase-out by 2001 in the U.S. based upon an
Ozone Depletion Potential (ODP) of 0.7.
The case of CH3Br is much more complex than the
halons or indeed of any other potential ozone-depleting
substance so far considered for regulation, because it is
produced by the biosphere and is emitted into the atmo-
sphere by natural processes. The atmospheric science of
CH3Br was reviewed in 1992 (Albritton and Watson,
1992) and many of the uncertainties associated with its
atmospheric distribution, sources, sinks, and involve-
ment in the removal of ozone in the stratosphere were
discussed. The present chapter is written against this
background and the 1992 methyl bromide review will be
referred to extensively (UNEP, 1992). A major objective
of the present chapter will be to describe more recent
progress towards defining a minimum and most likely
ODP for CH3Br and in highlighting the remaining un-
certainties in our knowledge of its behavior in the
atmosphere. ,
10.2 MEASUREMENTS, INCLUDING
INTERHEMISPHERIC RATIOS
Methyl bromide is a ubiquitous component of the
Earth's lower atmosphere. Over the past two decades,
sporadic measurements have been made largely in the
surface air but also in the free'troposphere and strato-
sphere. These latter measurements have been performed
using aircraft and balloon platforms. Here we provide a
synthesis of much of the recently available data, with
emphasis on the remote global atmosphere. In'most cas-
es, air samples are collected in pressurized stainless steel
canisters and analyzed after a period: of several days or
weeks. In some instances, especially on shipboard plat-
forms, this" sampling process is omitted and the air
sample is directly analyzed.
Many of the measurements have been made with a
technique involving sample preconcentration (100-1000
ml), gas chromatographic separation, and electron cap-
ture detection. Other measurements have involved mass
spectrometric detection; these are more specific and less
prone to artifacts. Substantial uncertainties in absolute
standards (±30%) probably still exist but no systematic
intercomparison studies have been performed to accu-
rately quantify the level of uncertainty that is present.in
the published measurements. The reported mean con-
centrations fluctuate between 15-30 pptv, but there
appears to be a convergence between 8-15 pptv in publi-
cations made since 1985. It is presumed that a large part
. of the differences in various measurements is due to the
uncertainties in calibration standards. However, there is
a distinct possibility that other sampling/analysis prob-
lems are also present, such as growth and decay in
sample containers and co-elution of other substances
with CH3Br that are detected by the electron capture
detector. 1
10.3
-------
METHYL BROMIDE
Table 10-1. Mean CH3Br mixing ratios (pptv) in the surface air of the Northern and
Southern Hemispheres.
NH SH
26 20
15 11
11 10
11 8
11 9
12.0 9.5
NH/SH
1.3
1.4
1.2*
1.4
1.2
1.3
Year
1981-1982
(December)
1982-1983
(November)
1985-1987
(Ann. Avg.)
. 1983-1992
(Ann. Avg.)
1992
(April/August)
1984-1993
(Spring/Fall)
Platform
(Region)
Ship
(Pacific)
Ship
(Atlantic)
Coastal
(Pacific)
Coastal
(Pacific)
Coastal
(Pacific)
Ship
(Pacific, Atlantic)
Latitude
Range
40°N -32°S
40°N - 75°S .
71°N-44°S
71°N-42°S
90°N - 45°S
60°N - 90°S
Ref. No.
(1)
(2)
(3)
(4)
(5)
(6)
(1) Singh et al. (1983) (2:)i Penkett et aL (1985)
(3) Cicerone et al (1988) (4) Khalil et aL (1993)
(5) Blake et aL (1993) (6) Schauffler et aL (1993a); Schauffler, personal communication.
* Note: the value of 1.15 has been corrected to 1.2 (Cicerone, 1994).
A number of campaigns have collected a body of
data largely in the surface marine boundary layer in both
hemispheres. In Table 10-1 we summarize the mean NH
(Northern Hemisphere) and SH (Southern Hemisphere)
surface air concentrations of CHaBr measured by several
different investigators.
As stated earlier, most of the recent measurements
show global mean concentrations in the vicinity of 8-15
pptv. In all cases a NH/SH gradient, which should be
independent of calibration uncertainties, is observed.
The higher NH mixing ratios have been ascribed to the
domination of anthropogenic CHsBr sources in the NH
. (c.g., Singh and Kanakidou, 1993; Reeves and Penkett,
1993). It is pertinent to note that the observed surface
NH/SH gradients are by no means uniform and a range
of 1.2 to 1.45 has been observed (Albritton and Watson,
1992). Figure 10-1 shows the variability that is inherent
in an extensive marine air data set collected by one set of
workers at different times and in different locations
(Schauffler et a/., 1993a) and it probably reflects the spa-
tial and temporal variability in the sources of CH^Br or,
alternatively, it may reflect experimental artifacts. An-
other factor in the calculation of interhemispheric ratios
from various data sets is the latitudinal range of the data.
This is sometimes restricted to a Southern Hemispheric
limit of 40°S which, as can be seen from the Schauffler et
al. data, would lead to a smaller ratio than a consider-
ation of the full range of 60°N to 90°S. Overall the most
likely value for the interhemispheric gradient is 1.3.
This contrasts with the interhemispheric gradient for
methyl chloride, which is close to 1 (Singh et al., 1983)
and strongly suggests a preponderance of Northern
Hemispheric sources of CH3Br that are very possibly
anthropogenic.
The major known removal process for CHsBr is its
reaction with OH, resulting in an atmospheric lifetime of
about 2 years. Theoretical studies suggest that such a
chemical should show a distinct seasonal cycle larger
than that observed for methyl chloroform, which would
be expected to have a smaller amplitude. However, in a
number of attempts so far, no distinct seasonal cycle has
been observed (Singh et al., 1983; Cicerone et al., 1988;
Khalil et al, 1993). Figure 10-2 shows an example of
this based on data collected in Tasmania and Oregon.
Inadequate measurement precision, seasonal variations
in the sources of CHsBr, or unidentified sinks may be
responsible for a lack of observed seasonal behavior.
10.4
-------
METHYL BROMIDE
-90 -60 -30 0 30
Latitude
,
20-
15-
n-
n,
« 10"
§
5-
0-
ilii
!
|p
i
1
1
!
A
D
a^i
i
A'
fipA
- D
!
A
A
**<
^4^
1 *t
!
)
'A' ^<"
A A^
R
S
i
A McMurdo 19S7Sepl-Oct
x Pacific cruise 1990 Feb-Mar
o Pacific cruise 1986 Nov-Dec
A Pacific cruise 198(5 Apr- July
A Pacific cruise 1993 April
D Atlantic cruise 1992 April
o South Pole 1984-1987
i
(j
!
A
A
^A A
^
A
4
;
'
*'-
i
•;.. '-,
-
.-
' ' -
90
Figure 10-1. Latitudinal transect measurements of methyl bromide in oceanic air (after Schauffler et al.,
1993a; Schauffler, personal communication). :
to
«<•)
X
1 1.5
10.5
9.5
c
o
- 8.5
c
o
O
7.5 :
6.5
JFMAMJ JAS
Time (month)
O N D
Oregon
Measured
Tasmanis:
Measured
Lines are
Calculated
Mid North &
Mid South
Figure 10-2. Seasonal cycle of methyl bromide in Tasmania (Southern Hemisphere) and Oregon (Northern
Hemisphere). (After Khalil era/., 1993.)
70.5-
-------
METHYL BROMIDE
13.5
D Oregon
O Hawaii
o Samoa
A Tasmania
1981
1984
1987
1990
1993
Time (Seasonal)
Figure10-3. Trends in methyl bromide at four locations over the period 1978 to 1992. (After Khalil et al., 1993.)
10.2.1 Vertical Profiles
The salient features of the vertical structure of
CHsBr are that its concentrations decrease with increas-
ing altitude at a slow rate in the troposphere (Blake et al.,
1993; Khalil et al., 1993), and then relatively rapidly in
the stratosphere (Lai et al., 1994). The slight decrease in
the troposphere is largely dictated by the surface source
of CHsBr.and a lifetime probably in excess of 1 year.
The rapid loss in the lower stratosphere suggests strong-
ly that CHaBr is a major source of bromine atoms in this
region. Co-measurements of CHsBr and CFC-11 in the
stratosphere by Schauffler et' al. (1993b) will allow an
accurate estimate of the stratospheric lifetime of CHaBr
for ODP purposes.
10.2.2 Trends
Only one set of internally consistent data is avail-
able to assess atmospheric trends of CHsBr. Figure 10-3
shows the nature of data reported by Khalil et al. (1993)
from four island sites in the NH and SH from 1978-1992.
An evident feature is the large variability in these mea-
surements that have no discernible seasonal character.
Based on these data, Khalil et al. calculate a positive glo-
bal trend of 0.3 (±0.1) pptv/year between 1988 and 1992
" (Figure 10-3). It appears that a significant trend may not
have existed prior to 1988. The positive trend in later
years is not inconsistent with the mean trend of 0.2 pptv
per year calculated from a consideration of increased
agricultural usage (Singh and Kanakidou, 1993). How-
ever, these studies did not take into account changes that
may have occurred in the potential source from gasoline
consumption and the large biomass contribution. The
variability in measured data is sufficiently large, and the
data base sufficiently sparse, that a quantitative rate of
increase cannot be reliably defined from these measure-
ments alone. It is also likely that experimental artifacts
associated with sample or standard storage would make
70.6
-------
a small trend impossible to detect. Overall it can be con-
cluded at present that no useful statement can be made
from a consideration of the available trend data.
10.2.3 Calibration Issues
At the time of writing there has been no attempt to
carry out an intercalibration exercise amongst the vari-
ous groups making and publishing CH^Br measurements
in the atmosphere. That such an exercise is clearly need-
ed is shown by the data in Table 10-1. Accurate
measurements of CH^Bi will allow limits to be set on the
source strength for comparison with independent esti-
mates (see Section 10.3). They will also allow a data
base to be built up in the future that could detect trends
and seasonal variations, etc.
10.3 SOURCES OF METHYL BROMIDE
10.3.1 The Oceanic Source
The oceans are a major natural reservoir of bro-
mine. They have generally also been regarded as a major
natural source of atmospheric CHaBr, based principally
on the measurements of Singh et al. (1983) that found
surface water concentrations in the eastern Pacific
Ocean to be 2.5 times the atmospheric equilibrium con-
centrations (/.«., 150% supersaturation). From this value
they calculated a net global oceanic source of about
300 Gg/yr. Singh and Kanakidou (1993) have recently
revised this net flux estimate downward to 40-80 Gg/yr
by correcting for large differences in calibration and by
weighting the calculations according to regional ocean
productivity differences. Taking only the correction for
calibration differences and using a .mean tropospheric
mixing ratio of 11 pptv, together with the same air-sea
exchange and solubility coefficients used by Butler
(1994), yields a global net flux of about 110 Gg/yr for a
supersaturation of 150%.
Most recently Khalil et al. (1993) have reported
the results of CHsBr measurements from two Pacific
Ocean expeditions in 1983 and 1987. They obtained a
range of surface water saturations for these expeditions
of 1.4 to 1.8 (i.e., 40-80% supersaturation), and calculat-
ed a net global flux of 35 Gg/yr (range: 30-40 Gg/yr) by
integrating the exchange fluxes as a function of latitude
and ocean area, without allowing for latitudinal
variations in the exchange coefficient or solubility. For
METHYL BROMIDE
purposes of comparison, if one uses this measured mean
supersaturation of 60%, together with the same 11 pptv
mean tropospheric mixing nitio and the same air-sea ex-
change and solubility coefficients used by Butler, the
resulting global net flux is 4:5 Gg/yr.
It is important to stress that these and other mea-
surements of the air-sea disequilibrium of CHsBr can
only be used to calculate net exchange fluxes across the
air-sea interface. If, as is discussed below, the oceans are
responsible for the chemical destruction of 50 Gg/yr of
tropospheric CHaBr, then the global net oceanic fluxes
reported above must be increased by this 50 Gg/yr to
obtain gross strengths for the oceanic source. It is also
important to stress that there; is a very large uncertainty
in the magnitude of the gross; oceanic source, which is a
necessary consequence of ,the disagreements among
measurements of the air-sea disequilibrium and the un-
certainties in the air-sea exchange rate and the oceanic
chemical destruction rate. These are active research top-
ics at the time of writing. ;
10.3.2 Agricultural Usage and Emission of
CH3Br
The use of CHaBr for agricultural purposes was
well covered in the UNEP 1992 Report (Albritton and
Watson, 1992), and the respective table showing CHaBr
sales over the period 1984-1990 is reproduced here with
updated values for 1991 and 1992 provided by Duafala
(personal communication, 1994) (Table 10-2). In 1990,
66.6 thousand tonnes were sold, with 3.7 thousand
tonnes being used as a chemical intermediate (1 metric
ton = 1 tonne = 103 kg). The resultant 63 thousand
tonnes were used in the environment in some manner,
with the amount in the column marked "Structural" re-
ferring to fumigation of builldings and containers, etc.
All of this will escape to the atmosphere, but the fraction
of the bulk of the CF^Br used for agricultural purposes
that escapes is not known with any certainty. A theoret-
ical analysis predicted that bstween 45 and 53% would
do so, resulting in an atmospheric source from agricul-
tural activities in the region of 30 thousand tonnes per
year (Albritton and Watson, 1992).
An earlier analysis carried out in 1982 (Rolston
and Glauz) compared measured concentrations of
CHsBr in soil after application with those calculated by
theory, with and without sheet covering at the time of
injection. Theory and measurement agreed well with the
10.7
-------
METHYL BROMIDE
Table 10-2. Methyl bromide sales, in thousands of tonnes.*
Year Pre-Planting Post-Harvesting Structural
Chemical
Intermediates**
Total.
1984
1985
1986
1987
1988
1989
1990
1991
1992
30.4
34.0
36.1
413
45.1
47.5
513
55.1
57.4
9.0
15
83
8.7
8.0
8.9
8.4
10.3
9.6
2.2
2.3
2.0
2.9
3.6
3.6
3.2
1.8
2.0 '
.4.0 .
4.5
4.0
2.7
3.8 '
25
3.T
4.1
' ' 2.6
:•;.,. :• :45;6; ,,
483
• 5"0.4
55.6 "'
••'•• ' 60.5
' : 62^ '
66.6
'•" 71.2
' ' '-716 >
**
production by companies based in Japan, Western Europe, and the U.S.
not released into the atmosphere
assumption that most of the CH3Br escaped to the atmo-
sphere, with 27% and 67% of the'applied CH3Br
escaping by 1 and 14 days, respectively, after fumiga-
tion. The work suggested that plastic barriers were
almost totally ineffective in preventing CH3Br release in
the long-term, but Rolston and Glauz appear to have
used unusually permeable tarping material.
More recently, an experimental .study was carried
out by Yagi et al. (1993) to compare the flux of CH3Br
released to the atmosphere with the amount applied.
They showed that 87% was released to the atmosphere.
Lower values have been obtained in unpublished studies
conducted recently both by workers at the University of
California at Davis and by Cicerone and co-workers,
who found that application in wet soil conditions greatly
reduced emissions to -35%. Soil pH and organic matter
parameters also influence rates of decomposition of
CHaBr, and thus the fraction that escapes. Further, the
depth and technique of injection are likely to exert some
influence. To date, these factors have not been investi-
gated thoroughly. Overall it is assumed here that 50% of
the CH3Br used for purposes such as pre-planting and
post-harvesting escapes to the atmosphere, leading to an
emission in the region of 35 thousand tonnes per year in
1991 and 1992 from a usage of approximately 70 thou-
sand tonnes per year for pre-planting, post-harvesting
purposes, and structural purposes.
10.3.3 Bibmass Burning
Recent measurements of gaseous emissions from
biomass burning in very diverse ecosystems indicate that
CH3Br is a significant combustion product (Mano and
Andreae, 1994). In addition,, satellite measurements
suggest that biomass burning (i.e., the burning of tropi-
cal, temperate, and boreal forests, savannas, grasslands,
and agricultural lands following the harvest) is much.
more widespread and extensive than,previously believed
(Levine, 1991; Gaboon et al., 1992; Andreae, 1993a).
Almost all biomass burning is initiated.or controlled by
human activities, and pyrogenic emissions must there-
fore be classified as an anthropogenic source. Wildfires
probably represent less than 10% of the biomass. com-
busted globally (Andreae, 1993b). About 80% of
biomass burning takes place in the tropics, mostly in
conjunction with savanna fires, deforestation, and bio-
mass fuel use. The, emissions in the Southern
Hemisphere, where the largest savanna areas are burned
and most deforestation takes place, are about twice as
large as those in the Northern Hemisphere (Andreae,
1993b;Haoefa/., 1990). . , , ,
Measurements of CH3Br emissions were obtained
from burning savanna grasslands in southern Africa and
boreal forests in Siberia (Mano and Andreae, 1994), and
from tropical forests in Brazil (Blake et al., 1993).. Mano
and Andreae (1994) reported a CH3Br tp CO? emission
ratio from the south African savanna fires in the range of
4.4 x 10'8 to 7.7 x IP'7,-with an average of 3.7, x 10'7
10.8
-------
METHYL BROMIDE
(The emission ratio is the ratio .of CH3Br in smoke minus
ambient atmospheric CH3Br to CO2 in smoke minus
ambient atmospheric CO2.) The CH3Brto CO2 emission
ratios from the boreal forest fires in Siberia were higher,
ranging from (1.1-13) x 10'7. The higher value from the
boreal forest fire is probably due to the fact that forest
fires usually have a lower combustion efficiency than
grass fires and, hence, a larger fraction of the smolder-
ing-phase compounds are produced. The emission ratio
for CH3Br to methyl chloride (CH3C1) from the south
African and boreal forest fires was found to be about 1%,
which is similar to the Br/Cl ratios found in plants
(0.1-1%). Mano and Andreae (1994) have estimated the
global emission of CH3Br from biomass burning based
on the CH3Br to CO2 and CH3Br to CH3C1 emission
ratios. The global emission of CO2 from biomass bum-
ing is in the range of 2.5-4.5 Pg C/yr (1 Petagram = 1015
grams) and the global emission of CH3C1 from biomass
burning is in the range of 0.65-2.6 Tg Cl/yr (1 Teragram
= 1012 grams) (Andreae, 1993b). Using these estimates
of pyrogenic CO2 and CH3C1 emissions and the corre-
sponding CH3Br emission factors, Mano and Andreae
(1994) estimate that the global production of CH3Br
falls in the range from 9-37, and from 22-50 thousand
tonnes CHsBr/yr, respectively. The range of emission
from this source is thus 10-50 thousand tonnes per year,
with perhaps a mid-range value of 30 thousand tonnes
per year.
10.3.4 Industrial Sources, including Gasoline
Engine Exhaust
Methyl bromide is used as an intermediate com-
pound in the manufacture of various industrial
chemicals, including pesticides. Assessments for the
preparation of the UNEP Methyl Bromide Technology
Report, which is proceeding simultaneously with this re-
port, suggest that approximately 2.1 thousand tonnes per
year is emitted by inadvertent production and in the
course of chemical processing.
Methyl bromide is also formed indirectly in the
internal combustion'engine from ethylene dibromide
added in conjunction with lead tetraethyl to gasoline.
According to a study conducted in 1989 (Baumann and
Heumann), between 22 and 44% of the bromine in gaso-
line is emitted in an identified organic form in the
exhaust, of which 64-82% is CH3Br.
Using these factors, an estimate for emissions of
CH3Br from motor vehicle exhaust worldwide has been
supplied for the year 1991-9'2 (M. Speigelstein, personal
communication, 1994). In this year about 24 thousand
tonnes of ethylene dibromide were used in the U.S. and
37 thousand tonnes in the rest of the world, making a
total of 61 thousand tonnes. This would allow a range of
between 8.6 and 22 thousand tonnes of CH3Br to be
emitted and a mean of 15 thousand tonnes.
The use of ethylene dibromide as a fuel additive
has declined rapidly since the 1970s in the U.S. This is
shown in Table 10-3.
In 1971, for instance, the amount of bromine used
for gasoline additives in the U.S. was 121 thousand
tonnes; this had declined to 100 thousand tonnes in 1978
. and very rapidly thereafter down to 24 thousand tonnes
in 1991. Obviously much more CH3Br would have been
emitted from this source using the above analysis in the
1970s than in the 1980s, with at least 30 thousand tonnes
being emitted from the U.S. i alone in 1971. The decline
in use of ethylene dibromidie, however, has been com-
pensated by the increase in use of bromine for a variety
of other purposes, including flame retardants (specified)
and most probably agricultural use of CH3Br, listed un-
der "other," so that the total bromine usage has remained
nearly constant (162 thousand tonnes in 1971 and 170
thousand tonnes in 1991). it is not impossible that the
growth in emission to the atmosphere from agricultural
usage could have compensated for the decline in emis-
sion from motor vehicle Łxhaust. No figures are
available for the time dependence of gasoline usage of
bromine in the rest of the world at the time of writing.
Emission of CH3Br from this source is thus highly un-
certain, but in the past it could have been dominant.
A recent study by the U.S. Environmental Protec-
tion Agency (W. Thomas, personal communication)
estimates that between 10 and 30 tonnes of CH3Br were
emitted from the 2 billion gallons of leaded gasoline
used in 1992 in the United States. The same study 'esti-
mated that about 100 billion gallons of leaded fuel are
used worldwide. Assuming the same ethylene dibro-
mide additive levels (0.04 gm per gallon) as in the
United States, and the same emission factors as found by
Baumann and Heumann, this would extrapolate to be-
tween 500 and 1500 tonnes of CH3Br emitted globally
from this source. These numbers are probably low esti-
mates, though, because the lead levels and hence
70.9
-------
METHYL BROMIDE
Table 10-3. U.S.: Bromine consumption by end-use, 1971 to 1991 (thousand tonnes).
Year
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
Gasoline
Additives
121
122
.115
109
100
109
103
100
91
73
54
45
39
34
35
-
30
-
32
25
'24
Sanitary
Preparations
11
11
17
17
17
18
18
16
26
21
26
27
16
16
-
-
14
-
24
-
9
Flame
Retardants
16
17
27
25
22
26
29
32
28
25
35
47
45
45
52
-
41
-
49
50
48
Other
14
14
6
14
16
25
20
23
35
16
35
46
48
68
85
-
67
-
70
-
89
Total
162
164
165
165
155
178
170
171
180
135
150
165
148
163
172
.
152 .
'
175 ,
-
170
[Source: Roskill Information Services Ltd., The Economics of Bromine, Sixth Edition, ISBN: 0 86214 383 7,
London, 1992.]
ethylene dibromidc levels used in gasoline in many
countries are likely to be significantly larger than in the
U.S.
To a large extent the discrepancy in emission of
CHsBr from gasoline additives between the estimates is
traceable to the quantities of ethylene dibromide as-
sumed to be used in the U.S. Table 10-3, for instance,
suggests that 24 thousand tonnes of bromine were being
used in 1991, whereas the U.S. EPA Survey (W. Thomas,
personal communication) estimated a usage of about 80
tonnes only.
10.3.5 Summary of CH3Br Emissions from
Individual Sources
So far, four major sources and two minor sources
have been identified for emission of CHsBr to the atmo-
sphere. Table 10-4 gives a summary of the most likely
contribution made by each source, with ranges, to the
atmospheric burden.
The uncertainty ranges in the estimates are also
shown in Table 10-4, and they show the very imperfect
state of knowledge with respect to sources of atmospher-
ic CHsBr at the present time. In the case of the ocean,
the newer, often unpublished, data indicate that it is an
active sink, and thus zero net emission cannot be dis-
counted. Agricultural emission estimates vary widely,
mostly in association with the care taken and conditions
prevailing at the time of application of the CHsBr. Bio-
mass burning estimates are also very uncertain,
reflecting the recent identification of this source and also
current uncertainties in the magnitude of biomass burn-
ing sources of many compounds. The uncertainties in
emission from structural purposes and those incurred
during industrial processing are likely to be small, but
the source of CH3Br associated with the inclusion of
70.70
-------
METHYL BROMIDE
Table 10-4. Emission of CH3Br in thousand tonnes/year (best estimates).
Source
Ocean*
Agriculture
Biomass Burning
Gasoline Additivest
Structural Purposes
Industrial Emissions
Totals
Strength
90
35
30
1
15
4
2
162
176
Range
60 - 160
20-60
10-50
0.5-1.5
9-22
4
2
97 - 278
105 - 298
Anthropogenic
0
35
25 ;
1 • i
15 ;
4 !
2 :'
67 1
81
Natural
90
0
5
0
0
0
0
95
95
The ocean source of 90 thousand tonnes per year is a gross source and is made up of two very
uncertain quantities, as explained in Section 10.3.1, and the most likely value and the range are
expected to change markedly as a result of new research.
The two values given for this source reflect the large difference in the two estimates discussed in the
text.
ethylene dibromide in leaded gasoline to prevent the ac-
cumulation of lead deposits in car engines could either
be large or insignificant. Even given these uncertainties,
however, it is very likely that the anthropogenic emis-
sions make up at least 40% of the total. This percentage
is heavily biased by the value given to the highly uncer-
tain ocean source.
10.4 SINK MECHANISMS
The residence time of CH3Br in the Earth's atmo-
sphere is controlled by various removal processes
occurring in the atmosphere, in the oceans, and on land.
The most quantitative information exists for tropospher-
.ic and stratospheric mechanisms involving chemical
reaction and photolysis. However, there are several deg-
radation processes that may be operative in oceanic
surface waters. This is now an accepted removal process
for CH3CCl3 (Kaye et ai, 1994) and both the hydrolysis
rate and the solubility of CH3Br are higher than those for
CH3CCl3. Finally, a quantitative assessment of any glo-
bal significance of the dry deposition of CH3Br on soils
or vegetation is yet to be made.
10.4.1 Atmospheric Removal Processes
The removal of CH3Br within the atmosphere oc-
curs primarily via its tropospheric reaction with the
hydroxyl radical (OH). The consistent body of laborato-
ry data for this reaction (Mellouki et al., 1992; Zhang et
al., 1992; Poulet, 1993) points to a tropospheric OH-
removal lifetime for CH3Br of slightly greater than two
years. Zhang et al. (1992) estimate a tropospheric life-
time with respect to OH of 2.1 years by a comparison
with the OH reactivity of CH3CQ3 (Talukdar et al.,
1992) coupled with the lifetime of the latter deduced
from observational data (Prather, 1993). The use of the
data from either of the other tv/o kinetic studies yields
the same value. Mellouki et al. (1992) used a coupled
dynamical/chemical two-dimensional-(2-D) model to
calculate a tropospheric lifetime with respect to OH of
1.83 years. The OH reactive loss process for CH3Br is
thought to dominate over reactions involving NO3 or Cl.
For example, tropospheric concentrations of NO3 are
highly variable, with nighttime values in continental air
masses ranging from 20-200 pptv (Wayne et al., 1991).
Assuming an average nighttime concentration of 50 pptv
over the continents in the lowest 2 km of the troposphere
(with negligible concentrations during daytime and over
the oceans), the lifetime for the removal of CH3Br by
NO3 is calculated to be greater than 28 years, using a
comparative estimate for the i reaction rate constant
(Wayn&et al., 1991). Given the jlarge uncertainty in this
calculation and the small estimated contribution (-5%)
to the tropospheric reactive lifetime, the NO3 reaction
will not be considered further in calculations of the over-
10.11
-------
METHYL BROJWDE, ,
Table 10-5. Oceanic loss mechanisms for
Process
Neutral
Hydrolysis
Reaction
CH3Br + H2O -»
CH3OH + HBr
Loss Rate, References
Elliott and Rowland (1993)
0.2-10 Elliott (1984)
Mabey and Mill (1978)
Robertson et al. (1959)
Laughton and Robertson (1956)
Basic
Hydrolysis
Nucleophilic
Displacement
UV
Photosensitization
Biological
Consumption
CH3OH
CH3Br
CH3C1 + Br"
CH3Br + hv -»
(CH3Br)*
(CH3Br)t + H2O
CH3OH + HBr
Uncertain
< 1 - 10
1-50
< 6 times neutral
hydrolysis
Uncertain
Gentile et al. (1989)
Mabey and Mill (1978)
Fells and Moelwyn-Hughes (1959)
Elliott and Rowland (1993)
Elliott (1984)
Swain and Scott (1953)
Gentile et al. (1989)
Castro and Belser (1981)
Rasche et al. (1990)
all lifetime. For the possible removal by atomic chlo-
rine, the lifetime is even more difficult to estimate since
there are no direct measurements of Cl in the tropo-
sphere and a mechanism for, maintaining concentrations
sufficient to have a significant impact (-105 cm'3) on a
global scale is not known. In fact, model calculations
support much lower global tropospheric Cl concentra-
tions, on the order of 102 - 103 cm'3, yielding Cl removal
lifetimes for CH3Br of 750 - 7500 years. A minor, but
clearly identified, removal process occurring in the at-
mosphere involves the transport of CH3Br to the
stratosphere followed by its reaction with OH and photo-
dissociation, with a lifetime of approximately 35 years
(Prather, 1993). Therefore, the overall lifetime of
CH3Br associated with identified atmospheric removal
processes alone is approximately 2 years with an overall
uncertainty of ±25%.
10.4.2 Oceanic Removal Processes
There is growing evidence that CH3Br is de-
stroyed in seawater by up to five processes of differing
efficiencies (Table 10-5). Three of these have been in-
vestigated to some extent in pure water and seawater,
allowing for rough estimates of the degradation rate of
CH3Br in the surface ocean (Table 10-5).
According to Elliott and Rowland (1993) the pre-
dominant reaction in seawater is chloride substitution,
which is significantly more effective than hydrolysis.
They further suggest that these reactions could be a fac-
tor of 10 times faster or slower at the oceanographic
extremes of 0°C and 30°C. The other two mechanisms
(photosensitization by ultraviolet light and destruction
by microorganisms) have not been studied under condi-
tions representative of natural systems, thereby not
permitting quantification of these rates at the present
time. However, there is a limit to the effect that aquatic
degradation can have on the atmospheric flux, since at
high loss rates, the flux will be restricted by air-sea ex-
change, as discussed in Section 10.5. These data can be
used to compute an area-weighted removal rate for
CH3Br in seawater of 10% per day (J. Butler, private
communication) with a probable range of 3-30% d'', de-
10.12
-------
METHYL BROMIDE
pending on the actual rates and their dependencies on
salinity, temperature, and (in the case of biological loss-
es) oceanic productivity. It must be stressed here,
however, that the ocean loss process has not been investi-
gated with the same thoroughness as the homogeneous
gas phase loss processes discussed above, and that the
absolute magnitude of this process is therefore not well
defined at present. The impact of oceanic loss of CH3Br
on the overall atmospheric lifetime is discussed later in
Sections 10.5 and 10.8.
10.4.3 Surface Removal Processes
Recent experiments have indicated the potential
for degradation of CH3Br in different environments.
Anaerobic degradation in salt marsh sediments (Orem-
land et aL, 1994b), has been attributed to nucleophilic
substitution reactions with sulfides of biological origin.
Time constants of 2-5 days for CH3Br consumption
were measured. Laboratory and field experiments have
also provided evidence for biodegradation by methan-
otrophic bacteria (Oremland et al., 1994a), with time
constants of a day or less. However, the degradation
time constant seems to be inversely related to the relative
concentrations of CHU and CH3Br in the experiments.
Because the smallest initial concentrations of CH3Br in-
jected in these studies were of the order of ppmv, it is
difficult to extrapolate time constants to the pptv levels
typical of the atmosphere. Other soil types may also
consume CH3Br; in the soil, CH3Br will be partitioned
between soil gas, liquid, and solid phases. The effective-
ness of soil sinks would depend on (a) the rate of
consumption by physical and/or biological processes in
the soil, and (b) the rate of exchange of soil gas with the
overlying atmosphere. Experiments to evaluate these
processes should be performed with CH3Br concentra-
tions as close as possible to those in the ambient
atmosphere because, for example, soil microbes may ex-
• hibit different activities in different concentration
ranges.
Given the lack of information on any of the indi-
vidual processes involved, further laboratory and field
measurements are required to quantify the role of any
land uptake and degradation of CH3Br, and it is not in-
cluded further in atmospheric lifetime calculations.
10.5 THE ROLE OF THE OCEANS
i
The oceans represent an important special case in
the global tropospheric budget of CH3Br. As indicated
previously, the oceans are not only likely to be the largest
natural source of tropospheric CH3Br, they have at the
same time been shown to be an important natural sink of
tropospheric CH3Br through chemical removal pro-
cesses hi the oceanic mixed layer. Because the exchange
time of tropospheric CH3B:r with the surface layer of the
ocean is of the same order of magnitude as its tropo-
spheric residence time with respect to photochemical
destruction, its time-dependent response must be evalu-
ated in the context of a1 coupled ocean-atmosphere
system.
Butler (1994) was the first to draw attention to the
t
relationships between the oceanic production and loss
mechanisms for CH3Br and the tropospheric lifetime of
CH3Br. To illustrate these relationships, we present here
a much-simplified tutorial that leans heavily on the work
of Butler and qualitatively amd quantitatively reproduces
the main characteristics of the coupled ocean-atmo-
sphere system. In our subsequent assessment of the
effect of the oceans on the atmospheric CH3Br lifetime
and its effect on the Ozone Depletion Potential, we rely
on Butler's (1994) published values.
t
10.5.1 A Simple Ocean-Atmosphere Model
Consider a simple two-bojc model representing the
average square meter of ocean surface (Figure 10-4).
Above this surface the equivalent volume of atmosphere,
calculated by dividing by the fraction of the Earth's sur-
face that is covered by ocean (0.71), corresponds to a
column height of 11.9 km calculated at 20°C and 1 atm.
The mean depth of the oceanic mixed layer below this
surface is taken as 75 m, but because the volume equilib-
rium partition coefficient (i.e., the Ostwald solubility
coefficient, S) favors the liquid phase by a factor of 3i9 at
20°C (Singh et al., 1983), the equivalent depth of the
mixed layer reservoir with respect to atmospheric
CH3Br is 3.9 x 75 m = 293 m. Here we have used the
same mixed layer depth as iButler (1994), but we have
done the calculation for a mpan solubility at 20°C rather
than the value at 25°C used by Butler. The effects of this
difference and other minor differences in the calcula-
tions are discussed below. '
10.13
-------
METHYL BROMIDE
Two-Box Model for Atmospheric CHSBr
In situ Oxidation
-^
and Land Sinks
Land-based
Sources
Anthropogenic
Sources
In situ
Production
Aquatic
Degradation
and Downward
Removal
Figure 10-4. A two-box model illustrating methyl bromide coupling between atmosphere and ocean (after
Butler, 1994).
In this simple system the effective volumes of the
two reservoirs for CHsBr differ by a ratio of 11,900 +•
293 = 41. That is, when the ocean mixed layer is at sol-
ubility equilibrium with the atmosphere, only about
2.5% of the atmospheric burden resides in the mixed lay-
er. The magnitude of "buffering" of the atmospheric
burden of CHaBr by the additional CHsBr in ocean sur-
face waters is therefore realistically limited to only about
2 or 3 percent.
Butler (1994) estimates that the mean atmospheric
exchange coefficient, or "piston velocity," for dissolved
oceanic CHaBr is about 4.1 m/d. That is, for a 75 m
mixed layer, the CH^Br mean residence time with respect
to atmospheric exchange is (75 m) *• (4.1 m/d) = 18.3 d.
As the exchange flux must be equal in both directions
and the atmospheric reservoir is 41 times larger than the
mixed layer reservoir, the residence time of atmospheric
CHsBr with respect to oceanic exchange is 41 x 18.3 d =
750 d, or 2.1 years.
The mean residence time of dissolved CHsBr in
the oceanic mixed layer with respect to the various
chemical destruction mechanisms listed in Table 10-5
has been estimated by Butler (1994) at about 10 d. For
purposes of illustration, consider first how the simple
two-box model would behave if the chemical destruction
rate in the surface ocean were infinite. In this case" the
mixed layer concentration would be zero, and the up-
ward component of the exchange flux would be reduced
to zero while the downward component would remain
unchanged. Thus, the atmospheric residence time with
respect to oceanic exchange of 2.1 years would also be
the atmospheric lifetime with respect to oceanic chemi-
cal destruction, and the air-sea exchange rate would
become the rate-limiting step. In other words, within the
uncertainties in the air-sea exchange rate, the atmospher-
ic lifetime with respect to oceanic chemical destruction
cannot be less than 2.1 years.
Consider now the balance that is achieved for
CHsBr in the oceanic mixed layer if the mean replace-
ment time by atmospheric exchange is 18.3 d, the mean
chemical destruction lifetime is 10 d, and there is no
oceanic production. If f is the fraction of the equilibrium
atmospheric CH3Br concentration in the mixed layer at
steady state, then the atmospheric replacement rate,
which is proportional to (1-f) -*• 18.3 d, must be equal to
the destruction rate, which is similarly proportional to
f * 10 d. Solving for f gives a value of 0.35. That is, for
the given ratio of the air-sea exchange and chemical de-
struction rate constants, and no oceanic production, the
mixed layer will be 65% undersaturated with respect to
atmospheric equilibrium. The corresponding atmo-
spheric CH3Br lifetime with respect to oceanic chemical
destruction then becomes 2.1 y * 0.65, or 3.2 years.
10.14
-------
METHYL BROMIDE
It is important to recognize that this 3.2-year atmo-
spheric lifetime of CH3Br with respect to oceanic
removal does not depend on whether the oceans are a net
source or sink for the atmosphere. This is because air-
sea exchange and oceanic chemical destruction are both
regarded as first-order processes. Any CH3Br produc-
tion in the oceans will be partly destroyed in situ and
partly exchanged with the atmosphere, where it will be
subjected to the same combination of atmospheric and
oceanic losses as CH3Br produced elsewhere, either nat-
urally or anthropogenically.
10.5.2 Oceanic Uptake and the Atmospheric
Lifetime
Butler (1994) carried out calculations similar to
the above tutorial, except that he included a relatively
small term for mixing between the oceanic mixed layer
and the underlying waters. The greatest difference be-
tween the two calculations is that Butler used the mean
solubility coefficient at 25°C rather than 20°C, which
leads to an increase of about 20% in the calculated atmo-
spheric lifetime with respect to oceanic destruction.
Although the mean ocean surface temperature is about
18°C, there is reason to weight the calculation toward the
higher temperature solubilities because the chemical re-
moval rates are much greater in warmer waters. Finally,
neither calculation takes into account that only the ~85%
of the atmosphere that is in the troposphere is able to
exchange with the oceans on this time scale. Correction
for this effect would shorten the atmospheric lifetime
with respect to oceanic destruction by about 15% in both
calculations.
Using the results reported by Butler (1994), the
best atmospheric mean lifetime for CH3Br with respect
to oceanic destruction is 3.7 years, with a large uncer-
tainty range of 1.3 to 14 years that depends principally
on the large uncertainties in the aquatic degradation rate
and the air-sea exchange rate. Assuming a mean tropo-
spheric CH3Br mixing ratio of 11 pptv, this corresponds
to an oceanic destruction of about 50 Gg/yr (range: 136 -
13 Gg/yr). If the atmospheric lifetime with respect to
atmospheric photochemical destruction alone is 2.0
years, then the corresponding best combined lifetime is
1.3 years (range: 0.8 - 1.7 years).
10.6 MODELED ESTIMATES OF THE GLOBAL
BUDGET
10.6.1 Introduction
j
In recent years, there have been several attempts to
determine the strength of the anthropogenic CH3Br
source by constraining model calculations with observed
atmospheric concentrations. ;These model calculations
have varied from 3-D and 2-D models to simple 2-box
models, but the principle, intrinsic to all these model
studies, has been to investigate the latitudinal gradient
exhibited in the observations and to account for the mag-
nitude of the average mixing ratio. The results of these
studies are summarized in Table 10-6. The atmospheric
lifetime of CH3Br, required for these studies, has largely
been estimated by combining modeled OH fields with
reaction kinetic data derived from laboratory studies,
and the individual lifetimes shown in the table reflect
differences in these quantities at the times of publication
of the modeling studies. This has the advantage, how-
ever, of considering a range of lifetimes including those
from gas phase processes aloiiie and including both at-
, mospheric and oceanic removal. None of the modeling
studies referred to here explicitly considered sources
such as biomass burning and motor vehicle exhausts or a
substantial ocean sink. Even so, conclusions concerning
the proportions of source type between natural and an-
thropogenic and the overall annual budget are probably
valid. |
10.6.2 Budget and the Anthropogenic
Contribution
Singh and Kanakidou (1993) used a simple model
made up of 2 boxes, each representing a hemisphere,
with an interhemispheric exchange rate of 1.1-1.2 years.
Assuming a lifetime of 1.7-1.9 years, no natural sources,
and injecting 93% of the anthropogenic emissions into
the Northern Hemispheric box,| an interhemispheric N/S
ratio of 1.6-1.8 was calculated. Their 2-D model also
produced a similarly high interhemispheric ratio. The 2-
D model of Reeves and Penkett (1993), calculates the
interhemispheric ratio of the surface concentrations to be
1.69 when no natural sources are assumed and all an-
thropogenic emissions are injected into the northern
midlatitudes (see Figure 10-5 for their relationship be-
tween the interhemispheric ratio and anthropogenic
-------
METHYL BROMIDE
Table 10-6. Modeled atmospheric CH3Br.
Reference
Singh and Kanakidou
(1993)
Khalil et al. (1993)
Singh and Kanakidou
(1993)
Reeves and Penkett
(1993)
Prather (Albritton and
Watson, 1992)
Model
2-box
4-box
2-D
2-D
3-D
Lifetime
(yr)
1.7-1.9
1.7-1.9
1.2
2.0
1.9
1.9
1.2
1.78
1.78
1.78
2.0
2.0
1.0
Source
(ktonnes
yr-l)
—
93
147
96
72
84
84
91
91
91
100
100
200
Atmos.
Burden0
(ktonnes)
167.4
167.4
176.4
150.0
136.8
159.6
100.8
162.0
162.0
162.0
200.0
200.0
200.0
Average
Cone.
(pptv)
-
12
12
9.3
-
11*
6-7a
11
11
11
12.5
12.5
12.5
Anthro-
pogenic
%
100
35 (20-50)
27 (20-35)
30-70b
100
29
29
100
54 (33-74)
25-48
100
25 (13-40)
6-20
N/S ratio
1.6-1.8
1.1-1.25
1.1-1.25
1.34
1.3?a
1.08*
1.18*
1.69
1.3+0.15
1.1-1.25
>2.0
1.3±0.15
1.3±0.15
Results given for 2- and 4-box models as tropospheric column averages and for the multi-dimensional models
as lowest layer averages, unless stated otherwise.
a Tropospheric column average.
b Includes unknown source in the tropics, possibly biomass burning, which amounts to up to 30% of the
total source.
c Calculated assuming steady state, i.e., production x lifetime
contribution to atmospheric CHaBr). Prather (Albritton
and Watson, 1992), using a 3-D model, calculated an in-
terhemispheric ratio greater than 2 when all emissions
were released in the Northern Hemisphere and a lifetime
of 2 years was assumed.
Some caution must be shown when comparing the
results of these modeling studies (Table 10-6), since
there are several inherent differences in the various sim-
ulations. For example, each box of the 2-box model
represents the tropospheric average, whilst the results
from the lowest layer are quoted for die 2-D models.
Both the 2-D models indicate a slight decrease in inter-
hemispheric ratio with increasing altitude (e.g., a ratio of
1.3 at 0-2.5 km, decreasing to 1.2 at 7.5-10 km [Reeves
and Penkett, 1993)). Consequently, the interhemispheric
ratio of the tropospheric averages should be lower than
that of the surface averages. Another difference is the
latitudinal and, in the case of the 3-D model, the longitu-
dinal distribution of the emissions within the Northern
Hemisphere.
Despite these differences, it is clear that the inter-
hemispheric ratios, calculated by all these models for a
Northern Hemispheric, presumably anthropogenic
source, are considerably higher man die observed sur-
face ratio of 1.3 ± 0.15 (Albritton and Watson, 1992).
This indicates the existence of a source releasing CHaBr,
at least in part, into the Soudiern Hemispheric atmo-
sphere, which could be oceanic or biomass burning
according to the discussion of sources in Section 10.3.
Both Reeves and Penkett (1993) and Singh and
Kanakidou (1993) then, by analogy to methyl chloride
(CH3C1), assumed an evenly distributed natural source
of CH3Br. Reeves and Penkett (1993) were best able to
fit their model results to the 1.3 ± 0.15 observed surface
ratio when the extra Northern Hemispheric contribution
was 54% (33-74%) of the total source (see Figure 10-5).
Singh and Kanakidou (1993) present 2-D results for
which the extra Northern Hemispheric contribution was
29% of the total.
70.76
-------
METHYL BROMIDE
1.7
33 Implied Range of 74
• Anthro. Contrib.-
--1.45
Range of Observed
Interhemi spheric
Ratios
"1 ' I ' r*-1—I—'—1—'—I—'—T^1—I—'—T
0 10 20 30 40 50 60 70 80 90 100"
Extra Northern Hemispheric Emission (% of Total)
Figure 10-5. Relationship between the extra Northern Hemispheric source contribution and the interhemi-
spheric ratio, as calculated in a 2-D global model (after Reeves and Penkett, 1993).
The 3-D modeling work of Prather (Albritton and
Watson, 1992) suggests that the observed concentrations
and interhemispheric ratio could be explained by an
emission rate 25 (13-40) thousand tonnes yr1 greater in
the Northern Hemisphere, with a total source strength of
about 100 thousand tonnes yr1. This implies an anthro-
pogenic contribution of 25% (13-40%) of the total
emissions.
Khalil el at. (1993) employed a 4-box model to
analyze their observed CH3Br concentrations. Each box
was of equal volume, 0°-30° and 30°-90° in each hemi-
sphere, with intrahemispheric transfer rates of 0.25 years
and an interhemispheric rate of 0.55 years giving a total
transport time across all latitudes of 1.05 years. By car-
rying out a budget analysis in each region, they deduced
that 60% of CH^Br emissions occur in the tropical re-
gions, with the rest mostly from the middle-to-high
northern latitudes. They also concluded that the total
emissions are around 100 thousand tonnes yr"1 and that
the ratio of emissions between the Northern and South-
ern Hemispheres is between 2 and 4. From their results
they conclude that the anthropogenic source is at least 30
thousand tonnes yr1, and based on their calculated oce-
anic source of 30-40 thousand tonnes yr1, of which 25
thousand tonnes yr1 is in the tropics, they identified an
unexplained tropical source. If this is biomass burning,
the total anthropogenic contribution will be 60-70 thou-
sand tonnes yr1 (see Table 10-4).
Employing a lifetime of 1.2 years to account for
deposition to the ocean, Singh and Kanakidou's 2-box
model results in an extra Northern Hemisphere fraction
of 27% (20-35%), whilst the interhemispheric ratio of
their 2-D model increased by about 15%. Using a life-
time of 1 year for CH3Br in the 3-D model, Prather
calculated a total emission source of 200 thousand
tonnes yr1, with an extra Northern: Hemispheric contri-
bution of 6-20%. i
Table 10-6 also shows the atmospheric CHaBr bur-
den for each of the model runs reported. These have
been calculated as the production (emission) rate multi-
plied by the lifetime, assuming steady state. Considering
those runs that attempted to reproduce realistic average
concentrations, the atmospheric burden varies from 160-
200 thousand tonnes, with an average of 177 thousand
tonnes. Linking this burden with the maximum source
allowed by Table 10-4 would piroduce a minimum atmo-
spheric lifetime of 0.6 years for CH3Br due to all sink
processes. i
70.77
-------
METHYL BROMIDE
Inorganic Bromine Cycling
!NO2
Figure 10-6. Stratospheric gas phase bromine cycle.
10 J STRATOSPHERIC CHEMISTRY:
MEASUREMENTS AND MODELS
The chemistry of bromine in the stratosphere is
analogous to that of chlorine and is shown schematically
in Figure 10-6. Upon reaching the stratosphere, the or-
ganic source gases photolyze or react with OH and
O('D) rapidly to liberate bromine atoms. Subsequent
reactions, predominantly with O3, OH, HO2, CIO, NO,
and NO2, partition inorganic bromine between reactive
forms (Br and BrO) and reservoir forms (BrONO2, BfCl,
HOBr, and HBr). However, unlike chlorine chemistry^
where reactive forms are a small fraction of the total in-
organic budget (except in the highly perturbed polar
regions in wintertime), reactive bromine is about half of
« the total inorganic bromine budget in the lower strato-
sphere. Therefore, bromine is more efficient in catalytic
destruction of ozone than is chlorine. In addition, the
gas phase photochemical partitioning between reactive
and reservoir forms of bromine is fairly rapid in sunlight,
on the order of an hour or less, such that direct heteroge-
neous conversion of HBr and BrONO2 to BrO is likely to
have little impact on the partitioning of bromine, except
perhaps in polar twilight (see later).
Mixing ratios of NOX, HOX, and C1OX increase
more strongly with altitude above 20 km than does BrO,
and the fractional contribution to ozone loss due to bro-
mine is greatest in the lower stratosphere (Avallone et
al, 1993a; Garcia and Solomon, 1994). There, where
oxygen atom concentrations are small, the O + BrO reac-
tion is relatively unimportant, and the three reaction
cycles listed below are primarily responsible for bro-
mine-catalyzed ozone loss, with Cycle III being of less
importance than Cycles I and II:
ClO + BrO + hv -» Br + Cl + O2
Br + O3 -> BrO + O2
Cl + O3 -» CIO + O2
(I)
BrO + HO2
HOBr + hv
Br + O3 -»
OH + O3 -
-> HOBr + O2
-* OH + Br
BrO + 02
> HO2 + O2
(H)
BrO + NO2 + M -» BrONO2 + M
BrONO2 + M -* Br + NO3
NO3 + hv -» NO + O2
Br + O3 -» BrO + O2
NO + O3 -» NO2 + O2
(HI)
In the polar regions, where NOX is reduced and
CIO is enhanced by heterogeneous reactions on sulfate
aerosols and polar stratospheric clouds, Cycle I domi-
nates the ozone loss due to bromine. At midlatitudes the
first two cycles contribute approximately equally to
ozone loss at 20 km, and Cycle II is the most important
near the tropopause, where the abundance of HO2 is sub-
stantial but where CIO abundances are negligible.
10.7.1 Observations
Measurements of organic bromine across the
tropopause indicate that mixing ratios of total bromine in
the stratosphere should be about 18 pptv, with CH3Br
providing 54% (Schauffler et al., 1993c). Both remote
and in situ measurements of BrO indicate mixing ratios
are between 4 and 10 pptv, generally increasing with al-
titude, in the lower stratosphere (Brune et ai, 1990;
Carroll et al., 1990; Toohey et al., 1990; Wanner et al.,
1990; Wahner and Schiller, 1992; Arpag et al., 1994).
Results from photochemical models are in good agree-
ment with in situ BrO profiles between 16 and 22 km
(Garcia and Solomon, 1994). However, profile informa-
tion above 22 km is limited because all in situ data to
JO. 18
-------
METHYL BROMIDE
date have been obtained with the NASA ER-2 aircraft, a
platform with an altitude ceiling of 22 km, and it is diffi-
cult to derive profile information above 20 km from
column measurements (Arpag et al., 1994).
Attempts to observe HBr directly by far-infrared
emission techniques have been hampered by the small
anticipated abundances, especially at high altitudes
where these techniques are most sensitive (Traub et al.,
1992). However, a systematic search of dozens of indi-
vidual spectra obtained at various altitudes from about
25 km to 35 km revealed a small positive signal that
could be attributed to an average HBr mixing ratio of
about 1 pptv of HBr (Traub 1993). Within the measure-
ment uncertainties, these observations are broadly
consistent with results from photochemical models that
include a small HBr branching ratio (less than 5%) for
the BrO + HO2 reaction and show no unexpected fea-
tures, suggesting that the major sources of HBr have
been accounted for adequately in ozone loss calcula-
tions. Supporting these observations are results from a
2-D model (Garcia and Solomon, 1994) indicating that
an HBr yield of greater than a few percent is also not
consistent with the in situ observations of the abun-
dances and latitudinal gradient of BrO at midlatitudes.
Thus, HBr likely represents a minor reservoir for reac-
tive bromine in the lower stratosphere and is unlikely to
exceed 2% of total bromine.
Information about inorganic bromine photochem-
istry is available from geographic and solar zenith angle
variations in BrO. Mixing ratios within the polar vorti-
ces are about twice as great as values observed at
midlatitudes under background sulfate aerosol condi-
tions, consistent with the differences in NOX abundances
in these regions (Toohey ef al., 1990). Higher BrO abun-
dances observed at midlatitudes following the eruption
of Mount Pinatubo (Avallone and Toohey, 1993; Arpag
et al., 1994) reflected the concurrent decreases in NOX
due to enhanced heterogeneous reaction of N2O5 on sul-
fate aerosols. Similar increases were observed in CIO
(Avallone et aL, 1993b). The results of ER-2 diurnal
studies (Toohey et al., 1990) and remote observations at
sunrise and sunset of both BrO and OC1O (Solomon et
al, 1990; Arpag et al, 1994), the latter a by-product of
the CIO and BrO reaction, indicate that reactive bromine
is tied up at night into photolytically labile reservoir
forms such as BrONO2 and BrCl. These results are con-
sistent with inferences that BrONO2 is a major inorganic
bromine reservoir. However, some measurements from
the NASA DC-8 aircraft at northern high latitudes reveal
non-zero BrO column abundances in darkness that can-
not be explained with standard photochemistry (Wahner
and Schiller, 1992).
10.7.2 Laboratory Studies
A breakdown of the contributions from the catalyt-
ic cycles above indicates ijhat Cycle I and Cycle II
account for most of the bromine-catalyzed ozone loss
and contribute about equally (Isaksen, 1993). At lower
altitudes, where bromine reactions contribute most to
ozone loss rates and the alpha factor is greatest (Garcia
and Solomon, 1994), temperatures are low (below
220K) and there are some uncertainties in BrO kinetics.
The reaction between BrO arid CIO is complex, but it has
been studied extensively under stratospheric conditions
and appears to be well understood (DeMoreefa/., 1992).
Remote observations of BrQ and OC1O, the latter pro-
duced by the side reaction BrO + CIO -» Br + OC1O and
itself not affecting ozone, and diurnal studies of BrO in
situ support the view that our understanding of the cou-
pled photochemistry between BrO and CIO is basically
sound at stratospheric pressures and temperatures
(Solomon et al, 1990; Wahner and Schiller, 1992).
Recent measurements of the rate constant for the
BrO and HO2 reaction indicate that at room temperature
it is about six times larger than previously reported, mak-.
ing Cycle II correspondingly more efficient (Poulet et
al, 1992;Bridierera/., 1993; Maguinetal, 1994). Fur-
thermore, it is now clear that liie major reaction products
are HOBr and ©2- A recent report of the upper limit to
the efficiency of the channel yielding HBr + 03 at room
temperature gave a value of less than 0.01 % (Mellouki et
al, 1994), which was established by investigating the
rate of the reverse reaction, namely, HBr + 03 -> BrO +
HO2. A direct determination at stratospheric tempera-
tures remains to be carried out.
10.7.3 Ozone Loss Rates
|. . •
Loss rates of ozone as calculated by a photochem-
ical model that best reproduces observations of ozone,
NOX, CIO, and BrO obtained at midlatitudes at the
spring equinox appear in Figure 10-7 (Garcia and
Solomon, 1994). Because bromine is released more rap-
idly with altitude than chlorine, and because a greater
10.19
-------
METHYL BROMIDE
Ox Loss Rote (mixing ratio/sec)
2K(CK»(BrO)
2K(0)(0^)
Total H0x-related
Total aOx-related
Total NOX- related
Figure 10-7. Midlatitude ozone loss rates associat-
ed with various removal cycles between 15 and
30 km (after Garcia and Solomon, 1994).
fraction of inorganic bromine remains in active forms,
catalytic destruction of ozone by bromine is more impor-
tant than chlorine on a mole-per-mole basis. As a
consequence, at about 20 km the bromine contribution to
the overall ozone loss rate is nearly as important as the
chlorine contribution. However, total ozone losses are a
result of continuous photochemical destruction as ozone
is transported from the source region in the tropics to
lower altitudes at higher latitudes (Rodriguez et al,
1994), so it is difficult to assess the overall contribution
to ozone column trends from instantaneous ozone loss
rates. However, 2-D model results indicate that at
present abundances of bromine and chlorine in the
stratosphere, a 5 pptv increase in inorganic bromine re-
sults in a column loss of ozone of 0.5% url.0%, with the
greater losses occurring at higher latitudes (Isaksen,
1993).
The importance of Cycle I in the lower strato-
sphere has been ascertained directly from simultaneous
in situ measurements of the abundances of BrO and CIO
and concurrent ozone decreases within the Antarctic
ozone hole (Anderson et al., 1990). Analyses using in
situ BrO and CIO data indicate that Cycle I contributed
approximately 25% to ozone losses observed over Ant-
arctica in 1987 (Anderson et al, 1990; Murphy, 1991)
and could contribute as much as 40% to total ozone loss
over the Arctic in winter (Salawitch et al., 1990, 1993).
On the other hand, because HO2 measurements
have a greater uncertainty (approx. 50%) relative to mea-
surements of BrO (approx. 35%) (Toohey et al, 1990)
and CIO (approx. 25%) (Anderson et al, 1990), and be-
cause the uncertainty in the rate constant for the BrO +
HOa reaction at low temperatures is greater than that for
the ClO + BrO reaction, the importance of the BrO +
HO2 reaction is less certain. Future simultaneous in situ
measurements of BrO and HO2 on the ER-2 aircraft, re-
ductions in BrO and HO2 measurement uncertainties,
and low-temperature kinetics studies will all contribute
to a better understanding of the importance of this reac-
tion in the atmosphere, leading to a better assessment of
ozone losses due to bromine. However, in the perturbed
polar regions where Cycle I dominates, uncertainties in
HOX kinetics and measurements are of little conse-
quence to estimates of the importance of bromine to
ozone losses.
10.8 THE OZONE DEPLETION POTENTIAL OF
METHYL BROMIDE
10.8.1 General Considerations
The concept of Ozone Depletion Potential (ODP)
has been extensively discussed in the literature (Wueb-
bles, 1983; Fisher et al, 1990; WMO, 1990, 1992;
Albritton and Watson, 1992; Solomon et al, 1992; So-
lomon and Albritton, 1992). A single time-independent
index has been introduced to quantify the steady-state
depletion of ozone by unit mass emission of a given trace
species, relative to the same steady-state ozone reduction
by unit mass emission of CFC-11. This index, the so-
called steady-state ODP, is approximately given by:
ODPCH]Br = \
fgC-ll TW O^-lglg-fy^
" I! c-
rCFC-l I y
H3Br ^CfC-ll
(10-1)
ODPCH3Br » [BLP][BEF]
where MCH3Br and MCFC-II denote ^ molecular
weight of CH3Br and CFC-11, FCH3B/FcFC-ll repre-
sents the bromine release from CH3Br relative to that of
chlorine release from CFC-11 in the stratosphere, a de-
70.20
-------
notes the efficiency of the released bromine in catalytic
removal of ozone, relative to chlorine; p is the decrease
in the mixing ratio of CH3Br at the tropical tropopause,
relative to the mixing ratio at the surface; .and < > de-
notes the spatial and seasonal averaging of the quantity
with the appropriate weighting given by the ozone distri-
bution.
The term in the first bracket represents the amount
of bromine delivered to the stratosphere by CH3Br rela-
tive to chlorine in CFC-11, per unit mass emission. This
is the so-called Bromine Loading Potential (BLP). The
second term, the Bromine Efficiency Factor (BEF), de-
notes the amount of stratospheric ozone removed per
unit mass of CH3Br delivered to the stratosphere, rela-
tive to CFC-11. Values for the parameters in Equation
10-1 can be obtained either from global models of the
atmosphere (and the ocean) or estimated from observa-
tions (Solomon et al., 1992; Solomon and Albritton,
1992). A fuller discussion of the usefulness of the OOP
is given in Chapter 13.
The time constants (lifetimes) TCFC-I i and TcH3Br
relate the change in (steady-state) atmospheric burden
(B) to a change in anthropogenic emission (S). This
therefore places limits on the range that can be chosen
for the atmospheric lifetimes (see Sections 10.3 and
10.6) and in the case of CH3Br:
ABCH3Br (kT) = ASCH3Br (kT/yr) TCH3Br (10-2)
This time constant (lifetime) can be obtained by consid-
ering all removal processes for the species in question,
both atmospheric and surface:
1
-L+_L+_L.+J.
(10-3)
where TQH denotes the time constant for removal by tro-
pospheric OH (2.0 years; see Section 10.4)'assuming the
rate constants of Mellouki et al. (1992), Zhang et al.
(1992), and Poulet (1993), and scaling to a lifetime of
6.6 years for methyl chloroform removal by OH; Tstrat is
the time constant for stratospheric removal (35 years;
Section 10.4); locean denotes the time constant for ocean
removal (about 3.7 years: Butler, 1994; Section 10.5);
and Bother denotes time constants for removal by other
sink mechanisms, such as reaction with Cl or biodegra-
dation, which are at this .point not well established and
METHYL BROMIDE
are therefore given a value 'of zero. Adopting the above
values for TQH. ^strat. and TOC^, we obtain a value of:
(10-4)
1-3 years
with an uncertainty range of 0.8 to 1.7 years.
Adopting the above value of TcH3Br and taking
TCFC-I i = 50 years (Kaye et al., 1994), a Bromine Load-
ing Potential of 0.013 is calculated from the expression
in the first brackets of Equation. 10-1. A Bromine Effi-
ciency Factor (BEF) of; 48 is. calculated by the
Atmospheric and Environmental Research, Inc. (AER)
2-D model, adopting heterogeneous chemistry on back-
ground aerosols and the kinetic recommendations of
DeMore et al., 1992. Using this value (48), the present
estimate for the OOP of CH3Br, taking into account un-
certainties in ocean removal., etc., is
10.8.2 Steady-State OOP: Uncertainties
The algorithm given by Equation 10-1 provides a
useful framework to estimate the uncertainties in the cal-
culated OOP of CH3Br due to uncertainties in the
different input parameters. Uncertainties in the input
parameters and their impact on the calculated ODP are
listed in Table 10-7. Uncertainties in the Bromine Load-
ing Potentials are directly calculated from Equation 10-1,
while the AER 2-D model has been used to calculate the
Bromine Efficiency Factor.
The largest uncertainties in ODP are due to the fol-
lowing: . ;
• Uncertainties in the lifetime of CH3Br. Values of
TCH3Br smaller than 1 year would be possible if
ocean uptake, removal by reactions with atomic
chlorine, and/or surface biodegradation were fast
enough. The value of (i is unlikely to be much less
than 1; recent measurements suggest a value of 0.9
(Blake et al., 1993). '
• Uncertainties in the kinetics of BrO + HO2. Atmo-
spheric measurements ' of BrO and (upper limits)
for HBr (Section 10.7); indicate that the branching
of the BrO + HO2 reaction to the HBr channel is
probably much less than 2%. Measurements of
HBr below 30 km would further constrain this pa-
rameter. There are at present no measurements of
either the rate or branching of the above reaction at
70.27
-------
METHYL BROMIDE
Table 10-7. Principal uncertainties in calculated steady-state OOP for CH3Br.
a
b
c
Parameter
TOH
^ocean
TCFC-H
FcHaBi^FcFC-ll
fcBrOtHO2
Branching of
BrO+HO2 -» HBr+O3
Value-Range
2.0 yr (± 25%)"
3.7 yr (± 1.3 - 14)*
50 yr (± 10%)^
1.08 (± 15%^
6.3(2.2- 18) x 10-n
cm3 s-1 e
0 (< 2%)f
Kaye et al., 1994; Prather, 1993 <*
Butler, 1994 (Section
Kaye et al., 1994
Table 10-8. Calculated
Time Horizon
(yr)
5.0
10
15
20
25
30
Infinite (steady state)
10.5) e
f
time-dependent ODPs.
TD-ODP
(Tocean = 3.7 yr)
16
5.3
3.1
2.2
1.8
1.5
0.6
tCH3Br BLP
(VTS)
1.1-1.5 0.011-0.015
0.78 - 1.7 0.0078 - 0.017
0.012-0.014
Pollock et al, 1992
DeMore et al, 1992, evaluated
Section 10.7
TD-ODP
(Tocean = 1-3 yr)
12
2.7
1.5
1.1
0.9
0.7
0.3
BEF OOP
48 0.52 - 0.76
48 0.37 - 0.80
48 0.55-0.67
41-55 0.52-0.70
32-50 0.41-0.64
30 - 48 0.38 - 0.61
at 215 K
TD-ODP
(tocean = 14 yr)
18
7.1
4.2
3.0
2.4
2.1
0.84
stratospheric temperatures. The uncertainties in
the rate of BrO + HO2 due to the lack of tempera-
ture information imply uncertainties in the
Bromine Efficiency Factor of the order of 50%.
At the same time, there is no single process whose
present estimated uncertainty could reduce the OOP of
CHaBr below 0.3. Smaller values could be possible if
two improbable situations occurred simultaneously and
several of the parameters were at the extremes of their
error limits.
The "semi-empirical",ODPs discussed by Sol-
omon et al (1992) provide a valuable constraint to the
model-based results, particularly if we are interested in
the ODP for a particular region of the atmosphere. Larg-
er uncertainties are introduced when steady-state ODPs
are derived from the semi-empirical approach, since the
necessary observations are usually not available for a
global coverage. This is particularly true for CHsBr,
where (a) coincidental measurements of the BrO, CIO,
and HO2 are sparse, particularly at midlatitudes, and (b)
the existing measurements have uncertainties of 25% for
CIO, 35% for BrO, and 50% for HO2 (see Section 10.7).
Overall, the lower limit for the ODP of CH3Br is
about 0.3 and its most likely value lies between 0.5 and
1.0 (0.6 with BEF = 48).
10.8.3 Time-Dependent ODPs
Previous studies (WMO, 1990, 1992; Albritton
and Watson, 1992; Solomon and Albritton, 1992) have
shown that species with short atmospheric lifetimes have
much larger ODPs over short time horizons than over
longer time horizons. Table 10-8 updates previous esti-
70.22
-------
METHYL BROMIDE
mates of the time-dependent ODP of CHaBr based on
the formulation of Solomon and Albritton (1992) with
particular respect to changes adopted in the ocean life-
time including a low value (1.3 years) which is outside
the limits set by the analysis of ocean sink processes in
Section 10.5.1. Uncertainties in the Bromine Efficiency
Factor would lead to the same scaling factors for each
time horizon as for the steady-state values in Table 10-7.
Over the period of any reasonable lifetime for CHsBr
(i.e., less than and up to 5 years), its ODP is in excess of
10, indicating that a cessation of emissions of GFTjBr
would have a rapid impact on the extent of stratospheric
ozone loss.
10.9 CONCLUSIONS
This review of the atmospheric science of Cf^Br
has revealed that many uncertainties still exist in both the
sources and sinks for this molecule, although its chemis-
try in the stratosphere and to a large extent in the
troposphere is now mostly resolved. The situation with
regard to sources and sinks is complicated by the role of
the ocean, which acts both as a source and a sink, with
the overall balance still in doubt. The research effort on
Ctt^Br has been somewhat limited and this is partly re-
sponsible for the uncertainties. In spite of this, there is
considerable confidence in our best current estimate of
0.6 for the ODP of CHsBr. Consideration of the existing
uncertainties indicates that it is improbable that this val-
ue would be less than 0.3 or larger than 0.8. Individual
points are addressed in more detail in the scientific sum-
mary for this chapter.
REFERENCES
Albritton, D.L., and R.T. Watson, Methyl bromide and
the ozone layer: A summary of current under-
standing, in Methyl Bromide: Its Atmospheric
Science, Technology, and Economics, Montreal
Protocol Assessment Supplement, edited by R.T.
Watson, D.L. Albritton, S.O. Anderson, and S.
Lee-Bapty, United Nations Environment Pro-
gramme, Nairobi, Kenya, 3-18, 1992.
Anderson, J.G., W.H. Brune, S.A. Lloyd, D.W. Toohey,
S.P. Sander, W.L. Starr, M. Loewenstein, and J.R.
Podolske, Kinetics of O;j destruction by CIO and
BrO within the Antarctic, vortex: An analysis based
on in situ ER-2 data, / Geophys. Res., 94, 11480-
11520,1990. I
Andreae, M.O., Global distribution of fires seen from
space, EOS Trans. AGU 74, 129-135, 1993a.
Andreae, M.O., The influence of tropical biomass burn-
ing on climate and the atmospheric environment,
in Biogeochemistry of Global Change: Radiatively
Active Trace Gases, edited by R.S. Oremland,
Chapman & Hall, New York, NY, 1993b.
Arpag, K.A., P.V. Johnson, H.L Miller, R.W. Sanders,
and S. Solomon, Observations of the stratospheric
BrO column over Colorado, 40°N, J. Geophys.
Res., 99, 8175-8181, 1994.
Avallone, L.M., and D.W. Toohey, ER-2 BrO results
from the Airborne Arctic Stratospheric Expedition
II, AASE-II CD-ROM, NASA, 1993.
Avallone, L.M., D.W. Tooheyj W.H. Brune, R.J. Sala-
witch, A.E. Dessler, j and J.G. Anderson,
Balloon-borne in situ measurements of CIO and
ozone: Implications for heterogeneous chemistry
and mid-latitude ozone loss, Geophys. Res. Lett.,
' 20,1795-1798, 1993a. i
Avallone, L.M., D.W. Toohey,; M.H. Proffitt, JJ. Mar-
gitan, K.R. Chan, and ;J.G. Anderson, In situ
measurements of CIO at mid latitudes: Is there an
effect from Mt. Pinatubo?, Geophys. Res. Lett., 20)
2519-2522, 1993b. ;
Baumann, H., and K.G. Heumann, Analysis of organo-
bromine compounds and HBr in motor car exhaust
gases with a GC/microwave plasma system, Frese-
niusZ. Anal., 186-192, 1989.
Blake, D.R., B. Sive, M. Zondlo, and F.S. Rowland, Esti-
mated methyl bromide emissions from biomass
burning, EOS Trans. AGU 74, C43), 134, 1993.
Bridier, I., V. Veyret, and R. Lesclaux, Flash photolysis
kinetic study of reactions of the BrO radical with
BrO and HO2, Chem. Phys. Lett., 201, 563-568,
1993.
Brune, W.H., J.G. Anderson, and K.R. Chan, In situ ob-
servations of BrO over Antarctica: ER-2 aircraft
results from 54°S to 72°S latitude, J. Geophys.
Res., 94, 16639-16647, 1990.
10.23
-------
METHYL BROMIDE
Butler, J.H., The potential role of the ocean in regulating
atmospheric CH^Br, Geophys. Res. Lett., 21, 185-
188,1994.
Cahoon, D.R., Jr., B.J. Stocks, J.S. Levine, W.R. Cofer,
m, and K.P. O'Neill, Seasonal distribution of Afri-
can savanna fires, Nature, 359, 812-815, 1992.
Carroll, M.A., R.W. Sanders, S. Solomon, and A.L.
Schmeltekopf, Visible and near-ultraviolet spec-
troscopy at McMurdo Station, Antarctica: 6.
Observations of BrO, J. Geophys. Res., 94, 16633-
16638,1990.
Castro, C.E., and N.O. Belser, Photohydrolysis ofwneth-
yl bromide and chloropicrin, J. Agric. Food
Chem., 29,1005-1008,1981.
Cicerone, RJ., Personal communication, 1994.
Cicerone, R J., L.E. Heidt, and W.H. Pollock, Measure-
ments of atmospheric methyl bromide and
bromoform, J. Geophys. Res., 93, 3745-3749,
1988.
DeMore, W.B., S.P. Sander, D.M. Golden, R.F. Hamp-
son, M J. Kurylo, CJ. Howard, A.R. Ravishankara,
C.E. Kolb, and M.J. Molina, Chemical Kinetics
and Photochemical Data for Use in Stratospheric
Modeling, Evaluation Number 10, JPL Publica-
tion 92-20, National Aeronautics and Space
Administration, Jet Propulsion Laboratory, Pasa-
dena, California, 1992.
Duafala,T., Personal communication, 1994.
Elliott, S.M., The Chemistry of Some Atmospheric Gases
in the Ocean, Ph.D. dissertation, University of
California, Irvine, 274 pp., 1984.
Elliott, S.M., and F.S. Rowland, Nucleophilic substitu- •
tion rates and solubilities for methyl halides in
seawater, Geophys. Res. Lett., 20, 1043-1046,
1993.
Fells, I., and E.A. Moelwyn-Hughes, The kinetics of the
hydrolysis of the chlorinated methanes, J. Chem.
Soc., 398-409, 1959.
Fisher, D.W., C.H. Hales, D.L. Filkin, M.K.W. Ko, N.-D.
Sze, P.S. Connell, D.J. Wuebbles, I.S.A. Isaksen,
and F. Stordal, Model calculations of the relative
effects of CFCs and their replacements on strato-
spheric ozone, Nature, 344, 508, 1990.
Garcia, R.R., and S. Solomon, A new numerical model
of the middle atmosphere 2. Ozone and related
species,./. Geophys. Res., 99, 12937-12951,1994.
Gentile, I. A., L. Ferraris, and S. Crespi, The degradation
of methyl bromide in some natural freshwaters:
Influence of temperature, pH, and light, Pestic.
ScL, 25, 261-272, 1989.
Hao, W.-M., M.-H. Liu, and P.J. Crutzen, Estimates of
annual and regional releases of CO2 and other
trace gases to the atmosphere from fires in the
tropics, based on the FAO statistics for the period
1975-1980, in Fire in the Tropical Biota: Ecosys-
tem Processes and Global Challenges, edited by
J.G. Goldammer, Springer-Verlag, Berlin, 1990.
Isaksen, I.S.A., presented at the Methyl Bromide State of
the Science Workshop, Washington, D.C., October
1993.
Kaye, J.A., S.A. Penkett, and F.M. Ormond (eds.), Re-
port on Concentrations, Lifetimes and Trends of
CFCs, Halons and Related Species, NASA Refer-
ence Publication No. 1339, 1994.
Khalil M.A.K., R.A. Rasmussen, and R. Gunawardena,
Atmospheric methyl bromide: Trends and global
mass balance, J. Geophys. Res., 98, 2887-2896,
1993.
Lai, S., B.H. Subbaraya, P. Fabian, and R. Borchers, Ver-
tical distribution of two methyl halides in the
atmosphere, Tellus, in press, 1994.
Laughton, P.M., and R.E. Robertson, Solvolysis in deu-
terium and hydrogen oxide, Can. J. Chem., 34,
1714-1718,1956.
Levine, J.S. (ed.), Global Biomass Burning: Atmospher-
ic, Climatic, and Biospheric Implications, MIT
Press, Cambridge, Massachusetts, 1991.
Mabey, W., and T. Mill, Critical review of hydrolysis of
organic compounds in water under environmental
conditions,/ Phys. Chem. Ref. Data, 7, 383-392,
1978.
Maguin, F., G. Laverdet, G. LeBras, and G. Pdulet, Dis-
charge-flow mass spectrometric study of the
reaction BrO + HO2, in press, 1994.
Mano, S., and M.O. Andreae, Emission of methyl bro-
mide from biomass burning, Science, 263,
1255-1257, 1994.
Mellouki, A., R.K. Talukdar, and CJ. Howard, Kinetics
of the reactions of HBr with O3 and HO2: The
yield of HBr from HO2 + BrO, J. Geophys. Res., in
press, 1994.
10.24
-------
METHYL BROMIDE
Mellouki, A., R.K. Talukdar, A.-M. Schmolter, T,
Gierczak, M.J. Mills, S. Solomon, and A.R. Ravi-
shankara, Atmospheric lifetimes and Ozone
Depletion Potentials of methyl bromide (CH3Br)
and dibromomethane (CH2Br2), Geophys. Res.
Lett., 19, 2059-2062, 1992.
Murphy, D.M., Ozone loss rates calculated along ER-2
flight tracks, J. Geophys. Res., 96, 5045-5053,
1991.
Oremland, R.S., L.G. Miller, P.M. Blunden, C.W. Cul-
bertson, M. Denise-Coulatkis, and L.L. Jahnke,
Degradation of methyl bromide and hydrochloro-
fluorocarbons by soil methanotrophic bacteria,
submitted to Env. Sci. and Tech., 1994a.
Oremland, R.S., L.G. Miller, and F.R. Strohmaier, Deg-
radation of methyl bromide in anaerobic
sediments, Env. Sci. and Tech., in press, 1994b.
Penkett, S.A., B.M.R. Jones, M.J. Rycroft, and D.A.
Simmons, An interhemispheric comparison of the
concentration of bromine compounds in the atmo-
sphere, Nature, 318,550-553, 1985.
Pollock, W.H., L.E. Heidt, R.A. Lueb, J.F. Vedder, M.J.
Mills, and S. Solomon, On the age of stratospheric
air and Ozone Depletion Potentials in the polar re-
gions,/. Geophys. Res., 97, 12993-12999, 1992.
Poulet, G., Atmospheric lifetimes of HCFCs and HFCs:
Current estimates and uncertainties, NASA/
NOAA/AFEAS Workshop on the Atmospheric
Degradation of HCFCs and HFCs, Boulder, Colo-
rado, November, 1993.
Poulet, G., M. Pirre, F. Maguin, R. Ramaroson, and G.
LeBras, Role of the BrO + HO2 reaction in the
stratospheric chemistry of bromine, Geophys. Res. •
Lett., 19, 2305-2308, 1992.
Prather, M.J., Atmospheric lifetimes of HCFCs and
HFCs: Current estimates and uncertainties,
NASA/NOAA/AFEAS Workshop on Atmospher-
ic Degradation of HCFCs and HFCs, November,
1993, Boulder, Colorado, 1993.
Rasche, M.E., M.R. Hyman, and D.J. Arp, Biodegrada-
tion of halogenated hydrocarbon fumigants by
nitrifying bacteria, Appl. Environ. Microbiol, 56,
2568-2571, 1990.
Reeves, C.E., and S.A. Penkett, An estimate of the an-
thropogenic contribution to atmospheric methyl
bromide, Geophys. Res. Lett., 20, 1563-1566
1993.
Robertson, R.E., R.L. Heppolette, and J.M.W. Scott, A
survey of thermodynamic parameters for solvoly-
sis in water, Can. J. Cham., 37, 803-824, 1959.
Rodriguez, J.M., M.K.W. Ko, N.-D. Sze, C.W. Heisey,
O.K. Yue, and M.P. McCormick, Ozone response
to enhanced heterogeneous processing after the
eruption of Mt. Pinatubo, Geophys. Res. Lett., 21,
209-212, 1994. J
Rolston, D.E., and R.D. Glau2:, Comparisons of simulat-
ed with measured transmit and transformation of
methyl bromide gas in soils., Pestic. Sci., 13, 653-
664,1982.
Salawitch, R.J., M.B. McElroy, J.H. Yatteau, S.C. Wofsy,
M.R. Schoeberl, L.R. Lait, P.A. Newman, K.R.
Chan, M. Loewenstein, J.R. Podolske, S.E. Stra-
han, and M.H. Proffitt, Loss of ozone in the Arctic
vortex for the winter of 1989, Geophys. Res. Lett.,
77,561-564,1990. |
Salawitch, R.J., S.C. Wofsy, E.W. Gottlieb, L.R. Lait,
P.A. Newman, M.R. Schoeberl, M. Loewenstein,
J.R. Podolske, S.E. Strahan, M.H. Proffitt, C.R.
Webster, R.D. May, D.W. Fahey, D. Baumgardner,
J.E. Dye, J.C. Wilson, K.K. Kelly, J.W. Elkins,
K.R. Chan, and J.G. Anderson, Chemical loss of
ozone in the Arctic pofcir vortex in the winter of
1991-1992, Science, 261, 1146-1149, 1993.
Schauffler, S.M., W.H. Pollock, and E.L. Atlas, Methyl
bromide measurements taken from the marine
boundary layer and during the 1991-1992 AASEII
flights, in Proceedings of the Methyl Bromide
State of the Science Workshop, Methyl Bromide
Coalition/NASA, Octobbr 28, 1993, SPA, Inc.,
Washington, D.C., 1993a.
Schauffler, S.M., W.H. Pollock, E.L. Atlas, L.E. Heidt,
T.M. Gilpin, G.P. Brasseur, R.A. Lueb, S. Solo-
mon, and J.F. Vedder, Correlations and
distributions of chemical tracers in the lower
stratosphere, EOS Trans.1 AGU, 74, 157, 1993b.
Schauffler, S.M., L.E. Heidt, W.H. Pollock, T.M. Gilpin,
J.F. Vedder, S. Solomon, ;R.A. Lueb, and E.L. At-
las, Measurements of halogenated organic
compounds near the tropical tropopause, Geophys.
Res. Lett., 20, 2567-2510, 1993c.
70.25
-------
METHYL BROMIDE
Singh, H.B., and M. Kanakidou, An investigation of the
atmospheric sources and sinks of methyl bromide,
Ceophys. Res. Lett., 20, 133-136, 1993.
Singh, H.B., LJ. Salas, and R.E. Stiles, Methyl halides
in and over the eastern Pacific (40°N-32°S), J.
Geophys. Res., 88,3684-3690, 1983.
Solomon, S., and D.L. Albritton, Time-dependent Ozone
Depletion Potentials for short- and long-term fore-
casts, Nature, 357,33-37, 1992.
Solomon, S., M. Mills, L.E. Heidt, W.H. Pollock, and
A.F. Tuck, On the evaluation of Ozone Depletion
Potentials,/. Geophys. Res., 97, 825-842,1992.
Solomon, S., R.W. Sanders, M.A. Carroll, and A.L.
Schmeltekopf, Visible and near-ultraviolet spec-
troscopy at McMurdo Station, Antarctica 5.
Observations of the diurnal variations of BrO and
OClO.y. Geophys. Res., 94, 11393-11403, 1990.
Speigelstein, M., The Economics of Bromine, Sixth
Edition, Roskill Information Services Ltd., ISBN:
0 86214 383 7, London, UK, 1992.
Swain, C.G., and C.B. Scott, Quantitative correlation of
relative rates: Comparison of hydroxide ion with
other nucleophilic reagents toward alkyl halides,
esters, epoxides, and acyl halides, J. Amer. Chem.
Soc., 75,141-147, 1953.
Talukdar, R.K., A. Mellouki, A.-M. Schmoltner, T. Wat-
son, S. Montzka, and A.R. Ravishankara, Kinetics
of the OH reaction with methyl chloroform and its
atmospheric implications, Science, 257, 227-230,
1992.
Thomas, W., Personal communication, 1994.
Toohey, D.W., J.G. Anderson, W.H. Brune, and K.R.
Chan, In situ measurements of BrO in the Arctic
stratosphere, Geophys. Res. Lett., 17, 513-516,
1990.
Traub, W.A., presented at Methyl Bromide State of the
Science Workshop, Washington, D.C., October,
1993.
Traub, W.A., D.G. Johnson, K.W. Jucks, and K.V.
Chance, Upper limit for stratospheric HBr using
far-infrared thermal emission spectroscopy, Geo-
phys. Res. Lett., 19, 1651-1654, 1992.
UNEP, Methyl Bromide: Its Atmospheric Science, Tech-
nology, and Economics, Montreal Protocol
Assessment Supplement, edited by R.T.: Watson,
D.L. Albritton, S,O. Anderson, and S. Lee-Bapty,
United Nations Environment Programme, Nairo-
bi, Kenya, 1992.
Wanner, A., J. Callies, H.-P. Dorn, U. Platt, and C.
Schiller, Near UV atmospheric measurements of
column abundances during Airborne Arctic Strato-
spheric Expedition, January-February 1989: 3.
BrO observations, Geophys. Res. Lett., 17, 517-
520,1990.
Wanner, A., and C. Schiller, Twilight variation of vertical
column abundances of OC1O and BrO in the north
polar region, J. Geophys. Res., 97, 8047-8055,
1992.
Wayne, R.P., I. Barnes, P. Biggs, J.P. Burrows, C.E.
Canosa-Mas, J. Hjorth,.G- LeBras, G.K. Moortgat,
D. Pemer, G. Poulet, G. Restelli, and H. Sidebot-
tom, The nitrate radical: Physics, chemistry, and
the atmosphere 1990, Atmos. Environ., 25A, 1-
203, 1991.
WMO, Scientific Assessment of Stratospheric Ozone:
1989, World Meteorological Organization Global
Ozone Research and Monitoring Project - Report
No. 20, Geneva, 1990.
WMO, Scientific Assessment of Ozone Depletion: 1991,
World Meteorological Organization Global Ozone
Research and Monitoring Project - Report No. 25,
Geneva, 1992.
Wuebbles, D.J., Chlorocarbon emission scenarios: Po-
tential impact on stratospheric ozone, J. Geophys.
Res., 88, 1433-1443, 1983.
Yagi, K., J. Williams, N.-Y. Wang, and R.J. Cicerone,
Agricultural soil fumigation as a source of atmo-
spheric methyl bromide, Proc. Natl. Acad. Sci.
U.S., 90, 8420-8423, 1993.
Zhang, Z., R.D. Saini, M.J. Kurylo, and R.E. Huie, A
temperature-dependent kinetic study of the reac-
tion of the hydroxyl radical with CHsBr, Geophys.
Res. Lett., 19, 2413-2416, 1992.
10:26
-------
CHAPTER 11
Subsonic and Supersonic Aircraft Emissions
Lead Authors:
A. Wahner
M.A. Geller
Co-authors:
F. Arnold
W.H. Brune
D.A. Cariolle
A.R. Douglass
C. Johnson
D.H. Lister
J.A. Pyle
R. Ramaroson
D. Rind
F. Rohrer
U. Schumann
A.M. Thompson
-------
-------
CHAPTER 11
SUBSONIC AND SUPERSONIC AIRCRAFT EMISSIONS
Contents :
[
SCIENTIFIC SUMMARY : !. 11.1
11.1 INTRODUCTION .....1 1!.3
11.2 AIRCRAFT EMISSIONS 11.4
11.2.1 Subsonic Aircraft ; 11.5
11.2.2 Supersonic Aircraft , 11.6
11.2.3 Military Aircraft 11.6
11.2.4 Emissions at Altitude 11.6
11.2.5 Scenarios and Emissions Data Bases 11.6
11.2.6 Emissions Above and Below the Tropopause | 11.7
11.3 PLUME PROCESSES 11.10
11.3.1 Mixing ;. 11.10
11.3.2 Homogeneous Processes 11.10
11.3.3 Heterogeneous Processes j : 11.12
11.3.4 Contrails .'.4 11.13
11.4 NOX/H2O/SULFUR IMPACTS ON ATMOSPHERIC CHEMISTRY ;. 11.13
11.4.1 Supersonic Aircraft 1 11.13
11.4.2 Subsonic Aircraft ', 11.14
11.5 MODEL PREDICTIONS OF AIRCRAFT EFFECTS ON ATMOSPHERIC CHEMISTRY 11.15
11.5.1 Supersonic Aircraft 11.15
11.5.2 Subsonic Aircraft ., 11.20
11.6 CLIMATE EFFECTS , L 11.22
11.6.1 Ozone , 11.23
11.6.2 Water Vapor I.. ..11.24
11.6.3 Sulfuric Acid Aerosols 11.24
11.6.4 Soot , 1 1J.24
11.6.5 Cloud Condensation Nuclei '. 11.24
11.6.6 Carbon Dioxide '. 11.24
11.7 UNCERTAINTIES '. 11.25
11.7.1 Emissions Uncertainties '. 11.25
11.7.2 Modeling Uncertainties .> 11.25
11.7.3 Climate Uncertainties 11.27
11.7.4 Surprises 1 11.27
ACRONYMS : 1 11.27
REFERENCES I '..-. 11.28
-------
-------
AIRCRAFT EMISSIONS
|
SCIENTIFIC SUMMARY j
Extensive research and evaluations are underway to assess the atmospheric effects of the present and future
subsonic aircraft fleet and of aprojected fleet of supersonic transports. Assessment of aircraft effects on the atmosphere
involves the following: j.
j
i) measuring the characteristics of aircraft engine emissions; j
ii) developing three-dimensional (3-D) inventories for emissions as a function of time; !
iii) developing plume models to assess the transformations of the aircraft engine emissions to the point where they are
governed by ambient atmospheric conditions;
iv) developing atmospheric models to assess aircraft influences on atmospheric composition and climate- and
v) measunng atmospheric trace species and meteorology to test the understanding of photochemistry and transport
as well as to test model behavior against that of the atmosphere. : .
Supersonic and subsonic aircraft fly in atmospheric regions that have quite different'dynamical and chemical
regimes. Subsonic aircraft fly in the upper troposphere and in the stratosphere near the tropopause, where stratospheric
residence times due to exchange with the troposphere are measured in months. Proposed supersonic aircraft will fly in
the stratosphere near 20 km, where stratospheric residence times due to exchange with the troposphere increase to years
fa the upper troposphere, increases in NOX typically lead to increases in ozone. In the stratosphere, ozone changes
depend on the complex coupling among HOX,NOX, and halogen reactions. j
1
• Emission inventories have been developed for the current subsonic and projected supersonic and subsonic aircraft
fleets. These provide reasonable bases for inputs to models. Subsonic aircraft flying in|the North Atlantic flight
corridor emit 56% of their exhaust emissions into the upper troposphere and 44% into the lower stratosphere on
an annual basis. i
i
Plume processing models contain complex chemistry, microphysics, and turbulence parameterizations Only a
few measurements exist to compare to plume processing model results.
• Estimates indicate that-present subsonic aircraft operations may have increased NOX concentrations at upper
troposphenc. altitudes in the North Atlantic flight corridor by about 10-100%, water vapor concentrations by
about 0.1% or less, SOX by about 10% or less, and soot by about 10% compared with the atmosphere in the
absence of aircraft and assuming all aircraft are flying below the tropopause. |
• Preliminary model results indicate that the current subsonic fleet produces upper tropospheric ozone increases as
much as several percent, maximizing at the latitudes of the North Atlantic flight corridor.
i
The results of these rather complex models depend critically on NOX chemistry. Since there are large uncertainties
in the present knowledge of the tropospheric NOX budget (especially in the upper troposphere), little confidence
should be put in these quantitative model results of subsonic aircraft effects on the atmosphere.
Atmospheric effects of supersonic aircraft depend on the number of aircraft, the altitude of operation, the exhaust
emissions, and the background chlorine and aerosol loading. Rough estimates of the impact of future supersonic
operations (assuming 500 aircraft flying at Mach 2.4 in the stratosphere and emitting 15 grams of nitrogen oxides
per kilogram of fuel) indicate an increase of the North Atlantic flight corridor concentrations of NOX up to about
250%, water vapor up to about 40%, SOX up to about 40%, H2SO4 up to about 200%, soot up to about 100% and
CO up to about 20%. .
II.1
-------
AIRCRAFT EMISSIONS
One result of two-dimensional model calculations of the impact of such a projected fleet in a stratosphere with a
chlorine loading of 3.7 ppbv (corresponding to the year 2015) implies additional annually averaged ozone column
decreases of 0.3-1.8% for the Northern Hemisphere. Although NOX aircraft emissions have the largest impact on
ozone, the effects from H2O emissions contribute to the calculated ozone change (about 20%).
Net changes in the column ozone from supersonic aircraft modeling result from ozone mixing ratio enhancements
in the upper troposphere and lower stratosphere and depletion at higher stratospheric altitudes.
There are important uncertainties in supersonic assessments. In particular, these assessment models produce
ozone changes that differ among each other, especially in the lower stratosphere below 25 km. When used to
calculate ozone trends, these same models predict smaller changes than are observed in the stratosphere below 25
km between 1980 and 1990. Thus, these models may not be properly including mechanisms that are important in
this crucial altitude range.
Increases in ozone at altitudes near the tropopause, such as are thought to result from aircraft emissions, enhance
the atmosphere's greenhouse effect. Research to evaluate the climate effects of supersonic and subsonic aircraft
operations is just beginning, so reliable quantitative results are not yet available, but some initial estimates indi-
cate that this effect is of the. same order as that resulting from the aircraft CO2 emissions.
11.2
-------
11.1 INTRODUCTION
Tremendous growth occurred in the aircraft indus-
try during the last several decades. Figure 11-1 shows
the increasing use of aircraft fuel as a function of time.
Aircraft fuel consumption has increased by about 75%
during the past 20 years and is projected to increase by
100 to 200% over the next 30 years. At the present time,
approximately 3% of the worldwide usage of fossil fuels
is by aircraft. Ninety-nine percent of this aircraft fuel is
burned by subsonic aircraft, of which a large proportion
occurs in the upper troposphere. Table 9.2 of the previ-
ous assessment (WMO, 1992) demonstrates that
subsonic aircraft emit a significant fraction of their ex-
haust products into the lower stratosphere. This depends
on factors such as latitude and season.
Despite the small percentage of the total fossil fuel
usage for aviation, the environmental effects of aircraft
should be closely examined for several reasons. One rea-
son is the rapid growth that has occurred and is projected
to occur in aircraft emissions, and another is that aircraft
emit their exhaust products at specific altitudes where
significant effects might be expected. For instance, an
environmental concern of the 1970s was the effect that
large fleets of supersonic aircraft would have on the
stratospheric ozone layer. The main concern was then
and still is that catalytic cycles involving aircraft-emitted
NOX (NO plus NO2) enhance the destruction of ozone.
1 ' 1
1990 2000
Yur
1
2010
Figure 11-1. Aviation fuel versus time. Data up to
1989 from the International Energy Agency (1990).
Extrapolations according Kavanaugh (1988) with
2.2% per year in a low-fuel scenario and with 3.6%
up to 2000 and 2.9% thereafter in a high-fuel sce-
nario. (Based on Schumann, 1994.)
AIRCRAFT EMISSIONS
Since supersonic aircraft engines may emit significant
amounts of NOX, the fear is that large fleets of superson-
ic aircraft flying at stratospheric levels, where maximum
ozone concentrations exist, might seriously deplete the
stratospheric ozone layer, leading to increased ultravio-
let radiation flux on the biosphere. Also, climate
sensitivity studies have shown that ozone changes in the
upper troposphere and lower stratosphere will have
greater radiative effects on changing surface and lower
tropospheric temperatures than would ozone changes at
other levels (see Chapter 8).
Also, in the 1950s, "smog reactions" were discov-
ered that implied the depletion of tropospheric ozone
when NOX concentrations are low and ozone production
when NOX concentrations are high. Thus, there is a con-
cern that new fleets of su{>ersonic aircraft flying in the
stratosphere would lead to harmful stratospheric ozone
depletion, while present and future subsonic aircraft op-
erations will lead to undesired enhanced levels of ozone
in the upper troposphere.
Development of any successful aircraft requires a
period of about 25 years, and each aircraft will have a
useful lifetime of about 25 years as well. Thus, even if an
environmentally motivated, decision is made to utilize
new aircraft technologies, it will take decades to fully
realize the benefits. j
One can get some perspective on possible atmo-
spheric effects of aircraft operations by noting the
following. Current subsonic aircraft operations in the
North Atlantic flight corridor are probably increasing
NOX concentrations at upper tropospheric altitudes by
about 10-100%, water vapor concentrations by about
0.1 % or less, and SOX by about 10% or less compared to
an atmosphere without aircraft. Future supersonic opera-
tions in the stratosphere might increase the North
Atlantic flight corridor concentrations of NOX up to
about 250%, water vapor up to about 40%, SOX up to
about 40%, H2SO4 up to about 200%, soot up to about
100%, and CO up to about 20%. Thus, present subsonic
aircraft perturbations in atmospheric composition are
now probably significant, and future large supersonic
aircraft fleet operations will also be significant in affect-
ing atmospheric trace gas concentrations.
These and other concerns have led to an increasing
amount of research into the atmospheric effects of cur-
rent and future aircraft operations. In the U.S., NASA's
Atmospheric Effects of Aviation Project is composed of
11.3
-------
AIRCRAFT EMISSIONS
two elements. The Atmospheric Effects of Stratospheric
Aircraft (AESA) element was initiated in 1990 to evalu-
ate the possible impact of a proposed fleet of high-speed
(i.fc, supersonic) civil transport (HSCT) aircraft. A Sub-
• sonic Assessment (Wesoky et al., 1994) was begun in
1994 to study the impact of the current commercial air-
craft fleet. In Europe, the Commission of the European
Communities (CEC) has initiated the Impact of NOX
Emissions from Aircraft upon the Atmosphere
(AERONOX) and Measurement of Ozone on Airbus In-
service Aircraft (MOZAIC) programs (Aeronautics,
1993) and Pollution from Aircraft Emissions in the
North Atlantic Flight Corridor (POLINAT) to investigate
effects of the emissions of the present subsonic aircraft
fleet in flight traffic corridors. In addition, there are also
several national programs in Europe and Japan looking
at various aspects of the atmospheric effects of aircraft
emissions.
Atmospheric models play a particularly important
role in these programs since there does not appear to be
any purely experimental approach that can evaluate the
global impact of aircraft operations on the atmosphere.
The strategy is to construct models of the present atmo-
sphere that compare well with atmospheric measurements
and to use these models to try to predict the future atmo-
spheric effects of changed aircraft operations. At the
present time, the subsonic and supersonic assessment
programs are in quite different stages of maturity and are
utilizing different approaches in both modeling and ob-
servations. Therefore, in this chapter the subsonic and
supersonic evaluations will be considered separately
since the chemical and dynamical regimes are quite dif-
ferent. In this context the "lower stratosphere" refers to
the region above the local tropopause where there are
lines of constant potential temperature that connect the
stratosphere and troposphere. In this region, strato-
sphere-troposphere exchange can occur by horizontal
advcction with no need to expend energy in overcoming
the stable stratification. In the stratosphere near 20 km,
where Mach 2.4 HSCT operate, no lines of constant po-
tential temperature connecting the stratosphere and
troposphere exist. Therefore residence times of tracers
are much larger (about 2 years) in the stratosphere at 20
km than in the lower stratosphere.
In this chapter, we will review what is known
about aircraft emissions into the atmosphere and discuss
the transformations that take place in the aircraft plume
as it adjusts from the physical conditions of the aircraft
exhaust leaving the engine tailpipe to those of the ambi-
ent atmosphere. Some of the atmospheric effects of the
different chemical families that are emitted by aircraft
are then considered, and finally, modeling studies of the
atmospheric effects of aircraft emissions on ozone are
presented, along with a discussion of possible climate
effects of aircraft operations. A discussion of the level of
uncertainty of these predictions, and some conclusions
are presented.
Further details of the NASA effort to assess the at-
mospheric effects of future supersonic aircraft
operations can be found in Albritton et al. (1993) and the
references therein. An external evaluation of these ef-
forts can be found in NRC (1994). No similar documents
exist at this time pertaining to the atmospheric effects of
subsonic aircraft operations.
11.2 AIRCRAFT EMISSIONS
The evaluation of the potential impact of, emis-
sions from aircraft on atmospheric ozone levels requires
a comprehensive understanding of the nature of the
emissions produced by all types of aircraft and a knowl-
edge of the operations of the total global aircraft fleet in
order to generate a time-dependent,' three-dimensional
emissions data base for use in chemical/dynamical at-
mospheric models.
Emissions from the engines, rather than those as-
sociated with the airframe, are considered to be
dominant (Prather et al., 1992). These are functions of
engine technology and the operation of the aircraft on
which the engines are installed. Primary engine exhaust
products are CO2 and H2O, which are directly related to
the burned fuel, with minor variations due to the precise
carbon-hydrogen ratio of the fuel. Secondary products
include NOX ( = NO + NO2), CO, unburned and partially
burnt fuel hydrocarbons (HC), soot particulates/smoke,
and SOX. NOX is a consequence of the high temperature
in the engine combustor; the incomplete combustion
products (CO, HC, and soot/smoke) are functions of the
engine design and operation and may vary widely be-
tween engines. SOX is directly related to fuel
composition. Currently, typical sulfur levels in aviation
kerosene are about 0.05% sulfur by weight, compared
with an allowed specification limit of 0.3% (ICAO,
1993).
11.4
-------
AIRCRAFT EMISSIONS
Table 11-1. Emission Index (grams per kilograms of fuel used) of various materials for subsonic and
supersonic aircraft for cruise condition. Values in parentheses are ranges for different engines and oper-
ating conditions, i
Species
(gmMW)
CO2 (44)
H20(18)
CO (28)
HC as methane (16)
S02(64)
NOX as NO2 (46)
Subsonic Aircraft*
Short range
3160
1230
5.9 (0.2-14)
0.9 (0.12-4.6)
1.1
9.3(6-19)
Long range
3160
1230
3.3(0.2-14)
0.56(0.12-4.6)
1.1
14.4(6-19)
Supersonic Aircraft*
i _ — .
3160
1230
1.5(1.2-3.0)
0.2 (0.02-0.5)
1.0
depends on design
(5-45)
* Mean (fuel-consumption weighted) emission indices for 1987 based on Boeing (1990). the values were calculated
from a data base containing emission indices and fuel consumptions by aircraft types. The difference between short
range (cruise altitude around 8 km) and long range (cruise altitude between 10 and 11 km)! reflects different mixes of
aircraft used for different flights. I
* Based on Boeing (1990) and McDonnell Douglas (1990). j
The measure of aircraft emissions traditionally
used in the aviation community is the Emissions Index
(El), with units of grams per kilogram of burnt fuel. Typ-
ical El values for subsonic and anticipated values for
supersonic aircraft engines are given in Table 11-1 for
cruise conditions. By convention, EI(NOX) is defined in
terms of NO2 (similarly, hydrocarbons are referenced to
methane).
Historically, the emissions emphasis has been on
limiting NOX, CO, HC, and smoke, mainly for reasons
relating to boundary layer pollution. Standards are in .
place for -control of these over a Landing/Take-Off
(LTO) cycle up to 915 m altitude at and around airports
(ICAO, 1993). Currently there are no regulations cover-
ing other flight regimes, e.g. cruise, though ICAO (1991)
is considering the need and feasibility of introducing
standards.
It is now recognized that the list of chemical spe-
cies (emitted from engines or possibly produced in the
young plume, also by reactions with ambient trace spe-
cies like hydrocarbons) that may be relevant to ozone
and climate change extends well beyond the primary
combustion species and NOX. A more complete set of
"odd nitrogen" compounds, known as NOy—including
NOX, N2O5, NO3, HNO3—and PAN (peroxyacetylni-
trate) should be considered, i along with SOX and soot
particles as aerosol-active species. HC and CO may also
play an important role in highi altitude HOX chemistry.
11.2.1 Subsonic Aircraft
Engine design is a compromise between many
conflicting requirements, amo
-------
AIRCRAFT EMISSIONS
would imply that the actual mass output should have de-
creased by about 77% for HC and 30% for CO, while
NOX mass output should have increased by about 110%.
Considerable further reductions of HC and CO will
come as older aircraft are phased out, but little change
can be expected for NOX without the introduction of
low-NOx technology engines.
The first steps to develop combustion systems pro-
ducing significantly lower NOX levels relative to existing
technology were made in the mid-1970s (CIAP 2, 1975).
These systems achieve at least a 30% NOX reduction,
and are now being developed into airworthy systems for
introduction in medium and high thrust engines.
11.2.2 Supersonic Aircraft
The first generation of civil supersonic aircraft
(Concorde, Tupolev TU144) incorporated turbojet en-
gines of a technology level typical of the early 1970s.
The second generation, currently being considered by a
number of countries and industrial consortia, will have
to incorporate technology capable of meeting environ-
mental requirements. A comprehensive study of the
scientific issues associated with the Atmospheric Effects
of Stratospheric Aircraft (AESA) was initiated in 1990
as part of NASA's High Speed Research Program
(HSRP; Prather et ai, 1992). No engines or prototypes
exist and designs are only at the concept stage. A range
of cruise EI(NOX) levels (45, 15, and 5) has been set as
the basis for use in atmospheric model assessments and
in developing engine technology. An EI(NOX) of 45 is
approximately what would be obtained if HSCT engines
were to be built using today's jet engine technology
without putting any emphasis on obtaining lower
EI(NOX) emissions. Jet engine experts have great confi-
dence in their ability to achieve an HSCT engine design
with EI(NOX) no greater than 15 and have set a goal of
designing an HSCT engine with EI(NOX) no greater than
5. Laboratory-scale studies of new engine concepts,
which appear to offer the potential of at least 70-80%
reduction in NOX compared with current technology, are
being pursued. Early results indicate that these systems
seem able to achieve the low target levels of EI(NOX) = 5
(Albritton et a/., 1993).
11.2.3 Military Aircraft
In contrast to the majority of civil aviation, mili-
tary aircraft do not operate to set flight profiles or
frequencies. Also, national authorities are reluctant to
disclose this information. Thus it is extremely difficult to
make realistic assessments of the contribution of mili-
tary aircraft in terms of fuel usage or emissions. Earlier
estimates (Wuebbles et ai, 1993) were that the world's
military aircraft used about 19% of the total aviation fuel
and emitted 13% of the NOX, with an average EI(NOX)
of 7.5. With the changes following the breakup of the
former Soviet Union, there has been considerable reduc-
tion in activity, and an estimate of about 10% fuel usage
may be more appropriate (ECAC/ANCAT, 1994).
11.2.4 Emissions at Altitude
As noted above, engines are currently only regulated
for some species over an LTO cycle. Internationally ac-
credited emissions data on these are available (ICAO,
1994). However, experimental data for other flight con-
ditions are sparse, since these can only realistically be
obtained from tests in flight or in altitude simulation test
facilities. Correlations, in particular for NOX, have been
developed from theoretical studies and combustor test
programs for prediction of emissions over a range of
flight conditions. A review of these is given elsewhere
(Prather et al., 1992; Albritton et al., 1993). Engine tests
under simulated altitude conditions are being carried out
within the AERONOX program (Aeronautics, 1993) and
should be useful to check this approach for subsonic en-
gines.
11.2.5 Scenarios and Emissions Data Bases
Air traffic scenarios have been developed as a ba-
sis for evaluating global distributions of emissions from
aircraft (Mclnnes and Walker, 1992; Prather et al., 1992;
Wuebbles et al., 1993; ECAC/ANCAT, 1994). The first
two based their traffic assessment on scheduled com-
mercial flight information from timetables and
supplemented these data with information from other
sources for non-scheduled charter, general aviation, and
military flights. The third is based on worldwide Air
Traffic Control data supplemented by timetable informa-
tion and other data as appropriate.
Mclnnes and Walker (1992) generated 2-D and
3-D inventories of NOX emissions from subsonic air-
craft, using relatively broad assumptions for numbers of
aircraft types, flight profiles/distance bands, and cell siz-
es. However, the evaluation did not include
non-scheduled, military, cargo, or general aviation, and
11.6
-------
AIRCRAFT EMISSIONS
both inventories accounted for only 51 % of the total esti-
mated fuel consumption of 166.5 x 109 kg for the year
1989 (IEA, 1990). The fuel consumption was simply
scaled to match the total estimated fuel consumption in
order to estimate the total NOX mass. Their average
EI(NOX) value of 11.6 is within the range quoted else-
where (NiiBer and Schmitt [1990] 6 - 16.4; Egli [1990]
11-30; and Beck et al. [1992] 17.9).
Wuebbles et al. (1993) generated for the HSRP/
AESA (Prather et al, 1992; Stolarski and Wesoky,
1993a) a comprehensive assessment of all aircraft types
to determine fuel, NOX, CO, and HC for general scenar-
ios comprising the 1990 fleet and projected fleets of
subsonic and supersonic aircraft (HSCTs) for the year
2015. A much better match (76%) of the calculated fuel
use with the total estimated fuel consumption for 1990
was achieved. The remainder is likely to be mainly at-
tributable to factors such as the non-idealized flight
routings and altitudes actually flown by aircraft due to
factors such as air traffic control, adverse weather, etc.,
as well as low-level unplanned delays and ground opera-
tions. However, scaling to match the total estimated fuel
consumption gave a total annual NOX mass (1.92 Tg)
similar to that" of Mclnnes and Walker. Illustrations of
the global NOX inventories as functions of latitude/longi-
tude, or altitude/latitude for both 1990 and 2015 are
given in Figures 11-2 and 11-3.
The European Civil Aviation Conference (ECAC)
Abatement of Nuisance Caused by Air Traffic (ANCAT)
work, carried out to complement the AERONOX pro-
gram, has also considered NOX emissions from subsonic
and supersonic fleets for the year 1992. Unlike the other
inventories, the traffic data have been compiled for four
equally spaced months throughout the year to provide
information on the seasonal variation. Preliminary re-
sults indicate a higher fuel burn, NOX annual mass, and
EI(NOX) than those of the other inventories and are like-
ly to represent upper bounds on the aircraft NOX
emission burden. The current grid scale is larger than
that of the HSRP/AESA inventory, but this may give a
more realistic representation of the NOX distribution
within the heavily traveled air traffic routes, such as the
North Atlantic, where there is known to be a significant
divergence of actual flight paths from the ideal great cir-
cle routes currently assumeid by all inventories. Further
work is being carried out to produce forecast inventories
for the years 2003 and 2015.
Considerable comparative analysis is being under-
taken between the ECAC/ANCAT and the HSRP/AESA
inventories in order to understand the reasons underlying
the differences (EI(NOX) 10.9 to 16.8; NOX mass 1.92 to
2.8 Tg) and to refine the inventories. For example, it is
already known that there is some double counting of
traffic in some geographically important areas of the
ECAC/ANCAT inventory. Another significant factor is a
large difference in the contribution from military air-
craft. A comparison summary of the inventories is given
in the table at the bottom of the page.
11.2.6 Emissions Above and Below the
Tropopause
In a global perspective, the North Atlantic, apart
from North America and Europe, contains the largest
specific subsonic traffic load. In 1990 the average daily
movements across the Atlantic (both directions) between
45° and 60°N amounted to 595 flights in July and 462
flights in November. One-recent study (Hoinka et al.,
1993) has assessed the aircraft fleet mix and the resulting
emissions for this flight corridor. By correlation of the
traffic data with the tropopa'use height from the Euro-
pean Centre for Medium-Range Weather Forecasts
Year
Grid size
Fuel match
El (NOX) global
NOX mass (Tg)#
Mclnnes and Walker,
1992
1989
7.5° x 7.5° x 0.5km
51%
11.6
1.91"
Wuebbles et al.,
1993
1990
l°xlkm
76%
10.9
1.92#
ECAC/ANCAT,
1994
1992
2J.8° x 2.8° x 1km
99%
16.8
# Note: all data for NOX mass have been scaled to 100% fuel match.
11.7
-------
AIRCRAFT EMISSIONS
-90
-180 -150 -120 -90 -60
0.01
-30 0 30
Longitude
60
120 150 180
50.01 100.01
Molecules/Year of NOX (xlO29)
150.01
-90
-180 -150 -120 -90 -60 -30 0 30
Longitude
60 90 120 150 180
0.01 10.01
20.01 30.01
Molecules/Year of NOX (xlO29)
40.01
50.01
Fiqure 11-2. Annual NOX emissions for proposed 2015 subsonic and Mach 24 (EI(NOX)=15) HSCT fleets as
functbn of latitude and longitude. Top panel shows emissions below 13 km pnmar, y subsonic traffic) whrte
bottom panel shows emissions above 13 km (primarily HSCT traffic). (Albntton. et al., 1993)
11.8
-------
AIRCRAFT EMISSIONS
OL
-90 -80 -70 -60 -50 -40 -30-20-10 0 10 20 30 40 50 60 70 80 90
Latitude ;
0.01 500.01
1000.01 1500.01 2000.01 2500.01 ; 3000.01 3500.01
Molecules/Year of NOX (xlO29) |
OL
-90 -80 -70 -60 -50 -40 -30 -20 -10 0 10 20 30 40 50 60 70 80 90
Latitude
0.01 500.01 1000.01 1500.01 2000.01 2500.01 3000.01 3500.01
Molecules/Year of NOX (xlO29) :
Figure 11-3. Annual NOX emissions as a function of altitude and latitude for 1990 subsonic fleet (Scenario A,
top panel) and for proposed 2015 subsonic and Mach 2.4 (EI(NOX)=15) HSCT fleets (bottom panel). (Albrit-
ton era/., 1993) ' i
11.9
-------
AIRCRAFT EMISSIONS
(ECMWF) data, it is estimated that 44% of the NOX
emissions are injected in the lower stratosphere and 56%
are injected in the upper troposphere.
11.3 PLUME PROCESSES
Plume processing involves the dispersion and con-
version of aircraft exhausts on their way from the scales
of the jet engines to the grid scales of global models. The
details of plume mixing and processing can be important
for conversion processes that depend nonlinearly on the
concentration levels, such as the formation of contrails,
the formation of soot, sulfur and nitric acid particles, and
nonlinear photochemistry. Also, the vertical motion of
the plumes relative to ambient air and sedimentation of
particles may change the effective distribution of emitted
species at large scales. Contrails may impact the mixing,
sedimentation, heterogeneous chemistry, and the forma-
tion of cirrus clouds, with climatic consequences.
11.3.1 Mixing
The aircraft wake can be conveniently subdivided
into three regimes (Hoshizaki et al., 1975): the jet, the
vortex, and the dispersion regimes. The vortex regime
persists until the vortices become unstable and break up
into a less ordered configuration. Thereafter, the disper-
sion regime follows, in which further mixing is
influenced by atmospheric shear motions and turbulence
depending on shear, stratification, and other parameters
(Schumann and Gerz, 1993). With respect to mixing
models, the jet and vortex regime, including the very ear-
ly dispersion regime, can be computed with models as
described by Miake-Lye et al. (1993). The engine
plumes grow by turbulent mixing to fill the vortex pair
cell. Due to rotation, centripetal acceleration causes in-
ward motions of the relatively warm jet plumes so that
the exhaust gases get trapped near the narrow well-
mixed core of the vortices. The radial pressure gradient
also causes adiabatic cooling and hence increases the
formation of contrails. These centripetal forces are much
larger for supersonic aircraft than for subsonic aircraft. It
should be noted, however, that these model results re-
main largely untested, observationally.
Details of the plume fluid dynamics depend criti-
cally on the aircraft scales. For a Boeing-747, one may
estimate that the jet regime lasts for about 10 s and the
following vortex regime for about 1 to 3 minutes. The
cross-section of the trailing vortex pair represents an up-
per bound for the mixed area of the plumes. However,
measurements of water vapor concentration and temper-
ature in the jet and vortex regime (>2 km behind a DC-9.
at cruising altitude) exhibit a spiky concentration field
within the double vortex system, indicating that the indi-
vidual jet plumes may not yet be homogeneously mixed
over the vortex cross-section at such distances (Bau-
mannera/., 1993).
The lift of the aircraft induces downward motion
of the double vortex structure at about 2.4 ± 0.2 m s-' for
a Boeing-747, which decreases when the vortices mix
with the environment at altitudes that may be typically
100 m lower than flight level. During this descent, parts
of the exhaust gases are found to escape the vortex cores.
In the supersonic case, the vortex pair has more
vertical momentum (descent velocity of about 5 mis),
and its vertical motion will continue (possibly in the
form of vortex rings) well after the vortex system has
broken up. This will lead to exhaust species deposition a
few hundred meters below flight altitude (Miake-Lye et
al., 1993). Radiation cooling of the exhaust gases may
contribute to additional sinking (Rodriguez et al., 1994),
in particular when contrails are forming. ;
Very little is known about the rate of mixing in the
dispersion range, and it is this rate of mixing that plays a
large role in determining the time evolution of the gas
composition of the plume (Karol et al., 1994). In fact, it
is yet unknown at what time scales the emissions be-
come indistinguishable from the ambient atmosphere.
Table 11-2 shows estimates of the concentration increas-
es due to aircraft emissions in a young exhaust plume
(vortex regime) and at the scales of the North Atlantic
flight corridor (Schumann, 1994). These are the scales in
' between which global models will be able to resolve the
concentration fields. The background concentration esti-
mates are taken from Penner et al. (1991) for NOX,
Mohler and Arnold (1992) for SO2, and Pueschel et al.
(1992) for soot. With respect to background, the concen-
tration increases in young plumes are of importance for
all aircraft emissions included in Table 11-2. A strong
corridor effect is expected for NOX and, at least in the
lower stratosphere, also for SOX and soot particles.
11.3.2 Homogeneous Processes
Several models have been developed to describe
the finite-rate chemical kinetics in the exhaust plumes
11.10
-------
AIRCRAFT EMISSIONS
'"* -,*••''• ' f
Table 11-2. Mean concentration increases in vortex regime (5000 m2 cross-section) of a B-747 plume,
and mean concentration increase in the North Atlantic flight corridor due to traffic exhaust emissions
from 500 aircraft. (Table adopted from Schumann, 1994.) ,
Species
C02
H2O
NOX(NO2)
S02
soot
El (g/kg)
3150
1260
18
1
0.1
Background
concentration
at 8 km
358 ppmv
20-400 ppmv
0.01-0.05 ppbv
50-300 pptv
3 ng/m3
Mean :
concentration
increase in
vortex regime
:
14 ppmv |
14 ppmv j
78 ppbv j
3 100 pptv ;
240 ng/m3 :
Mean
concentration
increase in
North Atlantic
flight corridor
0.02 ppmv
0.02 ppmv
. 0.1 ppbv
4 pptv
0.3 ng/m3
(Danilin et al., 1992; Miake-Lye et ai, 1993; Pleijel et
ai, 1993; Weibrink and Zellner, 1993). Most models fol-
low a well-mixed air parcel as a function of plume age or
distance behind the aircraft. The models are initialized
either with an estimate of emissions from the jet exit or a
separate model describing the kinetics after the combus-
tion chamber within the engine. Considerable deviations
from local equilibrium are predicted at the jet exit, in
particular for CO, NO, NO2, HNO3, OH, O, and H. In
the models, the air parcel grows in size as a prescribed
function of mixing with the environment, and the con-
centrations in the plume change according to mixing
with the ambient air and due to internal reactions in the
homogeneous mixture. The models differ in the treat-
ment of mixing, in the reaction set used to simulate the
exhaust plume finite-rate chemical kinetics, photolysis
rates, treatment of heterogeneous processes, and in the
prescription of the effective plume cross-section as a
function of time or distance. Since most of the NOX
emissions are in the form of NO, a rapid but local de-
struction of ozone is to be expected.
Besides some incidental measurements in flight
corridors or contrails (Hofmann and Rosen 1978; Doug-
lass et al., 1991), very few data exist at this time on the
gaseous emissions in aircraft plumes in the atmosphere.
Measurements of the gases HNO2, HNO3, NO, NO2,
and SO2 were recently made (Arnold et al., 1992,1994a)
in the young plume of an airliner at cruising altitude (see
Figure 11-4). The data imply that not more than about
1% of the emitted odd-nitrogen underwent chemical
conversion to longer living HNO;). Hence, most of the
emitted odd nitrogen initially remains in a reactive form,
which can catalytically influence ozone.
.10
o
(E
IU
m'
2L
I 8
F91-12 DC-9 TRAIL i
5 DEC 1991
-9
<
o:
-10 Z
-11.
12.27
12.28
12.29 12.30 12.31
UNIVERSAL TIME
12.32
12.33
Figure 11-4. Time plot of nitrous acid (HNC>2) and
nitric acid abundance measured during chase of a
DC-9 airliner at 9.5 km altitude and a distance of 2
km. Periods when the research aircraft was inside
the exhaust-trail of the DC-9 are marked by bars.
For these periods NO and MO2 abundance are also
given. (Arnold et al., 1992, 1994b; recalibration
changed conversion factors shown in figure to: NO x
0.006 and NO2 x 0.003.) j
11.11
-------
AIRCRAFT EMISSIONS
1—1—I—d—I—I—I—I
10
20 30 40
Elapsed time (s)
50
60
Rgure 11-5. Time series for NO, NOy, CO2, H2O,
and CN during the plume encounters on May 18,
1993. The approximate Greenwich Mean Time
(GMT) is noted in the,top panel. The scale on the
left side indicates the absolute value of each spe-
cies. The zero in the right scale is set to the
approximate background values,of each species.
At the ER-2 airspeed of 200 m s-1, the panel width
of 60 seconds corresponds to 12 km. (Based on
Faheyefa/., 1994.)
In situ measurements of NOy, NO, CO2, H2O,
condensation nuclei, and meteorological parameters
(Figure 11-5) have been used to observe the engine ex-
haust plume of the NASA ER-2 aircraft approximately
10 minutes after emission operating in the lower strato-
sphere (Fahey el al., 1994). The obtained EI(NOX) of 4 is
in good agreement with values scaled from limited
ground-based tests of the ER-2 engine. Non-NOx nitro-
gen species comprise less than about 20% of emitted
reactive nitrogen, consistent with model evaluations.
11.3.3 Heterogeneous Processes
New particles form in young exhaust plumes of jet
aircraft. This is documented by In situ condensation nu-
cleus (CN) measurements made (Hofmann and Rosen,
1978; Pitchford et ai, 1991; Hagen et al., 1992; White-
field et al., 1993) in plumes under flight conditions.
The molecular physics details of nucleation are
not well known and the theory of bimolecular nucleation
is only in a rudimentary state. For a jet engine exhaust
scenario, nucleation takes place in a non-equilibrium
mechanism, which further complicates a theoretical de-
scription. It seems, however, that jet aircraft may form
long-lived contrails composed of H2SO4-H2O aerosols
and soot particles covered with H2SC>4-H2O. Under
conditions of low ambient temperatures around 10 km
altitude, particularly in winter at high latitudes, contrails
composed of HNO3-H2O aerosols may also form (Ar-
nold et al., 1992). Even if HNO3-H2O nucleation does
not occur, some HNQs may become incorporated into
condensed-phase H2SC>4-H2O by dissolution at low tem-
peratures.
There are several potential effects of newly formed
CN and activated soot. Such CN may trigger water con-
trail formation, induce heterogeneous chemical
reactions, and serve as cloud condensation nuclei
(CCN). Thereby, jet aircraft-produced CN may have an
impact on trace gas cycles and climate. However, at
present this is highly speculative.
Numerical calculations with chemical plume mod-
els show that the impact of aircraft emissions on the
atmosphere in the wake regime critically depends on het-
erogeneous processes where considerable uncertainties
still exist (Danilin et al., 1992, 1994). Danilin et al.
(1992) have considered the heterogeneous reaction
N2C>5 + H2O —> 2HNO3 on ambient aerosol particles
only. They have found that this reaction does not play an
important role at time scales of up to one hour in the
wake, but may get important at larger time scales. Taking
contrail ice (or/and nitric acid trihydrate [NAT]) particle
formation into account, Danilin et al. (1994) estimate
that heterogeneous processes are more important at
lower temperatures, but their impact on heterogeneous
conversion is small during the first day after emission. In
contrast, Karol etal. (1994) found noticeable "heteroge-
neous impact" on the chemistry in the plume taking into
account the growth of ice particles.
Around 10 km altitude, there seems to exist a
strong CN source, which is not due to aircraft but to
H2SC>4 resulting from sulfur sources at the Earth's sur-
face (Arnold et al., 1994a). Hence, the relative
contribution of aircraft to CN production around 10 km
77.72
-------
AIRCRAFT EMISSIONS
Table 11-3. Estimates of stratospheric perturbations due to aircraft effluents of a fleet of approxi
mately 500 Mach 2.4 HSCTs (NOX El=15) relative to background concentrations. (Perturbations are
estimated for a broad corridor at northern midlatitudes.) (Expanded from Stolarski and Wesoky, I993b.)
Species
NOX
H2O
SOX
H2SO4
Soot
Hydrocarbons
CO
CO2
Perturbation
3-5 ppbv
0.2-0.8 ppmv
10-20 pptv
350-700 pptm
~7pptm
2 ppbv (NMHC)
~2ppbv
-1 ppbv
Background
2-16ppby
2-6 ppmv
50- 100 pptv
350-700 pptm
~7 pptm
1600 ppbv (CH4)
10-50 ppbv
350 ppmv
altitude needs to be determined. It is uncertain whether
CN production around 10 km actually has a significant
impact on trace gas cycles and CCN.
11.3.4 Contrails
Miake-Lye et al. (1993) have applied the analysis
of Appleman (1953) to the standard atmosphere as a
function of altitude and latitude. Their result shows that
much of the current high-flying air traffic takes place at
altitudes where the formation of contrails is very likely,
in particular in the northern winter hemisphere. A small
reduction of global mean temperature near and above the
tropopause, by say 2 K, would strongly increase the re-
gion in which contrails have to be expected. Also, a
slight change in the threshold temperature below which
contrails form has a strong effect on the area of coverage,
with contrails.
Except for in situ measurements by Knollenberg
(1972), little is known about the spatial structure and
microphysical parameters of contrails. Recent measure-
ments (Gayet et ai, 1993) show that contrails contain
more and smaller ice particles than natural cirrus, lead-
ing to about double the Optical thickness in spite of their
smaller ice content. Contrail observations from satellite
data, Lidar measurements, and climatological observa-
tions of cloud cover changes have been described by
Schumann and Wendling (1990). Large (1 to 10 km wide
and more than 100 km long) contrails are observed re-
gionally on about a quarter of all days within one year,
but the average contrail coverage is only about-0.4% in
mid-Europe. Lidar observations show that particles from
contrails sediment quickly ait approximately 10 km alti-
tude (Schumann, 1994). !
11.4 NOX/H20/SULFUR IMPACTS ON
ATMOSPHERIC CHEMISTRY
11.4.1 Supersonic Aircraft
i
The impacts of HSCT emissions on chemistry are
discussed in detail in Stolarski and Wesoky (1993b).
Here we give a short summary. Effects of emissions
from HSCTs (see Table 11-3) on ozone are generally
predicted to be manifested through gas phase catalytic
cycles involving NOX, HO*, C1OX, and BrOx. The
amounts of these radicals are changed by two pathways.
First, they are changed by chemistry, either addition of
or repartitioning within nitrogen, hydrogen, and halogen
chemical families. Predicted changes in ozone from this
pathway are initiated primarily by NOX chemistry! Sec-
ond, they are changed when HSCT emissions affect the
properties of the aerosols and the probability of polar
stratospheric cloud (PSC) formation. Changes in ozone
from this pathway are determined primarily by C1OX and
BrOx chemistry, with a contribution from HOX chemis-
try (see Chapter 6 for more detail).
Heterogeneous chemist^ on sulfate aerosols also
has a large impact on the pptential ozone loss. Most im-
portant is the hydrolysis of iN2O5: N2O5 + H2O ->
2 HNC-3. Several observations are consistent with this.
reaction occurring in the lower stratosphere (e.g., Fahey
et al., 1993; Solomon and Keys, 1992). Its most direct
11.13
-------
AIRCRAFT EMISSIONS
effect is to reduce the amount of NOX. Indirectly, it in-
creases the amounts of CIO and HO2 by shifting the
balance of CIO and C10NO2 more toward CIO during
the day and by reducing the loss of HOX into HNO3. As a
result, the HOX catalytic cycle is the largest chemical
loss of ozone in the lower stratosphere, with NOX sec-
ond, and both the C1OX and BrOx catalytic cycles have
increased importance compared to gas phase conditions.
The addition of the emissions from HSCTs will
affect the partitioning of radicals in the NOy, HOy, and
ClOy chemical families, and thus will affect ozone. The
NOX emitted from the HSCTs will be chemically con-
verted to other forms, so that the NOx/NOy ratio of these
emissions will be almost the same as for the background
atmosphere. As a result, the NOX emissions will tend to
decrease ozone, but less than would occur in the absence
of sulfate aerosols.
The increase in H2O will lead to an increase in
OH, because the reaction between O('D) that comes
from ozone photolysis and H2O is the major source of
OH; however, increases in NOy will act to reduce HOX
through the reactions of OH with HNO3 and HNO4. On
the other hand, HNO3, formed in the reaction of OH with
NOa, can be photolyzed in some seasons and latitudes to
regenerate OH. When all of these effects are considered,
the amount of HOX is calculated to decrease—HO2 by
up to 30% and OH by up to 10%. Thus, the catalytic de-
struction of ozone by HOX, the largest of the catalytic
cycles, is decreased.
Finally, C1OX concentrations decrease with the ad-
dition of HSCT emissions for two reasons. First and
most important, with the addition of more NO2, the day-
lime balance between CIO and C1ONO2 is shifted more
toward C10NO2. Second, with OH reduced, the conver-
sion of HC1 to Cl by reaction with OH is reduced, so that
more chlorine stays in the form of HC1. Thus, the catalyt-
ic destruction of ozone by CIOX is decreased.
The addition of HSCT emissions results in in-
creases in the catalytic destruction of ozone by the NOX
cycle that are compensated by decreases in the catalytic
destruction by C1OX and HOX. Because the magnitudes
of the changes in catalytic destruction of ozone are simi-
lar for the NOX, HOX, and C1OX cycles, compensation
results in a small increase or decrease in ozone. Model
calculations indicate a small decrease. The decreases in
the catalytic destruction of O3 by C1OX and HOX involve
the effects of increased water vapor and HNO3 on the
rates of heterogeneous reactions on sulfate and the prob-
.ability of PSC formation.
The addition of sulfur to the stratosphere from
HSCTs will increase the surface area of the sulfate aero-
sol layer. This change in aerosol surface area is expected
to be small compared to changes from volcanic erup-
tions, with a possible exception being the immediate,
vicinity of the aircraft wake. Model calculations by Bek-
ki and Pyle (1993) predict regional increases of the mass
of lower stratospheric H2SO4-H2O aerosols, due to air
traffic, by up to about 100%. The importance of sulfur
emissions from HSCTs in the presence of this large and
variable background needs to be assessed.
11.4.2 Subsonic Aircraft
The emissions from subsonic aircraft take place
both in the lower stratosphere and troposphere. The pri-
mary chemical effects of aircraft in the troposphere seem
to be related to their NOX emissions. The concentration
of ozone in the upper troposphere depends on transport
of ozone mainly from the stratosphere and on upper tro-
posphere ozone production or destruction. The impact of
subsonic aircraft occurs through the influence of NOX on
the tropospheric HOX cycle (see Chapter 5 for a fuller
discussion of tropospheric ozone chemistry).
The HOX cycle is initialized by the photolysis of
ozone itself, which results in the production of OH radi-
cals and destruction of ozone. OH radicals have two
possible reaction pathways: reaction with CO, CH4, and
non-methane hydrocarbons (NMHC) resulting in HO2
and RO2 radicals; or reaction with NO2, removing OH
and NOX from the cycle. The HO2 radicals that are pro-
duced also have two possible pathways: reaction with
ozone or reaction with NO. The first one removes ozone
from the cycle; the second one (also valid for RO2 radi-
cals) produces ozone and regains NO. Additionally, both
pathways regain OH radicals.
As a consequence, ozone is destroyed photochem-
ically in the'absence of NOX. Only in the presence of
NOX can ozone be produced. The net production/de-
struction depends on the combination of these processes.
Their relative importance is controlled mainly by the
NOX concentration. In a regime of low NOX, the ozone
concentration will be reduced photochemically. At high-
er NOX concentrations (on the order of 10 pptv NOX)
NOX will lead to a net ozone production. In both re-
gimes, additional MOX will result in higher ozone
11.14
-------
AIRCRAFT EMISSIONS
concentrations. Only when the concentration of NOx-is
so high (over a few hundred pptv NOX) that the OH con-
centration starts to decline, will additional NOX result in
a lower ozone production.
The impact of NOX emitted by aircraft depends,
therefore, on the background NOX concentration and on
the increase in NOX concentration. Measurements show
that background NOX concentrations (including NOX
emitted from subsonic aircraft) are in the range of 10-
200 pptv NOX. Therefore, airplane emissions take place
in the regime of increasing ozone production most of the
time, where increasing NOX results in increased local
ozone concentrations.
In this regime, the concentration of OH radicals is
enhanced also by additional NOX. First, enhanced ozone
means higher production of OH by photolysis of ozone.
Second, the partitioning in the HOX family is shifted to-
wards OH by the reaction of HO2 with NO. The loss
process of OH by reaction with NO2 is not yet important.
This enhancement of the OH concentration reduces the
tropospheric lifetime of many trace species like CHU
N0x,etc.
The emission of sulfur from aviation is much
smaller than from surface emissions and negligible in
terms of the resultant acid rain, but may be important if
emitted at high altitudes. Hofmann (1991) reported ob-
servations that show an increase of non-volcanic
stratospheric sulfate aerosol of about 5% per year. He
suggests that if about 1/6 of the Northern Hemisphere air
traffic takes place directly in the stratosphere and if a
small fraction of other emissions above 9 km would en-
ter the stratosphere through dynamical processes, then
the jet fleet appears to represent a large enough source to
explain the observed increase. On the other hand, Bekki
and Pyle (1992) conclude from a model study that al-
though aircraft may represent a substantial source of
sulfate below 20 km, the rise in air traffic is insufficient
to account for the observed 60% increase in large strato-
spheric aerosol particles over the 1979-1990 period.
Sulfate particles generated from SOX may also contrib-
ute to nucleation particles (Arnold et al., 1994a).
Whitefield et al. (1993) find a positive correlation be-
tween sulfur content and CCN efficiency of particles
formed in jet engine combustion.
The possible enhancement of aerosol surface area
may affect the nighttime chemistry of the nitrogen ox-
ides. The heterogeneous reaction of N2O5 (and possibly
NO3) on aerosol surfaces will reduce the concentration
of photochemically active NOX during the day, giving
rise to lower ozone and OH concentrations in the upper
troposphere (Dentener and Crutzen, 1993).
11.5 MODEL PREDICTIONS OF AIRCRAFT
EFFECTS ON ATMOSPHERIC CHEMISTRY
The first investigations concerning the potential
effects of supersonic aircraft on the ozone layer were
conducted in the 1970s. Early assessments were ob-
tained using one-dimensional (1-D) photochemical
models; more recent assessments rely on 2-D models
(e.g., Stolarski and Wesoky, 1993b). In addition, the
transport in 2-D models has been compared to 3-D mod-
el transport by examining die evolution of the
distribution of passive tracers.
11.5.1 Supersonic Aircraft
Evaluations of the effects of the emissions of the
HSCT on the lower stratosphere have used two-dimen-
sional (2-D) models. These are z:onally averaged (lati-
tude-height) models and are discussed in detail in
Chapter 6. For use in such 2-D models, both the source
of exhaust and the emission transport (both horizontal
and vertical) are zonally averaged. In fact, the source of
emissions is not zonally symmetric, as HSCT flight is
expected to be restricted to oceanic corridors. Further-
more, the transport processes through which trace spe-
cies are removed from the stratosphere are not well
represented by a zonally averaged model. Stratosphere-
troposphere exchange processes (STE) occur preferen-
tially near jet-systems, above frontal perturbations, and
during strong convection in tropical regions. The two
former processes may transport effluents released by
HSCTs irreversibly to lower levels and lead to tropo-
spheric sinks. Effluents may be rapidly advected also to
lower latitudes by large-scale;motions. Such processes
are poorly represented in 2-D models. The horizontal
scale for STE is small and can only be represented using
3-D models with high resolution. These small scales are
not explicitly resolved in most global 3-D models. Thus,
any use of a 3-D model to evaluate the use of a 2-D mod-
el for these assessments must include a critical evalua-
tion of the 3-D model STE. 2-D models do have the
practical advantage that it is possible to complete many
11.15'
-------
AIRCRAFT EMISSIONS
assessment calculations, using a reasonably complete.
representation of stratospheric chemistry, and also by
considering the sensitivity of the results to model param-
eters one can take some aspects of feedbacks among at-
mospheric processes into account.
Current 3-D models, though impractical for full
chemical assessments, are practical for calculations that
consider the transport of aircraft exhaust, which is treat-
ed as a passive tracer. Such calculations have been
compared directly with 2-D models (Douglass et al.,
1993; Rasch et al., 1993). Their results show that for
seasonal simulations, provided that the residual circula-
tion derived from the 3-D fields is the same as used in the
2-D calculation, the tracer is dispersed faster vertically
and has similar horizontal spread for 3-D compared with
2-D calculations. Although the tracer is also transported
upward more rapidly in 3-D than in 2-D (where vertical
upward transport is minimal), the more rapid downward
transport is the more pronounced effect. Accumulation
of aircraft exhaust in flight corridors is found in regions
of low wind speed, but only a small number of typical
corridors (North Atlantic, North Pacific, and tropical)
have been considered. The effect of such local accumu-
lation would be largest if a threshold chemical process
such as particle formation is triggered at high concentra-
tion of aircraft exhaust constituents. In 2-D models that
use residual mean formulation, transport to the tropo-
sphere takes place principally through two mechanisms:
advective transport by the residual mean circulation
(mostly at middle to high latitudes) and diffusive trans-
port across the tropopause (all latitudes). The latter is
largest where the 2-D model's tropopause height is dis-
continuous (to represent the downward slope' of the
tropopause from the tropics: to middle and polar lati-
tudes) (Shia et al., 1993). The difference in the character
of STE in 2-D and 3-D models leads to different sensitiv-
ities to the latitude at which exhaust is injected in the
models. For the 3-D model, the atmospheric lifetime of a
tracer species is relatively insensitive to the latitude of
injection. For the 2-D model, the tracer species lifetime
is much longer for injection at lower latitudes than at
higher latitudes, since transport to higher latitudes must
take place before most of the pollutant is removed from
the stratosphere.
Treatments of the transport and photochemistry
used in 2-D models have been examined through a series
of model intercomparisons and comparisons with obser-
vations (Jackman et al., 1989b; Prather and Remsberg,
1993). Model results for a "best" simulation, as well as
for various applications and constrained calculations,
were compared with each other and with observations.
There are significant differences in the models that lead
to differences in the model assessments as discussed be-
low. In addition, there are some features, such as the very
low observed values of N2O and CH4 in the upper tropi-
cal stratosphere, and the NOy/O3 ratio at tropical
latitudes, that are not well represented by all 2-D models.
There are also many areas of agreement between
models and observations that suggest that an evaluation
of the effects of the HSCT may be an appropriate use of
these models. For example, the models' total ozone
fields show general consistency when compared with
observed fields such as Total Ozone Mapping Spectro-
meter (TOMS) data, the overall vertical and latitudinal
distributions of such species as N2O, CH4, and HNOs,
and the ozone climatology that is based on Stratospheric
Aerosol and Gas Experiment (SAGE) and Solar Back-
scatter Ultraviolet (SBUV) observations. If the SAGE
results for O3 loss over the past decade at altitudes just
above the tropopause are correct (see Chapter 1), howev-
er, then the inability of present models to reproduce this
O3 decrease (see Chapter 6) casts doubt on their ability
to correctly model aircraft effects in this important re-
gion.
At the beginning of the NASA HSRP/AESA pro-
gram, the assessment models contained only gas phase
photochemical reactions. The importance of the hetero-
geneous reaction (temperature independent) N2O5 +
H2O ->"2 HNO3 on the surface of stratospheric aerosols
was noted by Weisenstein et al. (1991) and Bekki et al.
(1991) and has been further explored by Ramaroson and
Louisnard (1994). This process changes the balance be-
tween the reactive nitrogen species, NO and NO2 (NOX),
and the reservoir species, HNO3. For gas phase evalua-
tions, lower stratospheric ozone was most sensitive to
the amount of NOX from aircraft exhaust injected into
the lower stratosphere. For evaluations including this
heterogeneous process, the NOX levels in both the base
atmosphere and in the perturbed atmosphere are much
lower than in the gas phase evaluations, and the calculat-
ed ozone change is greatly reduced (Ko arid Douglass,
1993).
2-D models have also been used to examine other
processes that are of potential significance. For example.
11.16
-------
AIRCRAFT EMISSIONS
Table 11-4. Calculated percent change in the averaged column content of ozone between 40°N and 50°N.
Scenarios
I: Mach 1.6, NOXEI=5*
II: Mach 1.6, NOX EI=15*
III: Mach 2.4, NOXEI=5*
IV: Mach 2.4, NOX EI=15*
V: Mach 2.4, NOX EI=15**
VI: Mach 2.4, NOX EI=45*
AER
-0.04
-0.02
-0.47
-1.2
-2.0
-5.5
GSFC
-0.11
-0.07
-0.29
-0.86
-1.3
-4.1
LLNL
-0.22
-0.57
-0.58
-2.1
-2.7
-8.3
OSLO
+0.04
+0.15
-0.47
-1.3
-0.42
-3.5
GAMED
+0.69
+0.48
+0.38
-0.45
1-1-1
'-2.8
NCAR
-0.01
-0.60
-0.26
-1.8
-2.3
-6.9
Table 11-5. Calculated percent change in the averaged column content of ozone in the Northern Hemisphere.
Scenarios
I: Mach 1.6, NOX EI=5*
II: Machl.6,NOxEI=15*
IH: Mach 2.4, NOX EI=5*
IV: Mach 2.4, NOXEI=15*
V: Mach 2.4, NOX EI=15**
VI: Mach 2.4, NOX EI=45*
AER
-0.04
-0.02
-0.42
-1.0
-1.7
-4.6
GSFC
-0.12
-0.14
-0.27
-0.80
-1.2
-3.6
LLNL
-0.18
-0.48
-0.50
-1.8
-2.3
-7.0
OSLO
+0.02
+0.10
-0.39
-1.0
-0.43
-3.1
GAMED
+0.63
+0.63
+0.25
-0.26
-0.80
-2.1
NCAR
-0.04
-0.54
-0.25 ,
-1.5
-1.9
-5.1
* Relative to a background atmosphere with chlorine loading of 3.7 ppbv, corresponding to the year 2015
** Relative to a background atmosphere with chlorine loading of 2.0 ppbv, corresponding to the year 2060
if HSCT planes are flown, the lower stratospheric levels
of total odd nitrogen and water vapor are expected to
rise. In addition to a general increase over background
levels throughout the lower stratosphere, there is a possi-
bility for large enhancements in areas of high traffic (air
"corridors"). Peter et al. (1991) and Considine et al.
(1993) have considered the possibility that the increases
in H2O and in HNO3 (a consequence of the heteroge-
neous conversion of NOX) will lead to an increase in the
amount of nitric acid trihydrate (NAT) cloud formation.
They indeed find this to be so.
The evaluation of the effects of a future fleet of
supersonic aircraft on stratospheric ozone was made by
Johnston et al. (1989) and by Ramaroson (1993) using
gas phase models. The ozone loss for an injection at a
fixed level was found to increase nearly linearly as the
amount of NOX injected was increased. The ozone loss
was found to be larger for injection at higher levels be-
cause the ozone response time decreases with altitude,
and because the pollutant has a longer stratospheric life-
time when injected farther from the model tropopause.
Jackman et al. (1989a) used a 2-D model to test
the dependence of the supersonic aircraft assessments on
model dynamical inputs. As anticipated, the calculated
change in ozone is larger (smaller) for a slower (faster)
residual circulation because the circulation controls the
magnitude of the steady-state stratospheric NOX pertur-
bation. This paper also showed that the annual cycle of
the zonally averaged total ozone is sensitive to the annu-
al cycle in the residual circulation. A similar sensitivity
to the residual circulation has been demonstrated for a
3-D calculation using winds from a data assimilation
procedure for transport (Weaver et al., 1993).
The supersonic aircraft assessment scenarios dis-
cussed here are for Mach numbers 1.6 and 2.4, which
correspond to the two aircraft cruise altitudes 16 km and
20 km, respectively, and for three values for EI(NOX)
(see Stolarski and Wesoky [1993b] for specific details).
The emission indices are given in Table 11-1. The calcu-
lated total ozone changes are given for each participating
model in Table 11-4 for the calculated annually averaged
column ozone change in the latitude band where the air-
craft emissions are largest (40°-50°N), and in Table 11-5
for the Northern Hemisphere average. The model calcu-
lations use an aerosol background similar to that
observed in 1979 (e.g., before the Mt. Pinatubo erup-
tion). Some similarities and differences are seen among
the model results. For all of the models, the ozone
change for Mach 2.4 is more negative than that for Mach
1.6. The ozone change at Mach 2.4 is more negative as
the El is increased, but the change is more rapid than a
linear change. The complexity of the assessment is cap-
sulated by the change in ozone calculated at Mach 1.6
for the two different Els in Table 11-4. For all models,
11.17
-------
AIRCRAFT EMISSIONS
60
50
JO
30
20
10
NOy (Scenario IV) - AER
•90 -60
-30 0 30
LATITUDE (DEC)
60
NOy (Scenario IV) - GSFC
-60
-30 0 • 30
LATITUDE (DEG)
60
NOy (Scenario IV) - NCAR
90
90
-60
-30 0 30
LATITUDE (DEG)
60
90
NOy (Scenario IV) - GAMED
•90
-60
-30 0 30
LATITUDE (DEG)
NOy (Scenario IV) - LLNL
-90
-60
-30 0 30
LATITUDE (DEG)
60
50
40
! 30
•
/M
20
10
NOy (Scenario IV) - OSLO
-90 -60
•30 0 30
LATITUDE (DEG)
60
90
Figure 11-6. Calculated changes in the local concentration of NOy (ppbv) in June for Mach 2.4 (EI(NOX)=15)
case. The contour intervals are 1 ppbv, 2 ppbv, 3 ppbv, 4 ppbv, and 5 ppbv (Stoiarski and Wesoky, 1993b).
11.18
-------
AIRCRAFT EMISSIONS
O3 AER (Scenario IV) - June
-90 -60
-30 0 30
LATITUDE (DEG)
O3 GSFC (Scenario IV)
60 90
June
-60
-30 0 30
LATITUDE (DEG)
60
?: •
o : : •.
50 ^
40
I 30
20
10
60 90
O3 NCAR (Scenario IV) - June
-30 0 30
LATITUDE (DEG)
60
90
O3 GAMED^Scenario IV) - June
-30 0 30
LATITUDE (DEG)
O3 LLNL (Scenario IV) - June
-60
•30 ; 0 30
LATITUDE (OEG)
60 90
O3 OSLO (Scenario IV) - June
LATITUDE (DEG)
201 5
ton ef 1993)
1" 'otal °Z0?ne;°r J""eJ°r^ac.h_2-4 (EKNCy-15) fleet in the
' 0'5%'
11.19
%, 2%, 3%, 4% (Albrit-
-------
AIRCRAFT EMISSIONS
and for both cases at C1OX mixing ratios of 3.7 ppbv, the
changes are less than 1%. For three of the models (At-
mospheric and Environmental Research, Inc., AER;
Goddard Space Flight Center, GSFC; and the University
of Oslo, OSLO), the ozone change is less negative (more
positive) for El = 15 than for El = 5. For the other three
models (Lawrence Livermore National Laboratory,
LLNL-, the University of Cambridge and the University
of Edingburgh, CAMED; and the National Center for
Atmospheric Research, NCAR), the 'ozone change is
more negative (less positive) for the larger emission in-
dex.
The assessment initiated by the "Comite Avion-
Ozone" shows similar results. A 2-D model including
heterogeneous reactions on aerosol and PSC surfaces
and a similar emission scenario to that for the HSRP as-
sessments shows a global mean decrease of total ozone
of 0.3% (Ramaroson and Louisnard, 1994). The results
depend upon the prescribed background atmosphere
(<•.Ł., aerosol loading) used (see also: Tie et al., 1994;
Considine era/., 1994).
The change in NOy is given in Figure 11-6 for each
of the models for a scenario in which the HSCT fleet is
assumed to fly at Mach 2.4 with an EI(NOX) = 15 and a
background chlorine mixing ratio of 3.7 ppbv. This NOy
change indicates the sensitivity to the different transport.
LLNL has the largest change in NOy, and also the largest
global ozone changes in Tables 11-4 and 11-5. However,
• the calculated global changes are clearly not ordered by
the magnitude of the NOy change. The latitude height
change in ozone for each of the models is given in Figure
11-7. There are remarkably large differences in the local
ozone changes, particularly in the upper troposphere/
lower stratosphere region where the aircraft emissions
produce an increase in the ozone production as well as
an increase in the ozone loss. Although changes in NOX
have the largest impact on O3, the effects from H2O
emissions contribute to the calculated O3 changes (about
20%).
The assessment models' representation of upper
tropospheric chemistry was not considered as a part of
the Models and Measurements Workshop (Prather and
Remsberg, 1993). Further attention must be paid to the
upper tropospheric chemistry to understand the spread in
the results for these assessments. This subject- is dis-
cussed in the following section on the evaluation of the
impact of the subsonic fleet.
11.5.2 Subsonic Aircraft
The Chapter 7 discussions indicate that tropo-
spheric photochemical-dynamic modeling is much less
developed than is this type of stratospheric modeling;-
however, several types of models have been used to as-
sess the impact of subsonic aircraft emissions. These
include global photochemistry and transport models in
latitude-height dimensions ignoring the longitudinal
variation of emissions. This is an important drawback for
species with short lifetimes. Another type of model used
is the longitude-height model that addresses a restricted
range of latitudes. They neglect the effect of latitudinal
transport. Three-dimensional global dynamical models
are being developed to study the impact of aircraft emis-
' sions, but the results from these models are as yet
restricted to NOX and NOy species. The published results
from two-dimensional models have used a range of esti-
mates to represent present and future aircraft emissions,
and consequently, the results are not easily comparable.
There have been no organized efforts to intercompare
models for subsonic aircraft as there have been for the
supersonic aircraft problem.
.The sensitivity of modeled ozone concentrations
to changes in aircraft NOX emissions has been found to
be much higher than for surface emissions, with around
twenty times more ozone being created per unit NOX
emission for aircraft compared to surface sources
(Johnson et al, 1992). Several authors have investigated
the role of hydrocarbon and carbon monoxide emissions
from aircraft on ozone concentrations, but have found
small effects (Beck et al., 1992; Johnson and Henshaw,
1991; Wuebbles and Kinnison, 1990). The increase in
net ozone production with increasing NOX is steeper at
lower concentrations of NOX (Liu et al., 1987), and
therefore larger ozone sensitivities are expected for
emissions to the Southern Hemisphere, where NOX con-
centrations are lower (Johnson and Henshaw, 1991).
Beck et al. (1992) note the influence of lightning pro-
duction of NOX in controlling the sensitivity of ozone to
aircraft NOX emissions. These studies indicate the im-
portance of predicting a realistic background NOX
concentration, and underline the importance of measure-
ments in model testing.
Several recent publications (Johnson and Hen-
shaw, 1991; Wuebbles and Kinnison, 1990; Fuglestvedt
et al 1993; Beck et al, 1992; Rohrer et al., 1993) esti-
11.20
-------
AIRCRAFT EMISSIONS
mate the percentage increases in ozone concentrations
due to the impact of aircraft emissions. The results show
maximum increases at around 10km of between 12%
and 4% between 30° and 50°N.
NOX concentrations in the upper troposphere are
controlled by the transport of NOX downwards from the
stratosphere, by aircraft and lighting emissions, and by
the convection of NOX from surface sources (Ehhalt et
ai, 1992). The available measurements of NOX in the
free troposphere are discussed in Chapter 5. There are a
number of observations where the vertical NO profile is
strongly and unequivocally influenced by one or the oth-
er of these sources, e.g., lightning (Chameides et al,
1987; Murphy et al, 1993), aircraft emissions (Arnold et
al, 1992), fast vertical transport (Ehhalt et al, 1993),
which makes it clear that all these sources can and do
make a contribution to the NOX in the upper troposphere.
An example is given in Figure 11-8, which presents the
daytime NO distribution across the North Atlantic dur-
ing the period June 4-6,1984, of the Stratospheric Ozone
(STRATOZ III) campaign (Ehhalt et al, 1993). Large
longitudinal gradients of NO mixing ratio up to a factor
of 5 were observed at all altitudes in the free troposphere
in which the effects of an outflow of polluted air from the
European continent are seen. This tongue of high NO
over the Eastern Atlantic was accompanied by elevated
CO and CH4 mixing ratios and therefore was probably
due to surface sources. Figure 11-8 also illustrates the
variance superimposed by longitudinal gradients on av-
erage meridional cross sections. However, at present
there are not enough data to derive the respective global
contributions from atmospheric measurements alone. In-
dependent estimates of the various source strengths are
needed. Our lack of knowledge about the NOX budget in
the troposphere, especially in the upper troposphere,
makes model predictions for this region questionable.
Thus, at present, we can have little confidence in our
ability to correctly model subsonic aircraft effects on the
atmosphere.
Figure 11-9 shows published comparisons of
available NO measurements (Wanner et al, 1994) with
predictions from two-dimensional models (Berntsen and
Isaksen, 1992). Using a quasi-two dimensional longi-
tude-height model and considering estimates of all
important tropospheric sources of NOX (input from the
stratosphere, lightning, fossil fuel combustion, soil emis-
sions and aircraft) for the latitude band of 40°-50°N (see
'W 60'W 50'W 40W 30°W 20"W 10-W 0°
Longitude
Figure 11-8. Daytime NO mixing ratio distribution
(altitude vs. longitude) across the North Atlantic
during the period June 4-6,1984, of the STRATOZ
III campaign. (Based on Ehhalt et al., 1993.)
Figure 11-10), Ehhaltetal. (1992) could reproduce quite
reasonably the measured vertical profiles shown in Fig-
ure 11-9. The transport of polluted air masses from the
planetary boundary layer to the upper troposphere by
fast vertical convection is considered an important pro-
cess for NOX by these authors. However, Kasibhatla
(1993) suggests that the stratospheric source is a more
important source than that arising from rapid vertical
convection, but the calculations did not consider light-
ning, biomass burning, and soil emissions, and the
heterogeneous removal of N;.O5-
Despite considerable differences in model trans-
port characteristics and emission rates, all the studies
suggest that aircraft are important contributors to upper
tropospheric NOX and NOy concentrations. For example
Ehhalt et al (1992) suggest that aircraft emissions (esti-
mated for 1984) contribute around 30% to upper
tropospheric NOX (Figure 1 IrlO). Kasibhatla (1993) es- ,
timates that about 30% of the NOX in the upper
troposphere between 30° and 60°N are from aircraft. It is
clear from the results of Becker al. (1992) and Kasibhat-
la (1993) that despite large latitudinal variations in the
rate of aircraft emissions, the: impacts become manifest
over the entire zonal band, though not evenly. This be-
havior is in contrast to the behavior in the lower
troposphere, and is due to the; slower conversion of NOX
to form HNOs, and the slower removal rates for
which allow for reconversion back to NOX.
11.21
-------
AIRCRAFT EMISSIONS
40°-50°N , 60°W, near Halifax
f 10
111
Q
H '
_J 5
<
_L
J_
_L
STRATOZIII June 1984
(Drummond et al. 1988)
Model calculation June 1984
Ehhaltetal1992
Model calculation August
Bemtsenetal1992
TROPOZII January 1991
Model calculation January 1991
' Wanner et all 993
100 200 300 400
NO MIXING RATIO (ppt)
500
Figure 11-9. Comparisons of measured vertical profiles of NO (June 1984 and January 1991) with calcula-
tions from two-dimensional models. (Based on data from: Wahner ef al., 1994; Bemtsen and Isaksen, 1992;
Drumrnond et al, 1988.)
Several authors discuss the changes to OH concen-
tration consequent to the growth in ozone, and the
consequences to methane destruction. Beck et al. (1992)
predicts OH changes of+10% at around 10 km for the
region 30°-60°N. Similar values are suggested by Fug-
lestvedt and Isaksen (1992) (+20%) and Rohrer et al.
(1993) (+12%). These subsonic aircraft results should be
considered as being preliminary given the complexity of.
the models, the lack of model intercomparison exercises,
as well as the paucity of measurements to test against
model results.
11.6 CLIMATE EFFECTS
Both subsonic and supersonic aircraft emissions
include constituents with the potential to alter the local
and global climate. Species important in this respect in-
clude water vapor, NOX (through its impact on Os),
sulfur, soot, cloud condensation nuclei, and CO2- How-
ever, quantitative assessments of the climate effects of
aircraft operations are difficult to make at this time, giv-
en the uncertainty in the resulting atmospheric
composition changes, as well as uncertainties associated
with the climate effects themselves. Therefore, the fol-
lowing discussion will be on possible mechanisms by
which aircraft operations might affect climate, along
with some estimates of their relative importance.
Increases of CO2 and water vapor, and alterations
of ozone and cirrus clouds have the potential to alter in
situ and global climate by changing the infrared (green-
house) opacity of the atmosphere and solar forcing.
Sulfuric acid, which results-from SOX emissions, may
cool the climate through producing aerosols that give in-
creased scattering of incoming solar radiation, while
soot has both longwave and shortwave radiation im-
pacts. The direct radiative impact for the troposphere as
a whole is largest for concentration changes in the upper
troposphere and lower stratosphere, where the effective-
ness is amplified by the colder radiating temperatures.
However, the impact (including feedbacks) on surface
air temperature may be limited if changes at the tropo-
pause are not effectively transmitted to the surface (see
Chapter 8).
77.22
-------
E 10
.c
D)
65°
55"
45"
35°
25°
AIRCRAFT EMISSIONS
151 5° W '
0 0.1 0.2 0.30 0.1 0.2 0.30 0.1 0.2 0.30 0.1 0.2 0.30 0.1 0.2 0.30 0.1 0.! 0.30 0.1 0.2 0.3
stratosphere
NO/ppb June 1984
aircraft • lightning
surface
45°
35°
25°
5° W
0 0.1 0.2 0.30 0.1 0.2 0.30 0.1 0.2 0.30 0.1 0.2 0.30 0.1 0.2 0.30 0.1 O.J 0.30 0.1 0.2 0.3
NO/ppb January 1991 i
i
Figure 11-10 Calculations of vertical profiles of NO during summer (June, top panel) and winter Uanuarv
nŁ,%qf **"0*™™°^ longitude-height model for the latitude band oMO^N The'
^»9' ~*oe (fossi, fue, combustion and
11.6.1 Ozone
As has been discussed in Chapter 8, the impact of
ozone changes on the radiation balance of the surface-
troposphere system depends on the vertical distribution
of the ozone changes. Reduction in tropospheric and
lower stratospheric ozone tends to cool the climate, by
reducing the atmospheric greenhouse effect. Reduction
in middle and upper stratospheric ozone tends to warm
the climate, by allowing more shortwave radiation to
reach the surface (Lacis et al., 1990).
The preliminary assessments of the HSRP/AES A
program are that supersonic aircraft operations could de-
crease ozone in the lower stratosphere by less than 2
percent for an EI(NOX) of 15, while increasing it in the
upper troposphere by a similar percentage. When these
ozone changes were put into the NASA Goddard Insti-
tute for Space Studies (GISS) 3-D climate/middle
atmosphere model (Rind et al., 1988), the resulting
change in global average surface air temperature was
approximately -0.03°C. The net result is a consequence
of the net effect of varying influences: ozone reduction
in the stratosphere at 20 km, and ozone increases in the
upper troposphere produce jsurface warming,. while'
ozone reduction in the lower stratosphere produces sur-
face cooling. The net result provides the small
temperature changes found in this experiment.
Assuming a local ozone increase (8 to 12 km, 30°
to 50°N) of 4 - 7% due to doubling of the subsonic air-
craft NOX emission and incorporating these changes into
the Wang et al. (1991) model, the inference can be drawn
that a radiative forcing of 0.04 to 0.07 W m-2 will result
(Mohnen et al., 1993; Fortuin et at., 1994). This radia-
tive forcing is of the same order as that resulting from the
aircraft CO2 emissions (see Chapter 8.2.1). The estimat-
ed feedback on radiative i forcing from methane
11.23
-------
AIRCRAFT EMISSIONS
decreases (due to the OH increase from increasing NOX)
has been estimated to be small using two-dimensional
models (Johnson, 1994; Fuglestvedt et al., 1994).
11.6.2 Water Vapor
Water vapor is the primary atmospheric green-
house gas. Increases in water vapor associated with
aircraft emissions have the potential to warm the climate
at low tropospheric levels, while cooling at altitudes of
release, due to greater thermal emission. The effects are
largest when water vapor perturbations occur near the
tropopause (GraBl, 1990; Rind and Lacis, 1993), as is
likely to be the case.
High-speed aircraft may increase stratospheric
water vapor by up to 0.8 ppmv for a corridor at Northern
Hemisphere midlatitudes, with a Northern Hemispheric
effect perhaps 1/4 as large (Albritton et al., 1993). When
changes of this magnitude were used as input to the
stratosphere, the GISS climate/middle atmosphere mod-
el failed to show any appreciable surface warming, as the
radiative effect of the negative feedbacks (primarily
cloud cover changes) were as important as the strato-
spheric water forcing. In general, the stratosphere cooled
by a few tenths of a degree, associated with the increased
thermal emission.
Subsonic tropospheric emissions of water vapor
could possibly result in increases on the order of 0.02
ppmv. Shine and Sinha (1991) estimate that a global in-
crease of 1 ppmv for a 50 mbar slab between 400 and
100 mbar would increase surface air temperature by
0.02°C. Therefore the climate effects from subsonic wa-
ter vapor emission by aircraft seem to be very small.
11.6.3 Sulf uric Acid Aerosols
Subsonic aircraft, flying both in the troposphere
and stratosphere, are presently adding significant
amounts of sulfur to the atmosphere. Hofmann (1991)
has estimated that the current fleet may be contributing
about 65% of the background non-volcanic stratospheric
aerosol amount, whose optical thickness is approximate-
*ly 1 - 2 X 10-3; note however, that this view is a
controversial one as can be seen in Section 3.2.1 of
Chapter 6. This added optical thickness would imply a
contribution to the equilibrium surface air temperature
cooling on the order of 0.03°C due to aircraft sulfur
emissions (Pollack et al., 1993).
11.6.4 Soot
Particles containing elemental carbon are the re-
sult of incomplete combustion of carbonaceous fuel.
Such particles have greater shortwave absorbing charac-
teristics than do sulfuric acid aerosols, and thus a
different shortwave/longwave impact on net radiation.
Upper tropospheric aircraft emissions of soot presently
account for about 0.3% of the background aerosol
(Pueschelefa/., 1992).
The total soot source for the stratosphere is cur-
rently estimated as 0.001 teragrams/year (Stolarski and
Wesoky, 1993b), most likely coming primarily from
commercial air traffic. This accounts for about 0.01% of
the total stratospheric (background) aerosol loading
(Pueschel et al., 1992). It is estimated that the proposed
HSCT fleet would double stratospheric soot concentra-
tions for the hemisphere as a whole, while increases of
up to a factor of ten could occur in flight corridors (Tur-
cor 1992).
11.6.5 Cloud Condensation Nuclei
Contrails in the upper atmosphere act in a manner
somewhat similar to cirrus clouds, with the capability of
warming the climate by increasing longwave energy ab-
sorption in addition to the shortwave cooling effect.
Aircraft sulfur .emissions in addition to frozen droplets
are the most likely contributor to this "indirect" effect of
aerosols, but soot might also be important.
The impact of aircraft particle emissions on upper
tropospheric cloud amounts and optical processes is not
yet known, though it is likely to grow with increased air
traffic. Changes in cloud cover and cloud optical thick-
ness resulting from aircraft operations might be the most
significant aircraft/climate effect, but quantitative evalu-
ations of this are very uncertain. In a 2-D analysis,
increases in cirrus clouds of 5% between 20-70°N pro-
duced a warming of 1°C, due to increased thermal
absorption (Liou etal., 1990). For0.4% additional cloud
coverage by contrails and mid-European conditions, an
increase in surface temperature of about 0.05°C is esti-
mated (Schumann, 1994).
11.6.6 Carbon Dioxide
While aircraft CO2 emissions are at a different al-
titude from other anthropogenic emissions, the climate
11.24
-------
AIRCRAFT EMISSIONS
impact should be qualitatively similar, as CO2 is a rela-
tively well-mixed gas. Therefore the climate impact
from subsonic CC>2 emissions can be estimated to be ap-
proximately 3% of the total anthropogenic CO2 impact,
since subsonic aircraft fuel consumption is about 3% of
the global fossil fuel consumption.
11.7 UNCERTAINTIES
This chapter deals with the atmospheric effects of
both the present subsonic aircraft fleet and an envisioned
future supersonic aircraft fleet. The uncertainties in as-
sessing these two atmospheric effects are of a different
nature. For instance, there is a real uncertainty in the
present emissions data base that results from uncertain-
ties in the aircraft engine characteristics, engine
operations, and air traffic data. There are also uncertain-
ties relating to the models being used to examine the
atmospheric effects of these subsonic emissions. In the
supersonic case, assessments are being made for a hypo-
thetical aircraft fleet, so modeling uncertainties are the
main concern. The modeling uncertainties are probably
much greater than the emission uncertainties at the
present time.
11.7.1 Emissions Uncertainties
As was indicated previously, the evaluation of a
time-dependent emissions data base for use in atmo-
spheric chemical-transport models requires a rather
complete knowledge of the specific emissions produced
by all types of aircraft, as well as a knowledge of the
operations and routing of the aircraft fleet.
There has been very limited aircraft engine testing
under realistic cruise conditions for the present subsonic
aircraft fleet. At the present time, some engine tests are
being carried out under simulated altitude conditions to
see if the present method of determining NOX, for exam-
ple, from a combination of theoretical studies and
laboratory combustor testing can be validated.
A disagreement exists between the quantity of fuel
produced and predicted fuel usage by the data bases.
This discrepancy probably results from uncertainties in
emissions for the non-OECD (Organization for Eco-
nomic Cooperation and Development) countries and for
military traffic, and from the uncertain estimates of load-
ing and power settings of the aircraft fleet.
11.7.2 Modeling Uncertainties
There are two types of modeling uncertainties in
the aircraft assessment process. One is related to model-
ing of small-scale plume processes, while the other
relates to the global atmospheric modeling.
PLUME MODELING j
As was indicated earlier in this chapter, consider-
able modeling is required to characterize the evolution of
the aircraft exhaust leaving the engines' tailpipes to
flight corridor spatial scales and then to the scales that
are treated in the atmospheric models of aircraft effects.
These plume models must treat turbulent dynamics and
both gas phase and heterogeneous chemistry. Only one
such model presently exists that treats the full problem
and there exists no measurement program that is aimed
at the validation of this model (Miake-Lye et al., 1993).
There have been very few actual measurements in air-
plane exhaust wakes. There are the chemical
measurements at altitudes of about 10 km by Arnold et
al. (1992), and there were turbulence and humidity data
taken by Baumann et al. (1993) at the same time. Also,
there are the SPADE (Stratospheric Photochemistry,
Aerosols, and Dynamics Expedition) measurements tak-
en during crossings of the ER-2 exhaust plume (Fancy et
al., 1994). These measurements, while valuable, are not
sufficient to validate the plume processing model.
ATMOSPHERIC MODELING
The upper troposphere and lower stratosphere, the
regions of major interest in this chapter, are particularly
difficult regions to model. In 2-D models of supersonic
aircraft effects, the meridional transport circulation is
difficult to obtain since the radiative heating is com-
prised of a number of small teims of different sign. Thus,
small changes in any radiation term can have important
consequences for transport. Similarly, the time scales for
both transport and chemistry to modify the ozone distri-
bution are generally long and comparable. The complete
problem must be solved. The NOX, HOX, and C1OX
chemical processes are highly coupled in the strato-
sphere. Modeling the chemical balance correctly, in
regions where few measurements are available, presents
formidable difficulties. This situation is even worse in
the upper troposphere than in the stratosphere, given that
11.25
-------
AIRCRAFT EMISSIONS
the chemistry of the upper troposphere is more complex
and there are fewer existing observations of this region:
Supersonic aircraft have their cruising altitudes in
the middle stratosphere (near 20 km) while subsonic air-
craft have cruise altitudes that lie both in the troposphere
and lower stratosphere. Supersonic assessment calcula-
tions have been done using 2-D models up to the present
time, while it is generally appreciated that 3-D models
will be necessary for credible subsonic assessments.
Thus, separate discussions of modeling uncertainties
follow for aircraft perturbations in the stratosphere and
in the troposphere.
TRANSPORT
Two particular problems relating to atmospheric
transport are extremely important for the supersonic air-
craft problem. First, stratosphere-troposphere exchange,
which cannot be modeled in detail with great confidence
in global (2-D or 3-D) -models, is clearly of special sig-
nificance to the chemical distribution in these regions, to
the lifetime of emitted species, etc. More work on this
topic is essential. Second, the present 2-D assessment
models do not model well the details of the polar vortex,
although improvements are anticipated when these mod-
els include the Garcia (1991) parameterization for
breaking planetary waves. If the ideas of the polar vortex
as a "flowing processor" are correct (see Chapter 3), then
the correct modeling of polar vortex dynamics will have
a crucial impact on the distribution of species in the low-
er stratosphere, and present 2-D models would clearly be
performing poorly there. There is also the larger issue
that the uncertainty connected with the use of 2-D mod-
els to assess the inherently 3-D aircraft emission
problem needs to be evaluated further. Even when 3-D
models are available to model this problem, however, the
question will remain as to how well these 3-D models
simulate the actual atmosphere until adequate measure-
ment-model comparisons are done.
For modeling aimed at assessing the atmospheric
effects of both subsonic and supersonic aircraft, it is cru-
cial to properly model ambient NOX distributions in the
upper troposphere, and these, in turn, depend on proper-
ly modeling transport between the boundary layer and
the free troposphere, on proper modeling of the fast up-
ward vertical transport accompanying convection, and
on modeling the lightning source for NOX. Considerable
effort is needed to improve our capability in these areas.
It is also necessary to model stratospheric-tropospheric
transport processes carefully so that NOX fluxes and con-
centrations in the region near the tropopause are
realistic. This requires a substantial effort to improve our
understanding of stratosphere-troposphere exchange
processes.
CHEMICAL CHANGES
The effect of NOX emitted by subsonic aircraft de-
pends on the amount of NOX in the free troposphere. The
ambient NOX concentrations are not very well known,
and depend on several factors such as surface emission
from anthropogenic and natural biogenic sources, the
strength of the lightning source for NOX, and the trans-
port of stratospheric NOX into the troposphere (see
Chapter 2, Table 2-5). The inclusion of wet and dry dep-
osition processes and entrainment in clouds in
assessment models is at a very preliminary stage.
Heterogeneous chemistry is another important
area of uncertainty for models of the troposphere and
lower stratosphere. For example, the hydrolysis of N2O5
is important in both the troposphere and stratosphere, but
the precise rate for this reaction is not known. Observa-
tional studies are needed to elucidate the exact nature
and area of the reactive surfaces. Furthermore, at the
present time, heterogeneous chemistry is being crudely
modeled. Although there do exist models describing the
size distribution and composition of stratospheric aero-
sols, no aircraft assessment model presently exists that
incorporates and calculates aerosol chemistry.
In supersonic assessment models, it is important to
properly model the switch over (at some altitude) from
NOx-induced net ozone production to net ozone destruc-
tion. The precise altitude at which this switch over
occurs differs from model to model, and this can lead to
very different ozone changes in different models of su-
personic aircraft effects. The different responses of the
various models used in the HSCT/AESA assessment of
the impact of changed El (see Tables 11-4 and 11-5, for
example) point to important, unresolved differences in
these models that must be addressed before a satisfac-
tory assessment of the atmospheric effects of supersonic
aircraft can be made with confidence. Also, it is clear
from examining the modeled O3 changes "in Chapter 6
that the model results at altitudes below about 30 km dif-
fer significantly from one another. They also do not give
as large O3 losses as are observed (see Chapter 1). This
77.26
-------
AIRCRAFT EMISSIONS
problem is particularly acute if one accepts the SAGE
results indicating large decreases in ozone concentra-
tions just above the tropopause (see Chapter 1) as being
correct. Then, the fact that present stratospheric models
do not correctly give this effect casts doubt on present
assessment models to correctly simulate that atmospher-
ic region. Since it is in this region where effects from
aircraft operations are particularly significant, there is
the question of how well we can correctly predict atmo-
spheric effects in this altitude region. It may be that the
SAGE ozone trends in this region are in error, or it may
be that important effects in this region are not properly
included in present models.
11.7.3 Climate Uncertainties,
The study of the possible impact of aircraft on cli-
mate is now just beginning. One can make some
preliminary extrapolations based on existing climate re-
search, but one should appreciate that the complexity of
climate research, in general, implies that it will be some
time before great confidence can exist in estimates of air-
craft impacts on climate.
11.7.4 Surprises
Early assessments of the impact of aircraft on the
stratosphere varied enormously with time as understand-
ing slowly improved. Our understanding of the lower
stratosphere/upper troposphere region is still far from
complete and surprises can still be anticipated, which
may either result in greater or lesser aircraft effects on
the atmosphere.
ACRONYMS
AER
AERONOX
AESA
ANCAT
CAMED
CEC
CIAP
ECAC
ECMWF
El
GISS
GSFC
HSCT
HSRP
ICAO
IEA
LLNL
LTO
MOZAIC
NASA
NCAR
NRC
OECD
OSLO
POLINAT
SAGE
SBUV
SPADE
WMO
Atmospheric and Environmental
Research, Inc.
The Impact of lsfOx Emissions from
Aircraft upon the Atmosphere
Atmospheric Effects of Stratospheric
Aircraft
Abatement of Nuisance Caused by Air
Traffic ,
University of Cambridge and University
ofEdingburgh
Commission of the European
Communities
Climatic Impact Assessment Program
European Civil Aviation Conference
European Centre for Medium-Range
Weather Forecasts
Emission Index
NASA Goddard itnstitute for Space
Studies !
NASA Goddard Space Flight Center
High-Speed Civil! Transport
High Speed Research Program
International Civiil Aviation Organization
International Energy Agency
Lawrence Livermiore National Laboratory
Landing/Take-Off cycle
Measurement of Ozone on Airbus
In-service Aircraft
National Aeronautics and Space
Administration
National Center for Atmospheric
Research j
National Research Council
Organization for Economic Cooperation
and Development
University of Oslo
Pollution from Ail-craft Emissions in the
North Atlantic Flight Corridor
Stratospheric Aerosol and Gas
Experiment
Solar Backscatter Ultraviolet spectrometer
Stratospheric Photochemistry, Aerosols,
and Dynamics Expedition
World Meteorological Organization
77.27
-------
AIRCRAFT EMISSIONS
REFERENCES
Aeronautics, Synopsis of Current Aeronautics Projects
1993, Industrial and Materials Technologies Pro-
gramme—Area 3, Aeronautics, CEC Brussels
(ISBN 92-826-6442-2), 1993.
Albritton, D.L., et a/., The Atmospheric Effects of Strato-
spheric Aircraft: Interim Assessment of the NASA -
High-Speed Research Program, NASA Reference
Publication 1333, NASA Office of Space Science
and Applications, Washington, D.C., June 1993.
Appleman, H., The formation of exhaust condensation
trails by jet aircraft, Bull. Amer. MeteoroL Soc., 34,
14-20,1953.
Arnold, R, J. Scheldt, T. Stilp, H. Schlager, and M.E.
Reinhardt, Measurements of jet aircraft emissions
at cruise altitude I: The odd-nitrogen gases NO,
NO2, HNO2 and HNO3, Geophys. Res. Lett., 19,
2421-2424,1992.
Arnold, R, J. Scheidt, T. Stilp, H. Schlager, and M.E.
Reinhardt, Sulphuric acid emission by jet aircraft,
submitted to Nature, 1994a.
Arnold, R, J. Schneider, M. Klemm, J. Scheid, T. Stilp,
H. Schlager, P. Schulte, and M.E. Reinhardt, Mass
spectromctric measurements of SO2 and reactive
nitrogen gases in exhaust plumes of commercial
jet airliners at cruise altitude, in Impact of Emis-
sions from Aircraft and Spacecraft upon the
Atmosphere, Proc. Intern. Sci. Colloquium, Co-
logne, Germany, April 18-20, 1994, edited by U.
Schumann and D. Wurzel, DLR-Mitt. 94-06,323-
328,1994b.
Baumann, R., R. Busen, H.P. Fimpel, C. Kiemle, M.E.
Reinhardt, and M. Quante, Measurements on con-
trails of commercial aircraft, in Proceedings 8th
Symp. MeteoroL Obs. and Instrum., Amer. Met.
, Soc., Boston, 484-489,1993.
Beck, J.P., C.E. Reeves, RA.A.M. de Leeuw, and S.A.
Penkett, The effect of aircraft emissions on tropo-
spheric ozone in the North hemisphere, Atmos.
Environ., 26A, 17-29, 1992.
Bekki, S., and J.A. Pyle, 2-D assessment of the impact of
aircraft sulphur emissions on the stratospheric
sulphate aerosol layer, J. Geophys. Res., 97,
15839-15847,1992.
Bekki, S., and J.A. Pyle, Potential impact of combined
NOX and SOX emissions from future high speed
civil transport on stratospheric aerosols and ozone,
Geophys. Res. Lett., 20, 723-726, 1993.
Bekki, S., R. Toumi, J.A. Pyle, and A.E. Jones, Future
aircraft and global ozone, Nature, 354, 193-194,
1991.
Berntsen, T., and I.S.A. Isaksen, Chemical-dynamical
modelling of the atmosphere with emphasis on the
methane oxidation, Ber. Bunsenges. Phys. Chem.,
96, 241-251,1992.
Boeing, HSCTStudy, Special Factors, NASA Contractor
Report 181881, 1990.
Chameides, W.L., D.D. Davis, J. Bradshaw, M. Rodgers,
S. Sandholm, and D.B. Bai, An estimate of the
NOX production rate in electrified clouds based on
NO observations from the GTE/CITE 1 fall 1993
field campaign, J. Geophys. Res., 92, 2153-2156,
1987.
CIAP Monograph 2, Propulsion Effluents in the Strato-
sphere, Climate Impact Assessment Program,
DOT-TST-75-52, U.S. Department of Transporta-
tion, Washington, D.C., 1975.
Considine, D.B., A.R. Douglass, and C.H. Jackman, Ef-
fects of a parameterization of nitric acid trihydrate
cloud formation on 2D model predictions of
stratospheric ozone depletion due to stratospheric
aircraft, submitted to J. Geophys. Res., 1993.
Considine, D.B., A.R. Douglass, and C.H. Jackman,
Sensitivity of 2D model predictions of ozone re-
sponse to stratospheric aircraft, submitted to J.
Geophys. Res., 1994.
Danilin, M.Y., B.C. Kriiger, and A. Ebel, Short-term at-
mospheric effects of high-altitude aircraft
emissions, Ann. Geophys., 10. 904-911, 1992.
Danilin, M.Y., A. Ebel, H. Elbern, and H. Petry, Evolu-
tion of the concentrations of trace species in an
aircraft plume: Trajectory study, J. Geophys. Res.,
99,18951-18972,1994.
Dentener, F.J., and P.J. Crutzen, Reaction of N2O5 on
tropospheric aerosols: Impact on the global distri-
bution of NOX, O3, and OH, J. Geophys. Res., 98,
7149-7164, 1993.
Douglass, A.R., M.A. Carroll, W.B. Demore, J.R. Hoi-
ton, I.S.A. Isaksen, H.S. Johnston, and M.K.W.
Ko, The atmospheric effects of stratospheric air-
craft: A current, consensus, NASA Refer. Publ.
1251, 1991.
77.25
-------
Douglass, A.R., R.B. Rood, C.J. Weaver, M.C.
Cerniglia, and K.F. Brueske, Implications of three-
dimensional tracer studies for two-dimensional
assessments of the impact of supersonic aircraft on
stratospheric ozone, / Geophys. Res., 98, 8949-
8963, 1993.
Drummond, J.W., D.H. Ehhalt, and A. Volz, Measure-
ments of nitric oxide between 0 and 12 km altitude
and 67°N-60°S latitude obtained during STRA-
TOZin,7. Geophys. Res., 93, 15831-15849,1988.
ECAC, European Civil Aviation Committee, Abatement
of Nuisance Caused by Air Traffic, Databank,
1994.
Egli, R.A., NOX emissions from air traffic, Chimia, 44,
369-371, 1990.
Ehhalt, D.H., F. Rohrer, and A. Wanner, Sources and dis-
tribution of NOX in the upper troposphere at
northern mid-latitudes, /. Geophys. Res., 97,
3725-3738, 1992.
Ehhalt, D.H., F. Rohrer, and A. Wahner, The atmospheric
distribution of NO, O3, CO, and CUt above the
North Atlantic based on the STRATOZ Til flight,
in The Tropospheric Chemistry of Ozone in the
• Polar Regions, NATO ASI Series, Vol. 17, edited
by H. Niki and K.H. Becker, Springer-Verlag, Ber-
lin, Heidelberg, 171-187, 1993.
Fahey, D.W., S.R. Kawa, E.L. Woodbridge, P. Tin, J.C.
Wilson, H.H. Jonsson, J.E. Dye, D. Baumgardner,
S. Borrmann, D.W. Toohey, L.M. Avallone, M.H.
Proffitt, JJ. Margitan, R.J. Salawitch, S.C. Wofsy,
M.K.W. Ko, D.E. Anderson, M.R. Schoeberl, and
K.R. Chan, In situ measurements constraining the
role of nitrogen and sulfate aerosols in mid-lati
tude ozone depletion, Nature, 363, 509-514, 1993.
Fahey, D.W., E.R. Keim, E.L: Woodbridge, R.S. Gao,
K.A. Boering, B.C. Daube, S.C. Wofsy, R.P.
Lohmann, E.J. Hintsa, A.E. Dessler, C.R. Webster,
R.D. May, C. A Brock, J.C. Wilson, R.C. Miake-Lye,
R.C. Brown, J.M. Rodriguez, M. Loewenstein,
M.H. Profitt, R.M. Stimpfle, S. Bowen, and K.R.
Chan, In situ observations in aircraft exhaust
plumes in the lower stratosphere at mid-latitudes,
J. Geophys. Res., in press, 1994.
AIRCRAFT EMISSIONS
Fortuin, J.P.F., R. van Dorliand, W.M.F. Wauben, and H.
Kelder, Greenhouse effects of aircraft emissions as
calculated by a radiative transfer model, submitted
to Ann. Geophys., 1994.
Fuglestvedt, J.S, and I.S.A. Isaksen, in Regional and
Global Air Pollution from Aircraft, edited by F.
Stordal and U. Pedersen, Norsk Institutt for Luft-
forskning (NILU) Report O-1537, 1992.
Fuglestvedt, J.S., T.K. Bemtsen, and I.S.A. Isaksen, Re-
sponses in Tropospheric 03, OH, and CH4 to
Changed Emissions of Important Trace Gases,
Centre for Climate and Energy Research, Oslo,
Report 1993:4, 1993.
Fuglestvedt, J.S., I.S.A. Isaksen, and W.-C. Wang, Direct
and Indirect Global Warming Potentials of Source
Gases, Centre for Climate and Energy Research,
Oslo, Report 1994:1, March, 1994.
Garcia, R.R., Parameterization of planetary wave break-
ing in the middle atmosphere, /. Atmos. Sci., 48,
1405-1419, 1991.
Gayet, J.-F., G. Febvre, G. Brogniez, and P. Wendling,
Microphysical and optical properties of cirrus and
contrails: Cloud field study on 13 October 1989, in
EUCREX-2, Proc. 6th ICE/EUCREX Workshop,
Department of Meteorology, Stockholm Univ.,
Sweden, 24-26 June 1993, edited by H. Sund-
quist, 30-39, 1993.
GraBl, H., Possible climatic effects of contrails and addi-
tional water vapor, in Air Traffic and the
Environment, Lect. Notes in Engineering, 60, edit-
ed by U. Schumann, Springer-Verlag, Berlin,
124-137, 1990. !
Hagen, D.E., M.B. Trueblood, and P.O. Whitefield, A
field sampling of jet exhaust aerosols, Paniculate
Sci. & Techn., 10, 53-63, 1992.
Hofmann, D.J., Aircraft sulphur emissions, Nature, 349,
659, 1991. •,
Hofmann, D.J., and J.M. Rosen, Balloon observations of
a particle layer injected by a stratospheric aircraft
at 23 km, Geophys. Res. Lett., 5, 511-514, 1978.
Hoinka, K.Pt, M.E. Reinhardt, and W. Metz, North At-
lantic air traffic within the lower stratosphere:
Cruising times and corresponding emissions, /.
Geophys. Res., 98, 23113-23131, 1993.
11.29
-------
AIRCRAFT EMISSIONS
Hoshizaki, H., L.B. Anderson, R.J. Conti, N. Farlow,
J.W. Meyer, T. Overcamp, K.O. Redler, and V.
Watson, Aircraft wake microscale phenomena,
Chapter 2 in CIAP Monograph 3, Department of
Transportation, Washington, D.C., DOT-TST-75-
53,1975.
ICAO, International Civil Aviation Organization, Com-
mittee on Aviation Environmental Protection, 2nd
Meeting, Montreal, 1991.
ICAO, International Civil Aviation Organization, Air- .
craft emissions regulations, Annex 16, Vol. II 2nd
Edition, 1993.
ICAO, International Civil Aviation Organization, Air-
craft engine exhaust emissions databank, 1994.
EEA, International Energy Agency, Oil and Gas Informa-
tion (1987-89), OECD, 1990.
Jackman, C.H., A.R. Douglass, P.D. Guthrie, and R.S.
Stolarski, The sensitivity of total ozone and ozone
perturbation scenarios in a two-dimensional mod-
el due to dynamical inputs, J. Geophys. Res., 94,
9873-9887,1989a.
Jackman, C.H., R.K. Seals, and M.J. Prather (eds.), Two-
Dimensional Intercomparison of. Stratospheric
Models, NASA Conference Publication 3042,
Proceedings of a workshop sponsored by the Na-
tional Aeronautics and Space Administration,
Washington, D.C., Upper Atmosphere Theory and
Data Analysis Program held Sept. 11- 16,1988, in
Virginia Beach, VA,1989b.
Johnson, C.E., The Total Global Warming Impact of Step
Changes ofNOx Emissions, ARA/CS/18358017/
003, Harwell Laboratory, U.K., 1994.
Johnson, C.E., and J. Henshaw, The Impact ofNOx Emis-
sions from Tropospheric Aircraft, ARA-EE-0127,
Harwell Laboratory, U.K., 1991.
Johnson, C.E., J. Henshaw, and G. Mclnnes, Impact of
aircraft and surface emissions of nitrogen oxides
on tropospheric ozone and global warming, Na-
ture, 355.69-71,1992.
Johnston, H.S., D. Kinnison, and DJ. Wuebbles, Nitro-
gen oxides from high altitude aircraft: An update
of the potential effect on ozone, J. Geophys. Res.,
94, 16351-16363,1989.
Karol I.L., Y.E. Ozolin, and E.V. Rosanov, Small and
medium scale effects of high flying aircraft on the
atmospheric composition, accepted for publica-
tion in Ann. Geophysicae, 1994.
Kasibhatla, P.S., NOy from subsonic aircraft emissions:
A global three-dimensional model study, Geo-
phys. Res. Lett., 20, 1707-1710, 1993.
Kavanaugh, M., New estimates of NOX from aircraft:
1975-2025, in Proceedings SIst Annual Meeting
ofAPCA (Air Pollution Control Association), Dal-
las, Texas, June 19-24, paper 88-66.10, page 15,
1988.
Knollenberg, R.G., Measurements of the growth of the
ice budget in a persisting contrail, J. Atmos. Sci.,
29, 1367-1374,1972.
Ko, M., and A.R. Douglass, Update of model simula-
tions for the effects of stratospheric aircraft,
Chapter 4 in Atmospheric Effects on Stratospheric
Aircraft: A Third Program Report, edited by R.S.
Stolarski and H.L. Wesoky, 1993.
Lacis, A.A., D.J. Wuebbles, and J.A. Logan, Radiative
forcing of climate by changes in the vertical distri-
bution of ozone, J. Geophys. Res., 95, 9971-9981,
1990.
Liou, K.-N., S.-C. Ou, and G. Koenig, An investigation
of the climatic effect of contrail cirrus, in Air Traffic
and the Environment, Lect. Notes in Engineering,
60, edited by U. Schumann, Springer-Verlag, Ber-
lin, 154-169, 1990.
Liu, S.C., M. Trainer, EC. Fehsenfeld, D.D. Parrish, E.J.
Williams, D.W. Fahey, G. Huebler, and P.C. Mur-
phy, Ozone production in the rural troposphere
and the implications for regional and global ozone
distributions, J. Geophys. Res., 92, 4191-4207,
1987.
McDonnell Douglas Aircraft Co., Procedure for Gener-
ating Global Atmospheric Emissions Data for
Future Supersonic Transport Aircraft, NASA
Contractor Report 181882, 1990.
Mclnnes, G., and C.T. Walker, The Global Distribution
of Aircraft Air Pollution Emissions, Warren Spring
Lab., Report No. LR872 (AP), 1992.
Miake-Lye, R.C., M. Martinez-Sanchez, R.C. Brown,
and C.E. Kolb, Plume and wake dynamics, mix-
ing, and chemistry behind a high speed civil
transport aircraft, /. Aircraft, 30, 467-479, 1993.
Mohnen, V.A., W. Goldstein, and W.-C. Wang, Tropo-
spheric ozone and climate change, J. Air & Waste
Management Assoc., 43, 1332-1344, 1993.
77.30
-------
AIRCRAFT EMISSIONS
Mohler, O., and F. Arnold, Gaseous sulfuric acid and sul-
fur dioxide measurements in the Arctic
troposphere and lower stratosphere: Implications
for hydroxyl radical abundance, Geophys. Res.
Lett., 19, 1763-1766, 1992.
Murphy, D.M., D.W. Fahey, M.H. Proffitt, S.C. Liu,
K.R. Chan, C.S. Eubank, S.R. Kawa, and K.K.
Kelly, Reactive nitrogen and its correlation with
ozone in the lower stratosphere and upper tropo-
sphere,/. Geophys. Res., 98, 8751-8774, 1993.
NRC, National Research Council, Atmospheric Effects
, of Stratospheric Aircraft: An Evaluation of NASA's
Interim Assessment, National Academy Press,
Washington, D.C., 1994.
NiiBer, H.G., and A. Schmitt, The global distribution of
air traffic at high altitudes, related fuel consump-
tion and trends, in Air Traffic and the Environment,
Lect. Notes in Engineering, 60, edited by U. Schu-
mann, Springer-Verlag, Berlin, 1-11, 1990.
Penner, I.E., C.S. Atherton, J. Dignon, S.J. Ghan, and
J.J. Walton, Tropospheric nitrogen: A three-di-
mensional study of sources, distributions, and
deposition,/ Geophys. Res., 96, 959-990, 1991.
Peter, T., C. Briihl, and P.J. Crutzen, Increase in the PSC-
formation probability caused by high-flying
aircraft, Geophys. Res. Lett., 18, 1465-1468, 1991.
Pitchford, M., J.G. Hudson, and J. Hallet, Size and criti-
cal supersaturation for condensation of jet engine
exhaust particles, / Geophys. Res.,. 96, 20787-
20793, 1991.
Pleijel, K., J. Moldanova, and Y. Anderson-Skold, Chem-
ical Modelling of an Aeroplane Exhaust Plume-in
the Free Troposphere, Swedish Environmental Re-
search Institute, IVL Report B. 1105, 1993.
Pollack, J.B., D. Rind, A. Lacis, J. Hansen, M. Sato, and
R. Ruedy, GCM simulations of. volcanic aerosol
forcing I: Climate changes induced by steady state
perturbations,/. Climate, 7, 1719-1742, 1993.
Prather, M.J., et al., The Atmospheric Effects of Strato-
spheric Aircraft: A First Program Report, NASA
Reference Publication 1272, NASA Office of
Space Science and Applications, Washington,
D.C., January 1992.
Prather, M.J., and E.E. Remsberg (eds.), The Atmospher-
ic Effects of Stratospheric Aircraft: Report of the
1992 Models and Measurements Workshop,
NASA reference publication 1292, Vol. I-III,
1993.
Pueschel, R.F., D.F. Blake, K..G. Snetsinger, A.D.A.
Hansen, S. Verma, and K. Kato, Black carbon
(soot) aerosol in the lower stratosphere and upper
troposphere, Geophys. Res. Lett., 19, 1659-1662,
1992. \
Ramaroson, R., Etude des effets potentiels des future
avions supersoniques suf la couche d'ozone, La
Recherche Aerospatials, No. 3, 1993.
Ramaroson, R., and N. Loiiisnard, Potential effects on
ozone of future supersonic aircraft / 2D simula-
tion, Ann. Geophys., in press, 1994.
Rasch, P.J., X.X. Tie, B.A. Boville, and D.L. William-
son, A three-dimensional transport model for the
middle atmosphere, /. Geophys. Res., in press,
1993. •! ' '
Rind, D., and A. Lacis, The role of the stratosphere in
climate change, Surveys in Geophys., 14, 133-165,
1993. :
Rind, D., R. Suozzo, N.K. Balachandran, A. Lacis, and
G.L. Russell, The GISIS global climate/middle at-
mosphere model Part I: Model structure and
climatology, /. Atmos. ScL, 45, 329-370, 1988.
Rodriguez, J.M., R.-L. Shia, M.K.W. Ko, C.W. Heisey,
and O.K. Weisensteini Subsidence of aircraft en-
gine exhaust in the stratosphere: Implications for
calculated ozone depletions, Geophys. Res. Lett.,
21, 69-72, 1994. I
Rohrer, F., D.H. Ehhalt, E.S. Grobler, A.B. Kraus, and A.
Wahner, Model calculations of the impact of nitro-
gen oxides emitted by aircraft in the upper
troposphere at northern mid-latitudes, in Proceed-
ings of the 6th European Symposium on
Physico-chemical Behaviour of Atmospheric Pol-
lutants, Varese, in press, Oct., 1993.
Schumann, U., On the effect of emissions from aircraft
engines on the state of the atmosphere, Ann. Geo-
phys., 12, 365-384, 1994.
Schumann, U., and P. Wendlling, Determination of con-
trails from satellite data and observational results,
in Air Traffic and the Environment, Lect. Notes in
Engineering, Vol. 60, edited by U. Schumann,
Springer-Verlag, Berlin, 138-153, 1990.
11.31
-------
AIRCRAFT EMISSIONS
Schumann, U., and T. Gerz, Turbulent mixing in stably
stratified shear flows, submitted to J. Appl Met.,
1993.
Shia, R.-L., M.K. Ko, M. Zou, and V.R. Kotamarthi,
Cross-tropopause transport of excess >4C in a two
dimensional model, J. Geophys. Res., 98, 18599-
18606,1993.
Shine, K.P., and A. Sinha, Sensitivity of the earth's cli-
mate to height-dependent changes in the water
vapour mixing ratio, Nature, 354, 382-384, 1991.
Solomon, S., and J.G. Keys, Seasonal variations in Ant-
arctic NOX chemistry, J. Geophys. Res. 97,
7971-7978, 1992.
Stolarski, R.S., and H.L. Wesoky (eds.), The Atmospher-
ic Effects of Stratospheric Aircraft: A Second
Program Report, NASA Reference Publication
1293, NASA Office of Space Science and Appli-
cations, Washington, D.C., March 1993a.
Stolarski, R.S., and H.L. Wesoky (eds.), The Atmospher-
ic Effects of Stratospheric Aircraft: A Third
Program Report, NASA Reference Publication
1313, NASA Office of Space Science and Appli-
cations, Washington, D.C., November 1993b.
Tie, X.X., G. Brasseur, X. Lin, P. Friedlingstein, C.
Granier, and P. Rasch, The impact of high altitude
aircraft on the ozone layer in the stratosphere, J.
Atmos. Chem., 18,103-128, 1994.
Turco, R., Report of the Stratospheric Aerosol Science
Committee (SASC), May 19, 1992, in Engine
Trace Constituent Measurements Recommended
for the Assessment of the Atmospheric Effects of
Stratospheric Aircraft, edited by T.C. Miake-Lye,
WJ. Dodds, D.W. Fahey, C.E. Kolb, and S.R.
Langhoff, Engine Exhaust Trace Chemistry Com-
mittee, Atmospheric Effects of Stratospheric
Aircraft, NASA High Speed Research Program,
Report number ARI-RR-947, August 1992.
Wahner, A., F. Rohrer, D.H. Ehhalt, B. Ridley, and E.
Atlas, Global measurements of photochemically
active compounds, in Proceedings of 1st IGAC
Scientific Conference: Global-Atmospheric-Bio-
Spheric Chemistry, Eilat, Israel, in press, 205-222,
1994.
Wang, W.-C, G.-Y. Shi, and J.T. Kiehl, Incorporation of
the thermal radiative effect of CH4, N2O, CF2C12,
and CFC13 into the NCAR Community Climate
Model, J, Geophys. Res., 96, 9097-9103, 1991.
Weaver, C.J., A.R. Douglass, and R.B. Rood, Thermody-
namic balance of three-dimensional stratospheric
winds derived from a data assimilation procedure,
J. Atmos. Scl, 50, 2987-2993, 1993.
Weibrink, G., and R. Zellner, Modelling of plume chem-
istry of high flying aircraft with H2 combustion
engines, in Orbital Transport - Technical Meteoro-
logical and Chemical Aspects, edited by H. Oertel
and H. Korner, Springer-Verlag, Berlin, 439-448,
1993.
Weisenstein, D.K., M.K.W. Ko, J.M. Rodriguez, and
N.-D. Sze, Impact of heterogeneous chemistry on
model-calculated ozone change due to high speed
civil transport aircraft traffic, Geophys. Res. Lett.,
18, 1991-1994,1991.
Wesoky, H.L., A.M. Thompson, and R.S. Stolarski,
NASA Atmospheric Effects of Aviation project:
Status and plans, Proceedings of the International
Scientific Colloquium on the Impact of Emissions
from Aircraft and Spacecraft upon the Atmo-
sphere, Organized by DLR, Cologne, Germany,
April 1994,20-25, 1994.
Whitefield, P.O., M.B. Trueblood, and D.E. Hagen, Size
and hydration characteristics of laboratory stimu-
lated jet engine combustion aerosols. Paniculate
Sci. & Techn., 11, 25ff, 1993.
WMO, Scientific Assessment of Ozone Depletion: 1991,
World Meteorological Organization Global Ozone
Research and Monitoring Project—Report No. 25,
Geneva, 1992.
Wuebbles, D.J., and D.E. Kinnison, Sensitivity of strato-
spheric ozone to present and possible future
aircraft emissions, in Air Traffic and the Environ-
ment, Lect. Notes in Engineering, 60, edited by U.
Schumann, Springer-Verlag, Berlin, 107-123,
1990.
Wuebbles, D.J., S.L. Baughcum, M. Metwally, and R.K.
Seals, Jr., Emissions scenarios development:
Overview and methodology, in The Atmospheric
Effects of Stratospheric Aircraft: A Third Program
Report, edited by R. S. Stolarski and H. L. We-
soky, NASA Reference Publication 1313, NASA
Office of Space Science and Applications, Wash-
ington, D.C., 1993.
11.32
-------
i
CHAPTER 12
]
i
Atmospheric Degradation of
Halocarbon Substitutes
Lead Author:
R.A. Cox
Co-authors:
R. Atkinson
O.K. Moortgat
A.R. Ravishankara
H.W. Sidebottom
Contributors:
G.D. Hayman
C. Howard
M. Kanakidou
S.A. Penkett
J. Rodriguez
S. Solomon
O. Wild
-------
-------
CHAPTER 12 \
ATMOSPHERIC DEGRADATION OF HALOCARBON SUBSTITUTES
Contents
SCIENTIFIC SUMMARY ;. j
12.1 BACKGROUND i
*****"""*"""**""*"*""''"***"•'"""**""•••••*"•»••*••••••••••• J. Ł*J
12.2 ATMOSPHERIC LIFETIMES OF HFCS AND HCFCS j 12 3
12.2.1 Tropospheric Loss Processes ,2,
12.2.2 Stratospheric Loss Processes j " ,2 \
12.3 ATMOSPHERIC LIFETIMES OF OTHER CFC AND HALON SUBSTITUTES j 12 4
12.4 ATMOSPHERIC DEGRADATION OF SUBSTITUTES ! 12 5
J
12.5 GAS PHASE DEGRADATION CHEMISTRY OF SUBSTITUTES : 12 6
12.5.1 Reaction with NO i .-'
12.5.2 Reaction with NO2 ZZZZZZI : 127
12.5.3 Reaction with HO2 Radicals i 127
12.5.4 Hydroperoxides i "_
12.5.5 Haloalkyl Peroxynitrates ! 197
12.5.6 Reactions of Haloalkoxy Radicals j 12g
12.5.7 Halogenated Carbonyl Compounds ! 128
12.5.8 Aldehydes ZZZZZ 1 128
12.5.9 PeroxyacylNitrates \ J2 jQ
12.5.10 Carbonyl Halides Z..ZZZ... 12 10
12.5.11 Acetyl Halides ZZZZZZZZ i ' 1210
12.6 HETEROGENEOUS REMOVAL OF HALOGENATED CARBONYL COMPOUNDS 12.11
12.7 RELEASE OF FLUORINE ATOMS IN THE STRATOSPHERE i . 12 n
12.8 CF30X AND FC(O)OX RADICAL CHEMISTRY IN THE STRATOSPHERE—DO THESE
RADICALS DESTROY OZONE? ? .1213
12.8.1 CF3OX Radical Chemistry ZZZ......Z.Z '' 12 13
12.8.2 FC(O)OX Radical Chemistry ZZZZZZZI'ZZ 12 14
12.9 MODEL CALCULATIONS OF THE ATMOSPHERIC BEHAVIOR OF HCFCS AND HFCS ~ 12 15
12.9.1 The Models 12'15
12.9.2 Transport ofChlorine and Bromine from the Troposphere to the Stratosphere 12.15
12.16
12.16
12.16
12.9.3 Transfer of Cl to the Stratosphere by HCFC Molecules.
12.9.4 Modeling of Ozone Loss Due to CF3O Chemistry
12.9.5 Degradation Products That Have Other Potential Environmental Impacts
REFERENCES 12 {
-------
-------
HALOCARBON SUBSTITUTES
' i '
SCIENTIFIC SUMMARY |
The substitutes for long-lived halocarbons have been selected on the basis of either their susceptibility to oxida-
tion in the lower part of the atmosphere and minimization of their transport to the stratosphere, or by absence of chlorine
orbromine from the molecules. It has been assumed that the atmospheric degradation of the substitutes leads to
DroaUCtS that on nnf rniicp ciT/ino lr\co c.,^u«_ :» : , .., . ., , . . .
1 that the degradation products have no other deleterious
These assumptions are examined in this chapter by assessing three aspects of chlorofluorocarbon (CFC) and
halon substitutes: the factors that control their atmospheric lifetimes, the processes by which they are degraded in the
atmosphere, and the nature of their degradation products. The main findings
are:
If a substance containing Cl , Br, or I decomposes in the stratosphere, it will lead to ozone destruction Use of
hydroch orofluorocarbons (HCFCs) and other CFC substitutes containing Cl, Br, or I, which have short tropo-
sphenc lifetimes, will reduce the input of ozone-destroying substances to the stratosphere, leading to reduced
ozone loss. i
None of the proposed CFC substitutes that are degraded in the troposphere will lead to significant ozone loss in
the stratosphere via their degradation products. :
It is known that atomic fluorine itself is not an efficient catalyst for ozone loss and it is concluded that the
^-containing fragments from the substitutes (such as CF3OX) also do not destroy ozone.
Trifluoroacetic acid, formed in the degradation of certain HCFCs and hydrofluorocarbons (HFCs), will partition
into the aqueous environment where biological, rather than physico-chemical, remoyal processes may be effec-
The amount of long-lived greenhouse gases formed in the atmospheric degradation of HCFCs and HFCs appears
to be insignificant. , FF^"
f
Certain classes of compounds, some of which.have already been released to the atmosphere, such as perfluorocar-
bons, have extremely long atmospheric lifetimes and large global wanning potentials.
12.1
-------
-------
HALOCARBON SUBSTITUTES
ated compounds, such as CF3C1, from photolysis of
CF3C(0)C1,
CF3C(O)C1 + hv -» CF3C1 + CO (12-15)
is sufficiently low (Meller et al, 1993) that the ODP of
the parent compounds will be increased by <0.01.
12.6 HETEROGENEOUS REMOVAL OF
HALOGENATED CARBONYL COMPOUNDS
The carbonyl halides such as C(O)F2, HC(O)F,
and C(O)FCl, and acetyl halides, especially CF3C(O)Cl
and CF3C(O)F, are soluble in water. In aqueous solution
they undergo hydrolysis, forming halogenated carboxyl-
ic acids or hydrogen halides and carbon dioxide. They
are therefore likely to be removed from the troposphere
by heterogeneous processes such as rainout or uptake by
cloud droplets or surface waters, possibly followed by
hydrolysis (Wine and Chameides, 1990).
The rate of these removal processes is governed by
the rate of mass transfer of material between the gas and
the aqueous phase, the solubility in the liquid phase,
which is defined by the Henry's Law constant, H, and the
hydrolysis rate constant, kh.
CF3C(0)X (g)
H
CF3C(0)X (aq)
(12-16)
CF3C(0)X (aq) + H2O -»CF3C(O)OH (aq) + HX (aq)
(12-17)
or
H
(12-18)
C(0)X2(g) <==> COX2(aq)
kh .
COX2 (aq) + H20 -» CO2 (aq) + 2HX (aq) (12-19)
Both H and kh are required to assess their fate.
The Henry's Law constant controls aqueous phase up-
take, and the hydrolysis constant the rate of aqueous
phase destruction. For instance, if Iq, is low, then effi-
cient uptake into cloud droplets might not lead to
destruction because most cloud droplets are transient
and will evaporate on relatively short time scales, vapor-
izing unreacted absorbed carbonyl or haloacetyl halides
back into the atmosphere.
The uptake coefficients, g, reflect a convolution of
all processes at the interface that may influence the ef-
fective rate of mass transfer between gas and aqueous
phases. If the uptake coefficient is greater than -10'3,
the tropospheric uptake rate will not be determined by
the uptake coefficient, and removal time would be -1
week. Values of less than g = 5 x 10'3 have been ob-
tained for C(0)C12, C(O)F2, CC13C(O)C1, CF3C(O)F,
and CF3C(O)C1 by Worsnop et al. (1989), De Bruyn et
al. (1992a), Edney and Driscoll (1993), Ibusuki et al.
(1992), and George etal. (1993), and hence the removal
time of these compounds cam be larger than 1 week. Es-
timated lifetime values are given in Table 12- 2.
Trifluoroacetic acid (TEA), CF3C(O)OH, is the
hydrolysis product of both CF3C(O)F and CF3C(O)C1.
Currently it is believed that CF3C(O)OH, like other or-
ganic acids, is removed from the atmosphere primarily
by rainout (Ball and Wellington, 1993; Rodriguez et al.,
1993). Other processes, such as gas phase reactions with
OH (Carr et al., 1994) or surface photolysis (Meller et
al., 1993), are unlikely to lead to significant reduction in
the amount of CF3C(O)OH rained out. Although the en-
vironmental fate of TEA. cannot be defined yet (Edney et
al., 1992; Franklin, 1993), there are indications that
many natural organisms are capable of degrading it
(Visscherera/., 1994). '!'
The physical removal !of carbonyl compounds in
the troposphere is the key requirement for the eventual
removal of the degradation products from the atmo-
sphere. A comparison of the tropospheric lifetimes of
halogenated carbonyl compounds with respect to loss by
OH radicals, photolysis, and/or physical removal pro-
cesses is shown in Table 12-2.
The data in Table 12-2 indicate that the carbonyl
compounds C(O)F2, C(O)FC1, HC(O)F, CF3C(O)F, and
CF3C(O)OH have long tropbspheric lifetimes with re-
spect to photolysis or OH reaction. Consequently,
physical removal will be the most likely loss process that
.competes with transport into the stratosphere, where the
compounds are slowly photolyzed. The other chlorinat-
ed and brominated compounds will primarily undergo
photolysis in the troposphere. Depending on the loca-
tion, photolysis of CF3C(O)Cl will compete with wet
deposition. i
.}
12.7 RELEASE OF FLUORINE ATOMS !N THE
STRATOSPHERE ,
The atmospheric degradation of HFCs, HCFCs,
and PFCs can lead to the release of F atoms. For exam-
ple, the reaction of CF3O and;FC(O)O with NO leads to
FNO, which because of its strong absorption in the 290-
340 nm region (Johnston and Benin, 1959) will rapidly
photolyze to F atoms. In fact most CFCs also yield F
12.11
-------
HALOCARBON SUBSTITUTES
Table 12-2. Tropospheric lifetimes of halogenated carbonyl compounds.
Carbonyl halides
C(O)C12
C(0)F2
C(O)FC1
Formyl halides
HC(0)F
HC(O)C1
HC(0)Br
Acetyl halides
CF3C(0)F
CF3C(O)C1
CH3C(0)F
CH3C(O)C1
CC13C(O)C1
CC1H2C(O)C1
CC12HC(O)C1
Organic acids
CFtC(O)OH
16 years
> 1 x 10s years
> 1 x 107 years
> 1 x 108 years
3 years
4 days
1700 years
85 days
24 years
23 days
6 days
30 days
9 days
OH (*>)
> 30 years
—
—
> 8 years
> 36 days
no data
—
3 years
—
—
—
4 months
Heterogeneous (c)
< a few weeks
< a few weeks
no data
- 1 month
no data
no data
< a few weeks
< a few weeks
no data
no data
no data
no data
no data
< a few weeks
(a) Absorption cross sections have been measured by Libuda et al, (1991); Meller et al. (1991, 1993);
N611e etui. (1992, 1993); Rattigan et al., (1993). Photolysis processes become important in the lower
troposphere at wavelengths beyond 295 nm. .Unit quantum yields for the dissociation of the molecules
have been assumed for the calculation of the approximate troposphe.ric photolytic lifetimes near the
. boundary layer (2 km).
(b) An average OH concentration of 1 x 106 molecules cm-3 was used for the calculation of the
troposphcric lifetimes with respect to OH loss. Rate constant data are for 298 K, since temperature
dependencies are not available. The rate coefficients for the OH reactions are from Wallington and
Hurley (1993), Nelson et al. (1990), and Libuda et al. (1990). For compounds with no H atom, it can be
assumed that OH loss is negligible.
(c) There are considerable discrepancies in the values of the uptake rate coefficients measured in different
laboratories. Therefore, conservative upper limits for the heterogeneous removal rates are quoted.
(Behnke et al.. 1992; DeBruyn et al., 1992; Exner et al., 1992; Rodriguez et al., 1992; Ugi and Beck,
1961.)
72.72
-------
HALOCARBON SUBSTITUTES
12.1 BACKGROUND
Chlorofluorocarbons (CFCs) and halons deplete
stratospheric ozone because of their long atmospheric
lifetimes, allowing them to be transported to the strato-
sphere where they release chlorine and bromine,
resulting in catalytic destruction of ozone. The substi-
tute molecules have been selected on the basis of either
their shorter tropospheric lifetime due to their suscepti-
bility to oxidation in the lower part of the atmosphere
and minimization of their transport to the stratosphere
or, in some cases, by absence of chlorine or bromine
from the molecules. It has been assumed that the atmo-
spheric degradation of the substitutes leads to products
that have lifetimes shorter than transport times for deliv-
ery of chlorine or bromine to the stratosphere. Further, it
is assumed that the degradation products have no other
deleterious environmental effects. '
The purpose of this chapter of the 1994 WMO/
UNEP assessment is scientific evaluation of the above
assumptions concerning the substitute molecules. The
following lead questions will be addressed:
1) Is significant ozone-destroying halogen released
in the stratosphere from the substitute molecules
themselves?
2) Is significant ozone-destroying halogen transport-
ed into the stratosphere from the degradation
products formed in the troposphere?
3) Are ozone-depleting catalysts other than Cl or Br
released in the stratosphere?
4) Are there any products formed that have other po-
tential environmental impacts?
These questions are answered by examining three
aspects of CFC and halon substitutes: the factors that
control their atmospheric lifetimes, the processes by
which they are degraded in the atmosphere, and the na-
ture and behavior of their degradation products.
The atmospheric lifetime is the critical parameter
required for the calculation of the Ozone Depletion Po-
tential (ODP) and Global Warming Potential (GWP) of
the substitutes, as discussed in Chapter 13 of this docu-
ment. For the most part, these lifetimes are calculated
from models using laboratory data. The accuracy of the
calculated lifetimes, ODPs, and GWPs reflects the un-
certainties in the laboratory data and in the models, i.e.,
the treatment of transport, heterogeneous chemistry, etc.
Here, the hydrofluorocarbons (HFCs) and hydrochloro-
fluorocarbons (HCFCs), the liirgest classes of replace-
ments proposed to date, are treated first. Then, other
replacements', which do not fall into one single category,
are discussed. ,
12.2 ATMOSPHERIC LIFETIMES OF HFCS AND
HCFCS i
The atmospheric lifetimes of all the HFCs and
HCFCs are determined by the sum of their loss rates in
the troposphere and in the stratosphere. The processes
responsible for their losses in these two regions are
slightly different and, hence, are discussed separately.
12.2.1 Tropospheric Loss Processes
The major fraction of the removal of HFCs and
HCFCs occurs in the troposphere. Their reactions with
the hydroxyl (OH) radical have been identified as the
predominant tropospheric loss pathways. Reactions of
HFCs and HCFCs, which are saturated hydrocarbons,
with tropospheric oxidants such as NOj (Haahr et al.,
1991) and 03 (Atkinson and Carter, 1984) are very slow
and, hence, unimportant. Physical removal (i.e., dry and
wet deposition) of these compounds is negligibly slow
(WMO, 1990).
The evaluated rate coefficients for the reaction of
OH with the HFCs and HCFCs considered here are those
recommended by the National Aeronautics and Space
Administration (NASA) Panel (Dejvlore et al., 1992).
This Panel has reviewed the changes' in the data base
since the last evaluation. None; of the changes affects
significantly the calculated lifetimes and ODPs. Refer-
ences have been given where new data have been used.
In addition to their reaction with OH, these mole-
cules may be removed from the iroposphere via reaction
with chlorine atoms (Cl). The rate coefficients for the
reactions of Cl with a variety of HFCs and HCFCs have
been measured and found to be of the same order of
magnitude as the rate coefficients for their OH reactions
(DeMore et al., 1992; Atkinson, et al., 1992; Tuazon et
al., 1992; Wallington and Hurley, 1992; Sawerysyn et
al., 1992; Warren and Ravishankara, 1993; Thompson,
1993). Because the global tropospheric concentration of
Cl is likely to be less than 1% of that of OH, the only
effect of Cl atom reactions is the small reduction of at-
mospheric lifetimes; the products of reaction are similar
12.3
-------
HALOCARBON SUBSTITUTES
to those from OH reactions. The contributions of Cl re-
actions would be at most a few percent of those due to
OH reactions. Loss by Cl atom reaction will only reduce
the lifetimes in the atmosphere and the products of the
reactions are similar to those from the OH reactions.
125.2 Stratospheric Loss Processes
In addition to the reactions of OH free radicals, the
HFCs may be removed from the stratosphere by their re-
action with 0('D) atoms. In the case of HCFCs and
brominated compounds, ultraviolet (UV) photolysis can
also be important. The sum of the rates of these three
processes, i.e., OH reaction, O^D) reaction, and UV
photolysis, determines where and how rapidly these
molecules release ozone-depleting species in the strato-
sphere. In addition, the removal in the stratosphere also
contributes to the overall lifetimes of these compounds.
The rate coefficients for the reaction of O(' D) with
the HFCs and HCFCs have been evaluated by the NASA
and International Union of Pure and Applied Chemistry
(IUPAC) Panels (DeMore et al., 1992; Atkinson et aL,
' 1992). Inclusion of these reactions is unlikely to sub-
stantially reduce the calculated atmospheric lifetimes of
these species. The UV absorption cross sections needed
for these calculations have been reviewed previously
(WMO, 1990; Kaye et al., 1994) and there are no new
data that need to be considered here. In general, HCFCs
must have at least two Cl atoms for photolytic removal in
the stratosphere to be competitive with OH reaction.
The reactions of O('D) are important only for species
with lifetimes longer than a few decades, i.e., for mole-
cules such as HFC-23.
12.3 ATMOSPHERIC LIFETIMES OF OTHER
CFC AND HALON SUBSTITUTES
In addition to the HFCs and HCFCs, many other
substitutes for CFCs have been considered for use and
evaluated for their environmental acceptability. They
include the fluoroethers, perfluorocarbons (PFCs), sul-
fur hexafluoride (SF6), and trifluoromethyl iodide
(CF3I). The PFCs and SF6 are very long-lived species
with strong infrared absorption characteristics. Thus
they can be efficient greenhouse gases. On the other
hand CF3I is very short-lived. Yet, iodine in the strato-
sphere can be even more efficient than bromine in
destroying ozone and hence is of concern.
The rate coefficients for reactions of OH and
O('D) reaction with the fluoroethers have not so far been
reported. The H-containing fluoroethers are expected to
have reactivity with OH comparable to the HFCs, and
therefore their lifetimes will be similar due to tropo-
spheric degradation. The ether functional group does
not make photolysis an important loss process.
The major loss process for the PFCs, other than
CF4 and C2Fe, appears to be their photolysis in the upper
stratosphere and the mesosphere by the Lyman-a (121.6
,nm) radiation (Cicerone, 1979; Ravishankara et al.,
1993). The absorption cross sections at this wavelength,
needed for this evaluation, are given by Ravishankara et
al. (1993). Reaction with O('D) atoms has been shown
to be unimportant as a loss process for the PFCs. In the
case of some PFCs, such as the perfluorocyclobutane,
their reactions with ions in the ionosphere may also con-
tribute (Morris et al, 1994).
CF4 and C2F6 are already present in the atmo-
sphere as by-products of aluminum production. Their
loss processes through reaction with atmospheric ions
are slower than the heavier PFCs, giving lifetimes in ex-
cess of 300,000 years. The ion-molecule reactions are
the only identified loss processes for CF4 and C2F6; how-
ever, their breakdown in air used in combustion could
shorten the lifetimes of CF4 and C2F6 to 50,000 years
and 10,000 years, respectively (Morris et al, 1994).
Another long-lived compound is SFe, for which
the major loss processes appear to be Lyman-a photoly-
sis and electron attachment. Since it is not clear if SFg is
removed in the latter process, the estimated lifetime of
600 years is a lower limit.
CF3I has been considered as a substitute for halons
and CFCs. The major atmospheric loss process for this
molecule, as with all organic-iodine compounds, ap-
pears to be photolysis in the troposphere (Solomon et al,
1994). This process leads to an average atmospheric
lifetime of only a few days. Other loss processes, such
as reaction with OH, are unlikely to compete with the
photolytic removal of CF3I and can only marginally de-
crease the lifetime, even if they are very rapid.
The chemistry of iodine in the troposphere has
been described by Chameides and Davis (1980), Jenkin
et al. (1985), and more recently by Jenkin (1993), who
made use of the expanded kinetic data base that has been
evaluated by the IUPAC panel (Atkinson et al, 1992).
Recently Solomon et al. (1994) have considered the
12.4
-------
HALOCARBON SUBSTITUTES
STRATOSPHERE
Ozone
Loss
(years)
Transport to
Troposphere
TROPOSPHERE
Transport to
Stratosphere
impact of iodine compared to chlorine and bromine on
stratospheric ozone. They show that iodine is likely to
be at least as effective as bromine for ozone destruction,
and they note that several key chemical processes relat-
ing to iodine-catalyzed ozone destruction, notably IO +
CIO, IO + BrO and IO + O3, have not yet been quantified
in laboratory studies. These factors are taken into ac-
count in calculating the OOP for CF3I in Chapter 13.
The data base needed for the calculation of the
lifetimes of halons and their possible bromine-containing
substitutes has been evaluated in the past assessments
(WMO, 1990,1992) and there are no significant changes
in this data base.
12.4 ATMOSPHERIC DEGRADATION OF
SUBSTITUTES
A general flow diagram of the degradation of the
HFCs and HCFCs is shown in Figure 12-1, which shows
the approximate time scales for various processes. A key
question is: Could the degradation products of the sub-
stitutes generate species that can destroy ozbne in the
stratosphere? If long-lived chlorine-containing species
are produced, they can be transported into the strato-
sphere from the troposphere. ;In such a case, the
assumption that degradation in the troposphere essen-
tially stops transport of chlorine or bromine to the
stratosphere would be erroneous. Similarly, if ozone-
destroying radicals other than chlorine are released from
degradation products, erroneous QDPs will result. If
long-lived greenhouse gases are produced, their impact
on climate forcing becomes an.issuie, with potential feed-
back to the ozone depletion problem.
Laboratory studies to elucidate the atmospheric
degradation mechanisms and numerical atmospheric
model calculations have been carried out. The laborato-
ry studies include analysis of the end products in air and
direct measurements of the rate coefficients and prod-
ucts for some of the key reactions. From these studies, it
appears that the slowest step in.the: conversion of HFCs
and HCFCs to their ultimate stabile products (such as
CO2, H2O, HF, HC1, and in some cases, other products
such as trifluoroacetic acid) is the initiation by reaction
with OH. The time scale for this process ranges from
72.5
-------
HALOCARBON SUBSTITUTES
the RO radical lead to the formation of water-soluble end
products. Finally, it has been hypothesized that reactions
of oxygen with CF3O and FC(O)O could potentially lead
to destruction of O3 in the stratosphere.
The "broad-brush" picture of the degradation,
shown in Figure 12-1, will be discussed in detail in the
following sections. This picture shows where in the deg-
radation scheme the above questions arise. Research
carried out during the past few years has addressed these
issues and is discussed below.
12.5 GAS PHASE DEGRADATION CHEMISTRY
OF SUBSTITUTES
In the atmosphere, photolysis or OH radical reac-
tion (H-atom abstraction from a haloalkane, or OH
radical addition to a haloalkene) leads to the formation
of haloalkyl peroxy radicals (WMO, 1990, 1992). The
general degradation scheme, after formation of the halo-
alkyl radical, is shown in Figure 12-2 and is applicable to
both the troposphere and stratosphere, and leads to the
arrows.
weeks, for the shortest-lived substitutes, to hundreds of
years for the long-lived ones. In some cases, such as
with CF4 and C2F6, where the normal degradation pro-
cesses are inoperative, lifetimes are even longer, while
CF3I is removed by photolysis in a time scale of a few
days. The subsequent chemistry that leads to breakdown
is very rapid. However, the formation of shorter-lived,
but.potentially important atmospheric species needs to
be considered. The overall degradation of all the HFCs
and HCFCs and CF3I appears to go through the forma-
tion of the haloalkoxy (RO) radical. There are two
special reasons for the importance of the RO radical for-
mation. It can potentially lead to destruction of ozone in
the stratosphere, via reactions of species such as CF30
and FC(O)0, and in addition RO can lead to the forma-
tion of semi-stable species that are sufficiently
long-lived to be transported into the stratosphere. If
such a species contains an ozone-destroying Cl atom (or
CF3 group), the Ozone Depletion Potentials of the start-
ing HCFCs or HFCs would be larger than that calculated
by ignoring this transport. In addition, the reactions of
72.6
-------
HALOCARBON SUBSTITUTES
formation of the carbonyls C(O)X2, C(O)XY, CX3CHO,
and CX3C(O)Y from the CX3CYZ radical. There are
differences between the degradation of the carbonyls in
the troposphere and stratosphere caused by (a) the im-
portance of physical loss processes of carbonyls in the
troposphere and (b) increased intensity of short-wave-
length UV radiation in the stratosphere, leading to
increased importance of photolysis of carbonyl com-
pounds in the stratosphere.
12.5.1 Reaction with NO
Rate constants for the reactions of a number of
haloalkyl peroxy radicals with NO have been measured
(Wallington and Nielsen, 1991; Peelers and Pultau,
1994; Atkinson etai, 1992; Sehested etal, 1993). The
reactions are expected to produce NO2 and the ha-
loalkoxy radical, RO:
shown that the HO2 radical reaction with CH2F02 pro-
ceeds by two channels at room temperature:
CX3CYZ02 + NO
CX3CYZO + NO2
(12-1)
and, to date, there is no evidence for the formation of the
nitrates via the pathway:
M
CX3CYZ02 + NO -» CX3CYZON02 (12-2)
In any case, photolysis of the haloalkyl nitrates is
expected to occur with a close to unit quantum yield by
breakage of the O-NO2 bond (Atkinson et aL, 1992) to
produce the haloalkoxy radical, RO.
12.5.2 Reaction with NO2
The reactions of haloalkyl peroxy radicals with
NO2 have been evaluated by Atkinson et ai, (1992).
These reactions lead to the formation of peroxynitrates
CX3CYZOONO2.
12.5.3 Reaction with HO2 Radicals
Rate constants for reaction with the HO2 radical
have been measured for CF2C1CH2O2 and CF3CHFO2
radicals (Hayman, 1993), arid a product study has been
conducted for the CH2FO2 radical reaction (Wallington
et ai, 1994a). The two measured rate constants are sim-
ilar to those determined for the methyl and ethyl peroxy
radicals. However, Wallington et al. (1994a) have
CH2F02 + HO2 -> CH2FOOH + O2
(30%) '.
(12-3)
CH2F02 + H02 -> HC(0)F + O2 + H2O (12-4)
(70%)
As shown in Figure 12-2, this second reaction
channel bypasses the intermediate formation of the halo-
alkoxy radical, but forms the same carbonyl product.
12.5.4 Hydroperoxides
As discussed in WMO (1992), the hydroperoxides
CX3CYZOOH are expected to undergo photolysis,
reaction with the OH radical, and (in the. troposphere)
wet deposition. Photolysis leads to formation of the
alkoxy radical CX3CZYO plus OH or possibly to
X + CX2CZYOOH for X = Br and I. OH radical reaction
will lead to reformation of the haloalkyl peroxy
radical CX3CYZO2. For hydroperoxides of the structure
CX3CYZOOH with Z = H, OH reaction also yields
CX3C(0)Y: ;
OH + CX3CHYOOH -> H2O + CX3CYOOH (12-5)
! J,
1 CX3C(0)Y + OH
I (12-6)
i
To date, kinetic and photochemical data are only
available for methyl hydroperoxide and tert-butyl hy-
droperoxide (OH reaction rate constant only) and, based
on these limited data, the haloalkyl hydroperoxides are
expected to have tropospherie lifetimes of a few days
and hence a very low potential for transporting Cl or Br
into the stratosphere. '.
The fate of CX3CYZObH in aqueous solution
needs to be investigated. In particular, any transforma-
tion to yield a long-lived species (for example CX3H or
CX3Z), that is desorbed from isoludon back into the gas
phase may be important. j
12.5.5 Haloalkyl Peroxynitrates
As discussed in WMO (1992) and the IUPAC eval-
uation (Atkinson etaL, 1992), the haloalkyl peroxynitrates
thermally decompose back tol the peroxy radical and
NO2 (Figure 12-2). The thermal decomposition rates of
72.7
-------
HALOCARBON SUBSTITUTES
the peroxynitrates ROON02, where R = CF2C1, CFC12,
CC13, CF2C1CH2, and CFC12CH2, have been measured
(KSppenkastropandZabel, 1991;Kirchnerefa/., 1991).
The lifetimes due to thermal decomposition range from
<1 s for the C2 haloalkyl peroxynitrates and 3-20 s for
the GI haloalkyl peroxynitrates at 298 K, to approxi-
mately 2 days for the C2 haloalkyl peroxynitrates and
0.1-1 year for the C\ haloalkyl peroxynitrates in the up-
per troposphere and lower stratosphere.
By analogy with CH3OONO2 (Atkinson et al.,
1992), the haloalkyl peroxynitrates are also expected to
undergo photolysis in the troposphere, with lifetimes of
a few days, and transport of the haloalkyl peroxynitrates
to the stratosphere will be insignificant.
Hence, apart from those reaction paths noted
above and shown in Figure 12-2, the tropospheric degra-
dation reactions of the HCFCs and HFCs funnel through
the formation of the haloalkoxy radical, and the tropo-
sphcric reactions of the RO radicals then determine
tropospheric degradation products formed from the
HCFCs and HFCs (WMO, 1990,1992).
12.5.6 Reactions of Haloalkoxy Radicals
There are three potential reaction paths for the
haloalkoxy radicals formed from the HCFCs, HFCs and
halons:
C-Cl or C-Br bond cleavage:
CX3CYCIO -» CX3C(O)Y + Cl (Z = Cl) (12-7)
CX3CYBrO-»CX3C(O)Y + Br (Z = Br) (12-8)
C-C bond cleavage:
CX3CYZO -> CX3 + C(0)YZ (12-9)
H-atom abstraction:
CX3CHYO + 02 -» CX3C(0)Y + HO2 (Z = H)
(12-10)
The actual pathway followed and hence the particular
carbonyl product formed depend on the nature of X, Y,
andZ.
12.5.7 Halogenated Carbonyl Compounds
Halogenated carbonyl compounds are produced
from the atmospheric degradation of all halocarbons,
including CFCs, HCFCs, HFCs, halons, and the
halogenated aldehyde intermediates. The carbonyls fall
into the following categories:
Carbonyl halides C(O)X2 (X = ForCl)
Formylhalides HC(O)X (X = F, Cl, or Br)
Acetyl halides CX3C(O)Y (Y = ForCland
X = H,F,Cl,orBr)
Organic acids CX3C(O)OH (X = H, F, Cl, or Br)
Aldehydes CX3C(O)H (X = H, F, Cl, or Br)
The fate of these carbonyl compounds is depen-
dent on whether they are generated in the troposphere or
in the stratosphere. Removal in the stratosphere is large-
ly dominated by photolysis, whereas in the troposphere,
physical removal and hydrolysis processes may be im-
portant relative to photolysis or reaction with the OH
radical. Figure 12-2 and Table 12-1 show a summary of
the products formed from the tropospheric degradation
of HCFCs and HFCs. CF3 radicals are also formed from
several of the HCFCs and HFCs (Table 12-1), and their
atmospheric chemistry is considered below.
12.5.8 Aldehydes
In the troposphere, the aldehydes, CX3CHO, will
react with OH radicals and undergo photolysis. The rate
constants for the reaction with OH radicals have been
determined (Scollard et al., 1993; Atkinson, 1994) and
lead to lifetimes in the troposphere of 4-25 days (Scol-
lard et al., 1993). While the absorption cross sections
have been measured (Libuda et al, 1991; Rattigan et al,
1991, 1993), the photodissociation quantum yields are
not available. Assuming unit quantum yields, the pho-
tolysis lifetimes of the halogenated aldehydes and
CH3CHO are calculated to be 1-7 hours. Thus, the alde-
hydes are likely to have short tropospheric lifetimes, of
the order of a few hours to approximately one month,
depending on the magnitude of the photodissociation
quantum yields.
Assuming a photodissociation quantum yield sig-
nificantly less than unity, similar to that for CHsCHO,
photolysis of the halogenated aldehydes is still expected
to dominate as a tropospheric loss process, leading to
C-C bond cleavage.
CX3CHO + hv -* CX3 + HC(0)
(12-11)
12'.8
-------
HALOCARBON SUBSTITUTES
eri 1o?"HCFrir»nHyiPPr°dl!ptS fofmed/I°m the tropospheric degradation reactions of a
series of HCFCs and MFCs. (Formation of CF3 radicals is also noted.)
HCFC or HFC
methyl chloroform
chloroform
methylene chloride
HCFC-22
HCFC- 123
HCFC- 124
HCFC-141b
HCFC-142b
HCFC-225ca
HCFC-225cb
methyl bromide
HFC-23
HFC-32
HFC- 125
HFC-134
HFC-134a
HFC-143a
HFC-152a
HFC-227ea .
Chemical Formula
CH3CC13
CHC13
CH2C12
CHF2C1
CF3CHC12
CF3CHFC1
CFC12CH3
CF2C1CH3
CF3CF2CHC12
CF2C1CF2CHFC1
CH3Br
CHF3
CH2F2
CHF2CF3
CHF2CHF2
CH2FCF3
CH3CF3
CH3CHF2
CF3CHFCF3
Carbonyl and/or CF^
CC13CHO
C(0)C12
HC(0)C1
C(0)F2
CF3C(O)C1
CF3C(O)F
CFC12CHO
CF2C1CHO
CF3CF2C(0)C1
CF2C1CF2C(O)F
HC(O)Br
CF3
C(O)F2
C(0)F2 + CF3
C(0)F2
CF3C(0)F
HC(O)F + CF,
CF3CHO
CHF2CHO
C(0)F2
CF3C(O)F + CF3
(a)
(b)
(c)
(c)
(a)
From WMO (1990), Fxlney et al. (1991), Sato and Nakamura (1991), Hayman etaL (1991), Jemi-Alade
etal. (1991), Scollard et al. (1991), Edney and Driscoll (1992), Wallington et al. (1992), Nielsen et al
(1992a, b), Tuazon and Atkinson (1993a, b, 1994), Shi etal. (1993), Hayman (1993) Meller et al
(1991, 1993), Zellner et al. (1991, 1993), Rattigan et al. (1994). j
(b) -1% yield of C(O)FC1 also observed at room temperature and atmospheric pressure of air (Tuazon and
Atkinson, 1994); C(O)F2 also presumably formed as co-product with C(O)FC1. ,
(c)
CF3C(O)F and HC(O)F + CF3 yields are a function of temperature and O2 concentration (Wallington
et al., 1992; Tuazon and Atkinson, 1993a; Rattigan et al., 1994). !
72.9
-------
HALOCARBON SUBSTITUTES
The quantum yield for formation of CHF3 from
CF3CHO via
CF3CHO + hv (X>290 nm) -» CHF3 + CO
(12-12)
is too low to significantly enhance the GWP of the parent
compound (Meller et al., 1993).
The OH radical reactions proceed by H-atom ab-
straction to initiate a series of reactions such as shown in
Figure 12-3.
The initially formed acyl radical, CX3CO, has
been shown to either thermally decompose or react with
Oa to form the acyl peroxy radical, CX3C(O)OO
(Barnes et al, 1993; Tuazon and Atkinson, 1994):
CX, + HCO
CX3CO -» CX3 + CO
CX3CO + Oa -» CX3C(0)00
(12-13)
(12-14)
There is a monotonic trend towards decom-
position, at 298 K and atmospheric pressure of air,
with increasing number of Cl atoms in the CC1XF3.XCO
radical (Barnes et al., 1993; Tuazon and Atkins.cn,
1994). Only for CF3CO, CF2C1CO, and CFC12CO
is the Oj addition reaction important under atmo-
spheric conditions. This can lead to the formation
of the peroxyacylnitrates (CF3C(O)OONO2 from
HCFC-143a, CF2C1C(O)OONO2 from HCFC-142b,
and CFC12C(O)OONO2 from HCFC-141b) by adding to
N02. The alternative reaction pathways with NO or
HO2 lead to loss of the acyl group through formation of
RCO2, which decomposes to R + CO2.
12.5.9 Peroxyacyl Nitrates
By analogy with peroxyacetyl nitrate and methyl
peroxynitrate, the thermal decomposition lifetimes of
the halogen-containing peroxyacyl nitrates are expected
to be significantly longer than those for the haloalkyl
peroxynitrates, and this expectation'is borne out by the
data of Barnes et al (1993). Thermal decomposition
rates have been measured by Barnes et al. (1993) for
RC(O)OON02, with R = CF3, CF2C1, and CFC12.
The calculated thermal decomposition lifetimes of
these peroxyacyl nitrates range from approximately 2-3
hours at 298 K (ground level) to 6000-7000 years in the
upper troposphere (220 K). By analogy with peroxy-
acetyl nitrate (Atkinson el al, 1992), photolysis is likely
Figure 12-3. Oxidation of aldehydes formed from
HCFC -and HFC degradation. Stable species are
indicated by boxes; x = F or Cl.
to dominate as the loss process in the upper troposphere,
while still being slow enough that transport to the
stratosphere could be competitive. The potential for
transport of chlorine into the- stratosphere from
CF2C1C(O)OONO2 and CFC12(O)OONO2 is discussed
later.
12.5.10 Carbonyl HalideSj,
Carbonyl halides are produced in the stratosphere
from degradation of all halocarbons, including CFCs.
The photolysis of C(O)FC1 and C(O)F2 is slow in the
lower stratosphere and significant amounts of these deg-
radation products are present there, as shown from
infrared spectroscopic observation from space (Zander
et al., 1992) and from the ground (Reisinger et al,
1994). A fraction of these stratospheric carbonyls is
transported back to the troposphere, where efficient
physical removal takes place; when chlorine is removed
from the stratosphere in this way, e.g., as C(O)FC1 or
C(O)C12, the ODP of the precursor halocarbons can be
reduced because the assumption of complete Cl release
in the stratosphere is not valid.
12.5.11 Acetyl Halides
The acetyl halides released in the stratosphere will
behave similarly to the carbonyl halides, being removed
mainly by photolysis. The available evidence suggests
that the quantum yield .for formation of fully halogen-
72.70
-------
HALOCARBON SUBSTITUTES
atoms upon degradation in the stratosphere. Hence, the
possibility of the involvement of fluorine in catalytic de-
struction of 03 needs to be addressed.
The reaction of F atoms with 03 is much more rap-
id than the corresponding reaction of Cl atoms (DeMore
etal, 1992; Atkinson etal., 1992). Further, the reaction
of FO with O is also rapid, so that the catalytic cycle:
F
FO
O3
O
FO
F
02
02
(12-20)
(12-21)
Net: O
03
202
can occur rapidly. Other catalytic cycles involving F
atoms are also possible. However, the reactions of F at-
oms with CH* and H2O to form HF are also very fast and
can compete with the reaction between F and 03 (De-
More etal.,'1992; Atkinson etal., 1992). Therefore, any
catalytic cycle involving F atoms that destroys ozone
cannot have a large chain length, because F atoms are
efficiently removed to form HF.
Unlike the case of HC1, HBr, and HI, which can
react with various gas phase free radicals to regenerate
the corresponding halogen atoms, HF is inert to attack
by stratospheric free radicals, except for very reactive,
and hence very low abundance, species such as O^D)
atoms. Further, HF does not absorb at wavelengths long-
er than 165 nm and, consequently, is not photolyzed
efficiently in the stratosphere (Safary et al., 1951; Nee et
al., 1985). Lastly, HF cannot be converted to an active
F-containing species via heterogeneous reactions on ice
(Hanson and Ravishankara, 1992) and it is expected to
be very insoluble in sulfuric acid and unable to take part
in heterogeneous reactions. Therefore, release of fluo-
rine into the stratosphere from either CFCs or their
substitutes leads to the formation of stable HF and does
not lead to catalytic ozone destruction.
12.8 CF30X AND FC(O)OX RADICAL
CHEMISTRY IN THE STRATOSPHERE -
DO THESE RADICALS DESTROY OZONE?
12.8.1 CF3OX Radical Chemistry
As shown in Figures 12-2 and 12-3 and discussed
above, the trifluoromethyl radical is a major intermedi-
ate in the atmospheric degradation of HCFCs, HFCs,
and halons that contain the CF3 group. As discussed
previously for other haloalky 1 radicals, it is expected that
the CF3 radical will be quantitatively converted to CF3O,
by addition to O2 followed by reaction with NO. Halo-
methoxy radicals containing hydrogen, bromine, or
chlorine atoms are removed under atmospheric condi-
tions either by halogen atom elimination or by H atom
abstraction with molecular oxygen to give the corre-
sponding carbonyl or formyl species. In contrast, CF3O
does not undergo unimolecular elimination of a fluorine
atom because it is too endothermic, and reaction of
CF3O with O2 is too slow to be important (Bart and
Walsh, 1982, 1983; Schneider and Wallington, 1994;
Turnipseed et al, 1994). Hence, further degradation of
CFjO radicals must occur by reaction with atmospheric
trace gas species. • j
There has been speculation that CF3OX (CF3O and
radicals could participate in catalytic ozone de-
struction cycles in the stratosphere (Francisco et al.,
1987; Biggs etal, 1993). As discussed recently by Ko et
al. (1994), there are a number of potential catalytic
ozone destruction cycles involving CF3OX radicals that
are analogous to the corresponding HOX cycles. In the
lower stratosphere the cycle: ;
CF30 H
CF302 4
net:
- O3 -» CF:3O2 H
- O3 -> CF3O H
2O3-> 3O:>
H 02
h 2O2
(12-22)
(12-23)
could be important, whereas in the mid-stratosphere the
reaction sequence: j
j
CF3O + O3 -» CF3O2 + O2 (12-24)
CF3O2 + O -» CFjO + O2 (12-25)
net: O + O3 -> 2O;>
may also lead to ozone depletion. ^fhe reactions of CF3O
and CF3O2 radicals with ozone are chain-propagating
steps in the cycles, and the efficiencies of the chain pro-
cesses depend on the rate of these reactions relative to
those for the sink reactions of CF30X radicals.
The kinetics of the reaction of CF3O radicals with
ozone have recently been investigated using a number of
different techniques (Biggs et al., 1993; Nielsen and
Sehested, 1993; Wallington et al, 1993b; Maricq and
Szente, 1993; Fockenberg et al., 1994; Ravishankara et
al, 1994; Meller and Moortgat, 1994; O'Reilly et al,
1994; Turnipseed etal, 1994). With the exception of the
12.13
-------
HALOCARBON SUBSTITUTES
data of Biggs et al (1993), the data indicate that
k(CF3O + O3) < 5 x 10'14 cm3 molecule s'1 at 298 K.
For the reaction of CF302 with 03, only upper limits for
the rate constant have been estimated (Nielsen and Se-
hested, 1993; Maricq and Szente, 1993; Fockenberg et
al, 1994; Ravishankara et al., 1994; Meller and Moort-
gat, 1994; O'Reilly et al., 1994) and these studies
suggest k(CF3O2 + O3) < 1 x 10"14 cm3 molecule'1 s'1 at
298 K. The upper limits to the rate constants determined
for the reactions of CF3O and CF3O2 with O3 at 298 K
are similar to the measured rate coefficients for the anal-
ogous reactions of OH and HO2 radicals with O3
(Atkinson et al., 1992; DeMore et al., 1992).
In the stratosphere the main chain terminating pro-
cesses will be the reactions of CF3O with NO and CH}.
The reaction of CF3O radicals with NO over the pressure
range 1-760 Torr and at 298 K leads to stoichiometric
formation of C(O)F2 and FNO (Chen etal, 1992a, 1993;
Bevilacqua et al, 1993; Sehested and Nielsen, 1993):
around 1000-10,000 for the C1OX ozone loss cycle, sug-
gests that catalytic cycles involving CF3OX will be of
negligible importance. The permanency of the sink
mechanism further reduces its effectiveness.
In the troposphere the major fate of CF3O radicals
will be by reaction with hydrocarbons (Chen et al.,
1992b; Saathoff and Zellner, 1993; Kelly et al., 1993;
Sehested and Wallington, 1993; Bevilacqua et al., 1993;
Ravishankara etal, 1994; Bednarek etal, 1994;Barone
et al, 1994), H2O (Wallington et al, 1993a), CO
(Saathoff and Zellner, 1992; Ravishankara, private com-
munication, 1994), and NO (Chen etal, 1992a; Saathoff
and Zellner, 1992; Fockenberg et al, 1993; Sehested and
Nielsen, 1993; Ravishankara et al, 1994). As was the
case in the stratosphere, the ultimate fate of CF3O in the
troposphere is the formation of either CF3OH or CF2O.
Under tropospheric conditions, the most probable fate of
both CF3OH and CF2O is uptake by cloud, rain, or ocean
water to yield CO2 and HF (Franklin, 1993).
CF3O + NO -> C(O)F2 + FNO (12-26) -[2.6.2 FC(O)OX Radical Chemistry
The rate constant for this reaction has been shown
to be independent of both pressure and temperature
(Fockcnberg et al., 1993; Turnipseed et al, 1994).
These results suggest that the reaction of CF3O with NO
provides a permanent sink for CF3O. In contrast, the
sink mechanisms for C1OX and HOX generate only tem-
porary reservoirs for these O3-depleting species. The
reaction of CF3O with CHLj appears to involve a direct
hydrogen abstraction process with an activation energy
of approximately 3 kcal mol'1 (Bednarek et al, 1994;
Barone etal., 1994):
CF3O + CH4 -» CF3OH + CH3 (12-27)
CF3OH will be a temporary reservoir for CF3O
only if subsequent reactions in the stratosphere lead to
regeneration of CF3 or CF3O. The available evidence
indicates that photolysis or reaction with OH will be
negligible under stratospheric conditions (Wallington
and Schneider, 1994) and that circulation back into the
troposphere with loss by precipitation is the likely sink
for CF3OH (Ko et al., 1994). From the kinetic parame-
ters and the stratospheric concentrations of trace gas
species, the chain length of the catalytic cycles for O3
loss by reaction with CF3OX are estimated to be less than
unity. This value, compared with a chain length of
Atmospheric degradation of HCFCs and HFCs
gives rise to formation of HC(O)FCOFC1 and C(O)F2.
In the stratosphere, photolysis of HC(O)F and C(O)F2
may be a minor source of FC(O) radicals. Reaction of
FC(O) with O2 is rapid and leads to formation of
FC(O)O2 (Maricq etal, 1993; Wallington etal., 1994b).
It has been suggested that FC(O)OX radicals could par-
ticipate in a catalytic ozone destruction cycle (Francisco
et al, 1990) similar to that described for CF3OX,
FC(O)O2 -
FC(O)O -
t:
H O3 -
i- O3 -
203 -
•» FC(O)O ^
-» FC(O)O2 H
•* 302
- 2O2
- 02
(12-28)
(12-29)
Wallington et al. (1994b) have recently shown that
FC(O)O2 and FC(O)O both react rapidly with NO,
whereas the rate constant for reaction of FC(O)O with
O3 has an upper limit of 6 x 10'14 cm3 molecule'1 s'1.
Reaction of FC(O)O with NO gives FNO and CO2 and is
hence a permanent sink for FC(O)O. Use of these rate
parameters, together with the concentrations of NO and
O3 in the stratosphere, shows that the contribution to
ozone destruction for cycles involving FC(O)OX radicals
can have no significance.
12.14
-------
HALOCAFtBON SUBSTITUTES
12.9 MODEL CALCULATIONS OF THE
ATMOSPHERIC BEHAVIOR OF HCFCS
AND HFCS
The aim of this section is to review the state of
knowledge of the atmospheric behavior of the CFC sub-
stitutes as determined by calculations using 2- and
3-dimensional numerical models, which are formulated
on the basis of knowledge of atmospheric motions and
solar radiation, and on laboratory data related to atmo-
spheric chemistry. These models have been formulated
using global transport, validated against atmospheric
observations of chemically inert tracers such as CFCs,
85Kr, etc. Chemical schemes have been incorporated to
provide time-dependent fields of oxidizing species such
as OH, which allow the atmospheric loss by photochem-
ical oxidation of reactive substitutes and their oxidation
products to be calculated. This allows the evolving
distribution and concentration levels of a particular sub-
stitute molecule and its degradation products to be
calculated for a given emission scenario. Physical re-
moval in the precipitation and at the Earth's surface has
been incorporated in a parameterized way so that rainout
and hydrolysis of degradation products can be assessed,
and the distribution and fate of the degradation products
determined.
Some models include transport to and from the
stratosphere and allow a detailed treatment of strato-
spheric loss of these substitutes. This allows a treatment
of the delivery of halogen to the stratosphere, either di-
rectly by the halocarbon itself or by its degradation
products. This information has relevance for assessment
of the ODP of the substitutes, but the evaluation of these
comparative indices is dealt with in a later chapter in this
assessment. .It is unlikely that observations of the C^ car-
bonyls, peroxynitrates, or acids expected as degradation
products of HCFCs and HFCs will help validation of the
models, since the abundance of these molecules in the
troposphere will be extremely small; even with the fu-
ture anticipated buildup in the emission rates of the
substitutes, the abundance of these molecules will be too
small to detect with foreseeable technology. Analysis of
the model results allows determination of the atmospher-
ic lifetime of the various chemical species; assessment
of atmospheric lifetimes is dealt with in Chapter 13. In
this chapter the principal focus is the behavior of the
degradation products.
12.9.1 The Models
Three 2-dimensional models—from Harwell
(Hayman and Johnson, 1992), AER (Rodriguez et al.,
1993,1994) and Cambridge (Rattigan etal, 1992)—and
the Max-Planck-Institute 3-D MOGUNTIA model
(Kanakidou et al., 1993) have been employed for the as-
sessment of the atmospheric behavior of the degradation
products of HCFCs and HFCs. There are some differ-
ences in model domain; for iexample, only the AER and
Cambridge models provide full treatment of the strato-
sphere. All models have . detailed schemes for
tropospheric chemistry and degradation schemes for a
range of substitutes are included in all models except for
the AER model, which is restricted to HFC-134a, and
HCFC-123, and -124. The models all use different emis-
sion scenarios, and so calculated concentration fields
cannot be compared directly. However, the conclusions
drawn from analysis of model output can be compared.
Model calculations of the degradation of the pro-
posed CFC substitutes have: been carried out using the
mechanisms and photochemical kinetic data described
in the previous sections. The main questions addressed
by the modeling studies of the degradation of the pro-
posed CFC substitute molecules are:
• To what extent do any long-lived degradation
products of the substitutes transport chlorine and
bromine to the stratosphere, thereby enhancing
ozone depletion? I
• To what extent can the reactions of CF3
-------
HALOCARBON SUBSTITUTES
The effectiveness of the formyl, carbonyl, and
acetyl halides as chlorine and/or bromine carriers is re-
duced essentially to zero by their removal through
hydrolysis and removal in precipitation. The model cal-
culations of Rodriguez et al. (1993), Kanakidpu et al.
(1993), and Rattigan et al. (1992) show that the lifetimes
of these molecules is of the order of a few days, resulting
from removal at the surface, rainout, and loss in clouds.
In the upper tropbsphere the halogenated peroxy-
acetylnitrates CX3C(O)O2NO2 are relatively unreactive.
The oxidation of HCFC-141b and 142b in the tropo-
sphere produces the aldehydes CC12FCHO and
CC1F2CHO, which, following OH attack (in competition
with the photolysis of the aldehydes), may sometimes
form CC12FC(O)O2N02 and CC1F2C(O)O2NO2- Rod-
riguez et al. (1994) and Kanakidou et al. (1993) have
modeled the degradation of HCFC-141b (and 142b) us-
ing a variety of assumptions regarding the rate parameters
for the relevant photochemical reactions. Even when the
assumptions maximized the formation of peroxyacetylni-
trates, the calculated tropospheric concentrations of
CFC12C(O)O2NO2 and CC1F2C(O)O2NO2 were well be-
low the 1X 10~12 (pptv) level and comprised only a small
fraction (-1-2%) of the corresponding concentrations of
HFC-141b and 142b at the steady state. Thus it can be
concluded that transfer of Cl to the stratosphere in these
product molecules is insignificant.
The only other long-lived product containing chlo-
rine is the halocarbon CF3C1, possibly formed by
photolysis of CF3C(O)C1. Model studies of this process
in the atmosphere have not been performed, but the very
low quantum, yields of CF3C1 observed in laboratory
studies imply that it is of negligible importance in con-
veying Cl to the stratosphere.
12.9.3 Transfer of Cl to the Stratosphere by
HCFC Molecules
Although the HCFCs are removed predominantly
in the troposphere, there is some degradation and release
of Cl in the stratosphere by reaction with OH and by
photolysis. For example, Kanakidou et al. (1993) find
that stratospheric loss accounts for 7% for HCFC-22 and
10% for HCFC-14 Ib. Except for CF2HC1 (F22), these
arc the most important potential chlorine carriers; the
other HCFCs are a factor of 3-10 less effective in terms
of the fraction of their chlorine delivered to the strato-
sphere. These factors are taken into account in the ODP
calculations discussed further in Chapter 13.
12.9.4 Modeling of Ozone Loss Due to CF3O
Chemistry
The influence of additional 63 loss mechanisms
involving the CF3O reactions on the Ozone Depletion
Potentials of HCFCs and HFCs has been investigated in
model calculations (Ko et al., 1994; Ravishankara et al.,
1994). -.
In both studies the efficiency of CF3OX as a cata-
lyst for ozone depletion was calculated relative to the
efficiency of chlorine release from CFCs. Ravishankara
et al. (1994J showed that the new kinetics measurements
for the key reactions of CF3O lead to negligibly small
ODPs. For example, the best estimate of the ODP for the
key substitute HFC-134a is only (1-2) x 10'5. The re-
sults of Ko et al. (1994), which were based on estimates
for the relevant kinetic parameters, are consistent with
this conclusion.
12.9.5 Degradation Products That Have Other
Potential Environmental Impacts
Trifluoroacetic acid, formyl, and fluoride formed
from the degradation of HCFCs and HFCs have been
identified as a potential environmental concern because
of their toxicity.
Trifluoroacetic acid (TFA) is produced by hydrol-
ysis of CF3C(O)F formed in the degradation of
HFC-l34a and HCFC-124 and hydrolysis of CF3C(O)C1
from degradation of HCFC-123. The yield of CF3C(O)F
from HCFC-124 is almost 100%, but the competitive
pathway forming HC(O)F reduces the yield from HFC-
134a. Tropospheric photolysis of CF3C(O)C1 competes
with hydrolysis and consequently reduces the yield of
TFA from HFC-123.
Most interest has focused on the production of
TFA from HFC-134a (Rodriguez et al., 1993; Rattigan
et al, 1994; Kanakidou et al., 1993; Ball and Walling-
ton, 1993). Using the most recent laboratory data, cloud
hydrolysis of atmospheric CF3C(O)F is sufficiently rap-
id so that TFA production is equal to the rate of
CF3C(O)F production, and is therefore controlled by the
local rate of HFC-134a reaction with OH and by the
branching ratio for the competing reactions of
CF3CHFO:
72.76
-------
HALOCARBON SUBSTITUTES
CF3CHFO + O2
CF3CHFO
CF3C(O)F +. HO2 (12-30)
CF3 + HCOF (12-31)
Because of the temperature, total pressure, and O2
partial pressure dependence of this branching ratio, there
is significant latitude and altitude dependence in the
fraction of HFC-134a producing CF3C(O)F. For aver-
age atmospheric conditions, about 40% of HFC-134a is
degraded to TEA.
Rodriguez et al. (1993) have calculated the zonal-
ly averaged concentrations of TFA in rainwater, making
various assumptions regarding the extent to which the
gaseous acid is dry deposited at the surface after evapo-
ration from clouds. The results show considerable
latitudinal and seasonal variation in rainfall TFA, the
pattern depending on the assumptions made. The key
results of this study are:
Predicted global average concentrations of TFA in
rain are of the order of 1 mg/1 for a 1 Tg year"1
source of HFC-134a in the Northern Hemisphere.
These concentrations are relatively insensitive to
the parameters adopted for uptake of CF3C(O)F in
cloud droplets.
• The concentrations of TFA in rain are primarily
determined by the source strength of HFC-134a,
the relative yields of CF3C(O)F from the
CF3CHFO radical, and the loss processes for gas
phase TFA.
• Calculated local concentrations of TEA in rain
could be very sensitive to other loss processes of
CF3C(Q)F, as well as to rainfall patterns.
Calculations in the same study indicate a 50-100%.
yield of TFA in rain from degradation of HCFC-124 and
HCFC-123. The smaller values for HCFC-123 reflect the
removal of CF3C(O)C1 by photolysis in the troposphere.
The results from the other model studies of HFC- 134a ox-
idation are in broad agreement with these conclusions
concerning the formation of TFA. There are differences in
quantitative detail that may be a result of different model
formulation as well as uncertainties in the input data.
No laboratory data are available for the uptake and
hydrolysis rates of HC(O)F in aqueous solution. Its gas
phase loss processes are extremely slow in the troposphere
and, if the hydrolysis and uptake rates are also low, this
molecule could build up in the troposphere and be trans-
ported to the stratosphere (Kanakidou et al., 1993).
Stratospheric photolysis leads to FC(O)OX but, as dis-
cussed above, this will not lead to ozone depletion.
REFERENCES
Atkinson, R., and W.P.L. Carter, Kinetics and mecha-
nisms of the gas-phasie reactions of ozone with
organic compounds under atmospheric conditions,
Chem. Rev., 84,437-470, 1984.
Atkinson, R., D.L. Baulch, R.A. Cox, R.F. Hampson, Jr.,
J.A. Kerr,.and J. Troe, Evaluated kinetic and pho-
tochemical data for atmospheric chemistry,
Supplement IV, IUPAC subcommittee on gas ki-
netic data evaluation for atmospheric chemistry, J.
Phys. Chem. Ref. Data, 21, 1125-1568, 1992.
Atkinson, R., Gas-phase tropospheric chemistry of or-
ganic compounds, /. Phys. Chem. Ref. Data,
Monograph 2, 1-216, 1994.
Ball, J.C., and T.J. Wallington, Formation of trifluoro-
acetic acid from the atmospheric degradation of
hydrofluorocarbon 134a: A human health con-
cern?, J. Air Waste Manage. Assoc., 43, 1260-1262,
1993.
Barnes, I., K.H. Becker, F. Kirchner, F. Zabel, H. Richer,
and J. Sodeau, Formation and thermal decomposi-
tion of CC13C(0)02N02 and CF3C(O)O2NO2, in
Kinetics and Mechanisms for the Reactions of
Halogenated Organic Compounds in the Tropo-
sphere, STEP-HALOCSIDE/AFEAS Workshop,
March 23-25, Dublin, 1993.
Barone, S., A.A. Turnipseed, and A.R. Ravishankara,
Kinetics of the reactions of the CF3O radical with
alkanes, J. Phys. Chem., in press, 1994.
Batt, L., and R. Walsh, A reex.amination of the pyrolysis
of bis trifluromethyl peroxide, Int. J. Chem. Ki-
net., 14, 933-944 (1982) and 75, 605-607, (1983).
Bednarek, G., J.P. Kohlmann.J H. Saathoff, and R. Zell-
ner, Temperature dependence and product
distribution for the reaction of CF3O radicals with
methane, /. Phys. Chem., in press, 1994.
Behnke, W., M. Elend, H.-U. ICriiger, and C. Zetzsch, An
aerosol chamber technique for the determination
of mass accommodation coefficients on submi-
cron aqueous solution droplets of carbonyl
halides: Phosgene, in Atmospheric Wet and Dry-
Deposition of Carbonyl and Haloacetyl Halides,
Alternative Fluorocarbons Environmental Accept-
ability Study (AFEAS) Workshop Proceedings, 22
September 1992, Brussels, 68-75, 1992.
72.77
-------
HALOCARBON SUBSTITUTES
Bevilacqua, T.J., D.R. Hanson, and C.J. Howard, Chem-
ical ionization mass spectrometric studies of the
gas-phase reactions CF3O2 + NO, CF3O + NO and
CF30 + RH,J. Phys. Chem., 97,3750-3757,1993.
Biggs, P., C.E. Canosa-Mas, D.E. Shallcross, R.P.
Wayne, C. Kelly, and H.W. Sidebottom, The possi-
ble atmospheric importance of the reaction of
CF3O radicals with ozone, in Kinetics and Mecha-
nisms for the Reactions of Halogenated Organic
Compounds in the Troposphere, STEP-HALOC-
SIDE/AFEAS Workshop, March 23-25, Dublin,
177-181,1993.
Carr, S., J.J. Treacy, H.W. Sidebottom, R.K. Connel,
C.E. Canosa-Mas, R.P. Wayne, and J. Franklin,
Kinetics and mechanism for the reaction of hy-
droxyl radicals with TFA under atmospheric
conditions, Chem. Phys. Lett., in press, 1994.
' Chameides, W.L., and D.D. Davis, Iodine: Its possible
role in tropospheric photochemistry, /. Geophys.
Res., 85,7383-7398,1980.
Chen, J., T. Zhu, and H. Niki, FTIR product study of the
reaction of CF3O with NO: Evidence for CF3O +
NO -» CF2O + FNO, J. Phys. Chem., 96, 6115-
6117,1992a.
Chen, J., T. Zhu, H. Niki, and GJ. Mains, Long path
FTIR spectroscopic study of the reactions of CF3O
radicals with ethane and propane, Geophys. Res.
Lett., 19, 2215-2218, 1992b.
Chen, J., V. Young, T. Zhu, and H. Niki. Long path FTIR
spectroscopic study of the reactions of CF3OO and
CF3O radicals with NO2, in Kinetics and Mecha-
nisms for the Reactions of Halogenated Organic
Compounds in the Troposphere, STEP-HALOC-
SIDE/AFEAS Workshop, March 23-25, Dublin,
150-162,1993.
Cicerone, R.J., Atmospheric carbon tetrafluoride: A
nearly inert gas, Science, 206,59-61, 1979.
DeMore, W.B., S.P. Sander, D.M. Golden, R.F. Hamp-
son, M J. Kurylo, C J. Howard, A.R. Ravishankara,
C.E. Kolb, and M. Molina, Chemical Kinetics and
Photochemical Data for Use in Stratospheric
Modeling, NASA/Jet Propulsion Laboratory Pub.
No. 92-20,1992.
De Bruyn, W.J., S.X. Duan, X.Q. Shi, P. Davidovits,
D.R. Worsnop, M.S. Zahnizer, and C.E. Kolb, Tro-
pospheric heterogeneous chemistry of carbonyl
halides, Geophys. Res. Lett., 19, 1939-1942,
1992a.
De Bruyn, W.J., J.A. Shorter, P. Davidovits, D.R.
Worsnop, M.S. Zahnizer, and C.E. Kolb, The het-
erogeneous chemistry of carbonyl and haloacetyl
halides, in Atmospheric Wet and Dry Deposition of
Carbonyl and Haloacetyl Halides, Alternative
Fluorocarbons Environmental Acceptability Study
(AFEAS) Workshop Proceedings, 22 September
1992, Brussels, 12-17, 1992b.
Edney, E.O., and D.J. Driscoll, Chlorine initiated photo-
oxidation studies of hydrochlorofluorocarbons
(HCFCs) and hydrofluorocarbons (HFCs): Re-
sults for HCFC-22 (CHC1F.2); HFC-41 (CH3F);
HCFC-124 (CC1FHCF3); HFC-125 (CF3CHF2);
HFC-134a (CF3CH2F); HCFC-142b (CC1F2CH3);
and HFC-152a (CHF2CH3), Int. J. Chem. Kinet.,
24, 1067-1081, 1992.
Edney, E.G., and D.J. Driscoll, Laboratory investiga-
tions of the deposition of oxidation products of
hydrochlorofiuorocarbons (HCFCs) and hydro-
fluorocarbons (HFCs) to aqueous solutions, Water,
Air and Soil Poll., 66, 97-110, 1993.
Edney, E.G., D.J. Driscoll, E.W. Corse, and F.T. Blan-
chard, Atmospheric fate of trifluoroacetate during
evaporation of complex aqueous solutions, in At-
mospheric Wet and Dry Deposition of Carbonyl
and Haloacetyl Halides, Alternative Fluorocar-
bons Environmental Acceptability Study
(AFEAS) Workshop Proceedings, 22 September
1992, Brussels, 86-90, 1992.
Edney, E.O., B.W. Gay, Jr., and D.J. Driscoll, Chlorine
initiated oxidation studies of hydrochlorofluoro-
carbons: Results for HCFC-123 (CF3CHC12) and
HCFC-141b (CFC12CH3), J. Atmos. Chem., 12,
105-120, 1991.
Exner, M., H. Herrmann, J. Michel, and R. Zellner,
Hydrolysis rate of CF3COF and the decarboxyla-
tion of CF3COO" anions, in Atmospheric Wet and
Dry Deposition of Carbonyl and Haloacetyl Ha-
lides, Alternative Fluorocarbons Environmental
Acceptability Study (AFEAS) Workshop Proceed-
ings, 22 September 1992, Brussels, 50-56, 1992.
12.18
-------
Fockenberg, Ch., H. Saathoff, and R. Zellner, Paper pre-
sented at NOAA/NASA/AFEAS Workshop,
Boulder, Colorado, November 16-19, 1993.
Fockenberg, Ch., H. Saathoff, and R. Zellner, A laser
photolysis/LIF study of the rate constant for the
reaction CF3O + O3 -» products, Chem. Phys.
Lett., in press, 1994.
Francisco, J.S., Z. Li, and I.H. Williams, Dissociation
dynamics .of the trifluoromethoxy radical, Chem.
Phys. Lett., 140, 531-536, 1987.
Francisco, J.S., A.N. Goldstein, Z. Li, Y. Zhao, and I.H.
Williams, Theoretical investigation of chlorofluo-
rocarbon degradation processes: Structures and
energetics of XC(O)OX intermediates (X = F, Cl),
J. Phys. Chem., 94,4791-4795, 1990.
Franklin, J., The atmospheric degradation and impact
of 1,1,1,2-tetrafluoroethane (hydrofluorocarbon
134a), Chemosphere, 27, 1565-1601, 1993.
George, G, J.L. Ponche, and P. Mirabel, Experimental
determination of uptake coefficient for acid ha-
Hdes, in Kinetics and Mechanisms for the
Reactions of Halogenated Organic Compounds in
the Troposphere, STEP-HALOCSIDE/AFEAS
Workshop, March 23-25, Dublin, 196-202, 1993.
Haahr, N., J. Hjorth, and G. Ottobrini, An FTIR study of
the gas-phase reactions between NO3 radicals and
some HCFs and HCFCs and of the degradation of
halogenated PAN-type species in air, in Kinetics
and Mechanisms for the Reactions of Halogenated
Organic Compounds in the Troposphere, STEP-
HALOCSIDE/AFEAS Workshop, May 14-16,
Dublin, 27-30, 1991.
Hanson, D.R., and A.R. Ravishankara, The heteroge-
neous chemistry of HBr and HF, J. Phys.Chem
96, 9441-9446, 1992.
Hayman, G.D., M.E. Jenkin, T.P. Murrels, and S.J. Shal-
• liker, Kinetics and mechanistic studies associated
with the atmospheric degradation of HCFC-123,
in Kinetics and Mechanisms for the Reactions of
Halogenated Organic Compounds in the Tropo-
sphere, STEP-HALOCSIDE/AFEAS Workshop,
May 14-16, Dublin, 79-87, 1991.
HALOCARBON SUBSTITUTES
Hayman, G.D., and C.E. Johnson, Tropospheric model-
ing studies related to the degradation of the
replacement compounds, in Atmospheric Wet and
Dry Deposition of Carbonyl and Haloacetyl
Halides, AFEAS V/orkshop Proceedings, 22 Sep-
tember 1992, Brussels, 76-85, 1992.
Hayman, G.D., The kinetics and mechanisms of process-
es involved in the atmospheric degradation of
hydrochlorofluorocarbons and hydrofluorocar-
bons, in Kinetics and Mechanisms for the
Reactions of Halogenated Organic Compounds in
the Troposphere, STEP-HALOCSIDE/AFEAS
Workshop, March 23-25, Dublin, 1993.
Ibusuki, T, K. Takeuchi, and S. Kutsuna, Determination
of mass transfer coefficients of carbonyl and halo-
carbonyl halides' into water, in Atmospheric Wet
and Dry Deposition of Carbonyl and Haloacetyl
Halides, AFEAS Workshop Proceedings, 22 Sep-
tember 1992, Brussels, 62-67, 1992.
Jemi-Alade, A.A., P.D. Lightfoot, and R. Lssclaux, UV
absorption spectra of peroxy radical derivatives of
hydrohalocarbons, in Kinetics and Mechanisms
for the Reactions of Halogenated Organic Com-
pounds in the Troposphere, STEP-HALOCSIDE/
AFEAS Workshop, May 14-16, Dublin 67-72
1991.
Jenkin, M.E., R.A. Cox, and D.E. Candeland, Photo-
chemical aspects of tropospheric iodine behaviour,
J.Atm. Chem., 2, 359-375, 1985.
Jenkin, M. E., A comparative assessment of the role of
iodine photochemistry in tropospheric ozone de-
pletion, in The Tropospheric Chemistry of Ozone
in the Polar Regions > edited by H. Niki and K.-H.
Becker, NATO ASI Series, Vol 17, 405-416, 1993.
Johnston, H.S., and H.J. Bertiri, Absorption and emis-
sion spectra of nitrosyl fluoride, J. Mol.
Spectrosc., 3, 683-696, 1959.
Kanakidou, M., F.J. Dentener, and P.J. Crutzen, A global
3-dimensional study of the degradation of HCFCs
and HFC-134a in the troposphere, in Kinetics and
Mechanisms'for the Reactions of Halogenated
Organic Compounds \in the Troposphere, STEP-
HALOCSIDE/AFEAS Workshop, March 23-25,
Dublin, 113-130, 1993.
12.19
-------
HALOCARBON SUBSTITUTES
Kaye, J., S.A. Penkett, and EM. Ormond (eds.), Report
on Concentrations, Lifetimes and Trends ofCFCs,
Halons and Related Species, NASA Reference
Publication No. 1339, 1994.
Kelly, C., J.Treacy, H.W. Sidebottom, and O.J. Nielsen,
Chem. Phys. Lett., 207, 498-502, 1993.
Kirchner, R, F. Zabel, and K.H. Becker, Thermal stabili-
ty of CC1F2CH2O2NO2, CC12FCH2O2NO2 and
CC1F2C(O)O2NO2, in Kinetics and Mechanisms
for the Reactions of Halogenated Organic Com-
pounds in the Troposphere, STEP-HALOCSIDE/
AFEAS Workshop, May 14-16, Dublin, 73-78,
1991.
Ko, M.K.W., N.-D. Sze, J.M. Rodriguez, O.K. Weisen-
stein. C.W. Heisey, R.P. Wayne, P. Biggs, C.E.
Canosa-Mas, H.W. Sidebottom, and J. Treacy, CF3
chemistry: Potential implications for stratospheric
ozone, Geophys. Res. Lett., 21, 101-104, 1994.
KOppcnkastrop, D., and F. Zabel, Thermal decomposi-
tion of chlorofluoromethyl peroxynitrates, Int. J.
Chem. Kinet., 23, 1-15, 1991.
Libuda, H.G., F. Zabel, E.H. Fink, and K.H. Becker,
Formyl chloride: UV absorption cross sections
and rate constants for the reactions with Cl and
OH, /. Phys. Chem., 94,5860-5865, 1990.
Libuda, H.G., F. Zabel, and K.H. Becker, UV spectra of
some organic chlorine and bromine compounds of
atmospheric interest, in Kinetics and Mechanisms
for the Reactions of Halogenated Organic Com-
pounds in the Troposphere, STEP-HALOCSIDE/
AFEAS Workshop, May 14-16, Dublin, 126-131,
1991.
Maricq, M.M., and J J. Szente, Upper limits for the rate
constants of the reactions CF3O + 03 -> CF302 +
O2 and CF3O2 •*- O3 -» CF3O + 2O2, Chem. Phys.
Lett., 213,449-456, 1993.
Maricq, M.M., J J. Szente, G.A. Khitrov, and J.S. Fran-
cisco, Temperature dependent kinetics of the
formation and self reactions of FC(O)O2 and
FC(O)O radicals, J. Chem. Phys., 98, 9522-9531,
1993.
Meller, R., D. Boglu, and O.K. Moortgat, UV spectra of
several halogenated carbonyl compounds and
FTIR studies on the degradation of CF3COC1,
HCFC-123 and HFC-134a, in Kinetics and Mech-
anisms for the Reactions of Halogenated Organic
Compounds in the Troposphere, STEP-HALOC-
SIDE/AFEAS Workshop, May 14-16, Dublin,
110-115, 1991.
Meller, R., D. Boglu, and O.K. Moortgat, Absorption
cross-sections and photolysis studies of halogenat-
ed carbonyl compounds. Photo-oxidation studies
on CF3-containing CFC-substitutes, in Kinetics
and Mechanisms for the Reactions of Halogenated
Organic Compounds in the Troposphere, STEP-
HALOCSIDE/AFEAS Workshop, March 23-25,
Dublin, 1993.
Meller,.R., and O.K. Moortgat, Photolysis of CF3O2CF3
in the presence of O3 in oxygen: Kinetic study of
the reaction of CF3O and CF3O2 radicals with O3>
submitted to J. Photochem. Photobiol, 1994.
Morris, R.A., T.M. Miller, A.A. Viggiano, J.F. Paulson,
S. Solomon, and G. Reid, Effect of electron and
ion reactions on atmospheric lifetimes of fully
fluorinated compounds, /. Geophys. Res., in press,
1994.
Nee, J.B., M. Suto, and L.C. Lee, Photoabsorption cross
section of HF at 107-145 nm, /. Phys. B: Mol.
Phys., 18, L293-L294, 1985.
Nelson, L., I. Shanahan, H.W. Sidebottom, J. Treacy, and
O.J. Nielsen, Kinetics and mechanism for the oxi-
dation of 1,1,1-trichloroethane, Int. J. Chem.
Kinet., 22, 577-590, 1990.
Nielsen, O.J., and J. Sehested, Upper limits for the rate
constants of the reactions of CF3O2 and CF3O rad-
icals with ozone at 295 K, Chem. Phys. Lett., 213,
433-441, 1993.
Nielsen, O.J., T. Ellermann, E. Bartkiewicz, T.J. Wall-
ington, and M.D. Hurley, UV absorption spectra,
kinetics and mechanisms of the self-reaction of
CHF2O2 radicals in the gas phase at 298 K, Chem.
Phys. Lett., 192, 82-88, 1992a.
Nielsen, O.J., T. Ellermann, J. Sehested, and T.J. Wall-
ington, Ultraviolet absorption spectrum and
kinetics and mechanism of the self-reaction of
CHF2CF2O2 radicals in the gas phase at 298 K, J.
Phys. Chem., 96, 10875-10879, 1992b.
72.20
-------
HALOCARBON SUBSTITUTES
Nolle, A., H. Heydtmann, R. Meller, W. Schneider, and
O.K. Moortgat, UV absorption spectrum and
cross-section of COF2 at 296 K in the range 200-
230 nm, Geophys. Res. Lett., 19, 281-284, 1992.
Nolle, A., H. Heydtmann, R. Meller, and O.K. Moortgat,
Temperature dependent UV absorption spectra of
carbonyl chloro-fluoride, Geophys. Res. Lett., 20,
707-710, 1993.
O'Reilly, J., H.W. Sidebottom, M. Buck, C.E. Canosa-
'Mas, and R.P. Wayne, The reaction of O3 with
CF3O and CF3O2, submitted to Chem. Phys. Lett.,
1994.
Peelers, J., and V. Pultau, Reactions of hydrofluorocar-
bon- and hydrochlorofluorocarbon-derived peroxy
radicals with nitric oxide: Results for CF3CH2F
(HFC-134a), CF3CH3 (HFC-143k), CF2C1CH3
(HCFC-142b), CF3CC12H (HFC-123) and
CF3CFHC1 (HCFC-124), in Physico-Chemical
Behaviour of Atmospheric Pollutants, Vol. 1, Comm.
Eur. Communities, Rep. no. 15609, 372-378
1994.
Rattigan, O.V., R.A. Cox, and R.L. Jones, The UV ab-
sorption cross-sections of CF3COC1, CF3COF,
CH3COF and CC13CHO, in Kinetics and Mecha-
nisms for the Reactions of Halogenated Organic
Compounds in the Troposphere, STEP-HALOC-
SIDE/AFEAS Workshop, May 14-16, Dublin,
116-125, 1991.
Rattigan, O.V.,'0. Wild, K. Law, R.L. Jones, J. Pyle, and
R.A. Cox, Two-dimensional modeling studies of
HCFCs, HFCs and their degradation products, in
Atmospheric Wet and Dry Deposition of Carbonyl
and Haloacetyl Halides, AFEAS Workshop Pro-
ceedings, 22 September 1992, Brussels, 92-102,
1992.
Rattigan, O.V., O. Wild, R.L. Jones, and R.A. Cox, Tem-
perature-dependent absorption cross-sections of
CF3COC1, CF3COF, CH3COF, CC13CHO and
CF3COOH, J. Photochem. Photobiol. A: Chem.,
73, 1-9, 1993.
Rattigan, O.V., D.M. Rowley, O. Wild, R.L. Jones, and
R.A. Cox, Mechanism of the atmospheric oxida-
tion of 1,1,1,2 tetrafluoroethane (HFC 134a),
accepted for publication in J. Chem. Soc. Faraday
Trans., 1994.
Ravishankara, A.R., S. Solomon, A.A. Turnipseed, and
R.F. Warren, Atmospheric lifetimes of long-lived
halogenated species, Science, 259, 194-199, 1993.
Ravishankara, A.R., A.A. Turnipseed, N.R. Jensen, S.
Barone, M. Mills, C.J. Howard, and S. Solomon,
Do hydrofluorocarbons destroy stratospheric
ozone?. Science, 263, 71-75, 1994.
Reisinger, A.R., N.B. Jones, W.A. Matthews, and C.P.
Rinsland, Southern Hemisphere ground based
measurements of carbonyl fluoride (COF2) and
hydrogen fluoride(HF): Partitioning between fluo-
rine reservoir species, Geophys. Res. Lett., 21
797-800, 1994. •
Rodriguez, J.M., M.K.W. Ko, N.-D. Sze, and C.W.
Heisey, Model assessment of the impact of wet
and dry deposition o.n the tropospheric abundance
of acetyl halides and trifluoroacetic acid, in Atmo-
spheric Wet and Dry1 Deposition of Carbonyl and
Haloacetyl Halides, AFEAS Workshop Proceed-
ings, 22 September i;992t Brussels, 25-32, 1992.
Rodriguez, J.M., M.K.W. Ko, N.-D. Sze, and C.W.
Heisey, Two-dimensional assessment of the degra-
dation of HFC-134a: Tropospheric accumulations
and deposition of triiluoroacetic acid, in Kinetics
and Mechanisms for the Reactions of Halogenated
Organic Compounds in the Troposphere, STEP-
HALOCSIDE/AFEAS Workshop, March 23-25,
Dublin, 104-112, 1993.
Rodriguez, J.M., M.K.W. Ko, -N.-D. Sze, and C.W.
Heisey, Three-dimensional assessment of the deg-
radation of HCFCs and HFCs: Tropospheric
distributions and stratospheric inputs, in Atmo-
spheric Degradation of HCFCs and HFCs,
Proceedings NASA/NOAA/AFEAS Workshop,
Boulder, Colorado, 17-19 November 1993, 1994.
Saathoff, H., and R. Zellner, Spectroscopic identifica-
tion and kinetics of the CF3O (X2E) radical,
abstract of paper presented at the 12th Int. Symp.
on Gas Kinetics, Reading, U.K., July, 1992.
Saathoff, H., and R. Zellner, LIF detection of the CF3O
radical and kinetics of its reactions with CH} and
, C2H6, Chem. Phys. Lett., 206, 349-354, 1993.
Safary, E., J. Romand, and Ei. Vbdar, Ultraviolet absorp-
tion spectrum of gaseous hydrogen fluoride, /.
Chem. Phys., 19, 379-380, 1951.
72.27
-------
HALOCARBON SUBSTITUTES
Sato, H., and T. Nakamura, The degradation of hydro-
chlorofluorocarbons by reaction with Cl atoms
and Oa, Nippon Kagaku Kaishi, 548-551, 1991.
Sawerysyn, J.P., A. Talhaoui, B. Meriaux, and P.
Devolder, Absolute rate constants for elementary
reactions between chlorine atoms and CHF2C1,
CH3CFC12, CH3CF2C1, and CH2FCF3 at 297 ± 2
K, Chem, Phys. Lett., 198. 197-201, 1992.
Scollard, D., M. Corrigan, J. Treacy, and H. Sidebottom,
Kinetics and mechanisms for the oxidation of ha-
logcnated aldehydes, in Kinetics and Mechanisms
for the Reactions of Halogenated Organic Com-
pounds in the Troposphere, STEP-HALOCSIDE/
AFEAS Workshop, May 14-16, Dublin, 40-51,
1991.
Scollard, D.J., J.J. Treacy, H.W. Sidebottom. C. Bal-
cstra-Garcia, G. Laverdet, G. LeBras, H.
MacLeod, and S. Teton, Rate constants for the re-
actions of hydroxyl radicals and chlorine atoms
with halogenated aldehydes, J. Phys. Chem., 97,
4683-4688, 1993.
Schneider, W.F., and T.J. Wellington, Ab initio investiga-
tion of the heats of formation of several
trifluoromethyl compounds, submitted to J. Phys.
Chem., 1994.
Sehested, J. and O.J. Nielsen, Absolute rate constants for
the reaction of CF3O2 and CF3O radicals with NO
at 295 K, Chem. Phys. Lett.. 206, 369-375, 1993.
Sehested, J., O.J. Nielsen, and T.J. Wallington, Absolute
rate constants for the reaction of NO with a series
of peroxy radicals in the gas phase at 295 K,
Chem. Phys. Lett., 213, 457-464, 1993.
Sehested, J., and T.J. Wallington, Atmospheric chemis-
try of hydrofluorocarbon 134a: Fate of the alkoxy
radical CF3O, Environ. Sci. Technoi, 27, 146-152,
1993.
Shi, J., T.J. Wallington, and E.W. Kaiser, FTIR study of
the Cl-initiated oxidation of CaHsCl: Reactions of
the alkoxy radical CH3CHC1O, /. Phys. Chem.,
97, 6184-6192, 1993.
Solomon, S., J.B. Burkholder, A.R. Ravishankara, and
R.R. Garcia, On the ozone depletion and global
wanning potentials of CF3I, J. Geophys. Res., 99,
20929-20935, 1994.
Thompson, J.E., Kinetics ofO('D)and Cl(2P) Reactions
with Halogenated Compounds of Atmospheric In-
terest, M.S. thesis, University of Colorado at
Boulder, Boulder, Colorado, 1993.
Tuazon, E.G., R. Atkinson, and S.B. Corchnoy, Rate
constants of the gas-phase reactions of Cl atoms
with a series of hydrofluorocarbons and hydro-
chlorofluorocarbons at 298 ± 2 K, Int. J. Chem.
Kinet., 24, 639-648, 1992.
Tuazon, B.C., and R. Atkinson, Tropospheric degrada-
tion products of CH2FCF3 (HFC-134a), J. Atmos.
Chem., 16, 301-312, 1993a.
Tuazon, B.C., and R. Atkinson, Tropospheric transfor-
mation products of a series of hydrofluorocarbons
and hydrochlorofluorocarbons, /. Atmos. Chem.,
17, 179-199, 1993b.
Tuazon, B.C., and R. Atkinson, Products of the tropo-
spheric reactions of hydrochlorofluorocarbons
(HCFCs) -225ca, -225cb, -141b and -142b, sub-
mitted to Environ. Sci. Technoi., 1994.
Tumipseed, A.A., S. Barone, and A.R. Ravishankara,
Kinetics of the reactions of CF3OX radicals with
NO, O3, and O2, /. Phys. Chem., in press, 1994.
Ugi, I., and F. Beck, Reaktion von Carbonsaurehalo-
geniden mit Wasser und Aminen, Chem. Ber., 94,
1839-1850, 1961.
Visscher, P.T., C.W. Culbertson, and R.S. Oremland,
Degradation of trifluoroacetate in oxic and anoxic
sediments, Nature, 369, 729-731, 1994.
Wallington, T.J., and O.J. Nielsen, Pulse radiolysis study
of CF3CHFO2 radicals in the gas phase at 298 K,
Chem. Phys. Lett., 187, 33-39, 1991.
Wallington, T.J., and M.D. Hurley, A kinetic study of the
reaction of chlorine atoms with CF3CHC12,
CF3CH2F, CFC12CH3, CF2C1CH3, CHF2CH3,
CH3D, CH2D2, CHD3, CD4, and CD3C1 at 295 ± 2
K, Chem. Phys. Lett., 189, 437-442, 1992.
Wallington, T.J., M.D. Hurley, J.C. Ball, and E.W. Kai-
ser, Atmospheric chemistry of hydrofluorocarbon
134a: Fate of the alkoxy radical CF3CFHO, Envi-
ron. Sci. Technoi., 26, 1318-1324, 1992.
Wallington, T.J., and M.D. Hurley, Atmospheric chemis-
try of HC(O)F: Reaction with OH radicals,
Environ. Sci. Technoi., 27, 1448-1452, 1993.
72.22
-------
HALOCARBON SUBSTITUTES
Wallington, T.J., M.D. Hurley, W.F. Schneider, J. Sehes-
ted, and OJ. Nielsen, Atmospheric chemistry of
CF3O radicals: Reaction with H2O, J. Phys.
Chem., 97, 7606-7611, 1993a.
Wallington, T.J., M.D. Hurley, and W.F. Schneider, Ki-
netic study of the reaction CF3O + O3 -> CF3O2 +
O2, Chem. Phys. Lett., 213, 442-448, 1993b.
Wallington, T.J., M.D. Hurley, W.F. Schneider, J. Sehes-
ted, and OJ. Nielsen, Mechanistic study of the gas
phase reaction of CH2FO2 radicals with HO2> sub-
mitted to Chem. Phys. Lett., 1994a.
Wallington, T.J., T. Ellerman, OJ. Nielsen, and J. Se-
hested, Atmospheric chemistry of FCOX radicals:
UV spectra and self reaction kinetics of FCO and
FC(O)O2, and kinetics of some reactions of FCOX
with 02, 03 and NO at 296 K, J. Phys. Chem., in
press, 1994b.
Wallington, T.J., and W.F. Schneider, The stratospheric
fate of CF3OH, Geophys. Res. Lett., in press
1994.
Warren, R.F., and A.R. Ravishankara, Kinetics of C1(2P)
reactions with a few hydrochlorofluorocarbons
(HCFCs), Int. J. Chem. Kinet., 25, 833-844, 1993.
Wine, P.H., and W.L. Chameides, Possible atmospheric
lifetimes and chemical reaction mechanisms for
selected HCFCs, HFCs, CH3CC13 and their degra-
dation , products against dissolution and/or
degradation in sea water and cloud water, in Scien-
tific Assessment of Stratospheric Ozone: 1989,
Volume II, Appendix: AFEAS Report, World Me-
teorological Organization Global Ozone Research
and Monitoring Project - Report No. 20, Geneva
271-295, 1990.
WMO, Scientific Assessment of Stratospheric Ozone;
1989, World Meteorological Organization Global
Ozone Research and Monitoring Project - Report
No. 20, Volume II, Appendix: AFEAS Report,
Geneva, 1990.
WMO, Scientific Assessment of Ozone Depletion: 1991,
World Meteorological Organization Global Ozone
Research and Monitoring Project - Report No. 25
Geneva, 1992. !
Worsnop, D.R., M.S. Zahnizer, and C.E. Kolb, J.A..
Gardner, J.T. Jayne, L.R. Watson, J.M. VanDoren,
and P. Davidovito, Temperature dependence of
mass accommodation of SO2 and H2O2 on aque-
ous surfaces, /. Phys. Chem., 93, 1159-1172
1989. !
Zander, R., M.R. Gunson, C.B. Farmer, C.P. Rinsland,
F.W. Won, and E. Mahieu, The 1985 chlorine and
fluorine inventories in the stratosphere based on
ATMOS observations at 30°N latitude, /. Atmos
Chem., 15, 171-186, 1992.
Zellner, R., A. Hoffmann, D. Bingemann, V. Mors, and
J.P. Kohlmann, Time resolved product studies in
the oxidation of HCFC-22 and HFC-l34a under
simulated tropospheric conditions, in Kinetics and
Mechanisms for the Reactions of Halogenated
Organic Compounds in the Troposphere, STEP-
HALOCSIDE/AFEAS Workshop, May 14-16,
Dublin, 94-103, 1991.
Zellner, R., A. Hoffmann, iy. Mors, and W. Malms, Time
resolved studies of intermediate products in the
oxidation of HCFC's and HFC's, in Kinetics and
Mechanisms for the Reactions of Halogenated Or-
ganic Compounds in the Troposphere,
STEP-HALOCSIDE-/AFEAS Workshop, March
23-25, Dublin, 1993.
12.23
-------
-------
1
CHAPTER 13
Ozone Depletion Potentials,
Global Warming Potentials,
and Future Chlorine/Bromine Loading
Lead Authors:
S. Solomon
D. Wuebbles
Co-authors:
I. Isaksen
J. Kiehl
M.Lai
P. Simon
N.-D. Sze
Contributors:
D. Albritton
C. Briihl
P. Connell
J.S. Daniel
D. Fisher
D. Hufford
C. Granier
S.C. Liu
K. Patten
V. Ramaswamy
K. Shine
S. Pinnock
G. Visconti
D. Weisenstein
T.M.L. Wigley
-------
-------
CHAPTER 13 :
OZONE DEPLETION POTENTIALS, GLOBAL WARMING POTENTIALS
AND FUTURE CHLORINE/BROMINE LOADING
Contents
SCIENTIFIC SUMMARY.... i
[[[ • ........................................ 13.1
13.1 INTRODUCTION <
[[[ ii ......................... • .............. 13.3
13.2 ATMOSPHERIC LIFETIMES AND RESPONSE TIMES :
.................................... ........................................ 13.4
13.3 CI/Br LOADING AND SCENARIOS FOR CFC SUBSTITUTES :
13.3.1 Equivalent Tropospheric Chlorine Loading ............................. *" ................ ........................................ 13'7
13.3.2 Equivalent Effective Stratospheric Chlorine ............................... ..' ............. ....................................... 13"7
13.4 OZONE DEPLETION POTENTIALS ..................... 1
13.4.1 Introduction ........................................ _"'_" [[[ j ................... ; ................. 13'12
13.4.2 Relative Effectiveness of Halogens in Ozone Destruction .......................... ' ..................................... !MJ
13.4.2.1 Fluorine ....................... ......................... J ..................................... 13'13
13.4.2.2 Bromine ........ . ..................... ................................................ j ...................................... 13' 13
13.4.2.3 Iodine ..................................... ZZZZZ ................................... ! ..................................... 13'14
13.4.3 Breakdown Products of HCFCs and HFCs ..................... ".."... ..................... ..................................... 13'15
13.4.4 Model-Calculated and Semi-Empirical Steady-State ODPs ........................ ....... ............................ f?'!5
13.4.5 Time-Dependent Effects ....................... I ................... ' ................ J'16
[[[ i .................................... 13.18
13.5 GLOBAL WARMING POTENTIALS !
13.5.1 Introduction ....................................... ZZZZZ .......................................... ! .................................... 13'2°
13.5.2 Radiative Forcing Indices .................................. ............................. ! ................. ' .................. 13'2°
13.5.2.1 Formulation ........................... "ZZZZ" .................................... 1 ..................................... 13'21
13.5.2.2 Sensitivity to the State of the Atmosphere ..... ........................ ! .............. " ..................... !—.!
13.5.3 Direct GWPs .................................. .............................. .................................... I3'23
13.5.4 Indirect Effects ..................... ZZZZZ .................................... ................................... 13'24
13.5.4.1 General Characteristics ........... ..
13.5.4.2 Indirect Effects upon the GWP o
1
REFERENCES
-------
-------
ODPs, GWF»s and CI-Br LOADING
SCIENTIFIC SUMMARY
Scientific indices representing the relative effects of different gases upon ozone depletion and climate forcing are
presented. Several scenarios for future chlorine/bromine loading are described that are aimed at implementation of the
Copenhagen Amendments of the Montreal Protocol and the consideration of possible further options. Ozone Depletion
Potentials (ODPs) and Global Warming Potentials (GWPs) are evaluated with improved models and input data and
their sensitivities to uncertainties are considered in greater detail than in previous assessments. Major new findings are
as follows:
Peak levels of ozone-depleting compounds are expected at stratospheric altitudes in the late 1990s Because
current emission estimates suggest that the tropospheric chlorine/bromine loading will peak in 1994 further re-
ductions in emissions would not significantly affect the timing or magnitude of the peak stratospheric halogen
loading expected later this decade (i.e., about 3-5 years after the tropospheric peak).
Approaches to lowering stratospheric chlorine and bromine abundances are limited. Further controls on ozone-
depleting substances would be unlikely to change the timing or the magnitude of the peak stratospheric
halocarbon abundances and hence peak ozone loss. However, there are four approaches that would steepen the
initial fall from the peak halocarbon levels in the early decades of the next century:
i
(i) If emissions of methyl bromide from agricultural, structural, and industrial activities were to be eliminated
in the year 2001, then the integrated effective future chlorine loading above the 1980 level (which is related
to the cumulative future loss of ozone) is predicted to be 13% less over the. next 50 years relative to full
compliance with the Amendments and Adjustments to the Protocol. !
I
(ii) If emissions of hydrochlorofluorocarbons (HCFCs) were to be totally eliminated by the year 2004, then the
integrated effective future chlorine loading above the 1980 level is predicted to be 5% less over the next 50
years relative to full compliance with the Amendments and Adjustments to the Protocol.
(iii) If halons presently contained in existing equipment were never released to the atmosphere, then the inte-
grated effective future chlorine loading above the 1980 level is predicted to be 10% less over the next 50
years relative to full compliance with the Amendments and Adjustments to the Protocol.
(iv) If chlprofluorocarbons (CFCs) presently contained in existing equipment were never released to the atmo-
sphere, then the integrated effective future chlorine loading above the 1980 level is predicted to be 3% less
over the next 50 years relative to full compliance with the Amendments and Adjustments to the Protocol.
Failure to adhere to the international agreements will delay recovery of the ozone layer. If there were to be
additional production of CFCs at, for example, 20% of 1992 levels for each year through 2002 and ramped to zero
by 2005 (beyond that allowed for countries operating under Article 5 of the Montreal Protocol), then the integrat-
ed effective future chlorine loading above the 1980 level is predicted to be 9% more over the next 50 years relative
to full compliance with the Amendments and Adjustments to the Protocol. j
Production of CF3 from dissociation of CFCs, HCFCs, and hydrofluorocarbons (MFCs) is highly unlikely to
affect ozone. ODPs of HFCs containing the CF3 group (such as HFC-134a, HFC-23, and HFC-125) are highly
likely to be less than 0.001, and the contribution of the CF3 group to the ODPs of HCFCs (e.g., from HCFC-123)
and CFCs is believed to be negligible. i '
13.1
-------
ODPs, GWPs and CI-Br LOADING
• ODPs for several new compounds such as HCFC-225ca, HCFC-225cb, and CFjI have been evaluated using both
semi-empirical and modeling approaches, and estimated to be 0.03 or less.
• Both the direct and indirect components of the GWP of methane have been estimated using model calculations.
Methane's influence on the hydroxyl radical and the resulting effect on the methane response time lead to substan-
tially longer response times for decay of emissions than OH removal alone, thereby increasing the GWP. In
addition, indirect effects including production of tropospheric ozone and stratospheric water vapor were consid-
ered and are estimated to range from about 15 to 45% of the total GWP (direct plus indirect) for methane.
• GWPs including indirect effects of ozone depletion have been estimated for a variety of halocarbons (CFCs,
ftalons, HCFCs. etc.), clarifying the relative radiative roles of different classes of ozone-depleting compounds.
The net GWPs of halocarbons depend strongly upon the effectiveness of each compound for ozone destruction;
the halons are highly likely to have negative net GWPs, while those of the CFCs are likely to be positive over both
20- and 100-year time horizons.
• GWPs are not very sensitive to likely future changes in C02 abundances or major climate variables. Increasing
C02 abundances (from about 360 ppmv currently to 650 ppmv by the end of the 22nd century) could produce
20% larger GWPs for time horizons of the order of centuries. Future changes in clouds and water vapor are
unlikely to significantly affect GWPs for most species.
• GWPs for 16 new chemical species have been calculated, bringing the number now available to 38. The new
species are largely HFCs, which are being manufactured as substitutes for the CFCs, and the very long-lived fully
fluorinated compounds, SF$ and the perfluorocarbons.
13.2
-------
ODPs, GWPs; and CI-Br LOADING
13.1 INTRODUCTION
Numerical indices representing the relative im-
pacts of emissions of various chemical compounds upon
ozone depletion or global radiative forcing can be useful
for both scientific and policy analyses. Prominent
among these are the concepts of Chlorine/Bromine
Loading, steady-state and time-dependent Ozone Deple-
tion Potentials (ODPs), and Global Warming Potentials
(GWPs), which form the focus of this chapter. Detailed
descriptions of the formulations of these indices are pro-
vided later. Here we briefly review the broad definitions
of these concepts and cite some of their uses and limita-
tions:
Chlorine/Bromine Loading
Chlorine/bromine loading represents the amount
of total chlorine and bromine in the troposphere or
stratosphere. Stratospheric chlorine/bromine loading
depends upon the surface emissions of gases such as
chlorofluorocarbons (CFCs), hydrochlorofluorocarbons
(HCFCs), and halons (which are based in large part upon
industrial estimates of usage) and upon knowledge of the
reactivity and hence the atmospheric lifetimes and
chemical roles of those and related compounds. Recent
depletions in stratospheric ozone in Antarctica and in the
Arctic have been linked to anthropogenic halocarbon
emissions (see Chapter 3), and the weight of evidence
suggests that ozone depletions in midlatitudes are also
related to the emissions of these compounds (see WMO,
1992 and Chapter 4 of this document). Thus, the chlo-
rine/bromine loading is a key indicator of past and future
changes in ozone. However, it should be recognized that
chlorine/bromine loading is a measure only of changes
in halogen content. It does not account for additional
factors that could also affect the time-dependent changes
in atmospheric ozone or the linearity of their relationship
to chlorine/bromine loading (e.g., carbon dioxide trends
that can also affect stratospheric temperatures).
Ozone Depletion Potentials
Ozone Depletion Potentials (ODPs) provide a rel-
ative measure of the expected impact on ozone per unit
mass emission of a gas as compared to that expected
from the same mass emission of CFC-11 integrated over
time (Wuebbles, 1983; WMO, 1990, 1992; Solomon et
al, 1992). Their primary purpose is for comparison of
relative impacts of different gases upon ozone (e.g., for
evaluating the relative effects of choices among different
CFC substitutes upon ozone). As in prior analyses, the
ODP for each substance presented herein is based on the
mass emitted into the atmosphere, and not on the total
amount used. In some cases (such as emissions of
CHaBr in soil fumigant applications) not all of the com-
pound used may be emitted into the global atmosphere
(see Chapter 10). Steady-sitate ODPs represent the cu-
mulative effect on ozone over ari infinite time scale (also
referred to here as "time horizon"). Time-dependent
ODPs describe the temporal evolution of this ozone im-
pact over specific time horizons (WMO, 1990, 1992;
Solomon and Albritton, 1992; see Section 13.4.5). At-
mospheric models and seitni-empirical methods have
been used in combination to best quantify these relative
indices (Solomon et al, 1992; WMO, 1992). As a rela-
tive measure, ODPs are subject to fewer uncertainties
than estimates of the absolute percentage ozone deple-
tion, particularly when only the ODP differences among
various chlorinated gases are considered. Models used
to evaluate ODPs now include better representations of
midlatitude and polar vortex heterogeneous chemistry
processes than those used ejirlier. Comparisons of mod-
el and semi-empirical methods reduce the uncertainties
in ODPs. However, evaluations of ODPs are still subject
to uncertainties in atmospheric lifetimes and in the un-
derstanding of stratospheric chemical and dynamical
processes. The recent re-evaluation of the chemical rate
and products for the reaction of BrO + HO2 and resulting
effects on ODPs for bromocarbons provide a graphic ex-
ample of potential impacts of such uncertainties (see
Section 13.4). Like chlorine/bromine loading, ODPs do
not include other processes (such as changes in CO2 and
hence stratospheric temperatures) that could affect the
future impacts of different gases upon ozone.
Global Warming Potentials
.Global Warming Potentials provide a simple rep-
resentation of the relative radiative forcing resulting
from a unit mass emission of a greenhouse gas compared
to a reference compound. Because of its central role in
concerns about climate change, carbon dioxide has gen-
erally been used as the reference gas. However, because
of the complexities and uncertainties associated with the
13.3
-------
ODPs, GWPs and CI-Br LOADING
carbon cycle, extensive effort has been put into evaluat-
ing the effects on GWPs from uncertainties in the
time-dependent uptake of carbon dioxide emissions. As
described in Chapter 4 of IPCC (1994), calculations
made with climate models indicate that, for well-mixed
greenhouse gases at least, the relationship between
changes in the globally integrated adjusted radiative
forcing at the tropopause and global-mean surface tem-
perature changes is independent of the gas causing the
forcing. Furthermore, similar studies indicate that, to
first order, this "climate sensitivity" is relatively insensi-
tive to the type of forcing agent (e.g., changes in the
atmospheric concentration of a well-mixed greenhouse
gas such as CC>2, or changes in the solar radiation reach-
ing the atmosphere). GWPs have a number of important
limitations. The GWP concept is difficult to apply to
gases that are very unevenly distributed and to aerosols
(see, &Ł., Wang et al., 1991, 1993). For example, rela-
tively short-lived pollutants such as the nitrogen oxides
and the volatile organic compounds (precursors of
ozone, which is a greenhouse gas) vary markedly from
region to region within a hemisphere and their chemical
impacts arc highly variable and nonlinearly dependent
upon concentrations. Further, the indices and the esti-
mated uncertainties are intended to reflect global
averages only, and do not account for regional effects.
They do not include climatic or biospheric feedbacks,
nor do they consider any environmental impacts other
than those related to climate. The direct GWPs for a
number of infrared-absorbing greenhouse gases have
been analyzed in this report, with a particular emphasis
on a wide range of possible substitutes for halocarbons.
The evaluation of effects on other greenhouse gases re-
sulting from chemical interactions (termed indirect
effects) has been more controversial. Underlying as-
sumptions and uncertainties associated with both direct
and indirect GWPs are discussed briefly in Section 13.5.
13.2 ATMOSPHERIC LIFETIMES AND
RESPONSE TIMES
Atmospheric lifetimes or response times are used
in the calculation of both ODPs and GWPs. The list of
compounds considered in this assessment is an extension
of those in WMO (1992), primarily reflecting the con-
sideration of additional possible replacements for CFCs
and halons. Additional compounds, such as the unusually
long-lived perfluorocarbons and SFg, are also included
because of their potential roles as greenhouse gases and
because some have been suggested as CFC and halon
replacements.
After emission into the current or projected atmo-
sphere, the time scale for removal (i.e., the time interval
required for a pulse emission to decay to 1/e of its initial
perturbed value) of most ozone-depleting and green-
house gases reflects the ratio of total atmospheric burden
to integrated global loss rate. As such, the total lifetime
must take into account all of the processes determining
the removal of a gas from the atmosphere, including
photochemical losses within the troposphere and strato-
sphere (typically due to photodissociation or reaction
with OH), heterogeneous removal processes, and perma-
nent removal following uptake by the land or ocean. In a
few cases, the time scale for removal of a gas from the
atmosphere cannot be simply characterized or is depen-
dent upon the perturbation and/or the background
atmosphere and other sources; in those cases (chiefly
CX>2 and CHLt) we refer to removal of a pulse as the re-
sponse time or decay response.
Alternatively, atmospheric lifetimes can be de-
fined by knowledge of global source strengths together
with the corresponding mean atmospheric concentra-
tions and trends, but these are usually more difficult to
define accurately. The atmospheric lifetime may be a
function of time, due to changing photochemistry asso-
ciated, for example, with ozone depletion or temperature
trends, but these effects are likely to be small for at least
the next several decades and will not be considered here.
The total lifetimes of two major industrially pro-
duced halocarbons, CFC-11 and CH3CC13, have been
reviewed and re-evaluated in a recent assessment (Kaye
etal., 1994). The empirically derived lifetime for CFC-11
determined in that study is 50 (±5) years (as compared to
55 years in the previous WMO [1992] assessment). As
in previous assessments, the lifetimes presented here are
not based solely upon model calculations, but use infor-
mation from measurements to better constrain the
lifetimes of these and other gases. The lifetime of CFC-
11 is used here to normalize lifetimes for other gases
destroyed by photolysis in the stratosphere (based upon
scaling to the ratios of the lifetimes of each gas com-
pared to that of CFC-11 obtained in the models
discussed in Kaye et al, 1994). This approach could be
limited by the fact that different gases are destroyed in
13.4
-------
ODPs, GWPs and CI-Br LOADING
different regions of the stratosphere depending upon the
wavelength dependence of their absorption cross sec-
tions (weakening the linearity of their comparison to
CFC-11), particularly if stratospheric mixing is not rapid
(see Plumb and Ko, 1992), Depending on absolute cali-
bration factors used by different research groups, Kaye
et al. (1994) derived a total lifetime for CH3CC13 of ei-
ther 5.7 (±0.3) years for 1990 or 5.1 (±0.3) years,
respectively (see Prinn et al, 1992), compared to 6.1
years in the earlier WMO (1992) assessment. Because
of the current uncertainties in absolute calibration, we
use a lifetime for CH3CCl3 of 5.4 years with an uncer-
tainty range of 0.6 years in this report. From this total
atmospheric lifetime, together with the evaluated loss
lifetimes of CH3CC13 due to the ocean (about 85 years,
with an uncertainty range from 50 years to infinity; see
Butler et al., 1991) and stratospheric processes (40 ± 10
years), a tropospheric lifetime for reaction with OH of
6.6 years can be inferred (±25%). The lifetimes of other
key gases destroyed by OH (i.e., CHj, HCFCs, and
hydrofluorocarbons [HFCs]) can then be inferred rela-
tive to that of methyl chloroform (see, e.g., Prather and
Spivakovsky, 1990) with far greater accuracy than would
be possible from a priori calculations of the complete
tropospheric OH distribution. We note that a few of the
newest CFC substitutes (namely, the HFCs -236fa,
-245ca, and -43- lOmee) have larger uncertainties in life-
times since fewer kinetic studies of their chemistry have
been reported to date. It is likely that methane is also
destroyed in part by uptake to soil (IPCC, 1992), but this
process is believed to be relatively slow and makes a
small contribution to the total lifetime. Possible soil
sinks are not considered for any other species.
The special aspects of the lifetime of methane and
the response time of a pulse added to the atmosphere
were defined in Chapter 2 of IPCC (1994), based largely
upon Prather (1994). Those definitions are also em-
ployed here. Small changes in CH4 concentrations can
significantly affect the atmospheric OH concentration,
rendering the response time for the decay of the added
gas substantially longer than that of the ensemble (i.e.,
longer than the nominal 10-yr lifetime for the bulk con-
centration of atmospheric Ctit in the current atmosphere).
This is due to the nonlinear chemistry associated with
relaxation of the coupled OH-CO-CH4 system (see
Prather, 1994; Lelieveld et al., 1993; and Chapter 2 of
IPCC [1994]. for further details). This effect was also
discussed in IPCC (1990) and IPCC (1992) as an indirect
effect on OH concentrations, and thus is not new. It aris-
es through the fact that small changes in OH due to
addition of a small pulse of CH4 slightly affect the rate of
decay of the much larger amount of CH4 in the back-
ground atmosphere, thereby influencing the net removal
of the added pulse. It is critical to note that the exact
value of the CH4 pulse response time depends upon a
number of key factors, including the absolute amount of
CH4, size of the pulse, etc., making its interpretation
complex and case-dependent. Here we consider small
perturbations to the present atmosphere, and base the
definition of the methane pulse response time to be used
in calculation of the GWP upon the detailed explanation
of the effect as presented in Prather (1994) and in Chap-
ter 2 of IPCC (1994). •
Table 13-1 shows the recommended total atmo-
spheric lifetimes for all of the compounds considered
here except methyl bromide (the reader is referred to
Chapter 10 for a detailed discussion of the lifetime of
this important gas). The response time of methane is
also indicated. The lifetimes for many compounds have
been modified relative to values used in WMO (1992;
Table 6-2). The estimates, for the lifetimes of many of
the gases destroyed primarily by reaction with tropo-
spheric OH (e.g., HCFC-22, HCFC-141b, HCFC-142b,
etc.) are about 15% shorter than in WMO (1992), due
mainly to recent studies suggesting a shorter lifetime for
CH3CCl3 based upon improved calibration methods and
upon an oceanic sink (Butler et.al, 1991). Similarly, the
estimates for the lifetimes of gases destroyed mainly by
photolysis in the stratosphtbre (e.g., CFC-12, CFC-113,
H-1301) are about 10% shorter than in IPCC (1992) due
to a shorter estimated lifetime for CFC-11 and related
species. Lifetime estimates of a few other gases have
also changed due to improvements in the understanding
of their specific photocheniistry (e.g., note that the life-
time for CFC-115 is now estimated to be about 1700
years, as compared to about 500 years in earlier assess-
ments). Fully fluorinated species such as SF6, CF4, and
C2Fg have extremely long atmospheric lifetimes, sug-
gesting that significant production and emissions of
these greenhouse gases could ha.ve substantial effects on
radiative forcing over long time scales. In contrast,
CF3I, which is being considered for use as a fire extin-
guishant and other applications, has an atmospheric
lifetime of less than 2 days..
13.5
-------
ODPs, GWPs and CI-Br LOADING
Table 13-1. Lifetimes and response times recommended for OOP and GWP calculations.
Gas Lifetime or Response Reference
Time (yrs)
CFC-11
CFC-12
CFC-13
CFC-113
CFC-114
CFC-115
CC14
CHsCCls
CHCls
CH2Cl2
HCFC-22
HCFC-123
HCFC-124
HCFC-141b
HCFC-142b
HCFC-225ca
HCFC-225cb
CHsBr
CFsBr(H-1301)
CF2ClBr(H-1211)
HFC-23
HFC-32
HFC-125
HFC-134
HFC-134a
HFC- 143
HFC-143a
HFC-152a
HFC-227ea
HFC-236fa
HFC-245ca
HFC-43-10mee
HFOC-125E
HFOC-134E
SF6
CF4
C2F6
C6F14
C5F12
c-C4Fg
CFsI
N20
CH4 (pulse response)
50 (±5)
102
640
85
300
1700
42
5.4 (±0.4)
0.55
0.41
13.3
1.4
5.9
9.4
19.5
2.5
6.6
1.3
65
20
250
6,0
36
11.9
14
3.5
55
1.5
41
250
7
20.8
82
8
3200
50000
10000
3200
4100
3200
< 0.005
120
14.5 ± 2.5
2
3
1
3
1
1
3
2
4
4
4
4
4
4
4
4
4
Chapter 10
3
3
10
4
4
5
4
11
4
4
9
7
6
7
6
6
1
1
1
1
1
1
8
3 •
12
13.6
-------
ODPs,
Table 13-1. Notes.
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
GWPs and CI-Br LOADING
Ravishankara etal. (1993).
Prather, private communication 1993, based on NASA CFC report (Kaye etal., 1994) and other
considerations as described in text. '
Average of reporting models in NASA CFC report (Kaye et al, 1994). Scaled ro CFC-11 lifetime
Average of JPL 92-20 and IUPAC (1992) with 277 K rate constants for OH+halocarbon scaled against
OH+CH3CC13 and lifetime of tropospheric CH3CC13 of 6.6 yr. Stratospheric lifetime from WMO (1992)
DeMore etal (1993). Used 277 K OH rate constant ratios with respect to CH3CC13, scaled to tropospheric
lifetime of 6.6 yr for CH3CC13. j
Cooper etal. (1992). Lifetime values are estimates. !
W. DeMore (personal communication, 1994) with 277 K rate constants for OH+halocarbon scaled against
OH+CH 3CC13 and lifetime of tropospheric CH3CC13 of 6.6 yr. ;
Solomon etal. (1994). j
Briihl, personal communication based on data for the reaction rate constant with OH provided by Hoescht
Chemicals, 1993; Zhang etal (1994) and Nelson etal (1993) with 277 K rate constants for OH+halocarbon
scaled against OH + CH3CC13 and lifetime of tropospheric CH3CC13 of 6.6 yr. :
Schmoltner etal. (1993) with 277 K rate constants for OH+halocarbon scaled against OH+CH,CC1, and
lifetime of tropospheric CH0CCI} of 6.6 yr.
Barry et al. (1994) with 277 K rate constants for OH+halocarbon scaled against'OH+CH3CC1V and lifetime of
tropospheric CH3CC13 of 6.6 yr. ,
Prather (1994) and Chapter 2 of IPCC (1994). I
The basis for the recommended lifetimes is de-
scribed within the Table and its footnotes. These values
are used for all calculations presented in this chapter.
13.3 CHLORINE/BROMINE LOADING AND
SCENARIOS FOR CFC SUBSTITUTES
13.3.1 Equivalent Tropospheric Chlorine
Loading
For the purposes of this report, a detailed assess-
ment of those sources of tropospheric chlorine and
bromine loading relevant to stratospheric ozone destruc-
tion was carried out. The approach taken is similar to
that of Prather and Watson (1990) and previous assess-
ment reports (WMO, 1992). This analysis is more
complete in that it includes a description of the time
delay between consumption and emission of the ozone-
depleting substances. The time delays are based upon
uses (e.g., refrigeration, solvents, etc.). The procedure is
also discussed in Daniel et al. (1994). The best under-
standing of the past history of emissions of fourteen of
the most important halocarbons, together with current
estimates of the lifetimes of these gases (Table 13-1)
provides the input needed to evaluate past trends. The
longest and most complete record of CFC emissions is
contained in the industryrsponsored "Production, Sales
and Atmospheric Release of Fluorocarbons" report
(AFEAS, 1993). This report contains estimates of pro-
duction in countries not covered in the industry survey.
Recently, with declining global production in response
to the Montreal Protocol,' the fractional contribution to
the total of this "unreported" production, a portion of
which is in developing (Article 5) countries, has amount-
ed to about 25%. Estimates of unreported production
based on matching observed and calculated trends in the
relevant trace gases are consistent with AFEAS esti-
mates (see, e.g., the detailbd analysis in Cunnold et al,
1994). ;
Expected uses and the corresponding release times
for each of the gases are considered, in order to more
accurately determine yearly emission amounts (AFEAS,
1993; Fisher and Midgley, 1993; Gamlen et al, 1986;
McCarthy et al, 1977; McCulloch, 1992; Midgley,
1989; Midgley and Fisher, 1993). Possible time-depen-
dent changes in release times (e.g., for improved
technologies) are not considered. For methyl bromide, a
budget of natural and anthropogenic sources based upon
13.7
-------
ODPs, GWPs and CI-Br LOADING
Chapter 10 is adopted. Anthropogenic sources of methyl
bromide are assumed to be zero before 1931. A constant
anthropogenic emission is assumed from 1931 to 1994
of 73 ktonnes/year (see Chapter 10). As noted in Chap-
ter 10, it is possible that decreases in methyl bromide
emissions associated with the declining use of gasoline
additives could have offset some of the known increases
in agricultural use of this compound during the 1970s
and 1980s. However, precise information is not avail-
able. Although this assumption will affect the calculated
historical contribution of methyl bromide to equivalent
chlorine loading, because of the short lifetime of methyl
bromide, it has very little effect on projected contribu-
tions. Anthropogenic emission of methyl bromide does
not equal production, and this difference is explicitly
considered in all calculations of methyl bromide's atmo-
spheric loading and their impacts presented in this
chapter.
The calculated contributions of methyl bromide
and other bromocarbons to equivalent chlorine loading
are more uncertain than that of other compounds. For
the purpose of comparing the roles of chlorine- and bro-
mine-containing gases once they reach the stratosphere,
it is assumed that each bromine atom is 40 times more
damaging to ozone than chlorine (see Section 13.4), al-
lowing evaluation of an "equivalent tropospheric
chlorine" that includes an estimate of the net ozone im-
pact of bromocarbons. The enhanced effectiveness of
bromine (hereafter referred to as a)Depends in principle
upon the amount of active chlorine present, making it a
time-dependent quantity. However, in the next few de-
cades (i.e., until about 2020), the chlorine content of the
stratosphere is expected to change relatively little, mak-
ing a essentially constant during this period. Towards
the middle and latter parts of the twenty-first century,
decreases in chlorine abundances will likely lead to in-
creases in the value of a, at least in polar regions. This
follows from the fact that the reaction of CIO with itself
represents an important ozone loss process in the Antarc-
tic (and Arctic) that is dependent upon the square of the
stratospheric chlorine abundance, while the reaction of
CIO with BrO is linearly dependent upon the strato-
spheric chlorine abundance. Thus, as chlorine abundances
decline, the reaction of CIO with BrO will become more
important relative to CIO + CIO. This and other consid-
erations discussed in Section 13.4 (particularly the role
of the H©2 + BrO reaction in the lower stratosphere)
suggest that the adopted value of a of 40 is likely to be a
low estimate. A higher value of a would increase the
contributions of methyl bromide and the halons. The
adopted methyl bromide lifetime of 1.3 years includes an
ocean sink. If loss to the ocean were to be slower, the
lifetime would be longer and the anthropogenic methyl
bromide contribution would be larger. On the other
hand, a faster ocean sink would decrease the contribu-
tion. Similarly, a decrease (increase) in the fractional
emission of methyl bromide used for agricultural pur-
poses would decrease (increase) the calculated
contribution from that source. The budget of methyl
bromide and its uncertainties are discussed in detail in
Chapter 10 of this assessment.
Chlorinated solvents such as CH2C12, C2C14, and
C2HC13 were not explicitly considered in this analysis.
Based upon emission estimates, WMO (1992) suggests
that these species are present at about the 35, 32, and 1
pptv levels, respectively, within the current troposphere.
Wang et al. (1994) present observations of C2C14 show-
ing average abundances of only 7 pptv. The lifetimes of
these gases may be long enough to allow a fraction to
reach the stratosphere and thereby contribute to strato-
spheric chlorine loading. Schauffler et al. (1993) report
tropospheric measurements of CH2C12 of about 30 pptv
in 1992 and report direct measurements of this gas near
the tropical tropopause of about 15 pptv, suggesting
substantial transport to the stratosphere. While the abun-
dances of these gases are presently small, increasing use
would increase the abundances. At a growth rate of, for
example, 3%/year, CH2C12 and C2HC13 could reach
abundances of 0.1 ppbv in 36 and 156 years, respectively.
Thus, while these relatively short-lived gases probably
contribute little to contemporary stratospheric chlorine
loading, there is observational evidence of significant
transport to the stratosphere for some species, and con-
tinued growth would lead to a greater contribution to
stratospheric chlorine loading. On the other hand, a re-
cent survey (P. Midgley, personal communication)
indicates that industrial emissions of these gases in the
U.S., Europe, and Japan have steadily decreased since
1984, so that current emissions are more likely to be de-
creasing than increasing.
Water-soluble emissions such as sea salt or volca-
nic HC1 are effectively removed in clouds and rain (see,
e.g., Tabazadeh and Turco, 1993) and do not represent
significant sources of stratospheric chlorine. Short-lived
13.8
-------
ODPs, GWPs and CI-Br LOADING
as bromoform were also not consid-
Equivalent chlorine loading was evaluated for
eight cases to demonstrate impacts of various assump-
tions for future use of ozone-depleting substances A
complete description of the scenarios is provided in Ta-
ble 13-2. Global compliance to the Copenhagen
agreements is represented by case A. Estimates of future
emissions of hydrochlorofluorocarbons (HCFCs) and
hydrofluorocarbons (HFCs) are based on a detailed anal-
ysis of projected global demand for each gas carried out
20™m t *nV*0nmental Protection Agency (EPA) to
^ou (u. hturford, personal communication 1993) Es
tunates beyond 2030 will depend on agreements for
HCFC use in developing countries; no attempt is made
to account for potential use and emissions beyond 2030
Such use would increase the HCFC equivalent chlorine
ce±L!t"Łi^
n the EPA
A complete phase-out of HCFCs after 2030 is as-
sumed after which time a 2.5%/year increase in
HFC- 34a is adopted (intended to represent not only
HFC- 134a itself but the combined impact of a class of
hydrofluorocarbons that could be used as HCFC substi
tutes after 2030). These are important only insofar as
their radiative forcing is concerned, since they do not
significantly deplete stratospheric ozone (see Section
13.4). The use of shorter-lived or less infrared active
gases could reduce the estimated radiative forcing from
such compound, Figure 13-1 shows a steep incase in
the projected HFC concentrations in the latter part of the
twenty-first century; the effect of such increases on radi-
ative forcing is discussed further in Chapter 8
Cases B through G demonstrate impacts relative to
case A of continued CFC production outside internation-
al agreements, an accelerated HCFC phaseout, a methyl
bromide phaseout or a 2%/year increase in industrial
methyl bromide use, and complete recapture (as opposed
to recycling) of halons, CFCs-11, -12, and-113 banked
in existing equipment (i.e., refrigeration, air-condition- '
ing, fire extinguishants). Recapture illustrates the impact
of potential use of non-ozone depleting substitutes that
could reduce future emissions of these compounds. Case
H is presented in order to compare the current (Copen-
hagen) agreements to the earlier London Amendments
'HCFC and HFC Mixing Ratios forCaseA
lOOOi -
I/' ' \
-1"7 ' ' l-l I I ill |\ |MI |
O.I
1975 2000 2025 2050 2075 2100
Year
The bottom panel of Figure 13-2 shows the contri-
butions of the various gases considered here to the
equivalent tropospheric chlorine versus time for case A
It shows that anthropogenic sources of chlorine and bro-
mine are believed to have contributed much of the
equivalent chlorine in today's troposphere! Direct mea-
surements of chlorinated and brominated source gases
have been obtained near the inflow region at the tropical
U-opopause on recent aircraft missions (Schauffler et ai,
1993). These reveal abundance:'; of halocarbon source
gases very close to those shown in Figure 13-2 for 1992
Further, concurrent measurements on-board the same
aircraft confirm that HC1 emitted at low altitudes from
volcanoes, oceans, and other sources makes a very small
contribute to the total chlorine injected in the tropical
stratospheric inflow region (less than 0.1 ppbv Schauf-
fler etal, 1993). Figure 13-2 also shows that equivalent
chlorine is expected to maximized the troposphere in
1994 under current agreements, and would return to lev-
els near those believed to be present when Antarctic
ozone depletion first became statistically significant
compared to variability (i.e., near 1980) around the mid-
dle of the twenty-first century if the emissions
corresponding to case A are adoptbd. Since equivalent
13.9
-------
ODPs, GWPs and CI-Br LOADING
Table 13-2. Scenarios for future chlorine and bromine loading.
Case A
Global Compliance to Montreal Protocol as Amended and Adjusted in
Jopenhagen (Protocol): CFCs, carbon tetrachloride, and methyl
chloroform phased out in developed countries by 1996. Consumption
n 1992 for Article 5 countries is assumed to be 5% of 1992 global
production, growing to 10% of 1992 global production by 1996,
constant to 2002, and a linear decline to zero by 2006. HCFC
emissions based on U.S. EPA analysis as described in the test, and are
consistent with limits under the Protocol. The halons in existing
equipment (the "bank") as derived from McCulloch, 1992, are emitted
in equal amounts over the period 1993 - 2000 for halon-1211 and the
period 1993-2010 for halon-1301. Methyl bromide emissions are
assumed constant over the period 1994 - 2100.
Description
CaseB
Production and Consumption Outside Protocol: Assumes continued
production of CFC and carbon tetrachloride production at a rate equal
to about 20% of 1992 global production through 2002 and then a
linear decrease to zero by 2006. All other emissions as in case A.
CaseC
Destruction of Halon Bank: Assumes all halons contained in existing
equipment are completely recovered after 1994. All other emissions as
in case A
CascD
HCFC Early Phase-Out: Assumes that HCFC emissions cease on a
global basis in 2004. All other emissions as in case A.
CascE
Methyl Bromide Increase: Assumes a 2%/year increase in agricultural
emissions of methyl bromide until global agricultural emissions reach
a maximum value three times that of the present. All other emissions
as in case A. -
CaseF
Methyl Bromide Phase-Out: Assumes a 100% phase-out in all
anthropogenic sources of methyl bromide emission except biomass
burning (see Chapter 10) by 2001.' All other emissions as in case A.
CaseG
Destruction of CFC Bank: Assumes that all banked CFC-11 and CFC
12 in hermetically sealed and non-hermetically sealed refrigeration
categories are completely recovered in 1995 and hence never released
to the atmosphere. All banked CFC-113 is also assumed to be
completely recovered. All other emissions as in case A.
CaseH
London Amendments: Global compliance with the 1990 London
Amendments to the Montreal Protocol rather than the 1992
Copenhagen Amendments. ,
13.10
-------
ODPs, (SWF's and CI-Br LOADING
Equivalent Effective Stratospheric Chlorine
Case A
1940 I960 I960 3000 2030 2040 2060 2080 2100
Year
5000
4500
Tropospheric Chlorine Loading
Case A
1940 I960 1980 2000 2020 204O 2060 2080 2100
Year
Figure 13-2. Contributions of various gases to the
equivalent tropospheric (bottom) and stratospheric
(top) chlorine versus time for case A.
tropospheric chlorine loading is expected to maximize in
1994, further controls would not reduce peak concentra-
tions provided that global emissions continue to follow
the requirements of the Protocol and its Amendments.
However, consumption outside current Protocol agree-
ments could increase the concentration.
13.3.2 Equivalent Effective Stratospheric
Chlorine
Tropospheric chlorine loading alone does not de-
termine the impact of a compound upon ozone loss,
especially in the key region below about 25 km. Com-
pounds that dissociate less readily within the stratosphere
than others deliver less reactive chlorine, thereby de-
creasing their effectiveness from that indicated by their
tropospheric loading. Examples of this behavior include
HCFC-22 and HCFC-142b. Observations show that
about 65% of the input of these gases to the stratosphere
remains undissociated by the time they exit the strato-
sphere (see Solomon et ai, 1992), substantially reducing
their impact on stratospheric ozone as compared to gases
such as CCU, which undeirgo nearly complete dissocia-
tion while in the stratosphere. Here we evaluate the
chlorine release in the lower stratosphere (below 25 km),
since this is the region where most of the column-inte-
grated ozone loss in the present atmosphere is observed
to take place (WMO, 1992 and Chapter 1 of this docu-
ment). The dissociation of many key compounds
relative to a reference gas (CFC-11) in the lower strato-
sphere has been evaluated by Solomon et al. (1992) and
by Daniel et al. (1994) usi0g both observations and model
calculations and is used here to define the equivalent effec-
tive stratospheric chlorine (EESC). In addition, a 3-year
lag between tropospheric emission of halocarbons and
stratospheric ozone impact is assumed, based in part on
tracer studies (e.g.. Pollock et al., 1992). Using these
factors together with the estimate of a of 40 as discussed
above, we define an "equivalent effective stratospheric
chlorine" abundance that characterizes the impact of
each source gas upon lower stratospheric ozone (similar
to the "free halogen" defined in WMO, 1992). This def-
inition is the same as that used for time-dependent ODPs
discussed in Section 13.4.5.
The top panel of Figure 13-2 displays cumulative
equivalent effective stratospheric chlorine for case A.
Curves are lowered compared to tropospheric chlorine
loading due to incomplete dissociation of the com-
pounds. Peak chlorine loading occurs in 1997 as
determined by the peak trcipospheric loading that oc-
curred three years earlier (bottom panel), suggesting that
the maximum risk of ozone depletion has been deter-
mined by emissions occurring prior to 1995, assuming
case A emissions. 1
Figure 13-3 shows the equivalent effective strato-
spheric chlorine represented by case A (Copenhagen
Amendments) compared to the provisions of the original
1987 Montreal Protocol. The figure also illustrates what
could have happened with no international agreements
13.11
-------
ODPs, GWPs and CI-Br LOADING
Equivalent Effective Stratospheric Chlorine
15000
12000
90OO
a
6000
3000
No •
Protocol /
t
I.
/Montreal
/ Protocol
/ X
/ X
X
Copenhagen
_Amendrr>errts
i
1950 1975 2000 2025 2050 2075 2100
Year
Figure 13-3. Estimated equivalent effective strato-
spheric chlorine represented by case A (Copenhagen
Amendments) compared to the provisions of the
original 1987 Montreal Protocol, and a case with no
international agreements on ozone-depleting gas-
es (where a 3%/year increase in global emissions
of CFCs and methyl chloroform was assumed, less
than known trends up to that time).
on ozone-depleting gases (where a 3%/year increase in
global emissions of CFCs and methyl chloroform, was
assumed, less than known trends up to that time). The
figure shows that without international agreements,
equivalent effective stratospheric chlorine would likely
reach values about twice as large as today's levels by
2030 and about three times today's levels by about 2050.
Even with the provisions of the original Montreal Proto-
col, equivalent effective stratospheric chlorine would be
likely to double by about the year 2060. Instead, under
the current provisions, the stratospheric abundances of
ozone-depleting gases are expected to begin to decrease
within a few years.
One important measure of future ozone loss is the
time integrated equivalent effective chlorine (pptv-year)
to be expected from January 1, 1995, through the time
when ozone depletion is likely to cease (i.e., the integrat-
ed future ozone loss). Ozone depletion first became
observable in a statistically significant sense in about
1980, making the return to equivalent effective chlorine
for that year a reasonable proxy for the point where, all
other things being equal, ozone depletion is likely to
cease. For case A, for example, that point in time (re-
ferred to here as x) is expected to be reached in 2045.
Table 13-3 presents the corresponding years for the other
scenarios considered here. For evaluating cumulative
long-term ecological impacts due to ozone depletion, it
may also be useful to consider a similar integral begin-
ning not in 1995 but in 1980 (thus integrating over the
entire period when ozone depletion has been observed).
A similar definition was used in WMO (1992), except
that tropospheric values in 1985 were chosen as the ref-
erence point below which ozone depletion was assumed
.to cease, and the integral was performed from that point
onwards rather than from 1995 onwards. Table 13-3
compares the percent differences from the base case A
for each scenario for the following quantities: a) inte-
grated equivalent effective stratospheric chlorine
loading from 1995 until year x (the point when EESC
drops below 1980 levels) and b) integrated equivalent ef-
fective stratospheric chlorine loading from 1980 until
year \. Positive values denote integrated EESC levels
that exceed the base case, while negative values indicate
integrated EESC levels below the base Copenhagen sce-
nario. The magnitudes of natural sources of chlorine and
bromine (e.g., from CH3C1 and CH3Br) do not influence
these calculations, provided that they are not changing
with time.
13.4 OZONE DEPLETION POTENTIALS
13.4.1 Introduction
Understanding of atmospheric chemical processes
and the representation of these processes in models of
global atmospheric chemistry and physics have im-
proved since the WMO (1992) assessment. In particular,
prior modeling analyses of ODPs were based largely on
calculations including only gas phase chemistry, al-
though a few calculations were carried out that included
some of the chemistry occurring on background sulfuric
acid aerosols. Some of the models used in the analysis
presented here include representations of polar vortex
processes (albeit in highly parameterized fashions) as
well as most effects of heterogeneous chemistry on
background sulfuric acid (but not volcanic) aerosols.
The models still tend to underestimate the absolute
73.72
-------
ODPs, GWPs and CI-Br LOADING
Table 13-3. Results of scenario calculations: integrated EESC differences (from case A) and the year
when EESC drops below 1980 levels.
Scenario
A - Copenhagen
B - Production outside of
Protocol
C - Destruction of halon
bank
D- HCFC early phase-out
E - Methyl bromide
increase
F - Methyl bromide
phase-out
G- Destruction of CFC
bank
H - -London Amendments
Year (x) when EESC is
expected to drop below
1980 value
2045
2048
2043
2044
2057
2040
2044
2055
Percent difference in
X
/EESCdt from
1995 '
case A.
0.0
+9
ii
-10
I
•1
-5 . i
• +u i
1
-13
-3
+38
Percent difference in
J EESCdt from
1980
case A.
0.0
+7
-7
-4
+9
-10
-2
+30
ozone losses in the lowest part of the stratosphere (see
Chapter 6); these limitations can affect ODPs, especially
those for bromocarbons. The semi-empirical approach
developed by Solomon etal. (1992) implicitly accounts
for observed ozone destruction profiles both inside and
outside of the polar vortices that are believed to reflect
heterogeneous processes.. While the semi-empirical ap-
proach is based upon limited data at low latitudes and
high altitudes (above about 25 km), these limitations oc-
cur in regions that are believed to make relatively small
contributions to the globally averaged ozone loss and
hence to the ODP. Based upon these improvements in
understanding, we did not explicitly evaluate chlorine
loading potentials (a simpler but less complete index) in
this report (see WMO, 1992).
13.4.2 Relative Effectiveness of Halogens in
Ozone Destruction
A range of molecules are being considered as sub-
stitutes for the chlorofluorocarbons and halons. Some of
these are non-halogenated compounds that result in no
ozone loss, but others contain iodine or fluorine and
could in principle deplete stratospheric ozone. It is also
of interest to review the effectiveness of bromine relative
to chlorine for ozone loss, which is critical for the ODPs
of the halons and CH3Br.
13.4.2.1 FLUORINE
It has long been assumed that atomic fluorine re-
leased from chlorofluorocarbons would be tied up in the
form of HF and therefore unable to participate in catalyt-
ic cycles that significantly; deplete ozone.. For example,
Stolarski and Rundel (1975) concluded that the catalytic
efficiency/ foe ozone depletion by fluorine atoms is less
than 10-4 that of chlorine in the altitude range from 25 to
50 km. While recent estimates of the equilibrium con-
stant, Keq, for F + O2 « FO2 published in JPL (1992)
suggest that FO2 could have an appreciable thermal dis-
sociation lifetime of the order of 1 day or longer in the
stratosphere, it is unlikely that FOX compounds can lead
to significant ozone loss, as discussed in Chapter 12.
Direct observations of HF aod fluorine source gases
(e.g., Zander et al., 1992) support the view that there are
no large unrecognized reservoirs for fluorine. As in pre-
vious reports, we assume here that atomic fluorine and
related species do not cause significant ozone depletion.
In contrast to atomic fluorine, FO, and FO2, it has,
however, recently been suggested (Li and Francisco,
13.13
-------
ODPs, GWPs and CI-Br LOADING
1991; Biggs et ai, 1993) that the CF3OX group could be
stable enough to undergo catalytic cycles that deplete
ozone at a significant rate before being decomposed to
less stable products that form HE It has also been sug-
gested that the FC(O)OX group could undergo similar
chemistry (see Chapter 12). These free radical groups
are produced upon decomposition of a number of HFCs
and HCFCs, and even a few CFCs. Notably, it was sug-
gested that such processes could compromise the use of
HFC-134a as a substitute that does not damage the
ozone layer. Briefly, the key chemical reactions are:
CF3CFH2 (HFC-134a) + OH (multi-step) ->
+ other products (13-1)
(13-2)
(13-3)
(13-4)
50
M-»CF3O2
» CF3O + 2 O2
CF3O + O3 -> CF3O2 + O2
The last two reactions constitute a catalytic cycle analo-
gous to the OH and HO2 reactions with ozone, and could
in principle be an effective ozone loss cycle in the lower
stratosphere. The key factors in terminating this catalyt-
ic chain are reactions that can break down the CF3 group,
forming either stable products or products that rapidly
decompose to produce HF. Two such reactions have
been identified:
CF3O + NO-» CF20 + FNO (13-5)
CF3O + CH4 -» CF3OH + CH3 (13-6)
Chapter 12 discusses recent measurements of these and
other relevant kinetic rate constants in considerable de-
tail. Direct laboratory measurements coupled with
model calculations have shown that the chain-terminat-
ing reactions above are sufficiently fast, and the chain-
propagating reactions sufficiently slow, that the Ozone
Depletion Potentials relating to the presence of a CF3
group are essentially negligible. Recently, Ravishankara
et al. (1994) and Ko et al. (1994a) have examined the
implications of these processes for the effectiveness of
CF3 radical groups for ozone loss relative to chlorine.
Figure 13-4 shows the calculated efficiency of CF3 as
compared to chlorine from the Garcia-Solomon model
used in the study of Ravishankara et al. for midlatitudes
in winter. The figure illustrates that current laboratory
measurements imply that the CF3 group is at most about
45 -
40
a>
o
25
15
38°N, Winter
'
I I
C»
J L
;\ I (minimum)
\X I (max)
\\ \
•,v \
••\ \
'•\ \
:\ \
=4 \
\\ \
'• \ \
I '•.
I0"6 ICf4 IO"2 10° .I02 I04
Effectiveness for Ozone Destruction Relative to Chlorine
Figure 13-4. Calculated effectiveness of CF^, bro-
mine, and iodine in ozone destruction at midlatitudes
relative to chlorine (based on results from Garcia-
Solomon model as discussed in text).
1000 times less effective than chlorine for ozone de-
struction at 20 km in midlatitudes. While higher local
values might be obtained in polar winter (where NO
abundances are very, small), the impacts of CF3-related
reactions on the globally averaged ODPs of CF3-con-
taining chlorofluorocarbons (such as CF3Cl) and
hydrochlorofluorocarbons (such as CF3CHC12) are be-
lieved to be negligible, and the ODPs of HFCs such as
HFC-134a and HFC-23 are highly likely to be less than
IxlO-3 based upon current kinetic data (Ravishankara et
al., 1994).
13.4.2.2 BROMINE
The chemistry of atmospheric bromine is dis-
cussed further in Chapter 10. The understanding of the
relative roles of bromine and chlorine in depleting ozone
was discussed by Solomon et al. (1992), who noted that
in situ and remote sensing measurements of CIO, BrO,
and OC1O strongly suggest that bromine is about 40
times more efficient than chlorine for Antarctic ozone
loss. Assuming that the rate-limiting steps for ozone loss
in the Antarctic are the reactions CIO + CIO and CIO +
BrO, the value of a for Antarctic ozone loss can be de-
rived as follows:
13.14
-------
ODPs, C5WPs and CI-Br LOADING
a
2k(BrQ)(C10)/(Bry)
2k(C10)(C10) + 2k(BrO)(C10) / (Cly)
(13-7)
where the denominator represents the rate of ozone loss
due to chlorine compounds per atom of chlorine avail-
able (i.e., Cl released from all source gases, denoted here
as Cly) and the numerator represents the rate of ozone
loss due to bromine compounds per atom of bromine
available (Bry). Since the reaction CIO + CIO is believed
to account for about 75% of the Antarctic ozone loss
while.CIO + BrO accounts for about 25% (see Solomon
et ai, 1992 and references therein) and Cly is about 2.5
ppbv while Bry is about 15 pptv in this region, the value
of a for Antarctic ozone loss is about 40. Salawitch et al.
(1990, 1993) pointed out that the lower absolute abun-
dances of CIO observed in the Arctic as compared to
Antarctica implies that bromine will be more effective
for ozone loss there (i. e., CIO + BrO will be more impor-
tant compared to CIO + CIO).
Recent laboratory studies have confirmed and ex-
panded understanding of the important role of bromine.
Poulet et al. (1992) have shown that the kinetic rate con-
stant for the reaction of BrO + HO2 is about six times
faster than previously believed at room temperature; this
has been confirmed by the measurements of Bridier et
al. (1993). As noted in WMO (1992), the importance of
bromine for ozone loss could be substantially dimin-
ished if as much as 10% of the reaction between BrO +
HO2 were to yield HBr at the rate indicated by Poulet et
al. (1992), while it would be enhanced if less than a few
percent HBr is produced. The latter appears to be true
based upon the study of Mellouki et al. (1994), who
showed that the yield of HBr from this reaction is likely
to be below 0.1% even at stratospheric temperatures
based on new measurements and thermochemical data, a
result consistent with modeling studies of the BrO gradi-
ent (Garcia and Solomon, 1994). Figure 13-4 shows the
calculated effectiveness of bromine for ozone destruc-
tion relative to chlorine based upon the above
photochemistry from the model of Garcia and Solomon
(1994). The figure suggests that bromine is roughly 100
times more effective in the region of peak observed
ozone loss (near 20 km). Very similar results have also
been calculated with the Lawrence Livermore National
Laboratory (LLNL) two-dimensional model. The figure
illustrates that model calculations of the OOP for bro-
mine-bearing compounds'are likely to be quite sensitive
to the altitude profile of O2:one destruction. Since
present models tend to underestimate the observed
ozone losses in the lowest part of the stratosphere (see
Chapter 6), where bromine is particularly efficient for
ozone loss, this figure implies that the model-derived
globally averaged values of a (weighted by the ozone
loss distribution) will also be underestimates assuming
present photochemical schemes.
Bromine's effectiveness for ozone loss in the low-
er stratosphere is related to the fact that a large fraction
of the available Bry resides in the ozone-depleting forms
of Br and BrO. In contrast only a very small fraction of
available Cly resides in Cl and CIO except in the special
case of polar regions. Thus, since all halogen atoms are
very reactive (e.g., with atomic oxygen, HO2, and each
other), bromine chemistry's effectiveness relative to
chlorine will generally be driven by the fact that the BrO/
Bry ratio is on the order of 50-100 times larger than the
ClO/CIy ratio in the lower stratosphere outside of polar
regions. This in turn implies that the value of a is not
very sensitive to which reactions are the dominant rate-
limiting steps in ozone destruction, at least for current
photochemical schemes (e.g., CIO + BrO, HO2 + BrO,
H02 + CIO, etc.). !
13.4.2.3 IODINE
(
The ability of reservoir molecules to sequester
halogen radicals and thereby reduce their impact on
ozone is inversely related to the size of the halogen atom.
Thus fluorine rapidly forms HF, while chlorine forms
HC1 and C1ONO2. The bromine reservoirs (HBr and
BrONO2) are weakly bound, making BrO and Br effec-
tive ozone-destroying species as shown above. Iodine
reservoirs such as HI, IONO2, and others are known to
be very readily dissociated by photolysis or reaction
with OH, rendering any iodine that reaches the strato-
sphere at least as effective ais bromine for ozone loss and
very probably much more so. However, iodine source
gases are very short-lived because of the relatively weak
carbon-iodine bond. If the iodine source gases are short-
lived enough, then anthropogenic releases (particularly
at the surface at midlatitudes) may not reach the strato-
sphere in abundances sufficient to result in significant
ozone loss. In this case, compounds such as CF^l could
represent useful substitutes'for the halons.
13.15
-------
ODPs, GWPs and CI-Br LOADING
The chemistry of iodine in the troposphere was
discussed in detail by Chameides and Davis (1980). Re-
cently, Solomon et al. (1994a, b) have considered the
impact of iodine on stratospheric ozone compared to
chlorine, based mainly on the'iodine photochemistry
considered in the kinetic evaluation of Atkinson et al.
(1992). Solomon et al. (1994a, b) showed that current
photochemical schemes imply that iodine is at least as
effective as bromine for ozone destruction based upon
the measured rate for HC>2+IO (shown in Figure 13-4 as
Iodine [minimum]). In addition, Solomon etal. (1994b)
emphasized that several key chemical processes relating
to iodine-catalyzed ozone destruction have not yet been
quantified in laboratory studies, notably IO + CIO and
IO + BrO. If these reactions were to take place relatively
rapidly, iodine could be as much as 2000 times more ef-
fective than chlorine for ozone destruction near 20 km
(denoted as Iodine [max] in Figure 13-4). This proposed
chemistry does not significantly change the value of a,
for the reasons discussed above. In combination with
anthropogenic trends in CIO and BrO, as little as 1 pptv
of iodine in the lower stratosphere due to the very large
natural sources of compounds such as methyl iodide
could be significant for lower stratospheric ozone loss
(Solomon et al., 1994b). These considerations are taken
into account in the estimate of the OOP for CFjl present-
ed in Solomon et al. (1994a) and later in this chapter. In
spite of these large efficiencies, the very short lifetime of
CFal (less than 2 days; see Solomon et al., 1994a) results
in an estimated upper limit for the steady-state OOP for
surface emissions of this compound of only 0.008. Oth-
er iodine-bearing compounds, such as C2p5l, would
likely have similar ODPs.
13.4.3 Breakdown Products of HCFCs and
HFCs
In the calculation of the ODPs for HCFCs present-
ed here, it is assumed that chlorine atoms will be
promptly released (and hence able to participate in
ozone destruction) once the parent molecule is broken
down. Concern has been raised that the ODPs of some
HCFCs could be enhanced if the tropospheric break-
down products contain chlorine and have atmospheric
lifetimes comparable to or longer than the precursor
HCFC (WMO, 1990, 1992) and thus potentially be
transported to the stratosphere. Particular attention has
been focused on the carbonyl and PAN-like compounds.
The chemistry of these intermediates is discussed in de-
tail in Chapter 12, where it is shown that photolysis and
heterogeneous removal (in clouds and rain) likely makes
the tropospheric abundances of these intermediates too
small to affect ODPs or GWPs.
On the other hand, Kindler et al. (1994) showed
that the stratospheric lifetime of the phosgene (COC^)
produced by the dissociation of such compounds as CCU
and CHjCC\3 is long enough to imply a reduction of per-
haps 10-15% in the ODPs for CCLt and CH3CC13.
Similarly, fluorophosgene (COFC1) is a product of the
degradation of HCFC-14 Ib. The lifetime of this species
is also believed to be rather long in the stratosphere, sug-
gesting a similar reduction in the ODP of HCFC-14Ib.
These chemical processes have not been included in the
ODP estimates discussed below.
13.4.4 Model-Calculated and Semi-Empirical
Steady-State ODPs
Model-derived ODPs have been determined for a
range of compounds using the two-dimensional models
at LLNL (D. Wuebbles and K. Patten), Atmospheric and
Environmental Research, Inc. (AER; D. Weisenstein and
M. KoX and Universita' Degli Studi-L'Aquila (G. Vis-
conti and G. Pitari). In addition, the ODPs of some
bromocarbons were evaluated in the Oslo model (I. Isak-
sen et al.) and some HCFCs were considered in the
Indian Institute of Technology (IIT)/Delhi one-dimen-
sional model (M. Lai et al). The National Oceanic and
Atmospheric Administration/National Center for Atmo-
spheric Research (NOAA/NCAR) two-dimensional
model was used to analyze the ODPs for HFC-134a,
HFC-23, HFC-125, and CF3I (Ravishankara etal., 1994;
Solomon et al., 1994a). Each of these models used up-
dated kinetics (based primarily on JPL, 1992), with the
exception that the L'Aquila results do not include the
new BrO + HO2 rate. These models also account for the
effects of heterogeneous chemistry on background
stratospheric sulfate aerosols and most include a repre-
sentation of polar-vortex processes. The ODPs presented
in Table 13-4 use results from the models normalized to
the atmospheric lifetimes in Table 13-1. They agreed to
within 10% in most cases and within 30% in all cases
examined; the results from reporting models were aver-
aged. In the AER 2-D model (D. Weisenstein, private
13.16
-------
ODPs, GWPs and CI-Br LOADING
Table 13-4. Steady-state ODPs derived from 2-D models and from the semi-empirical approach
ODPs are normalized based on recommended atmospheric lifetimes in Section 13.2.
Trace Gas
CFC-11
CFC-12
CFC-113
CFC-114
CFC-115
CC14
CH3CC13
HCFC-22
HCFC-123
HCFC-124
HCFC-141b
HCFC-142b
HCFC-225ca
HCFC-225cb
HFC-134a
HFC-23
HFC- 125
CH 3Br(l. 3 yr lifetime)
CF3Br(H-1301)
CF2ClBr(H-l211)
CF3I
CH3C1
Model-Derived OOP
1.0
0.82
0.90
0.85
0.40
1.20
0.12
0.04
0.014
0.03
0.10
0.05
0.02
0.02
< 1.5x10-5
< 4xlO-4
< 3x10-5
0.64
12
5.1
0.02
i
Semi-Empirical OOP
1.0
0.9
0.9
0.12
0.05
0.02
0.1
0.066
0.025
! 0.03
[< 5x10-4
0.57
i 13
', 5
!j
1
< 0.008
>
communication, 1993J, the derived OOP for CH3Br in-
creased by 33% due to the change from the old to the
new kinetic rate constant for the reaction between BrO
and HC>2, illustrating the key role of this reaction as dis-
cussed above. The factors influencing the OOP for
CH3Brand their possible uncertainties are discussed fur-
ther in Chapter 10. The best estimate of the lifetime for
CH3Br is about 1.3 years as discussed in Chapter 10,
rather than the value of 2 years used in the WMO (1992)
report. Thus, the increased chemical effectiveness of
bromine for ozone loss is approximately cancelled by
the decreased lifetime in deriving an OOP for CH3Br.
Model-derived ODPs for the long-lived CFCs and ha-
lons shown in Table 13-4
-------
ODPs, GWPs and CI-Br LOADING
effectiveness of bromine relative to chlorine for ozone
loss in this analysis was assumed to be 40; as indicated in
Section 13.4, this value is likely to be too low in the re-
gion where bromine emissions are most effective in
destroying ozone at midlatitudes, suggesting that the
semi-empirical ODPs for CHsBr and the halons may be
underestimated. A value of a of 80 is plausible in the
lower stratosphere (see Chapter 10 and Garcia and So-
lomon, 1994), and would approximately double the
ODPs of these compounds.
13.4.5 Time-Dependent Effects
While steady-state Ozone Depletion Potentials de-
scribe the integrated impact of emission of a halocarbon
upon the ozone layer compared to CFC-11, it is also of
interest to consider the time dependence of these effects
(WMO, 1990, 1992; Solomon and Albritton, 1992).
Time-dependent ODPs can be used to provide insight
into the effect of a mix of compounds upon the short-
term future of the ozone layer (e.g., the next few decades,
when peak chlorine and bromine loading are expected to
occur), while steady-state ODPs indicate integrated ef-
fects over longer time scales. We describe below in more
detail than in previous reports the physical processes that
control the expected time dependence of ODPs for vari-
ous chemicals. We then present updated time-dependent
Ozone Depletion Potentials for several molecules of in-
terest based upon new kinetic information and lifetimes
as discussed in this report.
A simple semi-empirical framework for under-
standing the physical reasons for time-dependence of
ODPs was presented by Solomon and Albritton (1992),
who showed that the following equation can be used to
approximate the time-dependent ODP at any point in the
stratosphere:
The term in brackets, {Fx/Fcpc-i i )> denotes the fraction
of the halocarbon species, x, injected into the stratosphere
that has been dissociated compared to that of CFC-11
(obtained from measurements of both). Mx, MCFC-I i. /cx.
and TCFC-II indicate the molecular weights and atmo-
spheric lifetimes of species x and CFC-11, respectively,
while nx is the number of chlorine or bromine atoms in
the molecule (and note that CFC-11 contains 3 chlorine
atoms per molecule). Also, tj is the time required for a
molecule to be transported from the surface to the region
of the stratosphere in question, and t is time. In the fol-
lowing figures, the time refers to the time since reaching
the lower stratosphere at middle-to-high latitudes (which
is believed to be on the order of three years). In princi-
ple, the above equation should be integrated over the
entire stratosphere in order to derive the globally aver-
aged time-dependent ODP. In practice, however, the
ozone column depletion observed in the current atmo-
sphere is dominated by the region below 25 km. Further,
mixing processes imply compact linear correlations be-
tween many of the long-lived halocarbon source gases in
this region (Plumb and Ko, 1992), making the term in
brackets, {Fx/FcpC-!l K verv nearly a constant over
broad regions of the lower stratosphere (see Daniel et al.,
1994).
Using the above equation, together with the re-
vised lifetimes of Table 13-1, updated values of {Fx/
FCFC-I i) where available from Daniel et al. (1994), and
a value of a of 40 for bromocarbons and 2000 for io-
docarbons, semi-empirical time-dependent ODPs were
deduced. In addition, the instantaneous (i.e., not inte-
grated) relative ozone loss was also considered. Figure
13-5 shows instantaneous time-dependent relative ozone
loss rates (compared to CFC-11) for several molecules
of interest here. The time axis on the figure refers to the
time since reaching the stratosphere, not the total time
(which is about 3-5 years longer; see Pollock et al.,
1992). The instantaneous ozone loss rates relative to
CFC-11 for the first few years are determined largely by
the values of a for bromocarbons or iodocarbons and by
the values of {Fx/FCFC-ll} and nx f°r chlorocarbons.
Over longer time scales, the short-lived compounds are
removed from the atmosphere, and the slope of their de-
cay depends upon the relative values of Tx and TCFC-U-
Note, for example, that HCFC-141b (which contains 2
chlorine atoms) initially destroys roughly 2/3 as much
ozone as CFC-11. It has a lifetime of about 10years,and
therefore its instantaneous ozone loss drops to very
small values within a few decades. The ozone-depleting
effects of pulsed injections of compounds with shorter
lifetimes (such as HCFC-123) decay much faster. A
13.18
-------
ODPs, GWPs and CI-Br LOADING
100
U>
o _
go
o u.
10
c o>
o OC
.01
.001
CH3Br
S. ""V
"-—**-. .^. VHCFC-l4lb
\ ~~"\^
HCFC-123 \ \
1 \ *N
» \HCFC-22
' ' ' • '• Ai
Jb 1
10 100
Time (years)
1000
Figure 13-5. Instantaneous time-dependent rela-
tive ozone loss rates (compared to CFC-11) for
several compounds of interest. Note that the x-axis
refers to the time since reaching the stratosphere,
not the total time.
compound with a lifetime longer than that of CFC-11
(such as CFC-113) has an impact on the ozone layer rel-
ative to CFC-11 that grows for time scales longer than
the 50-year lifetime of CFC-11, because of the decay of
the reference gas. The behavior of CH3Br is qualitative-
ly similar to that of HCFC-123, but it has a very large
initial ozone impact because of the value of a, making its
relative ozone loss in the first few years close to 10 times
that of CFC-11 (approximately o/3).
The time-dependent Ozone Depletion Potentials
are simply the time integrals of the instantaneous relative
ozone loss rates shown in Figure 13-5. These are illus-
trated in Figure 13-6. Note, for example, the growth of
the OOP for CFC-113 for time scales longer than about
100 years, at which time more CFC-113 remains to de-
stroy ozone than the reference gas, CFC-11. The
time-dependent OOP for a very short-lived gas such as
HCFC-123 has large values for the first five years. How-
ever, by the end of the first five years, HCFC-123 is
destroying very little ozone (Figure 13-5), because it has
been nearly completely removed from the atmosphere.
The reference gas, CFC-11, is continuing to destroy
ozone, so that the cumulative value of the denominator
in Equation 13-8 continues to increase. It is this slow
increase in the denominator that controls when the ODPs
for short-lived gases such as HCFC-123 reach their
steady-state values. The steady-state ODP for HCFC-
123 therefore asymptotes to a value below 0.02 in about
100 years. A calculation of the time-dependent ODPs
for CH3Br using the Oslo model gave values of 5.6, 2.3,
and 1.5 for time scales of 10, 20, and 30 years, respec-
tively, very similar to the siemi -empirical values shown in
Figure 13-6. In the above calculations, a lifetime of 2.0
years was used for CH3Br. The ODPs for this gas would
be about 30% smaller over long time scales if a lifetime
of 1.3 years was employed.
Figure 13-6 includes an upper-limit estimate of the
time-dependent ODP for surface releases of CF3I, based
on the framework described in Solomon et al. (1994a).
The calculated upper limit to the ODP for this gas is
about 0.08 in the first five years and asymptotes to a
value below 0.01 in about 100 years.
Although the ODP concept has primarily been ap-
plied to the relative Effects of halocarbons on
stratospheric ozone, there have also been several recent
attempts to determine ODPs for emissions of other
gases. For example, Ko et al. (1994b) have evaluated an
ODP for chlorine emitted directly into the stratosphere
from launch of the U.S. Space Shuttle. They derive a
time-dependent ODP that]is quite large initially (but is
also dependent on the definition of what constitutes a
mass emission, the choice being emission of HC1 only or
a
"c
CD
O
a.
f—
o
CD
a.
a>
Q
CD
C
O
O
inn
i \j\j
i n
i \j
1
I
. I
.01
001
: ' 1 1 II III) ""I 1 — 1 — 1 I 1 1 1 j 1 1 — | i MI
:__. i CF3Br '•
1 ^^ ' j
' r^> or CF-ClBr :
\s n -j or t
r j — — . ^ j..-- -
cSŁ~^~" — - : CFC-II3 ;
"' "^"<~ ii:s!--r- -L. "~-- HCFC-l4lb
— ^*^ ^^" *x "^™* _ *^"^"— ^""^^ —
\. '"**'"'-•.. ^ — . __HC_FC-22 j
CF3I<>^:'':;-'^;:'=: :
'^ -^ ^HCFC^lZ~j ^_
HCFC-225ca :
1 1 — 1 Mill! ll 1 — 1 i i . i i i i i
10 ; 100
Tin.ie (years)
1000
Figure 13-6. Time-dependent Ozone Depletion
Potentials for several compounds of interest. Note
that the x-axis refers to the time since reaching the
stratosphere, not the total time.
13.19
-------
ODPs, GWPs and CI-Br LOADING
the total fuel load). The effect from the Space Shuttle
decays quite rapidly due to removal of the emitted HC1
from the stratosphere.
Since the ozone layer is believed to respond rela-
tively rapidly to changes in chlorine and/or bromine
loading (time scale of about 3-5 years or less), time-
dependent Ozone Depletion Potentials provide an appro-
priate measure of the expected ozone response to
changing inputs of source gases relative to the reference
molecule. On the other hand, steady-state Ozone Deple-
tion Potentials may be applicable to evaluation of
associated long-term biological impacts, where the eco-
system response may take place over many decades of
exposure to changes in ultraviolet radiation resulting
from ozone changes.
13.5 GLOBAL WARMING POTENTIALS
13.5.1 Introduction
This section addresses the numerical indices that
can be used to provide a simple representation of the rel-
ative contribution of an atmospheric trace gas to
greenhouse warming, drawing heavily on the informa-
tion in the earlier ozone assessments (WMO, 1990,
1992), the climate-system reports of the Intergovern-
mental Panel on Climate Change (IPCC, 1990, 1992,
1994), and recent journal publications. The major objec-
tive of the text that follows is to update the information
on radiative forcing indices. To this end, we describe the
calculations of the indices contained herein, discuss the
sensitivity of the results to some of the specifications and
assumptions, and present the resulting numerical indices
and their uncertainties.
As in the case of ODPs, calculating the relative al-
teration in radiative forcing due to the change in
greenhouse gas A compared to that due to a change in
greenhouse gas B can be evaluated more accurately than
the absolute climate response due a change in a single
greenhouse gas alone. In the following, we briefly dis-
cuss some key factors that contribute to GWPs.
Common to all greenhouse gases are three major
factors - two technical and one user-oriented - that de-
termine the relative contribution of a greenhouse gas to
radiative forcing and hence are the primary input in the
formulation, calculation, and use of radiative forcing
indices:
Factor I: The strength with which a given species
absorbs longwave radiation and the spectral location of
its absorbing wavelengths. Chemical species differ
markedly in their abilities to absorb longwave radiation.
Overlaps of the absorption spectra of various chemical
species with one another (especially H2O, CO2, and, to a
lesser extent, 03) are important factors. In addition,
while the absorption of infrared radiation by many
greenhouse gases varies linearly with their concentra-
tion, a few important ones display nonlinear behavior
(e.g., CO2, CH4, and N2O). For those gases, the relative
radiative forcing will depend upon concentration and
hence upon the scenario adopted for the future trace-gas
atmospheric abundances. A key factor in the greenhouse
role of a given species is the location of its absorption
spectrum relative to the region in the absorption of atmo-
spheric water vapor through which most outgoing
planetary thermal radiation escapes to space. Conse-
quently, other things being equal, chemical species that
have strong absorption band strengths in the relatively
weak water-vapor "window" are more important green-
house gases than those that do not. This is illustrated in
Figure 13-7, which shows how the instantaneous radia-
io-<0
'QI |0
_ie
'Ł io-'2
5
o, |0-I3
'o
Ł ID"14
.i
.1j |0'15
°% IO"16
o IO"17
•^
1 1 1—| 1 1 Hl| 1 1 1 1 1 lll| 1 r~ TT 1 1
C2F6
"~-~^ ^-^ Lifetime~IO,OOOyr
X
^ •. KI .-,
f \ \ ^S^20yr
\ \
[ \ \
\ \ rn
*
-,
\ T~~ -V-2- i
'•_ \ \HFC-l34a^
[ HCFC-225ca\ \~l4yr E
: ~2.5yr \ \ J
[ \ ». ;
1 | .\iiilll 1" 1 1 1 1 HI
10 100
Time (yr)
1000
Figure 13-7. Instantaneous radiative forcing
(W rrr2 kg'1) versus time after release for several
different greenhouse gases. The CO2 decay re-
sponse function is based upon the Bern carbon
cycle model with fixed CO2 concentrations.
13.20
-------
ODPs, GWPs and CI-Br LOADING
live forcings due to the pulse emission of one kilogram
of various long-lived gases with differing absorption
properties change as the concentrations decay away in
time after they have become well mixed (e.g., about a
year after injection into, the atmosphere). The relevant
point here is on the left-hand scale at t = 1, namely, that
the radiative forcing of an equal emission of the various
gases can differ by as much as four orders of magnitude.
Laboratory studies of molecular radiative properties are
a key source of the basic information needed in the cal-
culation of radiative forcing indices. The status of such
spectroscopic data of greenhouse gases is discussed in
detail in Chapter 8 and in Chapter 4 of IPCC (1994).
Factor 2. The lifetime of the given species in the
atmosphere. Greenhouse gases differ markedly in how
long they reside in the atmosphere once emitted. Clear-
ly, greenhouse gases that persist in the atmosphere for a
long time are more important, other things being equal,
in radiative forcing than those that are shorter-lived.
This point is also illustrated in Figure 13-8. As shown,
the initial dominance of the radiative forcing at early
times can be overwhelmed by the lifetime factor at later
times.
The relative roles of the strength of radiative ab-
sorption and lifetimes on GWPs, as shown in Figures
13-7 and 13-8, parallel those of chemical effectiveness
and lifetimes on ODPs, as illustrated in Figures 13-5 and
13-6.
Factor 3. The time period over which the radiative
effects of the species are to be considered. Since many
of the responses of the Earth's climate to changes in radi-
ative forcing are long (e.g., the centennial-scale wanning
of the oceans), it is the cumulative radiative forcing of a
greenhouse gas, rather than its instantaneous value, that
is of primary importance to crafting a relevant radiative
forcing index. As a consequence, such indices involve
an integral over time. Rodhe (1990) has noted that the
choice of time interval can be compared to cumulative-
dosage effects in radiology. IPCC (1990, 1992) used
integration time horizons of 20, 100, and 500 years in
calculating the indices. Figure 13-8 shows the integrals
of the decay functions in Figure 13-7 for a wide range of
time horizons. It illustrates the need for the user of the
radiative forcing indices to select the time period of con-
sideration. A strongly absorbing, but short-lived, gas
like HCFC-225ca will contribute more radiative forcing
in the short term than a weaker-absorbing, but longer-
o
~ 10000
g
'c
"5
c
'E
o
a
.a
^
O
1000
100
10
^2.6
Lifetime ~ 10,000 yr
-I 1—l_U-uJ 1—I i I i i nl i i ! i i i
10 100
Time Horizon (yr)
IOOO
Figure 13-8. Global Warming Potentials (GWPs)
for a range of greenhouse gases with differing life-
times, using CO2 as the reference gas.
lived, gas like N2O; however, in the longer term, the re-
verse is true. Methane is a key greenhouse gas discussed
extensively below; its integrated radiative forcing would
lie below that of N2O and reach a plateau more quickly
because of its shorter lifetime.
The spread of numerical values of the radiative
forcing indices repotted in Section 13.5.2 below largely
reflects the influence of these three major factors. In ad-
dition to these direct radiative effects, some chemical
species also have indirect effects on radiative forcing
that arise largely from atmospheric chemical processes.
For example, important products of the oxidative remov-
al of CH4 are water vapor in the stratosphere and ozone
in the troposphere, both of which are greenhouse gases.
These are discussed in Section 13.5.4.
13.5.2 Radiative Forcing Indices
i
13.5.2.1 FORMULATION
The primary radiative forcing indices used in sci-
entific and policy assessments are the Global Warming
Potential (GWP) and Absolute Global Warming Poten-
tial (AGWP). Other possible formulations are described
and contrasted with those in IPCC (1994).
13.21
-------
ODPs, GWPs and CI-Br LOADING
Global Warming Potential
Based on the major factors summarized above, the
relative potential of a specified emission of a greenhouse
gas to contribute to a change in future radiative forcing,
i.e., its GWP, has been expressed as the time-integrated
radiative forcing from the instantaneous release of 1 kg
of a trace gas expressed relative to that of 1 kg of a refer-
ence gas (IPCC, 1990):
_CV[x(t)]dt
(13-9)
where TH is the time horizon over which the calculation
is considered; ax is the climate-related radiative forcing
due to a unit increase in atmospheric concentration of the
gas in question; [x(t)] is the time-decaying abundance of
a pulse of injected gas; and the corresponding quantities
for the reference gas are in the denominator. The adjust-
ed radiative forcings per kg, a, are derived from infrared
radiative transfer models and are assumed to be indepen-
dent of time. The sensitivity of these factors to some
climate variables (HaO, clouds) is discussed later. As
noted above, ar is a function of time when future changes
in CC>2 arc considered. Time-dependent changes in ax or
lifetimes are not explicitly considered here. The trace
gas amounts, [x(t)] and [r(t)], remaining after time t are
based upon the atmospheric lifetime or response time of
the gas in question and the reference gas, respectively.
The reference gas has been taken generally to be
C02, since this allows a comparison of the radiative
forcing role of the emission of the gas in question to that
of the dominant greenhouse gas that is emitted as a result
of human activities, hence of the broadest interest to pol-
icy considerations. However, the atmospheric residence
time of COa is among the most uncertain of the major
greenhouse gases. Carbon dioxide added to the atmo-
sphere decays in a highly complex fashion, showing an
i initial fast decay over the first 10 years or so, followed by
a more gradual decay over the next 100 years or so, and a
very slow decline over the thousand-year time scale,
mainly reflecting transfer processes in the biosphere,
ocean, and deep ocean sediments, respectively. Because-
of these different time constants, the removal of COi
from the atmosphere is quite different from that of other
trace gases and is not well described by a single lifetime
(Moore and Braswell, 1994). Wuebbles et at. (1994b)
and Wigley (1993) have also noted the importance of un-
certainties in the carbon cycle for calculations of GWPs
when CC>2 is used as the reference. Furthermore, CO^ is
also recirculated among these reservoirs at an exchange
rate that is poorly known at present, and it appears that
the budget of CC>2 is difficult to balance with current
information. As a result, when CC>2 is used as the refer-
ence gas, the numerical values of the GWPs of all
greenhouse gases are apt to change in the future (perhaps
substantially) simply because research will improve the
understanding of the removal processes of CC>2. While
recognizing these issues, Caldeira and Kasting (1993)
discuss feedback mechanisms that tend to offset some of
these uncertainties for GWP calculations.
Absolute Global Wanning Potential
Wigley (1993; 1994a, b) has emphasized the un-
certainty in accurately defining the denominator for
GWP calculations if CC>2 is used as the reference mole-
cule, and suggested the use of "Absolute" or AGWPs
given simply by the integrated radiative forcing of the
gas in question:
. AGWP(x) = J ax • [x(t)] dt W • yr • kg'1 • m'2
(13-10)
The advantage of this formulation is that the index is
specific only to the gas in question. An important disad-
vantage is that the absolute value of radiative forcing
depends upon many factors that are poorly known, such
as the distributions and radiative properties of clouds
(e.g.. Cess etal., 1993).
Based upon the recommendation of the co-authors
of Chapter 1 from IPCC (1994), we use the results from
the carbon cycle model of Siegenthaler and co-workers
("Bern" model) for the decay response of CC>2 for the
GWP calculations presented here. The fast initial (first
several decades) decay of added CC>2 calculated in cur-
rent carbon cycle models reflects rapid uptake by the
biosphere and is believed to be an important improve-
ment compared to that used in IPCC (1990, 1992). This
change in decay decreases the integrated radiative forc-
13.22
-------
ODPs, GWPs and CI-Br LOADING
ing of CO2 and thereby acts to increase the estimated
GWPs of all gases (see IPCC, 1994). We present AG-
WPs for CC>2 needed for conversion of the results to
other units and other CC>2 decay functions (e.g., to show
the impact of the choice of the denominator on GWP
values).
13.5.2.2 SENSITIVITY TO THE STATE OF THE ATMOSPHERE
To provide realistic evaluations of GWPs for spec-
ified time horizons and estimate their uncertainties,
future changes in the radiative properties of the atmo-
sphere must be considered. Some of these changes to the
present state can be estimated based upon scenarios
(e.g., CC>2 concentrations), while others are dependent
upon the evolution of the entire climate system and are
poorly known (e.g., clouds and water vapor). In IPCC
(1990), the composition of the background atmosphere
used in the GWP calculations was the present-day abun-
dances of CO2, CHLj, and nitrous oxide (N2O), which
were assumed constant into the future. However, likely
changes in CC>2, CRj, or N22 are particularly sensitive to changes in
concentration, since the large optical depth of CC>2 in the
current atmosphere makes its radiative forcing depend
logarithmically on concentration (see WMO, 1992 and
Chapter 8 of this document). Thus, the forcing for a par-
ticular incremental change of CC>2 will become smaller
in the future, when the atmosphere is expected to contain
a larger concentration of the gas. In the case of CFfy and
N2O, there is a square-root dependence of the forcing on
their respective concentrations (IPCC, 1990); hence, just
as for CO2, the forcings due to a specified increment in
either gas are expected to become smaller for future sce-
narios. For the other trace gases considered here, the
present and likely future values are such that the direct
radiative forcing is linear with respect to their concentra-
tions and hence is independent of the scenario.
IPCC (1994) showed in detail that the dependence
of the AGWP of CC>2 upon choice of future atmospheric
CC>2 concentrations is not a highly sensitive one. A con-
stant atmosphere at pre-industrial values (280 ppmv)
would yield values different by less than about 20% for
all time horizons. Similarly, the increasing CC>2 concen-
trations in a future scenario stabilizing at 650 ppmv
would yield GWP values that are smaller by 15% or less.
The decreases in the radiative forcing per molecule due
to the increasing CC>2 atmospheric abundance appear to
be opposite in sign to those due to the changed CC>2 de-
cay response (see Caldeira and Kasting, 1993, and
Wigley, 1994a).
IPCC (1994) and this report also considered the
possible evolution of the radiative forcing of CH4 and
N2O and the interplay between the spectral overlap of
these two gases using the IS92a scenario published in the
Annex of IPCC (1992). If the calculations were made
with the IS92a CH4 and N2O scenarios rather than with
the constant current values, the direct GWPs of CH4
would decrease by 2 to 3%, and the 20-, 100-, and 500-yr
GWPs of N2O would decrease by 5, 10, and 15%, re-
spectively. The impact of the adopted future scenarios
for CC>2, CFij, and N2O on the radiative forcing of other
trace species was not considered.
Water Vapor \
While it is likely that water vapor will change in a
future climate state, the effect of such changes upon the
direct GWPs of the great majority of molecules of inter-
est here is expected to be quite small. For example, the
model of Clerbaux et al. (1993) was used to test the sen-
sitivity of the direct GWP for Clit to changes in water
vapor. Even for changes as large as 30% in water vapor
concentration, the calculated GWP of CH4 changed by
only a few percent (C. Granier, personal communication,
1993). For many other gases whose radiative impact
occurs largely in the region where water vapor's absorp-
tion is relatively weak, similar or smaller effects are likely.
Clouds
i
Clouds composed of water drops or ice crystals
possess absorption bands in virtually the entire terrestri-
al infrared spectrum. By virtue of this property, they
modulate considerably the: infrared radiation escaping to
space from the Earth's surface and atmosphere. Since
cloud tops generally have lower temperatures than the
Earth's surface and the lower part of the atmosphere,
they reduce the outgoing infrared radiation. This reduc-
tion depends mainly on cloud height and optical depth. .
The higher the cloud, the lower is its temperature and the
greater its reduction in infrared emission. On the other
hand, higher clouds (in particular, high ice clouds) tend
13.23
-------
ODPs, GWPs and CI-Br LOADING
to have low water content and limited optical depths.
Such clouds are partially transparent, which reduces the
infrared trapping effect.
The absorption bands of several trace gases over-
lap significantly with the spectral features of water drops
and ice crystals, particularly in the "window" region.
Owing to the relatively strong absorption properties of
clouds, the absolute radiative forcing of many trace mol-
ecules is diminished in the presence of clouds. However,
it is important to note that the impact of changes in
clouds upon GWPs depends upon the difference be-
tween the change in radiative forcing of the gas
considered and that of the reference gas, not the absolute
change in radiative forcing of the gas alone. IPCC
(1994) shows that the model calculations of Granier and
co-workers suggest that the presence or absence of
clouds results in changes of the relative radiative forc-
ings of the molecules considered here of at most about
12%. Thus, uncertainties in future cloud cover due to
climate change are unlikely to substantially impact GWP
calculations.
13.5.3 Direct GWPs
New direct GWPs of many gases were calculated
for IPCC (1994) and for this report with the radiative
transfer models developed at the National Center for At-
mospheric Research - NCAR (Briegleb, 1992; Clerbaux
et at., 1993), Lawrence Livermore National Laboratory
- LLNL (Wuebbles et al., 1994a, b), the Max Planck In-
stitutfiirChemie-Mainz(C.Bruhle/a/., 1993; Roehler
al, 1994), the Indian Institute of Technology (Lai and
Holt, 1991, updated in 1993), and the University of Oslo
(Fuglestvedtcra/., 1994). The radiative forcing a-factors
adopted are those given in Chapter 8 of this report and in
IPCC (1994). Some of these values are apt to be amend-
ed in the near future (see Chapter 4 of IPCC, 1994).
Table 13-5 presents a composite summary of those re-
sults. In addition, it presents results from the studies of
Ko et al. (1993) and Stordal et al. (personal communica-
tion, 1994) for SFfi, and from Solomon et al. (1994a) for
CFsI. With the exception of CF3I, all of the molecules
considered have lifetimes in excess of several months
and thus can be considered reasonably well-mixed; only
an upper limit rather than a value is presented for CF3I.
For those species addressed in IPCC (1992), a majority
of the GWP values are larger, typically by 10-30%.
These changes are largely due to (i) changes in the CO2
reference noted above and (ii) improved values for atmo-
spheric lifetimes.
Several new gases proposed as CFC and halon
substitutes are considered here for the first time, such as
HCFC-225ca, HCFC-225cb, HFC-227ea, and CF3I.
Table 13-5 also includes for the first time a full evalua-
tion of the GWPs of several fully fluorinated species,
namely SFg, CF4, C2F6, and C6F14. SF6 is used mainly
as a heat transfer fluid for electrical equipment (Ko et al.,
1993), while CF4 and Ctff, are believed to be produced
mainly as accidental by-products of aluminum manufac-
ture. C6F)4 and other perfluoroalkanes have been
proposed as potential CFC substitutes. The very long
lifetimes of the perfluorinated gases (Ravishankara et
al., 1993) lead to large GWPs over long time scales.
The uncertainty in the GWP of any trace gas other
than CO2 depends upon the uncertainties in the AGWP
of CO2 and the AGWP of the gas itself. The uncertain-
ties in the relative values of AGWPs for various gases
depends upon the uncertainty in relative radiative forc-
ing per molecule (estimated to be about 25% for most
gases, as shown in Chapter 8) and on the uncertainty in
the lifetimes of the trace gas considered (which are likely
to be accurate to about 10% for CFC-11 and CH3CC13
and perhaps 20-30% for other gases derived from them).
Combining these dominant uncertainties (in quadrature)
suggests uncertainties in the direct AGWPs for nearly all
of the trace gases considered in Table 13-5 of less than
±35%. Uncertainties in the AGWPs for CO2 depend
upon uncertainties in the carbon cycle (see Chapter 1 of
IPCC, 1994) and on the future scenario for CO2. The
effect of the latter uncertainty is likely to be relatively
small, as shown in Chapter 5 of IPCC (1994).
The reference gas for the GWPs in Table 13-5 is
the CO2 decay response from the "Bern" carbon cycle
model (Chapter 1 of IPCC, 1994). The GWPs calcula-
tions were carried out with background atmospheric
trace gas concentrations held fixed at 354 ppmv.
The direct GWPs given in Table 13-5 can be readi-
ly converted to other frameworks such as AGWPs,
GWPs for a changing atmosphere, and GWPs using as
reference either a specific carbon cycle model or the
three-parameter fit employed in IPCC (1990, 1992).
Table 13-6 presents the relevant factors to carry out such
conversions:
• To convert to AGWP units, the numbers in Table
13-5 should be multiplied by the AGWP for the
13.24
-------
ODPs,, GWPs and CI-Br LOADING
Table 13-5. Global Warming Potentials (mass basis), referenced to the AGWP for the adopted carbon
cycle model COa decay response and future CO2 atmospheric concentrations held constant at cur-
rent levels. Only direct effects are considered, except for methane.
Species
H-1301
Chemical
Formula
Global Warming Potential
(Time Horision)
20 years 100 years 500 years
CFC-11
CFC-12
CFC-13
CFC-113
CFC-114
CFC-115
HCFCs. etc.
Carbon tetrachloride
Methyl chloroform
HCFC-22 (ftt)
HCFC-141b (ttt)
HCFC-142b (ftt)
HCFC-123 (tt)
HCFC-124 (tt)
HCFC-225ca (tt)
HCFC-225cb (tt)
CFC13
CF2C12
CC1F3
c2F3a3
c2F4a2
c2F5a
CC14
CH3ca3
CF2HC1
C2FH3Q2
C2F2H3a
C2F3HC12
C2F4HC1
C3F5HC12
C3F5HC12
5000
7900
8100
5000
6900
6200
2000
360
4300
1800
4200
300
1500
550
1700
4000 i
8500
1 1700,
5000
9300 ;
9300 i
1400 '
110 I
1700 :
630 ;
2000 ,
93 |
480 j
170 i
530
1400
4200
13600
2300
8300
13000.
500
35
520
200
630
29
150
52
170
CF3Br
6200
5600
2200
Other
HFG-23 (t)
HFC-32 (ttt)
HFC-43-10mee (t)
HFC- 125 (tt)
HFC- 134 (t)
HFC-134a (ttt)
HFC-l52a(tt)
HFC- 143 (t)
HFC-143a(tt)
HFC-227ea (t)
HFC-236fa (t)
HFC-245ca (t)
Chloroform (tt)
Methylene chloride (tt)
Sulfur hexafluoride
Perfluoromethane
Perfluoroethane
Perfluorocyclo-butane
Perfluorohexane
Methane*
Nitrous oxide
Trifluoroiodo-methane
CHF3
CH2F2
C4H2F10
C2HF5
CHF2CHF2
CH2FCF3
C2H4F2
CHF2CH2F
CF3CH3
C3HF7
C3H2F6
C3H3F5
CHC13
CH2C12
SF6
CF4
C2F6
c-C4F8
CeF|4
CH4
N20
CF3I
* Includes direct and indirect components (see Section
(ttt) Indicates HFC/HCFCs in production
(tt) Indicates HFC/HCFCs in production
9200
1800
3300
4800
3100
3300
460
950
5200
4500
6100
1900
15
28
16500
4100
8200
6000
4500
62 ±20
290
<5
13.5.4.2).
now and likely to be widely
now for specialized end use
12100
580 ;
1600 !
3200
1200 ;
1300 .
140 |
290 :
4400 ;•
3300 ;
8000 :
610 1
5 !
9 ]
24900
6300
12500
9100
6800 j.
24.5 ± 7.5
320 i
« 1 ' 1
used (see Chapter
(see Chapter '4 of
9900
180
520
1100
370
420
44
90
1600
1100
6600
190
1
3
36500
9800
19100
13300
9900
7.5 ± 2.5
180
<« 1
4 of IPCC, 1994).
IPCC, 1994).
(t) Indicates HFC/HCFCs under consideration for specialized end use (see Chapter 4 of IPCC, 1994).
13.25
-------
ODPs, GWPs and CI-Br LOADING
Table 13-6. Absolute GWPs (AGWPs) (W m-2 yr pprmrV
Case
,
CO 2, Bern Carbon Cycle Model, fixed CO 2 (354 ppmv)
CO 2, Bern Carbon Cycle Model, S650 scenario
COa, Wigley Carbon Cycle Model, S650 scenario
002, Enting Carbon Cycle Model, S650 scenario
CO2, LLNL Carbon Cycle Model, S450 scenario
CO2, LLNL Carbon Cycle Model, S650 scenario
002, LLNL Carbon Cycle Model, S750 scenario
CO2-like gas, IPCC (1990) decay function, fixed CO2 (354 ppmv)
Time Horizon
20 year
0.235
0.225
0.248
0.228
0.247
0.246
0.247
0.267
100 year
0.768
0.702
0.722
0.693
0.821
0.790
0.784
• 0.964
500 year
2.459
2.179
1.957
2.288
2.823
2.477
2.472
2.848
*Multiply these numbers by 1.291 x 10-'3 to convert from per ppmv to per kg.
adopted Bern carbon cycle model, fixed CO2 (354
ppmv) scenario (i.e.. Line 1 in Table 13-6) and
multiplied by 1.291 x 10-13 to convert the AGWP
of CO2 from per ppmv to per kg.
• To convert to GWP units using one of the other
indicated carbon cycle models and/or trace-gas fu-
ture scenarios, the numbers in Table 13-5 should
be multiplied by the AGWP for the adopted Bern
carbon cycle model, fixed CO2 (354 ppmv) sce-
nario (Line 1) and divided by the AGWP value in
Table 13-6 for the carbon cycle model and/or sce-
nario in question.
• To convert to GWPs that are based on the same
reference as was used in IPCC (1990,1992), the
numbers in Table 13-5 should be multiplied by the
AGWP for the adopted Bern carbon cycle model,
fixed CO2 (354 ppmv) scenario (Line 1) and divid-
ed by the AGWP value in Table 13-6 for the
C02-Uke gas, IPCC (1990) decay function, fixed
C02 (354 ppmv) (i.e., last line in Table 13-6).
13.5.4 Indirect Effects
13.5.4.1 GENERAL CHARACTERISTICS
In addition to the direct forcing caused by injec-
tion of infrared-absorbing gases to the atmosphere, some
compounds can also modify the radiative balance
through indirect effects relating to chemical transforma-
tions. When the full interactive chemistry of the atmo-
sphere is considered, a very large number of possible
indirect effects can be identified (ranging from the pro-
duction of stratospheric water vapor as an indirect effect
of H2 injections to changes in the HC1/C1O ratio and
hence in ozone depletion resulting from CH4 injections).
The effects arising frorr^such processes are diffi-
cult to quantify in detail (see Chapter 2 of IPCC, 1994),
but many are highly likely to represent only small pertur-
bations to the direct GWP and, to global radiative
forcing. As noted above for ODPs, recent work has
shown that the production of products such as fluoro-
and chlorophosgene and organic nitrates from the break-
down of CFCs and HCFCs is unlikely to represent a
substantial indirect effect on the GWPs of those species,
due to the rapid removal of these water-soluble products
in clouds and rain (see Chapter 12 and Kindler et ai,
1994). Similarly, the addition of HCFCs and MFCs to
the atmosphere can, in principle, affect the oxidizing ca-
pacity of the lower atmosphere and hence their lifetimes,
but the effect is completely negligible for reasonable
abundances of these trace gases.
Table 13-7 summarizes some key stratospheric
and tropospheric chemical processes that do represent
important indirect effects for GWP estimates. The cur-
rent state of understanding of these processes is
examined in detail in Chapters 2 and 5 of IPCC (1994).
It is particularly difficult to calculate GWPs of short-
13.26
-------
ODPs, GWPs and CI-Br LOADING
Table 13-7. Important indirect effects on GWPs.
Species
CH4
CFCs, HCFCs,
Bromocarbons
CO
NOX
NMHCs
Indirect Effect
Changes in response times due to changes in tropospheric OH ,
Production of tropospheric O 3
Production of stratospheric H2O i
Production of CO2 (for certain sources)
Depletion of stratospheric 03
Increase in tropospheric OH due to enhanced UV
Production of tropospheric 03
Changes in response times due to changes in tropospheric OH
Production of tropospheric CO 2 i
Production of tropospheric Oj
Production of tropospheric 03 ;
Production of tropospheric CO 2
Sign of Effect
on GWP
+
-
+
+
. +
lived gases with localized sources, such as NOX and non-
methane hydrocarbons. Further, lack of detailed
knowledge of the distributions of these and other key tro-
pospheric gases complicates calculations of indirect
effects relating to tropospheric ozone production (see
Chapters 5 and 7). It is, however, important to recognize
that ozone processes in the upper troposphere are more
effective for radiative forcing than those near the surface
(see Chapter 8), emphasizing chemical processes occur-
ring in the free troposphere.
We present here the indirect GWP effect of tropo-
spheric ozone production only for CH4. Additional
GWP quantification (e.g., for tropospheric ozone precur-
sors such as CO, non-methane hydrocarbons (NMHCs),
and NOX) must await further study of the model inter-
comparisons described in Chapter 2 of IPCC (1994) and
improved field, laboratory, and theoretical characteriza-
tion of the processes involved in tropospheric ozone
production. Reliable radiative forcing indices for gases
that form atmospheric aerosols (e.g., sulfur dioxide,
SO2)- cannot currently be formulated meaningfully,
chiefly because of the lack of understanding of many of
the processes involved (e.g., composition of the aerosols,
radiative properties, etc.) and because of uncertainties
regarding the climate response to the inhomogeneous
spatial distributions of the aerosols (see Chapter 3 of
IPCC, 1994). For the first time, an estimate of the effects
from depletion of ozorie on halocarbon GWPs is also
presented in this chapter, drawing upon (i) the extensive
discussion on ODPs and photochemical considerations
behind them in Section 13.4, (ii) the discussion of the
relationship between radiative forcing due to ozone
change and climate sensitivity in Chapter 8, and (iii) the
available scientific literature.
13.5.4.2 INDIRECT EFFECTS UPON THE GWP OF CH4
Recent research studies of the indirect effects on
i
the GWP of methane include those of Hauglustaine et al.
(1994a, b), Lelieveld and Crutzen (1992), Lelieveld et
al., (1993), and Briihl (l;993). In this report, we consider
those results together with inputs from Chapters 2 and 4
of IPCC (1994). The relative radiative forcing for meth-
ane itself compared to CO2 on a per-molecule basis is
, given in Table 4.2a oflPCC (1994) and is used here.
Eight multi-dimensional models were used to study the
chemical response of the atmosphere to a 20% increase
in methane, as discussed in Section 2.9 of IPCC (1994).
The calculated range of ozone increases from the full set
of tropospheric models considered in that study provides
insight regarding the likely range in ozone production.
13.27
-------
ODPs, GWPs and CI-Br LOADING
Uncertainties in these calculations include those related
to the NOX distributions employed in the various models,
formulation of transport processes, and other factors dis-
cussed in detail in Chapter 2 of IPCC (1994). The
estimated uncertainty in the indirect GWP for CH4 from
troposphcric ozone production given below is based
upon the calculated mid-to-upper tropospheric ozone re-
sponse of the models to the prescribed methane
perturbation at northern midlatitudes and consideration
of the current inadequacies in the understanding of many
relevant atmospheric processes. The calculated ozone
changes from the model simulations derived for a 20%
increase in methane imply an indirect effect that is about
25 ± 15% of the direct effect of methane (or 19 ± 12% of
the total), using the infrared radiative code of the LLNL
model. A similar number is estimated in Chapter 4 of
IPCC (1994). This upper end of this range is close to
that presented in IPCC (1990).
Release of CUt leads to increased stratospheric
water vapor through photochemical oxidation; estimates
of this indirect effect range are on the order of 5% or less
of the direct effect of methane (4% of the total) based on
the discussion in Chapter 4 of IPCC (1994); current re-
sults from the LLNL, NCAR, and Mainz radiative/
photochemical two-dimensional models; and the pub-
lished literature (e.g., Lelieveld and Crutzen, 1992;
Lelieveld etal., 1993; Briihl, 1993; Hauglustaine et al,
1994a. b). We adopt 5% of the direct effect in the table
below, which is smaller than the value quoted in IPCC
(1990).
Each injected molecule of CHj ultimately forms
COa, representing an additional indirect effect that
would increase the GWPs by approximately 3 for all
time horizons (see IPCC, 1990). However, as noted by
Lelieveld and Crutzen (1992), this indirect effect is un-
likely to apply to biogenic production of QHLj from most
sources (e.g., from rice paddies), since the ultimate
source of the carbon emitted as CUt in this case is CC>2,
implying no net gain of carbon dioxide. While non-bio-
genic methane sources such as mining operations do
lead indirectly to a net production of CC>2, this methane
is often included in national carbon production inven-
tories. In this case, consideration of CC>2 production in
the GWP could lead to "double-counting," depending
upon how the GWPs and inventories are combined. As
shown in IPCC (1994), most human sources of methane
are biogenic, with another large fraction being due to
coal mines and natural gas. Thus, the indirect effect of
CC>2 production does not apply to much of the CH4 in-
ventory, and is not included in the table below (in
contrast to IPCC (1990), where this effect was included).
As in Table 13-5, the GWPs were calculated rela-
tive to the CO2 decay response of the Bern carbon cycle
model with a constant current CC>2 and CHLt atmosphere.
Table 13-8 summarizes the composite result for methane
GWPs, its uncertainty, and considers the breakdown of
the effects among various contributing factors. The
ranges in CH4 GWPs shown in Table 13-8 reflect the
uncertainties in response time, lifetime, and indirect ef-
fects, as discussed below. We assume a lifetime of
methane in the background atmosphere of 10 ± 2 years
(which is consistent with the budget given in IPCC,
1994). However, the response time of an added pulse is
assumed to be much longer (12-17 years based upon
Chapter 2 of IPCC, 1994). The total GWPs reported in
IPCC (1990) including indirect effects are within the
ranges shown in Table 13-8. The longer response time
adopted here for methane perturbations is responsible
for a large part of the change in methane GWP values
compared to the nominal values including direct effects
only in the IPCC (1992) report (although the fact that
indirect effects were likely to be comparable to the direct
effect was noted). This change is based entirely on the
analysis presented in Chapter 2 of IPCC (1994) used to
define the methane response time for this report (see
Prather, 1994). The decay response has been thoroughly
tested only for small perturbations around a background
state and continuing input flux approximately represen-
tative of today's atmosphere. It would be different if, for
example, large changes in methane emissions were to
occur in the near future. It is also believed to be sensitive
to other chemical factors such as the sources of carbon
monoxide. The GWP determined in this manner is sim-
ilarly valid for relatively small perturbations, e.g., those
that would be required to stabilize concentrations at cur-
rent levels rather than continuing the small trend (order
1%/year) observed in the past decade (see Chapter 2).
However, the GWP shown in Table 13-8 cannot be used
to estimate the radiative forcing that occurred since pre-
industrial times, when methane concentrations more
than doubled.
13.28
-------
ODPs, GWPs and CI-Br LOADING
Table 13-8. Total GWP for CH4, including indirect effects, referenced to the AGWP computed for the
COa decay response of the Bern carbon cycle model and future COa atmospheric concentrations
held constant at current levels.
GWP
Total CrLt GWP, including indirect effects and 12-17 year
response time
Fraction of total GWP due to tropospheric O 3 change
Fraction of total GWP due to stratospheric H2O change
Time Horizon
20 year 100 year 500 year
42-82
19 ±12%
4%
17-32
19 i: 12%
4%
5-10
19112%
4%.
13.5.4.3 NET GLOBAL WARMING POTENTIALS FOR
HALOCARBONS
Chlorofluorocarbons effectively absorb infrared
radiation and have been estimated to have accounted for
as much as about 25% of the anthropogenic direct radia-
tive forcing of the Earth's climate system over the period
from 1980 to 1990 (IPCC, 1990). Improved understand-
ing of the impact of ozone depletion on global radiative
forcing has, however, markedly altered this picture
(WMO, 1992; IPCC, 1992). It is now clear that the large
ozone depletions observed in the lower stratosphere are
likely to influence temperatures near the tropopause
(Lacis etal., 1990; Ramaswamy etai, 1992), implying
that in addition to their direct greenhouse warming, the
indirect effect of ozone depletion is significant for
estimating the GWPs of ozone-destroying gases. Ra-
maswamy et al. (1992) and WMO (1992) concluded that
the globally averaged decrease in radiative forcing at the
tropopause due to ozone depletion approximately bal-
anced the globally averaged increase in direct radiative
warming in the troposphere related to the direct forcing
due to halocarbons during the decade of the 1980s.
While changes in ozone have been reported in the upper
troposphere (see Chapter 1), these are probably due to
factors other than halocarbon increases (e.g., changes of
CO, NOy, etc.) and do not affect the inference of halocar-
bon GWPs so long as the vertical profile of ozone
depletion can be characterized. If such changes were to
mask the vertical extent of halocarbon-induced ozone .
loss, then the cooling tendency ascribed to halocarbons
could be underestimated. Updated estimates of halocar-
bon radiative forcing are provided in Chapter 8 of this
report, IPCC (1994), and Schwarzkopf and Ramaswamy
(1993). Daniel et al. (1994) have considered the indirect
effects of ozone depletion;in analyses of the GWPs for
halocarbons. They concluded that the indirect effect var-
ies greatly for different kinds of halocarbons (e.g.,
halons, CFCs, HCFCs), a result that will be discussed
further below.
Several recent studies have addressed the degree to
which the radiative heating due to additions of a quas'i-
uniformly distributed tropospheric gas such as a CFC
may be equated with the spatially inhomogeneous cool-
ing at the tropopause due; to ozone depletion for the
purposes of evaluating a net climate response (e.g., Mol-
nar et al., 1994). Some! studies suggest that ozone
depletion may result in important dynamical changes
that modulate the realized climate response (Molnar ei
al., 1994). For the purposes of the present analysis, it
will be assumed that the indirect and direct radiative ef-
fects of halocarbons can be compared to one another in a
globally averaged sense, an assumption that is currently
being tested with detailed trtree-dimensional models (see
Chapter 8 and IPCC, 1994).
Model calculations show that radiative cooling is a
strong function of the vertical profile of the ozone loss
(Schwarzkopf and Ramaswamy, 1993; Wang et al.,
1993). This implies that it: will be difficult to calculate
these effects using a fully jinteractive two-dimensional
chemistry-dynamics model \ since these tend to underes-
timate the ozone losses observed in the critical lowest
part of the stratosphere (see, e.g., Hauglustaine et ai,
1994a). Satellite and ozonesonde observations (see
Chapter I) can, however, be used to characterize the
shape of the ozone loss profile fairly well. It has been
shown by Schwarzkopf and Ramaswamy (1993) that the
uncertainty in the globally averaged ozone cooling is on
13.29
-------
ODPs, GWPs and CI-Br LOADING
Indirect Cooling Partitioning
1990
Direct Heating Partitioning
1990
HCFCs
CH3CCI3'
Figure 13-9. Contributions of various gases to the total estimated radiative cooling (indirect) and heating
(direct) due to halocarbons in 1990 (Adapted from Daniel et al., 1994). The adopted value of a for these
calculations is 40.
the order of ±30% for a broad range of assumptions re-
garding the magnitude of the ozone depletion observed
during the 1980s in the lowest part of the stratosphere
(/.«., below the region where satellite data exist). This
estimate does not, however, include the enhanced ozone
depletions that have been obtained in 1992 and 1993, nor
does it consider the large changes in ozone observed by
the Stratospheric Aerosol and Gas Experiment (SAGE)
near the tropical tropopause (see Chapter 1). Insofar as
these may be halocarbon-induced, these effects would
tend to increase the global cooling and hence decrease
the GWPs of ozone-depleting gases shown below.
Daniel et al. (1994) combined estimates of radia-
tive cooling for the 1980s and their uncertainties (from
the work of Schwarzkopf and Ramaswamy, 1993) with
the detailed evaluation of past and future equivalent ef-
fective stratospheric chlorine for each halocarbon
described in Section 13.3 to examine the net radiative
forcing that can be attributed to each halocarbon. They
emphasized that both Antarctic and midlatitude total
ozone depletions appear to be~quite small prior to about
1980, but to increase rapidly after that time, suggesting
that a "threshold" for ozone destruction may have been
reached. They assumed that the indirect radiative cool-
ing for each halocarbon depends linearly upon its
contribution to the total equivalent effective stratospher-
ic chlorine whenever the latter lies above this threshold
value. Possible nonlinearities associated, for example,
with temperature feedbacks between ozone depletion
and polar stratospheric cloud frequencies have therefore
been neglected in this study. The impact of changing
UV radiation due to ozone depletion upon OH and hence
tropospheric chemistry has also not been considered
here.
Insofar as significant ozone loss likely occurs only
for total equivalent effective stratospheric chlorine levels
above a certain threshold, the total indirect radiative
cooling caused by any halocarbon depends upon the
abundances of others and cannot be specified indepen-
dent of scenario. This implies that GWPs for halocarbons
based upon the indirect effects estimated for injection of
an infmitesimally small amount of added gas can no
longer be used to directly calculate the net radiative im-
pact of the true amount of that gas in the Earth's
atmosphere; this limitation is similar to that for methane
discussed above.
Figure 13-9 shows an estimate of the contributions
of various gases to the total estimated radiative cooling
(indirect) and heating (direct) due to halocarbons in
1990 (Daniel et al., 1994). A key point noted by Daniel
et al. (1994) is that the CFCs are likely to be responsible
for a much larger fraction of the estimated heating than
13.30
-------
ODPs, GWPs and CI-Br LOADING
CFC-12 Global Warming Potentials
0.
>
^*
O
3UUU
8OOO
7000
60OO
5000
4000
3000
20OO
1000
n
/s^^ Direct
/ ^^— -—
^ Xnet
-
-
-
-
-
i i i i i
Halon 1301 Global Warming Potentials
1990 2010 2030 2050 2070
Year
2090
-20000 -
-25000
1990 20IQ 2030 2050 2070 2090
i Year
Figure 13-10. Calculated time-dependent GWPs for CFC-12 and halon-1301; adapted from the study of
Daniel et at. (1994), for the basic Copenhagen scenario described in Section 13.3 (case A) and assuming a
value of a of 40. The denominator used in these calculations is based upon the carbon cycle model as
discussed in the text.
of the cooling, while for compounds such as the halons
.and anthropogenic CF^Br, the situation is reversed.
This is due to the enhanced effectiveness of brominated
compounds compared to chlorinated species for ozone
loss (see Section 13.4.2), by about a factor of 40. CCLt
and CH3CC13, while not as effective as the bromocar-
bons for ozone destruction, contain several chlorine
atoms per molecule and release them readily in the
stratosphere, making them relatively effective ozone de-
stroyers (and hence cooling agents) as well. This
introduces a new factor that would have to be dealt with
in the use of such indices in policy decisions, underscor-
ing the difficulty of considering gases with multiple, and
very different, environmental impacts using a single
simple index. Multiple impacts could require more so-
phisticated policy tools.
Figure 13-10 shows calculated GWPs for CFC-12
and halon-1301 as a function of time horizon adapted
from the study of Daniel et al. (1994), for the base
Copenhagen scenario (case A) described in Section 13.3,
assuming a value of a of 40, and using the Bern et al.
carbon cycle model results for the denominator as in
IPCC (1994). As suggested by Figure 13-10, the net
GWP of CFC-12 remains positive while that of halon-
1301 becomes large and negative when indirect effects
are considered in this framework. Daniel et al. (1994)
considered the following key uncertainties in deriving
the GWPs for halocarboris: (i) variations in the scenario
for future concentrations of ozone-depleting gases, as in
the scenarios of Section! 13.3, (ii) uncertainties in the
globally-averaged relative efficiency of bromine for
ozone loss as compared ito chlorine (a, assumed to lie
between 40 and 200), and (iii) uncertainties in the mag-
nitude of the cooling in !the lower stratosphere due to
uncertainties in the ozone loss profile (estimated to be
about ±30% as noted above). They found that the GWPs
were not as sensitive to the adopted range of possible
scenarios for future concentrations of halocarbons nor to
the exact values of the thresholds or scenarios assumed
as to the uncertainties in i:he absolute value of the cool-
ing and the value of a. This is consistent with the rather
small differences in key aspects of the various scenarios
shown in Table 13-3. The GWPs for bromocarbons were
found to be extremely sensitive to the chosen value of a,
while those for CFCs were quite sensitive to the adopted
uncertainty in the total absolute radiative cooling in the
1980s. Table 13-9 shows|the range of 20- and 100-year
net GWPs derived for the halocarbons including, indirect
13.31
-------
ODPs, GWPs and d-Br LOADING
Table 13-9. Net GWPs per unit mass emission for halocarbons including indirect effects (adapted
from Daniel et al., 1994). Relative to CO2 using Bern model for decay function (as in IPCC, 1994).
compound
CFC-11
CFC-12
CFC-113
HCFC-22
HCFC-142b
CH3Br
H-1301
H-1211
HCFC-141b
CH3CC13
CO,
HCFC-123
HCFC-124
HFC-134a
Time Horizon = 2010
Uncertainty
in scenario, a
min max
1900 2900
6300 6900
3200 3800
3900 4000
3800 3900
-18600 -4900
-97200 -22400
-92400 -21500
910 1200
-780 -450
-1800 -520
120 170
1300 1400
3300 3300
Uncertainty
in cooling
min max
1300 3000
6100 6900
2800 3800
3800 4000
3700 4000
-6400 -3300
-31000 -13800
-29600 -13400
690 1200
-1100 -420
-2500 -430
67 180
1300 1370
3300 3300
Direct
5000
7900
5000
4300
4200
6200
1800
360
2000
300
1500
3300
Time Horizon = 2090
Uncertainty
in scenario, a
min max
1400 1800
6900 7100
3300 3500
1500 1500
1800 1800
-5700 -1500
-87300 -21600
-50600 -13600
270 370
-260 -150
-1500 -1100
37 52
410 430
1300 1300
Uncertainty
in cooling
min max
640 2200'
6500 7400
2800 3800
1500 1600
1700 1800
-2000 -1000
-31200 -14200
-18800 -8900
180 390
-360 -140
-2400 -630
20 54
390 430
1300 1300
Direct
4000
8500
5000
1700
2000
5600
630
110
1400
93
480
1300
effects from these sensitivity studies and compares them
to GWPs for the direct effect only (adapted from Daniel
et al., 1994 for the denominator used here).
The range of values in the table underscores the
uncertain nature of these estimates due to uncertainties
in a and in the total absolute radiative cooling (i.e.,
ozone loss distribution), but also illustrates systematic
differences between various broad classes of compounds
that are more robust. In particular, the CFCs and HCFCs
are highly likely to be net wanning agents. CC14 and
CHaCCls are likely to be nearly "climate neutral," while
halons and methyl bromide are believed to be net cool-
ing agents. The impact of the implementation of the
Copenhagen Amendments on radiative forcing and
hence on climate change will depend upon the time-
dependent mix of these gases and their substitutes in the
future (see Chapter 8).
REFERENCES
AFEAS, Alternative Fluorocarbons Environmental
Acceptability Study, Production, Sales and Atmo-
spheric Release of Fluorocarbons through 1992,
Grant Thornton Company, Washington, D.C.,
1993.
Atkinson, R., D.L. Baulch, R.A. Cox, R.F. Hampson,
J.A. Kerr, and J. Troe, Evaluated kinetic and pho-
tochemical data for atmospheric chemistry,
supplement IV, ILJPAC subcommittee on gas ki-
netic data evaluation for atmospheric chemistry, J.
Phys. Chem. Ref. Data, 21, 1125-1568, 1992.
Barry, J., H.W Sidebottom, J. Treacy, and J. Franklin,
Kinetics and mechanism for the atmospheric oxi-
dation of 1,1,2-trifluoroethane, Int. J. Chem. Kin.,
in press, 1994.
Biggs, P., et al., paper presented at STEP-HALOCIDE
AFEAS workshop, University College, Dublin,
Ireland, 23 to 25 March, 1993.
Bridier, I., B. Veyret, and R. Lesclaux, Flash photolysis
kinetic study of reactions of the BrO radical with
BrO and HO2, Chem. Phys. Lett., 201, 563-568,
1993.
13.32
-------
ODPs, GWPs and CI-Br LOADING
Briegleb, B., Longwave band model for thermal radia-
tion in climate studies, J. Geophys. Res., 97,
11475-11486, 1992.
Briihl, C., The impact of the future scenarios for meth-
ane and other chemically active gases on the GWP
of methane, Chemosphere, 26, 731-738, 1993.
Butler, J.H., J.W. Elkins, T.M. Thompson, B.D. Hall,
T,H. Swanson, and V. Koropalov, Oceanic con-
sumption of CH3CC13: Implications for
tropospheric OH, /. Geophys. Res., 96, 22347-
22355, 1991.
Caldeira, K., and J.F. Kasting, Global warming on the
margin, Nature, 366, 251-253, 1993.
Cess, R.D., etal.. Uncertainties in carbon dioxide radia-
tive forcing in atmospheric general circulation
models, Science, 262, 1252-1255, 1993.
Chameides, W.L., and D.D. Davis, Iodine: Its possible
role in tropospheric photochemistry, /. Geophys.
Res., 85, 7383-7398, 1980.
Clerbaux, C., R. Colin, P.C. Simon, and C. Granier, In-
frared cross sections and global warming
potentials of 10 alternative hydrohalocarbons, /.
Geophys. Res., 98, 10491-10497, 1993.
Cooper, D.L., T.P. Cunningham, N.L. Allen, and A.
McCulloch, Tropospheric lifetimes of potential
CFC replacements: Rate coefficients for reaction
with the hydroxyl radical, Atmos. Env., 26A, 1331-
1334, 1992.
Cunnold, D., P. Fraser, R. Weiss, R. Prinn, P. Simmonds,
B. Miller, F. Alyea, and A. Crawford, Global
trends and annual releases of CFC13 and CF2C12
estimated from ALE/GAGE and other measure-
ments from July 1978 to June 1991, J. Geophys.
Res., 99, 1107-1126, 1994.
Daniel, J.S., S. Solomon, and D.L. Albritton, On the
evaluation of halocarbon radiative forcing and glo-
bal warming potentials, accepted for publication
in / Geophys. Res., 1994.
DeMore, W.B., Rate constants for the reactions of OH
with HFC-134a and HFC-134, Geophys. Res.
Lett., 20, 1359-1362, 1993.'
Fisher, D.A., and P.M. Midgley, The production and re-
lease to the atmosphere of CFCs 113, 114 and 115,
Atmospheric Environment, 27A, 271-276, 1993.
Fuglestvedt, J.S., J.E. Jonson, and I.S.A. Isaksen, Effects
of reduction in stratospheric ozone on tropospher-
ic chemistry through changes in photolysis rates,
accepted for publication in Tellus, 1994.
Gamlen, PH., B.C. Lane, P.M. Midgley, and J.M. Steed,
The production iind release to the atmosphere of
CC13F and CCl^ (chlorofluorocarbons CFC 11
and CFC 12), Atmospheric Environment, 20,
1077-1085, 1986,
Garcia, R.R., and S. Solomon, A new numerical model
of the middle atmosphere 2. Ozone and related
species,/ Geophys. Res., 99, 12937-12951, 1994.
Hauglustaine, D.A., C.: Granier, G.P. Brasseur, and G.
Megie, The importance of atmospheric chemistry
in the calculation of radiative forcing on the cli-
mate system, /. \Geophys. Res., 99, 1173-1186
1994a.
Hauglustaine, D.A., C, Granier, and G.P. Brasseur, Im-
pact of increased methane emissions on the
atmospheric composition and related radiative
forcing of the climate system, submitted to Envi-
ron. Monitor. Assoc., 1994b.
IPCC, Intergovernmental Panel on Climate Change, Cli-
mate Change: The Scientific Assessment, edited by
J.T. Houghton, G.J. Jenkins, and J.J. Ephraums,
Cambridge University Press, Cambridge, U.K.,
1990.
IPCC, Intergovernmental Panel on Climate Change, Cli-
mate Change: The Supplementary Report to the
IPCC Scientific Assessment, edited by J.T. Hough-
ton, B.A. Callander, and S.K. Varney, Cambridge
University Press, Cambridge, U.K., 1992.
IPCC, Intergovernmental Panel on Climate Change, Cli-
mate Change: The Scientific Assessment, in
preparation, 1994.J
IUPAC, International Union of Pure and Applied Chem-
istry, Evaluated kinetic and photochemical data for
atmospheric chemistry, /. Phys. Chem. Ref. Data^
21, 6, 1992. I
Johnson, C., J. Henshaw, and G. Mclnnes, Impact of air-
craft and surface emissions of nitrogen oxides on
tropospheric ozone and global warmiAg, Nature,
355,69-71, 1992.!
JPL, Jet Propulsion Laboratory, Chemical Kinetics and
Photochemical Data for Use in Stratospheric
Modelling, JPL report number 92-20, 1992.
13.33
-------
ODPs, GWPs and Cl-Br LOADING
Kayc, J., S. Pcnkett and P.M. Ormond, Report on Con-
centrations, Lifetimes and Trends of CFCs,
Halons, and Related Species, NASA office of Mis-
sion to Planet Earth, NASA Reference Publication
1339, 1994.
Kindler, T.P., W.L. Chameides, P.H. Wine, D.M. Cun-
nold, EN. Alyea, and J.A. Franklin, The fate of
atmospheric phosgene and the ozone depletion po-
tentials of its parent compounds: CCLj, €2014,
C2HC13, CH3Ca3, CHC13, submitted to J. Geo-
phys. Res., 1994.
Ko, M.K.W., N.-D. Sze, W.-C. Wang, G. Shia, A. Gold-
man, FJ. Murcray, D.G. Murcray, and C.P.
Rinsland, Atmospheric sulfur hexafluoride:
Sources, sinks, and greenhouse warming, J. Geo-
phys. Res., 98. 10499-10507, 1993.
Ko, M.K.W., N.-D. Sze, J.M. Rodriguez, O.K. Weisen-
stein, C.W. Heisey, R.P. Wayne, P. Biggs, C.E.
Canosa-Mas, H.W. Sidebottom, and J. Treacy, CF3
chemistry: Potential implications for stratospheric
ozone, Geophys. Res. Lett., 21, 101-104, 1994a.
Ko, M.K.W., N.-D. Sze, and MJ. Prather, Protection of
the ozone layer: Adequacy of existing controls,
Nature, 367, 545-548, 1994b.
Lacis, A.A., D.J. Wuebbles, and J.A. Logan, Radiative
forcing by changes in the vertical distribution of
ozone,/. Geophys. Res., 95. 9971-9981, 1990.
Lai, M., and T. Holt, Ozone depletion due to increasing
anthropogenic trace gas emissions: Role of strato-
spheric chemistry and implications for future
climate, Clint. Res., 1, 85-95, 1991.
Lelieveld, J., and P.J. Crutzen, Indirect chemical effects
of methane on climate wanning, Nature, 355, 339-
342. 1992.
Lelieveld, J., P.J. Cmtzen, and C. Briihl, Climate effects
of atmospheric methane, Chemosphere, 26, 739-
768, 1993.
Li, Z., and J.S. Francisco, Laser-induced fluorescence
spectroscopic study of the 2Ai-2E transition of tri-
fluoromethoxy radical, Chem. Phys. Lett., 186,
336, 1991.
McCarthy, R.L., F.A. Bower, and J.P. lesson, The fluoro-
carbon-ozone theory—I. Production and release:
World production and release of CC13F and
CCI2F2 (fluorocarbons 11 and 12) through 1975,
Atmospheric Environment, 11, 491-497, 1977.
McCulloch, A., Global production and emissions of
bromochlorodifluoromethane and bromotrifluoro-
methane (halons 1211 and 1301), Atmospheric
Environment, 26A. 1325-1329, 1992.
Mellouki, A., R.K. Talukdar, and C.J. Howard, Kinetics
of the reactions of HBr with O3 and HC>2: The
yield of HBr from HO2 + BrO, 7. Geophys. Res.,
99, 22949-22954, 1994.
Midgley, P.M., The production and release to the
atmosphere of 1,1,1 -trichloroethane (methyl chloro-
form), Atmospheric Environment, 23, 2663-2665,
1989.
Midgley, P.M., and D.A. Fisher, The production and re-
lease to the atmosphere of chlorodifluoromethane
(HCFC 22), Atmospheric Environment, 27A, 271-
279, 1993.
Molnar, G.I., M.K.W. Ko, S. Zhou, and N.-D. Sze, Cli-
matic consequences of the observed 1980s ozone
loss: Relevance to the greenhouse problem, J.
Geophys. Res., in press, 1994.
Moore, B., Ill, and B.H. Braswell, Jr., The lifetime of
excess atmospheric carbon dioxide, Global Bio-
geochem. Cycles, 8, 23-28, 1994.
Nelson, D.D., M.S. Zahniser, and C.E. Kolb, OH reac-
tion kinetics and atmospheric lifetimes of
CF3CFCHCF3 and CF3CH2Br, Geophys. Res.
Lett., 20, 197-200, 1993.
Plumb, R.A., and M.K.W. Ko, Interrelationships be-
tween mixing ratios of long-lived stratospheric
constituents,/ Geophys. Res., 97, 10145-10156,
1992.
Pollock, W.H., L.E. Heidt, R.A. Lueb, J.F. Vedder, M.J.
Mills, and S. Solomon, On the age of stratospheric
air and ozone depletion potentials in polar regions,
7. Geophys. Res.. 97, 12993-12999, 1992.
Poulet, G.M., M. Pirre, F. Magain, R. Ramaroson, and G.
Le Bras, Role of the BrO + HOa reaction in the
stratospheric chemistry of bromine, Geophys. Res.
Lett., 23, 2305-2308, 1992.
Prather, M.J., and R.T. Watson, Stratospheric ozone de-
pletion and future levels of atmospheric chlorine
and bromine, Nature, 344, 729-734, 1990.
Prather, M.J., and C.M. Spivakovsky, Tropospheric OH
and the lifetimes of hydrochlorofluorocarbons
(HCFCs), 7. Geophys. Res., 95, 18723-18729,
1990.
13.34
-------
ODPs,
GWPs and CI-Br LOADING
Prather, M.J., Lifetimes and Eigenstates in atmospheric
chemistry, Geophys. Res. Lett., 21, 801-804, 1994.
Prinn, R., D. Cunnold, P. Simmonds, F. Alyea, R. Boldi,
A. Crawford, P. Eraser, D. Gutzler, D. Hartley, R.
Rosen, and R. Rasmussen, Global averaged con-
centration and trend for hydroxyl radicals deduced
from ALE/GAGE trichloroethane (methyl chloro-
form) data for 1978-1990, J. Geophys. Res., 97,
2445-2462, 1992.
Ramaswamy, V., M.D. Schwarzkopf, and K.P. Shine,
Radiative forcing of climate from halocarbon-in-
duced global stratospheric ozone loss, Nature,
355, 810-812, 1992.
Ravishankara, A.R., A.A. Turnipseed, N.R. Jensen, S.
Barone, M. Mills, C.J. Howard, and S. Solomon,
Do hydrofluorocarbons destroy stratospheric
ozone?, 5dence, 263, 71-75, 1994.
Ravishankara, A.R., S. Solomon, A A. Turnipseed, and
R.F. Warren, Atmospheric lifetimes of long-lived
species, Science, 259, 194-199, 1~993.
Rodhe, H., A comparison of the contributions of various
gases to the greenhouse effect, Science, 248, 1217-
1219, 1990.
Roehl, C.M., D. Boglu, C. Briihl, and O.K. Moortgat,
Infrared band intensities and global warming po-
tentials of CF4, C2F6, C3F8, C4F,0, C5FI2, and
CgF^, submitted to Geophys. Res. Lett., 1994.
Salawitch, R.J., etai. Loss of ozone in the Arctic vortex
for the winter of 1989, Geophys. Res. Lett., 17,
561-564, 1990.
Salawitch, R.J., etal.. Chemical loss of ozone in the Arc-
tic polar vortex in the winter of 1991-1992,
Science, 261, 1146-1149, 1993.
Schauffler, S.M., L.E. Heidt, W.H. Pollock, T.M. Gilpin,
J. Vedder, S. Solomon, R.A. Lueb, and E.L. Atlas,
Measurements of halogenated organic compounds
near the tropical tropopause, Geophys. Res. Lett.,
20, 2567-2570, 1993.
Schwarzkopf, M.D., and V. Ramaswamy, Radiative forc-
ing due to ozone in the 1980s: Dependence on
altitude of ozone change, Geophys. Res. Lett., 20,
205-208, 1993.
Schmoltner, A.M., R.K. Talukdar, R.F. Warren, A. Mel-
louki, L. Goldfarh, T. Gierczak, S.A. McKeen, and
A.R. Ravishankara, Rate coefficients for reactions
of several hydrofluorocarbons with OH and O('D)
and their atmospheric lifetimes, /. Phys. Chem.,
97, 8976-8982, 1993.
Solomon, S., J. Burkholder, A.R. Ravishankara, and
R.R. Garcia, Ozone depletion and global warming
potentials for CF3I, /. Geophys. Res., 99, 20929-
20935, 1994a. <
Solomon, S., R.R. Garcia, and A.R. Ravishankara, On
the role of iodine in ozone depletion, /. Geophys.
Res., 99, 20491-20499, 1994b. '
Solomon, S., M.J. Mills, L.E. Heidt, W.H. Pollock, and
A.F. Tuck, On the evaluation of ozone depletion
potentials, /. Geophys. Res., 97, 825-842, 1992.
Solomon, S., and D.L. Albritton, A new analysis of time-
dependent ozone depletion potentials, Nature,
357, 33, 1992. ;
Stolarski, R.S., and R.D. Rundel, Fluorine photochemis-
try in the stratosphere, Geophys. Res. Lett., 2,
443-445, 1975. ;
Tabazadeh, A., and R.P. Turco, Stratospheric chlorine
injection by volcanic eruptions: HC1 scavenging
and implications for ozone, Science, 260, 1082-
1086, 1993.
Wang, W.C., M.P. Dudek, X.Z. Liang, and J.T. Kiehl,
Inadequacy of effective CO2 as a proxy in simulat-
ing the greenhouse effect of other radiative active
gases, Nature, 350. 573-577, 1991.
Wang, W.C., Y.C. Zhuang, and R.D. Bojkov, Climate
implications of obsierved changes in ozone vertical
distributions in middle and high latitudes of the
Northern Hemisphere, Geophys. Res. Lett., 20,
1567-1570, 1993.
Wang, C.J.L., D.R. Blake, and F.S. Rowland, Seasonal
variations in the atmospheric distribution in re-
mote surface locations of reactive chlorine
compounds, tetrachloroethylene (C2Cl4), submit-
ted to Geophys. Res. Lett., 1994.
Wigley, T.M.L., Balancing the carbon budget: Implica-
tions for projections of future carbon dioxide
concentration changes, Tellus, 45B, 409-425,
1993.
13.35
-------
ODPs, GWPs and CI-Br LOADING
Wigley, T.M.L., The effect of carbon cycle uncertainties
on global wanning potentials, submitted to Geo-
, phys. Res. Lett., 1994a.
Wigley, T.M.L., How important'are carbon cycle model
uncertainties?, in Climate Change and the Agenda
for Research, edited by T. Hanisch, Westview
Press, Boulder, Colorado, 169-191, 1994b.
WMO, Scientific Assessment of Stratospheric Ozone:
1989, World Meteorological Organization Global
Ozone Research and Monitoring Project—Report
No. 20, Geneva, 1990.
WMO, Scientific Assessment of Ozone Depletion: 1991,
World Meteorological Organization Global Ozone
Research and Monitoring Project—Report No. 25,
Geneva, 1992.
Wuebbles, D.J., Chlorocarbon emission scenarios: Po-
tential impact on stratospheric ozone, J. Geophys.
Łes., 88. 1433-1443, 1983.
Wuebbles, D J., J.S. Tamaresis, and K.O. Patten, Quanti-
fied estimates of total GWPs for greenhouse gases
taking into account tropospheric chemistry,
UCRL-ID-115850, LLNL report, 1994a.
Wuebbles, D.J., A.K. Jain, K.O. Patten, and K.E. Grant,
Sensitivity of direct global warming potentials to
key uncertainties, Climatic Change, in press,
1994b.
Zander, R., M.R. Gunson, C.B. Farmer, C.P. Rinsland,
F.W. Won, and E. Mahieu, The 1985 chlorine and
fluorine inventories in the stratosphere based on
ATMOS observations at 30° north latitude, J. At-
mos. Chem., 15, 171-186, 1992.
Zhang, Z., S. Padmaja, R.D. Saini, R.E. Huie, and M.J.
Kurylo, Reactions of hydroxyl radicals with
several hydrofluorocarbons: The temperature
dependencies of the rate constants for
CHF2CF2CH2F (HFC-245ca), CF3CHFCHF2
(HFC-236ea), CF3CHFCF3 (HFC-227ea), and
CF3CH2CH2CF3 (HFC-356ffa), J. Phys. Chem.,
95,4312-4315,1994.
13.36
-------
APPENDICES
,i
List of International Authors,
Contributors, and Reviewers
B
Major Acronyms and Abbreviations
Chemical Formulae and Nomenclature
-------
-------
APPENDIX A
LIST OF INTERNATIONAL AUTHORS,
CONTRIBUTORS, AND REVIEWERS
CO-CHAIRS
Daniel L. Albritton
Robert T. Watson
Piet J. Aucamp
National Oceanic and Atmospheric Administration, Boulder,1 Colorado US
Office of Science and Technology Policy, Washington, D.C.. US
Department of National Health, Pretoria | South Africa
AUTHORS AND CONTRIBUTORS
Susan Solomon
F. Sherwood Rowland
COMMON QUESTIONS ABOUT OZONE
Coordinators
NOAA Aeronomy Laboratory
University of California at Irvine
US
US
PART 1. OBSERVED CHANGES IN OZONE AND SOURCE: GASES
CHAPTER 1: OZONE MEASUREMENTS '
Neil R.P. Harris
Gerard Ancellet
Lane Bishop
David J. Hofmann
James B. Ken-
Richard D. McPeters
M. Margarita Prendez
William J. Randel
Johannes Staehelin
B.H. Subbaraya
Andreas Volz-Thomas
Joseph M. Zawodny
Christos S. Zerefos
Marc Allaart
James K. Angell
Chapter Lead Author
European Ozone Research Coordinating Unit
Co-authors
Centre National de la Recherche Scientifique
Allied Signal, Inc.
NOAA Climate Monitoring and Diagnostics Laboratory
Atmospheric Environment Service
NASA Goddard Space Flight Center
Universidad de Chile
National Center for Atmospheric Research
Eidgenossische Technische Hochschule Zurich
Physical Research Laboratory
Forschungszentrum Jiilich
NASA Langley Research Center
Aristotle University of Thessaloniki
Contributors
Koninklijk Nederlands Meteorologisch Instituut
NOAA Air Resources Laboratory
UK
France
US
US
Canada
US
Chile
US
Switzerland
India
Germany
US
Greece
The Netherlands
US
A. I
-------
AUTHORS, CONTRIBUTORS, AND REVIEWERS
Rumen D. Bojkov
Kenneth P. Bowman
GJ.R. Coetzee
Malgorzata Deg6rska
John J. DeLuisi
Dirk De Muer
Terry Deshler
Lucien Froidevaux
Reinhard Furrer
Brian G. Gardiner
Hartwig Gemandt
James F. Gleason
Ulrich GSrsdorf
Kjell Henriksen
Emest Hilsenrath
Stacey M. Hollandsworth
0ystein Hov
Hennie Kelder
Volker Kirchhoff
UlfK8hler
Walter D. Komhyr
Janusz W. KrzyScin
Zcnobia Litynska
Jennifer A. Logan
Pak Sum Low
W. Andrew Matthews
A.J. Miller
Samuel J. Oltmans
Walter G. Planet
J.-P. Pommereau
Hans-Eckhart Scheel
Jonathan D. Shanklin
Paula SkFivdnkova'
Herman Smit
Joe W. Waters
Peter Winkler
World Meteorological Organization
Texas A&M University
Weather Bureau
Polish Academy of Sciences
NOAA Air Resources Laboratory
Institut Royal Meteorologique de Belgique
University of Wyoming
California Institute of Technology/Jet Propulsion Laboratory
Freie Universitat Berlin
British Antarctic Survey
Alfred Wegener Institut
NASA Goddard Space Flight Center
Deutscher Wetterdienst
University of Troms0
NASA Goddard Space Flight Center
Applied Research Corporation
Universitetet I Bergen
Koninklijk Nederlands Meteorologisch Instituut
Instituto Nacional de Pesquisas Espaciais
Deutscher Wetterdienst
NOAA Climate Monitoring and Diagnostics Laboratory
Polish Academy of Sciences
Centre of Aerology
Harvard University
United Nations Environment Programme Ozone Secretariat
National Institute of Water and Atmospheric Research
NOAA National Meteorological Center
NOAA Climate Monitoring and Diagnostics Laboratory
National Oceanic and Atmospheric Administration, NESDIS
Centre National de la Recherche Scientifique
Fraunhofer Institut fur AtmosphSrische Umweltforschung
British Antarctic Survey
Czech Hydrometeorological Institute
Forschungszentrum Jiilich
California Institute of Technology/Jet Propulsion Laboratory
Deutscher Wetterdienst
Switzerland
US
South Africa
Poland
US
Belgium
US
US
Germany
UK
Germany
US
Germany
Norway
US
US
Norway
The Netherlands
Brazil
Germany
US
Poland
Poland
US
Kenya
New Zealand
US
US
US
France
Germany
UK
Czech Republic
Germany
US
Germany
Eugcnio Sanhueza
Paul J. Fraser
Rudi J. Zander
CHAPTER 2: SOURCE GASES: TRENDS AND BUDGETS
Chapter Lead Author
Instituto Venezolano de Investigaciones Cientificas
Co-authors
CSIRO Division of Atmospheric Research
University of Liege
Venezuela
Australia
Belgium
A.2
-------
AUTHORS, CONTRIBUTORS, AND REVIEWERS
Fred N. Alyea
Meinrat O. Andreae
James H. Butler
Derek N. Cunnold
J. Dignon
Ed Dlugokencky
Dieter H. Ehhalt
James W. Elkins
D. Etheridge
David W. Fahey
Donald A. Fisher
Jack A. Kaye
M.A.K. Khalil
Paulette Middleton
Paul C. Novelli
Joyce Penner
Michael J^ Prather
Ronald G. Prinn
William S. Reeburgh
J. Rudolph
P. Simmonds
L. Paul Steele
Michael Trainer
Ray F. Weiss
Donald J. Wuebbles
Contributors
Georgia Institute of Technology
M'ax-Planck-Institut fur Chemie
NOAA Climate Monitoring and Diagnostics Laboratory
Georgia Institute of Technology
Lawrence Livermore National Laboratory
NOAA Climate Monitoring and Diagnostics Laboratory
Forschungszentrum Jiilich
NOAA Climate Monitoring and Diagnostics Laboratory
CSIRO Division of Atmospheric Research
NOAA Aeronomy Laboratory
E.I. DuPont de Nemours and Company
NASA Goddard Space Flight Center
Oregon Graduate Institute of Science and Technology
Science and Policy Associates, Inc.
University of Colorado
Lawrence Livermore National Laboratory
University of California at Irvine
Massachusetts Institute of Technology
University of California at Irvine
Forschungszentrum Jiilich
University of Bristol
CSIRO Division of Atmospheric Research
NOAA Aeronomy Laboratory
Scripps Institution of Oceanography
University of Illinois
US
Germany
US
US
US
US
Germany
US
Australia
US
US
US
US
US
US
US
US
US '
US
Germany
UK
Australia
US
US
US
David W. Fahey
Geir Braathen
Daniel Cariolle
Yutaka Kondo
W. Andrew Matthews
Mario J. Molina
John A. Pyle
Richard B. Rood
James M. Russell HI
Ulrich Schmidt
Darin W. Toohey
PART 2. ATMOSPHERIC PROCESSES RESPONSIBLE
FOR THE OBSERVED CHANGES IN OZONE
CHAPTER 3: POLAR OZONE i
Chapter Lead Author
NOAA Aeronomy Laboratory
Co-authors (
Norsk Institutt for Luftforskning j
Meteo-France, Centre National de Recherches Meteorologiques
Nagoya University ,!
National Institute of Water and Atmospheric Research
Massachusetts Institute of Technology
University of Cambridge
NASA Goddard Space Right Center
NASA Langley Research Center
Forschungszentrum Julich
University of California at Irvine
US
Norway
France
Japan
New Zealand
US
UK
US
US
Germany
US
A.3
-------
AUTHORS, CONTRIBUTORS, AND REVIEWERS
Joe W. Waters
Christopher Webster
Steven CWofsy
Terry Deshler
James E. Dye
T. Duncan A. Fairlie
William A. Grose
Gloria L. Manney
Paul A. Newman
Alan R. O'Neill
R. Bradley Pierce
William J.Randel
Aldan E. Roche
Charles R. Trepte
California Institute of Technology/Jet Propulsion Laboratory
California Institute of Technology/Jet Propulsion Laboratory
Harvard University
Contributors
University of Wyoming
National Center for Atmospheric Research
NASA Langley Research Center
NASA Langley Research Center
California Institute of Technology/Jet Propulsion Laboratory
NASA Goddard Space Flight Center
University of Reading
NASA Langley Research Center
National Center for Atmospheric Research
Lockheed Corporation
NASA Langley Research Center
US
US
US
US
US
US
US
US
US
UK
US
US
US
US
Roderic L, Jones
Linnea Avallone
Lucien Froidevaux
Sophie Godin
L.J. Gray
Stefan Kinne
Michael E. Mclntyre
Paul A. Newman
R. Alan Plumb
John A. Pyle
James M. Russell in
Margaret A. Tolbert
RalfToumi
Adrian F. Tuck
Paul Wennberg
Richard P. Cebula
Sushil Chandra
Eric L. Fleming
Lawrence E. Flynn
Stacey M. Hollandsworth
Charles H. Jackmah
Lament R. Poole
CHAPTER 4: TROPICAL AND MIDLATITUDE OZONE
Chapter Lead Author
University of Cambridge
Co-authors
University of California at Irvine
California Institute of Technology/Jet Propulsion Laboratory
Centre National de la Recherche Scientifique
Rutherford Appleton Lab
NASA Ames Research Center
University of Cambridge
NASA Goddard Space Flight Center
Massachusetts Institute of Technology
University of Cambridge
NASA Langley Research Center
University of Colorado
University of Cambridge
NOAA Aeronomy Laboratory
Harvard University
Contributors
Hughes STX
NASA Goddard Space Flight Center
Applied Research Corporation
Software Corporation.of America
Applied Research Corporation
NASA Goddard Space Flight Center
NASA Langley Research Center
UK
US
US
France
UK
US
UK
US
US
.UK
US
US
UK
US
US
US
US
US
US
US
US
US
A.4
-------
AUTHORS, CONTRIBUTORS, AND REVIEWERS
Andreas Volz-Thomas
Brian A. Ridley
Meinrat O. Andreae
William L. Chameides
Richard G. Derwent
Ian E. Galbally
Jos Lelieveld
Stuart A. Penkett
Michael-O. Rodgers
Michael Trainer
Geraint Vaughan
Xiu Ji Zhou
Elliot Atlas
Carl Brenninkmeijer
Dieter H. Ehhalt
Jack Fishman
Frank Flocke
Daniel J. Jacob
Joseph M. Prospero
Franz Rohrer
Rainer Schmitt
Herman G.J. Smit
Anne M. Thompson
CHAPTER 5: TROPOSPHERIC OZONE
Chapter Lead Authors
Forschungszentrum Jiilich
National Center for Atmospheric Research
Co-authors
Max-Planck-Institut fur Chemie
Georgia Institute of Technology
UK Meteorological Office
CSIRO Division of Atmospheric Research
Wageningen University
University of East Anglia
Georgia Institute of Technology
NOAA Aeronomy Laboratory
University of Wales
Academy of Meteorological Science
Contributors
National Center for Atmospheric Research
National Institute of Water and Atmospheric Research
Forschungszentrum Julich
NASA Langley Research Center
Forschungszentrum Jiilich
Harvard University
University of Florida
Forschungszentrum Julich
Meteorologie Consult GmbH
Forschungszentrum Julich
NASA Goddard Space Flight Center
PART 3. MODEL SIMULATIONS OF GLOBAL OZONE
CHAPTER 6: MODEL SIMULATIONS OF STRATOSPHERIC OZONE
Malcolm K.W. Ko
Abdel M. Ibrahim
Ivar S.A. Isaksen
Charles H. Jackman
Franck Lefevre
Michael J. Prather
Philip J. Rasch
Ralph Toumi
Guido Visconti
Germany
US
Germany
US
UK
Australia
The Netherlands
UK
US
US
UK
China
US
New Zealand
Germany
US
Germany
US
US
Germany
Germany
Germany
US
Chapter Lead Author
Atmospheric and Environmental Research, Inc.
Co-authors
Egyptian Meteorological Authority
Universitetet I Oslo
NASA Goddard Space Flight Center i
Meteo-France, Centre National de Recherches Meteorologiques
University of California at Irvine j
National Center for Atmospheric Research
University of Cambridge
Universita' degli Studi-l'Aquila
US
Egypt
Norway
US
France
US
US
UK
Italy
A.5
-------
AUTHORS, CONTRIBUTORS, AND REVIEWERS
Slimanc Bekki
Guy P. Brasseur
Christoph Briihl
Peter S. Connell
D. Considine
Paul J. Crutzen
E. Fleming
J. Gross
Linda Hunt
D. Kinnison
S. Palermi
Thomas Peter
Giovanni Pitari
Karen Sage
Tom Sasaki
XueX.Tie
D. Weisenstein
Donald J.Wuebbles
Contributors
University of Cambridge
National Center for Atmospheric Research
Max-Planck-Institut fur Chemie
Lawrence Livermore National Laboratory
NASA Goddard Space Flight Center
Max-Planck-Institut fur Chemie
NASA Goddard Space Flight Center
Max-Planck-Institut fur Chemie
NASA Langley Research Center
Lawrence Livermore National Laboratory
Universita' degli Studi-l'Aquila
Max-Planck-Institut fur Chemie
Universita' degli Studi-l'Aquila
NASA Langley Research Center
Meteorological Research Institute
National Center for Atmospheric Research
Atmospheric and Environmental Research, Inc.
University of Illinois
UK
US
Germany
US
US
Germany
US
Germany
US
US
Italy
Germany
Italy
US
Japan
US
US
US
Frodc Stordal
CHAPTER 7: MODEL SIMULATIONS OF GLOBAL TROPOSPHERIC OZONE
Chapter Lead Author
Norsk Institutt for Luftforskning
Richard G. Derwent
Ivar S.A. Isaksen
Daniel J. Jacob
Maria Kanakidou
Jennifer A. Logan
Michael J. Prather
T. Bemtsen
Guy P. Brasseur
Paul J. Crutzen
J.S. Fuglestvedt
D.A. Hauglustaine
Colin E. Johnson
K.S. Law
Jos Lelieveld
J. Richardson
M. Rocmer
A. Strand
Donald J. Wuebbles
Co-authors
UK Meteorological Office
Universitetet I Oslo
Harvard University
Centre National de la Recherche Scientifique
Harvard University
University of California at Irvine
Contributors
Universitetet I Oslo
National Center for Atmospheric Research
Max-Planck-Institut fur Chemie
Center for International Climate and Energy Research
Centre National de la Recherche Scientifique
UK Meteorological Office/AEA Technology
University of Cambridge
Wageningen University
NASA Langley Research Center
TNO Institute of Environmental Sciences
Universitetet I Bergen
University of Illinois
Norway
UK
Norway
US
France
US
US
Norway
US
Germany
Norway
France
UK
UK
The Netherlands
US
The Netherlands
Norway
US
A.6
-------
AUTHORS, CONTRIBUTORS, AND REVIEWERS
Keith P. Shine
Karin Labitzke
V. Ramaswamy
Paul C. Simon
Susan Solomon
Wei-Chyung Wang
Christoph Briihl
J. Christy
Claire Grander
A.S. Grossman
James E. Hansen
D.A. Hauglustaine
Huiting Mao
A.J. Miller
S. Pinnock
M.D. Schwarzkopf
R. Van Dorland
PART 4. CONSEQUENCES OF OZONE CHANGE-
CHAPTER 8: RADIATIVE FORCING AND TEMPERATURE TRENDS
Chapter Lead Author
University of Reading
Co-authors
Freie Universita't Berlin
NOAA Geophysical Fluid Dynamics Laboratory/Princeton University
Institut d' Aeronomie Spatiale de Beigique
NOAA Aeronomy Laboratory
Atmospheric Sciences Research Center, SUNY
Contributors
Max-Planck-Institut fur Chemie
University of Alabama in Hunts ville
National Center for Atmospheric Research
Lawrence Livermore National Laboratory
NASA Goddard Institute for Space Studies
Centre National de la Recherche Scientifique
Atmospheric Sciences Research Center, SUNY
NOAA National Meteorological Center
University of Reading
NOAA Geophysical Fluid Dynamics Laboratory
Koninklijk Nederlands Meteorologisch Instituut
UK
Germany
US
Belgium
US
US
Germany
US
US
US
US
France
US
US
UK
US
The Netherlands
Richard L. McKenzie
M. Blumthaler
C.R. Booth
Susana B. Diaz
John E. Frederick
Tomoyuki Ito
Sasha Madronich
G. Seckmeyer
Sergio Cabrera
Mohammad Ilyas
James B. Ken-
Colin E. Roy
Paul C. Simon
CHAPTER 9: SURFACE ULTRAVIOLET RADIATION
Chapter Lead Author
National Institute of Water and Atmospheric .Research
Co-authors
University of Innsbruck
Biospherical Instruments
Austral Center of Scientific Research (CADIC/CONICET)
University of Chicago
Japan Meteorological Agency
National Center for Atmospheric Research
Fraunhofer Institut fur Atmospharische Umweltforschung
Contributors
Universidad de Chile
University of Science Malaysia
Atmospheric Environment Service
Australian Radiation Laboratory
Institut d'Aeronomie Spatiale de Beigique
New Zealand
Austria
US
Argentina
US
Japan
US
Germany
Chile
Malaysia
Canada
Australia
Belgium
A.7
-------
AUTHORS, CONTRIBUTORS, AND REVIEWERS
David I. Wardle
Atmospheric Environment Service
Canada
PART 5. SCIENTIFIC INFORMATION FOR FUTURE DECISIONS
CHAPTER 10: METHYL BROMIDE
Stuart A. Penkett
James H. Butler
Michael J. Kurylo
C.E. Reeves
Jose M. Rodriguez
Hanwant B. Singh
Darin W. Toohey
Ray R Weiss
Meinrat O. Andrcae
N.J. Blake
Ralph J. Cicerone
Tom Duafala
Amram Golombek
M.A.K. Khalil
Joel S. Levine
Mario J. Molina
Susan M. Schauffler
Chapter Lead Author
University of East Anglia
Co-authors
NOAA Climate Monitoring and Diagnostics Laboratory
NASA Headquarters/NIST
University of East Anglia
Atmospheric and Environmental Research, Inc.
NASA Ames Research Center
University of California at Irvine
Scripps Institution of Oceanography
Contributors
Max-Planck-Institut fur Chemie
University of California at Irvine
University of California at Irvine
Methyl Bromide Global Coalition
Israel Institute for Biological Research
Oregon Graduate Institute of Science and Technology
NASA Langley Research Center
Massachusetts Institute of Technology
National Center for Atmospheric Research
UK
US
US
UK
US
US
US
US
Germany
US
US
US
Israel
US
US
US
US
Andreas Wahner
Marvin A. Geller
Frank Arnold
William H. Bruno
Daniel A. Cariolle
Anne R. Douglass
Colin E. Johnson
Dave H. Lister
John A. Pyle
Richard Ramaroson
David Rind
Franz Rohrer
CHAPTER 11: SUBSONIC AND SUPERSONIC AIRCRAFT EMISSIONS
Chapter Lead Authors
Forschungszentrum Jiilich
State University of New York at Stony Brook
Co-authors
Max-Planck-Institut fur Kemphysik
Pennsylvania State University
M6tŁo-France, Centre National de Recherches Meteorologiques
NASA Goddard Space Flight Center
UK Meteorological Office/AEA Technology
Defence Research Agency/Aerospace and Propulsion Department
University of Cambridge
Office National d'Etudes et Recherches Aerospatiales
NASA Goddard Institute for Space Studies
Forschungszentrum Jiilich
Germany
US
Germany
US
France
US
UK
UK
UK
France
US
Germany
A.8
-------
AUTHORS, CONTRIBUTORS, AND REVIEWERS
Ulrich Schumann
Anne M. Thompson
DLR Institut fur Physik der Atmosphare
NASA Goddard Space Flight Center
Germany
US
CHAPTER 12: ATMOSPHERIC DEGRADATION OF HALOCARBON SUBSTITUTES
R.A. Cox
Roger Atkinson
Geert K. Moortgat
A.R. Ravishankara
H.W. Sidebottom
G.D. Hayman
Carleton J. Howard
Maria Kanakidou
Stuart A. Penkett
Jose M. Rodriguez
Susan Solomon
Oliver Wild
Chapter Lead Author
National Environmental Research Council Headquarters
Co-authors
University of California at Riverside
Max-Planck-Institut fur Chemie
NOAA Aeronomy Laboratory
University College, Dublin
Contributors
Harwell Laboratory/AEA Environment and Energy
NOAA Aeronomy Laboratory
Centre National de la Recherche Scientifique
University of East Anglia ,
Atmospheric and Environmental Research, Inc.
NOAA Aeronomy Laboratory
University of Cambridge
UK
US
Germany
' US
Ireland
UK
US
France
UK
US
US
UK
CHAPTER 13: OZONE DEPLETION POTENTIALS, GLOBAL WARMING POTENTIALS,
AND FUTURE CHLORINE/BROMINE LOADING
Susan Solomon
Donald J. Wuebbles
Ivar S.A. Isaksen
Jeffrey T. Kiehl
Murari Lai
Paul C. Simon
Nien-Dak Sze
Daniel L. Albritton
Christoph Briihl
Peter S. Connell
John S. Daniel
Donald A. Fisher
D. Hufford
Claire Granier
Chapter Lead Authors
NOAA Aeronomy Laboratory
University of Illinois
Co-authors
Universitetet I Oslo
National Center for Atmospheric Research
Indian Institute of Technology
Institut d' Aeronomie Spatiale de Belgique
Atmospheric and Environmental Research, Inc.
Contributors
NOAA Aeronomy Laboratory
Max-Planck-Institut fur Chemie
Lawrence Livermore National Laboratory
NOAA Aeronomy Laboratory/CIRES
E.I. DuPont de Nemours and Company
U.S. Environmental Protection Agency
National Center for Atmospheric Research
US
US
Norway
US
India
Belgium
US
US
Germany
US
US
US
US
US
A.9
-------
AUTHORS, CONTRIBUTORS, AND REVIEWERS
Shaw C. Liu
Ken Patten
S. Pinnock
V. Ramaswamy
Keith P. Shine
Guido Visconti
D. Weisenstein
Tom M.L. Wigley
NOAA Aeronomy Laboratory
Lawrence Livermore National Laboratory
University of Reading
NOAA Geophysical Fluid Dynamics Laboratory/Princeton University
University of Reading
Universita' degli Studi-l'Aquila
Atmospheric and Environmental Research, Inc.
University Corporation for Atmospheric Research
US
US
UK
US
UK
Italy
US
US
REVIEWERS
Daniel L. Albritton
Gerard Ancellet
Mcinrat O. Andreae
Roger Atkinson
PietJ.Aucamp
Hclmuth Bauer
Slimane Bekki
Tlbor Screes
Lane Bishop
Donald R. Blake
G. Bodcker
Rumen D. Bojkov
Byron Boville
Guy P. Brasseur
Christoph Briihl
William H. Brune
James H. Butler
Sergio Cabrera
Bruce A. Callander
Daniel A. Cariolle
William L. Chameides
Maric-Lisc Chanin
Ralph J. Cicerone
R. A. Cox
Paul J. Crutzen
John S. Daniel
Frank Dentener
Susana B. Diaz
Russell Dickerson
Tom Duafala
Christine A. Ennis
David W. Fahey
Jack Fishman
P.M. de F. Forster
NOAA Aeronomy Laboratory
Centre National de la Recherche Scientifique
Max-Planck-Institut fur Chemie
University of California at Riverside
Department of National Health
Forschungszentrum fur Umwelt und Gesundheit
University of Cambridge
Hungarian Academy of Sciences
Allied Signal, Inc.
University of California at Irvine
University of Natal/NIWA
World Meteorological Organization
National Center for Atmospheric Research
National Center for Atmospheric Research
Max-Planck-Institut fur Chemie
Pennsylvania State University
NOAA Climate Monitoring and Diagnostics Laboratory
Universidad de Chile
UK Meteorological Office
Metdo-France, Centre National de Recherches Meteorologiques
Georgia Institute of Technology
Centre National de la Recherche Scientifique
University of California at Irvine
National Environmental Research Council Headquarters
Max-Planck-Institut fur Chemie
NOAA Aeronomy Laboratory/CIRES
Wageningen Agricultural University
Austral Center of Scientific Research (CADIC/CONICET)
University of Maryland
Methyl Bromide Global Coalition
NOAA Aeronomy Laboratory/CIRES
NOAA Aeronomy Laboratory
NASA Langley Research Center
University of Reading
US
France
Germany
US
South Africa
Germany
UK
Hungary
US
US
South Africa
Switzerland
US
US'
Germany
US
US
Chile
UK
France
US
France
US
UK
Germany
US
The Netherlands
Argentina
US
US
US
US
US
UK
A.10
-------
AUTHORS, CONTRIEJUTORS, AND REVIEWERS
James Franklin
Paul J. Fraser
Lucien Froidevaux
Brian G. Gardiner
Marvin A. Geller
Amram Golombek
Thomas E. Graedel
Claire Granier
William B. Grant
Alexander Gruzdev
James E. Hansen
Neil R.P. Harris
Shiro Hatekeyama
Sachiko Hayashida
David J. Hofmann
James R. Holton
Lon L. Hood
Robert D. Hudson
Abdel M. Ibrahim
Mohammad Ilyas
Ivar S.A. Isaksen
Tomoyuki Ito
Charles H. Jackman
Daniel J. Jacob
Harold S. Johnston
P.V. Johnston
Roderic L. Jones
Torben S. J0rgensen
Igor L. Karol
Prasad Kasibhatla
Jack A. Kaye
Hennie Kelder
James B. Kerr
M.A.K. Khalil
Vyacheslav Khattatov
Volker Kirchhoff
Malcolm K.W. Ko
Antti Kulmala
Michael J. Kurylo
Murari Lai
G. LeBras
Yuan-Pern Lee
Jos Lelieveld
Robert Lesciaux
Joel Levy
J.B. Liley
Peter Liss
SolvayS.A. '
CSIRO Division of Atmospheric Research j
California Institute of Technology/Jet Propulsion Laboratory
British Antarctic Survey j
State University of New York at Stony Bropk !
Israel Institute for Biological Research j
AT&T Bell Laboratories j
National Center for Atmospheric Research i
NASA Langley Research Center j
Russian Academy of Sciences 1
NASA Goddard Institute for Space Studies j
European Ozone Research Coordinating Unit <
National Institute for the Environment '
Nara Women's University i
NOAA Climate Monitoring and Diagnostics Laboratory j
University of Washington . j
University of Arizona
University of Maryland
Egyptian Meteorological Authority
University of Science Malaysia |
Universitetet I Oslo *
Japan Meteorological Agency • j
NASA Goddard Space Flight Center !
Harvard University ,
University of California at Berkeley i
National Institute of Water and Atmospheric Research j
University of Cambridge J
Danish Meteorological Institute j
A.I. Voeikov Main Geophysical Observatory
Georgia Institute of Technology [
NASA Goddard Space Flight Center j
Koninklijk Nederlands Meteorologisch Instituut j
Atmospheric Environment Service j
Oregon Graduate Institute of Science and Technology j
Central Aerological Observatory i
Institute Nacional de Pesquisas Espaciais
Atmospheric and Environmental Research, Inc.
World Meteorological Organization j
NASA Headquarters/NIST j
Indian Institute of Technology j
Centre National de la Recherche Scientiflque j
National Tsing Hua University I
Wageningen University j
Universite de Bordeaux 1 j
NOAA Office of Global Programs
National Institute of Water and Atmospheric Research
University of East Anglia
Belgium
Australia
US
UK
US
Israel
US
US
US
Russia
US
UK
Japan
Japan
US
US
US
US
Egypt
Malaysia
Norway
Japan
US
US
US
New Zealand
UK
Denmark
Russia
US
US
The Netherlands
Canada
US
Russia
Brazil
US
Switzerland
US
India
France
Taiwan
The Netherlands
France
US
New Zealand
UK
A.ll
-------
AUTHORS, CONTRIBUTORS, AND REVIEWERS
Nicole Louisnard
Pak Sum Low
Daniel Lubin
Sasha Madronich
Jerry Mahlman
W. Andrew Matthews
Konrad Mauersberger
Archie McCulloch
Mack McFarland
Richard L. McKenzie
G6rard MŁgie
AJ. Miller
Igor Mokhov
Hideaki Nakane
Samuel J. Oltmans
Alan R. O'Neill
Michael Oppenheimer
Juan Carlos Pelaez
Stuart A. Penkett
Thomas Peter
Leon F. Phillips
Ken Pickering
Michel Pure
Giovanni Pitari
Michael J. Prather
M. Margarita Pr6ndez
John A. Pyle
Lian Xiong Qiu
V. Ramaswamy
William J.Randel
A.R. Ravishankara
Curtis P. Rinsland
Henning Rodhe
Jose M. Rodriguez
F. Sherwood Rowland
Jochen Rudolph
Nelson Sabogal
Ross Salawitch
Eugenio Sanhueza
Ulrich Schmidt
Keith P. Shine
Paul C. Simon
Susan Solomon
Johannes Staehelin
Knut Stamnes
Leopoldo Stefanutti
Richard S. Stolarski
Office National d'Etudes et Rech.erch.es Aerospatiales
United Nations Environment Programme Ozone Secretariat
University of California at San Diego
National Center for Atmospheric Research
NOAA Geophysical Fluid Dynamics Laboratory
National Institute of Water and Atmospheric Research
Max-Planck-Institut fur Kernphysik
ICI Chemicals and Polymers Limited
E.I. DuPont de Nemours and Company
National Institute of Water and Atmospheric Research
Centre National de la Recherche Scientifique
NOAA National Meteorological Center
Institute of Atmospheric Physics
National Institute for Environmental Studies
NOAA Climate Monitoring and Diagnostics Laboratory
University of Reading
Environmental Defense Fund
Institute de Meteorologia
University of East Anglia
Max-Planck-Institut fur Chemie
University of Canterbury
NASA Goddard Space Flight Center
Centre National de la Recherche Scientifique
Universita' degli Studi-l'Aquila
University of California at Irvine
Universidad de Chile
University of Cambridge
Academia Sinica
NOAA Geophysical Fluid Dynamics Laboratory/Princeton University
National Center for Atmospheric Research
NOAA Aeronomy Laboratory
NASA Langley Research Center
Stockholm University
Atmospheric and Environmental Research, Inc.
University of California at Irvine
Institut fur Chemie und Dynamik der Geosphare
United Nations Environment Programme
Harvard University
Institute Venezolano de Investigaciones Cientificas .
Forschungszentrum Jiilich
University of Reading
Institut d'Aeronomie Spatiale de Belgique
NOAA Aeronomy Laboratory
Eidgenossische Technische Hochschule Zurich
University of Alaska
Istituto di Riccrea sulle Onde Elettromagnetiche del CNR
NASA Goddard Space Flight Center
France
Kenya
US
US
US
New Zealand
Germany
UK
US
New Zealand
France
US
Russia
Japan
US
UK
US
Cuba
UK
Germany
New Zealand
US
France
Italy
US
Chile
UK
China
US
US
US
US
Sweden
US
US
Germany
Kenya
US
Venezuela
Germany
UK
Belgium
US
Switzerland
US
Italy
US
A. 12
-------
Frode Stordal
B.H. Subbaraya
Anne M. Thompson
Margaret A. Tolbert
Ralf Toumi
Karel Vanicek
Andreas Volz-Thomas
Andreas Wahner
David A. Warrilow
Robert T.Watson
E.C. Weatherhead
Ray F. Weiss
Howard Wesoky
Paul H. Wine
Donald J. Wuebbles
Vladimir Yushkov
Ahmed Zand
Reinhard Zellner
Christos Zerefos
AUTHORS, CONTRIBUTORS,
Norsk Institutt for Luftforskning ;
Physical Research Laboratory
NASA Goddard Space Flight Center ,
University of Colorado
University of Cambridge ;
Czech Hydrometeorological Institute ,j
Forschungszentrum Julich
Forschungszentrum Julich 'i
UK Department of the Environment
Office of Science and Technology Policy ;
NOAA Air Resources Laboratory ;
Scripps Institution of Oceanography
National Aeronautics and Space Administration i
Georgia Institute of Technology ;
University of Illinois |
Central Aerological Observatory \
Tehran University
Universitat Gesamthochschule Essen
Aristotle University of Thessaloniki
AND REVIEWERS
Norway
India
US
US
UK
Czech Republic
Germany
Germany
UK
US
US
US
US
US
US
Russia
. Iran
Germany
Greece
OZONE PEER-REVIEW MEETING
Daniel L. Albritton
Meinrat O. Andreae
Piet J. Aucamp
Helmuth Bauer
Lane Bishop
Rumen D. Bojkov
Byron Boville
Guy P. Brasseur
William H. Brune
Bruce A. Callander
Marie-Lise Chanin
Ralph J. Cicerone
R. A. Cox
Paul J. Crutzen
John S. Daniel
Susana B. Diaz
Tom Duafala
Christine A. Ennis
David W. Fahey
Paul J. Fraser
Marvin A. Geller
Les Diablerets, Switzerland
July 18-22, 1994 (
NOAA Aeronomy Laboratory i
Max-Planck-Institut fur Chemie i
,1
Department of National Health |
Forschungszentrum fur Umwelt und Gesundheit •
Allied Signal, Inc. !
World Meteorological Organization
National Center for Atmospheric Research 1
National Center for Atmospheric Research j
Pennsylvania State University :
UK Meteorological Office i
Centre National de la Recherche Scientifique :
University of California at Irvine j
National Environmental Research Council Headquarters [
Max-Planck-Institut fur Chemie <
NOAA Aeronomy Laboratory/CIRES
Austral Center of Scientific Research (CADIC/CONICET)
Methyl Bromide Global Coalition
NOAA Aeronomy Laboratory/CIRES
NOAA Aeronomy Laboratory
CSIRO Division of Atmospheric Research !
State University of New York at Stony Brook
i
I
A.13 j
1
1
US
Germany
South Africa
Germany
US
Switzerland
US
US
US
UK
France
US
UK
Germany
US
Argentina
US
US
US
Australia
US
-------
AUTHORS, CONTRIBUTORS, AND REVIEWERS
Amram Golombek
Neil R.P. Harris
Sachiko Hayashida
David J. Hofmann
James R. Holton
Abdel M. Ibrahim
Mohammad Ilyas
Ivar S.A. Isaksen
Tomoyuki Ito
Charles H. Jackman
Daniel J. Jacob
Rodcric L. Jones
Igor L. Karol
Hennie Kelder
James B. Kerr
M.A.K. Khalil
Malcolm K.W. Ko
Antti Kulmala
Michael J. Kurylo
MurariLal
Joel Levy
Pak Sum Low
Sasha Madronich
W. Andrew Matthews
Konrad Mauersberger
Mack McFarland
Richard L. McKenzie
Gdrard Mdgie
Hideaki Nakane
Samuel J. Oltmans
Alan R. O'Neill
Michael Oppenheimer
Juan Carlos Pelaez
Stuart A. Penkett
Michael J. Prather
M. Margarita Prdndez
Lian Xiong Qiu
V. Ramaswamy
A.R. Ravishankara
R Sherwood Rowland
Nelson Sabogal
Eugenio Sanhueza
Keith P. Shine
Paul C. Simon
Susan Solomon
Johannes Staehelin
Richard S. Stolarski
Israel Institute for Biological Research
European Ozone Research Coordinating Unit
Nara Women's University
NOAA Climate Monitoring and Diagnostics Laboratory
University of Washington
Egyptian Meteorological Authority
University of Science Malaysia
Universitetet I Oslo
Japan Meteorological Agency
NASA Goddard Space Flight Center
Harvard University
University of Cambridge
A.I. Voeikov Main Geophysical Observatory
Koninklijk Nederlands Meteorologisch Instituut
Atmospheric Environment Service
Oregon Graduate Institute of Science and Technology
Atmospheric and Environmental Research, Inc.
World Meteorological Organization.
NASA Headquarters/NIST
Indian Institute of Technology
NOAA Office of Global Programs
United Nations Environment Programme Ozone Secretariat
National Center for Atmospheric Researcb
National Institute of Water and Atmospheric Research
Max-Planck-Institut fur Kernphysik
E.I. DuPont de Nemours and Company
National Institute of Water and Atmospheric Research
Centre National de la Recherche Scientifique
National Institute for Environmental Studies
NOAA Climate Monitoring and Diagnostics Laboratory
University of Reading
Environmental Defense Fund
Institute de Meteorologia
University of East Anglia
University of California at Irvine
Universidad de Chile
Academia Sinica
NOAA Geophysical Fluid Dynamics Laboratory/Princeton University
NOAA Aeronomy Laboratory
University of California at Irvine
United Nations Environment Programme
Institute Venezolano de Investigaciones Cientificas
University of Reading
Institut d'Aeronomie Spatiale de Belgique
NOAA Aeronomy Laboratory
Eidgenossische Technische Hochschule Zurich
NASA Goddard Space Flight Center
Israel
UK
Japan
US
US
Egypt
Malaysia
Norway
Japan
US .
US
UK
Russia
The Netherlands
Canada
US
US
Switzerland
US
India
US
Kenya
US
New Zealand
Germany
US
New Zealand
France
Japan
US
UK
US
Cuba
UK
US
Chile
China
US
US
US
Kenya
Venezuela
UK
Belgium
US
Switzerland
US
A. 14
-------
AUTHORS, CONTRIBUTORS, AND REVIEWERS
Frode Stordal Norsk Institutt for Luftforskning | Norway
B.H. Subbaraya Physical Research Laboratory I India
Margaret A. Tolbert University of Colorado j US
Andreas Volz-Thomas Forschungszentrum Jiilich Germany
Andreas Wahner Forschungszentrum Jiilich ! Germany
David A. Warrilow UK Department of the Environment UK
Robert T. Watson Office of Science and Technology Policy US
Ray F. Weiss Scripps Institution of Oceanography j US
Donald J. Wuebbles University of Illinois < US
Vladimir Yushkov Central Aerological Observatory Russia
Ahmed Zand Tehran University i jjan
Christos Zerefos Aristotle University of Thessaloniki j Greece
Sponsoring Organizations Liaisons :
Rumen D. Bojkov World Meteorological Organization Switzerland
K.M. Sarma United Nations Environment Programme Kenya
Daniel L. Albritton National Oceanic and Atmospheric Administration US
Michael J. Kurylo National Aeronautics and Space Administration US
1
Coordinating Editor !
Christine A. Ennis NOAA Aeronomy Laboratory/CIRES US
ij
Editorial Staff '
Jeanne S. Waters NOAA Aeronomy Laboratory US
i
Publication Design and Layout '
University of Colorado at Boulder Publications Service:
Elizabeth C. Johnston |
Patricia L. Jensen
Andrew S. Knoedler
j
Conference Coordination and Documentation j
Rumen D. Bojkov World Meteorological Organization Switzerland
Marie-Christine Charriere World Meteorological Organization ; France
Christine A. Ennis NOAA Aeronomy Laboratory/CIRES US
Jeanne S. Waters NOAA Aeronomy Laboratory/CIRES US
Conference Support
Flo M. Ormond Birch and Davis Associates, Inc. US
Kathy A. Wolfe Computer Sciences Corporation US
A. 15
-------
-------
APPENDIX B
MAJOR ACRONYMS AND ABBREVIATIONS
AAOE
AASE
ABLE2A
ABLE 3B
AEA
AER
AERONOX
AESA
AFEAS
AGU
AGWP
AL
ALE/GAGE
ANCAT
ASL
ATLAS
ATMOS
BEF
BLP
BM
CADIC/COCINET
CCMS
CCN
CEC
CHEMRAWN
CFC
CIAB
CIAP
CIRES
CITE
CLAES
CLP
CMDL
CN
CNRM
CNRS
CSIRO
CTM
Airborne Antarctic Ozone Experiment
Airborne Arctic Stratospheric Expedition !
Amazon Boundary Layer Experiment 2A I
Arctic Boundary Layer Expedition 3B
Atomic Energy Authority (United Kingdom) ;
Atmospheric and Environmental Research, Inc. (United States)
Impact of NOX Emissions from Aircraft upon the Atmosphere
Atmospheric Effects of Stratospheric Aircraft
Alternative Fluorocarbons Environmental Acceptability Study
American Geophysical Union
Absolute Global Warming Potential
Aeronomy Laboratory (NOAA)
Atmospheric Lifetime Experiment/Global Atmospheric Gases Experiment
Abatement of Nuisance Caused by Air Traffic
above sea level ;
Atmospheric Laboratory for Applications and Science
Atmospheric Trace Molecule Spectroscopy
Bromine Efficiency Factor
Bromine Loading Potential
Brewer-Mast (ozonesonde)
Austral Center of Scientific Research/National Council of Scientific and Technological
Research (Argentina)
Committee on the Challenges of Modem Society I
cloud condensation nuclei ;
Commission of the European Communities ',
Chemical Research Applied to World Needs
chlorofluorocarbon i
Coal Industry Advisory Board '!
Climatic Impact Assessment Program
Cooperative Institute for Research in Environmental Sciences (United Stales)
Chemical Instrumentation Test and Evaluation |
Cryogenic Limb Array Etalon Spectrometer |
Chlorine Loading Potential |
Climate Monitoring and Diagnostics Laboratory (NOAA) \
condensation nuclei i
Centre National de Recherch.es Meteorologiques (France) !
Centre National de la Recherche Scientifique (France)
Commonwealth Scientific and Industrial Research Organization (Australia)
chemistry transport model
B.I
-------
ACRONYMS
DIAL Differential Absorption Laser
DNA deoxyribonucleic acid
DoY Day-of-Year
DU Dobsonunit
EASOE European Arctic Stratospheric Ozone Expedition
ECAC European Civil Aviation Conference
ECC electrochemical concentration cell (ozonesonde)
ECMWF European Centre for Medium-Range Weather Forecasts (United Kingdom)
EESC equivalent effective stratospheric chlorine
El Emissions Index
EMEP MSC-W European Monitoring and Evaluation Programme, Meteorological Synthesizing Centre — West
EMEX Equatorial Mesoscale Experiment
ENSO El Nino-Southern Oscillation
EPA Environmental Protection Agency (United States)
ES A European Space Agency
ETBL equivalent tropospheric bromine loading
ETCL equivalent tropospheric chlorine loading
FDH Fixed Dynamical Heating
FTIR Fourier transform infrared spectrometer
GAGE Global Atmospheric Gases Experiment
GCM general circulation model
GFDL Geophysical Fluid Dynamics Laboratory (NOAA)
GISS Goddard Institute for Space Studies (United States)
GIT Georgia Institute of Technology (United States)
GMT Greenwich Mean Time
GSFC Goddard Space Flight Center (NASA)
GWP Global Warming Potential
HALOE Halogen Occultation Experiment
HC hydrocarbon
HCFC hydrochlorofluorocarbon
HFC hydrofluorocarbon
HSCT High Speed Civil Transport
HSRP High Speed Research Program
ICAO International Civil Aviation Organization
EEA International Energy Agency
IIT Indian Institute of Technology
INPE Institute Nacional de Pesquisas Espaciais (Brazil)
IOTP International Ozone Trends Panel
IPCC Intergovernmental Panel on Climate Change
IR infrared
ISAMS Improved Stratospheric and Mesospheric Sounder
IUPAC International Union of Pure and Applied Chemistry
IVIC Institute Venezolano de Investigaciones Cientificas (Venezuela)
B.2
-------
JPL
KNMI
LIMS
LLNL
LRC
LTO
MIPAS
MLOPEX
MLS
MOZAIC
MPI
MPIA
MPIC
MRI
MSU
NACNEMS
NAD
NASA
NAT
NCAR
NCSU
NESDIS
NH
NILU
NIR
NIST
NIWA
NMC
NMHC
NOAA
NPP
NRC
NSF
NYU
ODP
ODW
OECD
OSE
OTP
Jet Propulsion Laboratory (California Institute of Technology; United States)
Koninklijk Nederlands Meteorologisch Instituut
Limb Infrared Monitor of the Stratosphere
Lawrence Livermore National Laboratory (United States)
Langley Research Center (NASA)
Landing/Take-Off cycle
Michelson Interferometric Passive Atmosphere Sounder
Mauna Loa Observatory Photochemistry Experiment
Microwave Limb Sounder
Measurement of Ozone on Airbus In-service Aircraft
Max-Planck-Institute (Germany)
Max-Planck-Institute for Aeronomy (Germany)
Max-Planck-Institute for Chemistry (Germany)
Meteorological Research Institute (Japan)
Microwave Sounder Unit
North American Cooperative Network of Enhanced Measurement Sites
nitric acid dihydrate
National Aeronautics and Space Administration (United States)
nitric acid trihydrate
National Center for Atmospheric Research (United States)
North Carolina State University (United States)
National Environmental Satellite, Data, and Information Service (NOAA)
Northern Hemisphere
Norsk Institutt for Luftforskning (Oslo)
near infrared
National Institute of Standards and Technology (formerly NB.S; United States)
National Institute of Water and Atmospheric Research, Ltd. (New Zealand)
National Meteorological Center (United States)
non-methane hydrocarbon
National Oceanic and Atmospheric Administration (United States)
net primary productivity i
National Research Council (United States)
National Science Foundation (United States)
New York University (United States)
Ozone Depletion Potential
Ozone Data for the World
Organization for Economic Cooperation and Development (Paris)
ozonesonde instrument used in former East Germany; similar to Brewer-Mast
Ozone Trends Panel
ACRONYMS
B.3
-------
ACRONYMS
PAN
PEL
PFCs
POLINAT
ppbm
ppbv
ppmv
pptv
PSCs
PV
QBO
RAF
RB
SAGE
SAM II
SAMS
SAOZ
SAT'
SBUV
SH
SOS/SONIA
SPADE
SPEs
SSA
SSBUV
STE
STEP
STP
STRATOZ
SUNY
SUSIM
SZA
TFA
TIROS
TNO
TOMS
TOR
TOYS
TROPOZII
peroxyacetyl nitrate
planetary boundary layer
perfluorocarbons
Pollution from Aircraft Emissions in the North Atlantic Flight Corridor
parts per billion by mass
parts per billion by volume
parts per million by volume
parts per trillion by volume
polar stratospheric clouds
potential vorticity
quasi-biennial oscillation
Radiation Amplification Factor
Robertson-Berger (UV irradiance meter)
Stratospheric Aerosol and Gas Experiment
Stratospheric Aerosol Measurement
Stratospheric and Mesospheric Sounder
Systeme d' Analyse par Observation Z6nithale
sulfuric acid tetrahydrate
Solar Backscatter Ultraviolet spectrometer
Southern Hemisphere
Southern Oxidants Study/Southeast Oxidant and Nitrogen Intensive Analysis
Stratospheric Photochemistry, Aerosols and Dynamics Expedition
solar proton events
stratospheric sulfuric acid aerosol
Shuttle Solar Backscatter Ultraviolet spectrometer
stratosphere-troposphere exchange
Stratosphere-Troposphere Exchange Project
standard temperature and pressure
Stratospheric Ozone expedition
State University of New York (United States)
Solar Ultraviolet Spectral Irradiance Monitor
solar zenith angle
trifluoroacetic acid
Television and Infrared Observation Satellite
Netherlands Organization for Applied Scientific Research
Total Ozone Mapping Spectrometer
Tropospheric Ozone Research
TIROS Operational Vertical Sounder
Tropospheric Ozone II expedition
B.4
-------
ACRONYMS
UARS Upper Atmosphere Research Satellite
UCI University of California at Irvine (United States)
UEA University of East Anglia (United Kingdom)
UKMO United Kingdom Meteorological Office
UNEP United Nations Environment Programme
UV ultraviolet
UV-A ultraviolet-A
UV-B ultraviolet-B
VOC volatile organic compound
WCRP World Climate Research Programme
WMO World Meteorological Organization
WODC World Ozone Data Center
B.5
-------
-------
APPENDIX C
CHEMICAL FORMULAE AND NOMENCLATURE
HALOGEN-CONTAINING SPECIES '
Cl
CIO
OC1O
C1202
C1ONO
C1ONO2
HC1
HOC1
F
FO
HF
SF6
HALOCARBONS
atomic chlorine
chlorine monoxide
chlorine dioxide
dichlorine peroxide (CIO dimer)
chlorine nitrite
chlorine nitrate
hydrogen chloride (hydrochloric acid)
hypochlorous acid
atomic fluorine
fluorine monoxide
hydrogen fluoride (hydrofluoric acid)
sulfur hexafluoride
Chlorofluorocarbons (CFCs)
CFC-10
CFC-11
CFC-12
CFC-13
CFC-14
CFC-113
. CFC-114
CFC-115
CFC-116
ecu
CC13F
CC12F2
CC1F3
CF4
CC12FCC1F2
CC1F2CCIF2
CC1F2CF3
CF3CF3
Br
BrO
BrNO2
BrONO2
HBr
HOBr
I
10
HI
IONO2
atomic bromine
bromine monoxide
•
bromine nitrite
bromine nitrate
hydrogen bromide
hypobroitnous acid
atomic iodine
iodine monoxide
hydrogen iodide
iodine nitrate
Hydrochlorofluorocarbons (HCFCs)
, HCFC-21
HCFC-22
HCFC-30
HCFC-40
HCFC-123
HCFC-124
HCFC-141b
HCFC-142b
HCFC-225ca
HCFC-225cb
CHC12F
CHF2CL
CH2C12
CH3C1
CF3CHC12
CF3CHFCl
' CFC12CH3
CF2C1CH3
CF3CF2CHC12
CF2C1CF2CHFC1
Hydrofluorocarbons (HFCs)
HFC-23
HFC-32
HFC-41
HFC- 125
HFC- 134
HFC-l34a
HFC-143
HFC-143a
CHF3
CH2F2
CH3F
CHF2CF3
CHF2CHF2
CH2FCF3
CHF2CH2F
CH3CF3
HFC-152a
HFC-227ea
HFC-236cb
HFC-236ea
HFC-236fa
HFC-245ca
HFC-43-10mee
CH3CHF2
CF3CHFCF3
CF3CF2CH2F
CF2CHFCHF2
CF3CH;>CF3I
CHF2CF2CFH2
CF3CH1FCHFCF2CF3
!
C.I
-------
CHEMICAL FORMULAE
Halons
halon-1211
halon-1301
halon-2402
Others
CH3C1
CH2C12
CHC13
CCLj
CH3CC13
C2HC13
C2C14
COC12
CF4
CHFj
TFA
CHClBf2
CF3Br
CR^CII
CF3I
CF2CIBr
CF3Br
methyl chloride
methylene chloride, dichloromethane
chloroform, trichloromethane
carbon tetrachloride
methyl chloroform
trichloroethylene
tetrachloroethylene
phosgene, carbonyl chloride
perfluoromethane
perfluoroethane
perfluoropropane
perfluorocyclobutane
perfluorohexane
fluoroform, trifluoromethane
trifluoroacetic acid (CF3C(O)OH)
dibromochloromethane
trifluorobromomethane (halon- 1301)
chloroiodomethane
trifluormethyl iodide
iodopentafluoroethane
CH3Br
CH2Br2
CHBr3
methyl bromide
methylene bromide, dibromomethane
bromoform, tribromomethane
ethylene dibromide;
1,2 dibromoethane
CH3I
methyl iodide
COFC1
fluorophosgene
OTHER CHEMICAL SPECIES
o
02
o3
0('D)
ox
atomic oxygen
molecular oxygen
ozone
atomic oxygen (first excited state)
odd oxygen (O, O(tD), O3)
H atomic hydrogen
H2 molecular hydrogen
OH hydroxyl radical
HO2 hydroperoxyl radical
H2O water
H2O2 hydrogen peroxide
HOX odd hydrogen (H, OH, HO2, H2O2)
C.2
-------
CHEMICAL FORMULAE
N
N2
N20
NO
NO2
NO3
N205
C1ONO2
HN02,HONO
HN03
RONO2
NO3-
S
SO2
SOX
H2S04
SAT
S04=
Be
Pb
Sr
C
CO
C02
HC
NMHC
VOC
CH4
C3H8
C2H2
C5H8
CFCs
HCFCs
HFCs
atomic nitrogen
molecular nitrogen
nitrous oxide
nitric oxide
nitrogen dioxide
nitrogen trioxide, nitrate radical
dinitrogen pentoxide
chlorine nitrate
nitrous acid
nitric acid
alkyl nitrates
nitrate ion
atomic sulfur
sulfur dioxide
sulfur oxides
sulfuric acid
sulfuric acid tetrahydrate
(H2S04-4H20)
sulfate ion
beryllium
lead
strontium
carbon
carbon monoxide
carbon dioxide
hydrocarbon
non-methane hydrocarbon
volatile organic compound
methane
ethane
propane
ethylene, ethene
acetylene, ethyne
isoprene (2-methyl 1,3 butadiene)
benzene
chlorofluorocarbons*
hydrochlorofluorocarbons*
hydrofluorocarbons *
HO2NO2
ROONO2
PAN
NOV
NOX
NAD
NAT
SF6
CS2
COS, OCS
Kr
Rn
CH20
CH3OH
RO
CH3OOH
CH3COO
R02
CH3C(0)00
peroxynitric acid
peroxynitrates
peroxyacetylaitrate
(CH3C(O)OON02)
odd nitrogen (usually including
NO, N02, N03, N205, C10N02,
HNo4, HNO3)
oxides of nitrogen (NO + NO2)
nitric acid dihydrate
(HNO3-2H2O)
nitric acid trihydrate
(HNO3-3H:2O)
sulfur hexaifluoride
carbon disulfide
carbonyl sulfide
krypton
radon
formaldehyde
methanol !
alkoxy radicals
methyl hydroperoxide
methyl peroxy radical
organic peroxy radical
acetyl peroxy radical
Family of compounds; see above for individual species
C.3
-------
-------
-------
------- |