Diapositive 1

1
MOZAIC Measurement of Ozone, Water Vapor, Carbon
Monoxide, Nitrogen Oxide by Airbus In-Service
Aircraft
Valérie Thouret (CNRS, Toulouse, France) (
valerie.thouret_at_aero.obs-mip.fr ) with
contributions from the MOZAIC PIs Jean-Pierre
Cammas, Philippe Nédélec, Andreas Volz-Thomas,
Herman Smit, Fernand Karcher, Vincent-Henri
Peuch, Kathy Law. and from several MOZAIC CoIs
(Europe, USA, China, Africa) http//mozaic.aero.o
bs-mip.fr/web/
2
MOZAIC INSTRUMENTATION (on commercial aircraft)
PITOT TUBES ON THE AIRCRAFT FUSELAGE
INSTRUMENTATION BELOW THE COCKPIT O3/CO/H2O/NOy/
DAS
3
MOZAIC Consortium
Forschunszentrum Jülich
? The transport is free of charge !!
4
MOZAIC Measurement of Ozone, Water Vapor, Carbon
Monoxide, Nitrogen Oxide by Airbus In-Service
Aircraft

• Why ? Comprehensive and continuous observations
  are needed to contribute to the assessment of
  climate change and of the impact of aircraft
• Ozone and water vapor in UT/LS have a key role
  in climate
• Large natural variability in space and time

• The general aim of MOZAIC is to collect data for
  studies on
• ? Atmospheric transport and chemical processes
• ? Budgets of ozone and water vapor in the UT/LS
  (and CO and NOy)
• ? Inter-annual variability and long term
  tendencies of the atmospheric composition
• Validation of global chemistry-transport models
  and regional air quality models
• Validation of new satellite observations
• Assessment of aircraft impact on the atmosphere

Passenger aircraft are a very powerful observing
platform daily flights
5
MOZAIC I, II III (1994-2006) as EU project
MOZAIC Aug. 1994 June. 2009
1 record every 4s ? High resolution vertical
profiles during take-off and landing ( 20m) High
resolution horizontal profiles at 9-12 km
altitude during inter-continental flight ( 1
km) Regularity of measurements with 5 aircraft
flying almost every day
10
3
9
32 000 Flights 230 000 Hours

• 1994-2002 20 000 flights with O3/H2O
• 2002-2009 12 000 flights with additional CO
  (4 aircraft) and with additional CO/NOy (1
  aircraft)

6
Status of MOZAIC
From 5 to 3 aircraft

• MOZAIC fleet in October 2009
• Lufthansa 2 aircraft (one with the NOy
  instrument)
• Air Namibia 1 aircraft since 2006 (transport
  maintenance costs CNRS and FZJ) will probably
  stop at the end of the year
• Air France MOZAIC instrumentation dropped off
  in 2004
• Austrian MOZAIC instrumentation dropped off
  in 2006

• MOZAIC database
• Data base opened to co-investigators under a data
  protocol
• On-line request to access the data
• http//mozaic.aero.obs-mip.fr/web/

7
MOZAIC

• Work by Principal Investigators and
  co-investigators (gt60)
• Climatology, inter-annual variability,
  tendencies
• Stratosphere-Troposphere Exchanges
• Transport and transformation of pollution on
  global scale
• Validation of chemistry-transport models and
  regional air quality models
• Validation of satellite sensors for the
  atmospheric chemistry
• Aircraft impact on the atmosphere
• Regional air quality
• Literature
• gt 150 publications in international journal
  since 1997

8
Selection of the MOZAIC  best  results

• Thanks to the definition of the program
• Statistics
• Specific locations (Africa, Asia, UTLS)
• Accuracy and precision of in situ data (CO
  profiles)
• Different from results based on
• Aircraft campaign
• Satellite data
• Sondes network
• A few topics
• H2O, O3, CO, NOy global distributions
• Impact of boreal fires in the UTLS
• New data over Africa and China
• European Summer Heatwave in 2003 (Tressol et al.,
  ACP 2008)
• MOZAIC use in GEMS, MACC and future GAS (Ordonez
  et al., ACPD 2009)
• Tropopause and STEs (Brioude et al., JGR 2006,
  2008)
• Focus on O3 CO  trends  in the UTLS

9
Quasi global climatologies

• In the troposphere and in the UTLS
• H20, O3, CO, NOy

10
MOZAIC Relative Humidity in UT over North
Atlantic comparison with ECMWF
ECMWF-MOZAIC

• MOZAIC provided the first climatology of UTH
• UT is much wetter than assumed before MOZAIC
• ECMWF does not reproduce ice super saturation

Gierens et al., 1997, 1998, 1999, 2004
Spichtinger et al., 2002, 2003 MOZAIC H20 PI
Herman Smit at FZJ julich, Germany
11
MOZAIC is watching on both sides of the tropopause
Example over New York City, for the year 1997
5 flight levels 196, 216, 238, 262, 287 hPa
- 150 hPa
LS
- 45 hPa
- 15 hPa
Tropopause, PV2
15 hPa
UT
75 hPa
Thouret et al. ACP, 2006
North of 25N
12
Thouret et al. ACP, 2006

• Seasonal maximum in spring in the LS (O3gt150
  ppb), in summer in the UT (O3lt100 ppb)
• Intermediate seasonal cycle and O3 concentration
  at the tropopause

13
O3 in the Upper TroposphereClimatology 1994 -
2004
Winter (DJF)
Summer (JJA)

• O3 summer maximum in the UT
• Higher concentrations of both O3 and CO in the
  eastern hemisphere
• The Black Sea region is characterized by an O3
  maximum and a CO minimum (strong stratospheric
  influence ?)

Adapted from Thouret et al., 2006
14
O3 in the Lower Stratosphere Climatology 1994 -
2004
Summer (JJA)
Winter (DJF)

• Spring maximum in the LS

15
CO in the Upper Troposphere
Winter (DJF)
Summer (JJA)

• CO maximum in spring or in summer (if boreal
  fires) in the UT
• Higher concentrations of both O3 and CO in the
  eastern hemisphere
• The Black Sea region is characterized by an O3
  maximum and a CO minimum (strong stratospheric
  influence)
• CO maxima over burning regions (West Africa in
  DJF, Central Africa in JJA)

Thouret et al., in preparation
16
CO in the Lower Stratosphere
Winter (DJF)
Summer (JJA)

• no clear seasonal cycle in the LS

17
NOy in the Upper Troposphere (MOZAIC 2001-2008)
Enhanced NOy over USA and Europe in summer is due
to convective transport and lightning Up to 3 ppb
NOy (average!) Not reproduced by CTMs !
No significant enhancement in flight corridor
during winter lt 0.5 ppb NOy
Volz-Thomas et al., 2009 in preparation
18
Impact of boreal fires in the UTLS

• High CO concentrations in the UTLS

19
Biomass fires over Northeastern Asia in Spring
2003 evidence of daily extreme CO
concentrations in the upper troposphere with
MOZAIC observations
Nédélec et al., GRL, 2005
20
Seasonal signature of plumes of biomass fires
injected in the upper troposphere over
Northeastern Asia in 2003 comparison with Europe
Nédélec et al., GRL, 2005
- Very high CO in the UT between April and
August, much higher than what observed over
Europe even during the summer heat wave. - 2003
is clearly anomalous
21
Interannual variability of plumes of biomass
fires injected in the upper troposphere over
Northeastern Asia 2002-2006

• Direct impact on O3 distribution ?
• - Global effect ?

CO
O3

• CO anomalies in the UT
• Aug-Dec 2002
• Apr-July 2003
• Feb 2006 ???
• CO anomalies in the LS
• - July 2002,03,05 only

22
Injection in the lower stratosphere of biomass
fire emissions followed by long-range transport
a MOZAIC case study
300
MOZAIC time series shows an ozone-rich
upper-level trough with embedded CO peaks over
Eastern Atlantic
200
100
Backtrajectories on 5 days indicate that a likely
source of CO plumes is over the region west of
Great Bear Lake over Northwest Territories of
Canada on 25-27 June 2004
MODIS image, June 24, 2004
Meso-NH runs did reproduce injection in LS
Cammas et al., ACP, 2009
23
Regions of interest not (well) sampled before

• I-) Africa

24
MOZAIC over Africa since 1997
Some of the main African O3 precursors biomass
burning, biogenic VOCs, and LiNOx
Dec. Jan. Feb.
Jul. Aug. Sep.
ATSR biomass fire detection
Lagos 400 MOZAIC profiles
Brazzaville 200 MOZAIC profiles
LIS lightning detection
MOZAIC transect by Air Namibia Daily flight
since 2006
Sauvage et al., ACP 2005, JGR 2006, ACP 2007
25
Ozone distribution in the upper troposphere
MOZAIC measurements over West Africa (9-12 km)
Seasonal mean JJA over the period 1997-2004
Ozone (ppbv) HR () V(m/s)
Sauvage et al., GRL 2007
Ozone meridional gradient with min. at the
latitude of the ITCZ (7N) What are the relative
contributions - of the photochemical
production of ozone and of the transport ? -
of the surface sources and of the source of NO
from Lightning ? gt 2D study in the
latitude-altitude reference to assess the role of
Lightning
Bilan O3P - Mai 2009
25
26
Meridional Distribution Analysis with the meso
scale model Meso-NH
sources of O3 (ppb/hour)
without LNOx
with LNOx

• Chemical Production maximun where NOx maximum
• Convection source of O3 poor air in the UT gt
  minimum
• LNOX gt keep high P(O3) in the branches gt
  gradient

Saunois et al, JGR 2008
27
Africa continent of large ozone variability !!
Three years of daily transects (50N-20S) in the
UT over Africa
Sept. - Oct. - Nov.
Large ozone variability in the upper troposphere,
observed at all seasons. Variability of sources,
convective transport, large-scale transport
pathways
Cammas et al., in preparation
28
Long-range transport of biomass burning plumes
over Windhoek (Namibia)
Influence of South American fires In the UT over
austral Africa
Biomass Burning Plume
MOZAIC vertical profile over Windhoek, 05/09/2006
Retroplume of air parcels sampled inside the
biomass burning plume over Windhoek (see MOZAIC
vertical profile on left figure) within the lower
troposphere (0-3 km altitude) and during the
period August 29-23, 2006. Grey circles biomass
fires detected by MODIS during the period of
interest.
Cammas et al., in preparation
29
Regions of interest not (well) sampled before
II-) China
30
Tropospheric ozone climatology over Beijing
Comparisons of mean profiles O3 profiles in the
lower troposphere over Beijing 1995-1999 and
2000-2005. The solid lines are annual mean
values, and the dashed lines on the right hand
side represent the data collected in summer
afternoons at local times 15h-16h in
May-June-July
Ding et al., ACP, 2008
31
5 to 20 ppbv more in the PBL probably due to the
increase of precursors emmisions in the city
of Beijing Example 3 times more cars in 10
years !
No similar  trend  over Paris, New York or
Tokyo ! Effect of emissions regulations ?
Ding et al., ACP, 2008
Collaboration established to further study data
over Hong Kong and Shangai
32
CO vertical profiles over Beijing and Tokyo
(compared to Frankfurt)
1000 ppb
2000 ppb

• Still High CO in China
• High CO in the UT over Tokyo
• Export from continental Asia (boreal fires ?)

• Very high CO up to 4-6 km altitude
• over Beijing.
• Tokyo, quite similar to Frankfurt

33
Focus on UTLS, longer term variabililty and
trends

• Already published manuscripts
• Bortz et al., JGR 2007 (Upper Tropical
  Troposphere O3 increase of 1 ppb/yr)
• Thouret et al., ACP 2006 (UTLS North Atlantic,
  O3)
• Koumoutsaris et al., JGR 2008 (Geos-Chem CTM)
• Schnadt et al., ACP 2009 (GASP, MOZAIC and
  sondes, better see following talk by Johannes
  Staehelin)
• An update with CO and O3 recent MOZAIC data

34
Characteristics of the first 10 years of MOZAIC
measurements In both UTLS The 1998-1999 anomaly
Europe
1.00 /- 0.99 /yr
1.99 /- 1.21 /yr
UT
LS
Thouret et al., ACP 2006
35
Same behaviour over the eastern US in both UTLS
The 1998-1999 anomaly
Eastern US
0.99 /- 0.82 /yr
0.80 /- 1.06 /yr
UT
LS
Thouret et al., ACP 2006
36
O3 increase of about 1/year in both the UT and
the LS Causes global anthropogenic pollution
and large scale modes of atmospheric variability.
Transport and circulation anomalies in 1998-1999
? Correlations with NAO in the UT and NAM in the
LS
Strong interannual variability partly linked to
the variability of the tropospheric and
stratospheric circulation, same anomalies ?
strong coupling between UT and LS
Thouret et al., ACP 2006
37
The 1998-1999 anomaly as seen by GEOS-Chem over
the period 1987-2005
GEOS chem over US East

• A non negligeable influence of the extreme El
  Nino event in 1997 through
• STEs
• Asian pollution export and transport toward
  Europe because of a change in convective activity
  in East Asia and a strenghtening of the
  subtropical jet stream

Koumoutsaris et al., JGR 2008
38
Regions quite well documented to assess the
interannual variabilities and first decennal
trends of O3 (1995-2009) CO (2002-2009)
O3 (CO)
O3 (CO)
Work in progress
39
EU west
USA East
LS
UT
consistent with Mace Head time series. !!! By
chance or does it Mean something ?
2 periods 1995-2000 (O3 increase) and 2001-2009
(levelling off) In both regions, in both UT LS
40
East Asia
Black sea
LS
Not consistent with caucasian site. Comparable ?
UT
Different behaviour O3 increase in the UT
almost no trend in the LS Max of O3 increase in
the UT over Asia since 2001 Still 1998-1999
anomaly in the LS (global NH feature)
41
USA East
CO in the UT
EU west
East Asia
Black sea
CO from 2002 general decrease in the UT, even
over Asia (not significant More the effect of
spring-summer 2003). Largest decrease over the
US. Overall min over Black Sea and max over Asia.
42
Summary Linear Fit (/year) ?
1994-2003
Thouret et al., ACP 2006
1994-2007
Are not statistically significant
Thouret et al., EGU 2009
43
Interannual variabilities and first decennal
trends over selected regions

• O3
• General increase in the UT 0.7 to 1.6 /yr
  (dep. on the lenghth of the time series and
  regions)
• Largest and linear increase over the Western
  Medit., Black Sea and East Asia
• Levelling off over EU west and US east after 2001
• No significant signal in the LS (zonal
  homogeneity) before or after 2000-2001 ?
• 1998-1999 anomalies still characterize the time
  series (influence of the strong El-Nino in 1997
  at hemispheric scale ?) in both UT and LS
• CO in the UT
• General decrease (even over Asia) -1/yr
  how come ?
• Largest decrease over USA East
• Positive anomalies 2002 (global, biomass
  burning), over Asia in 2003 and over Med West in
  2007
• How to go further and attribute (quantify) causes
  ?

44
Conclusion

• MOZAIC (1993-2009) has demonstrated
• High quality data of regular measurements
• Case studies and Statistical analysis
• Complementarity to other networks of in situ data
  (sondes, research aircraft campaigns, surfaces
  sites)
• Validation of Models and Satellites
• Long term feasability and  low cost  data base
  (500 /flight including 2 profiles)
• MOZAIC will become IAGOS from 2010 on !
• New measurements (Cloud droplets, NOx, CO2, CH4,
  Aerosols), more aircraft (20), NRT transmission
  of data
• Longer time scale (20 years), new sampled regions
  (Pacific with China Airline and Cathay Pacific
  involved)

45
In-service Aircraft for a Global Observing System
European Research Infrastructure
IAGOS will form a cornerstone of IGACO/GEOSS and
of the European GMES initiative
www.iagos.org
46
Thanks for your attention
thov _at_ aero.obs-mip.fr http //mozaic.aero.obs-mip
.fr http//www.iagos.org
47
Scientific Value
Objectives

• Changes in the Tropopause region (O3 and H2O
  trends)
• Spatial and temporal resolution
• Higher resolution than satellites (UTLS)
• Higher frequency than Ozone-sondes
• Validation of Climate Models and Satellite
  Retrievals
• Tropospheric profiles of CO, CO2, NOx, aerosol,
  ..
• Global Air Quality
• Profiles of NOx, Aerosols
• Influence of developing regions
• International Transfer Standard
• Same systems everywhere
• Regular Quality Assurance
• Cost effective
• 500 /flight including 2 profiles

• Equip 10-20 longhaul aircraft with scientific
  instruments for
• chemical composition (H2O, O3, CO, CO2, NOx,
  NOy),
• aerosol and cloud particles
• Longterm deployment
• Global coverage (same instrument)
• Open data policy
• Near-realtime data provision to GMES-GAS and Met
  Services

48
(No Transcript)
49
IAGOS instrumentation
Package 1 (72 kg) Basic mandatory system
O3 CO Relative humidity Cloud droplets and ice
crystals Meteorological parameters Data
acquisition and transmission
Package 2 (50 kg) Optional extra 2a NOy 2b
(NOx NO NO2) 2c Aerosol (gt2012 ?) 2d
CO2 CH4 (gt2012 ?)
50
Ozone Instrument validation MOZAIC ozone
sounding at Hohenpeissenberg (48N, 11E)

• MOZAIC O3 instrument improved version of
  commercial dual-beam UV absorption instrument
  (Thermo-Electron, model 49-103),
• Accuracy ? 2 ppbv / precision ? 2 / response
  time 4 sec
• -Sensor replacement pre- and post-calibration
  in laboratory every 6-10 months, with NIST
• Automatic calibration in flight (zero Ozone
  generator)
• To make sure there is no derive !
• Marenco et al., JGR, 1998

Seasonal variations of ozone monthly averages
for the Hohenpeissenberg sounding station
1980-93 compared with MOZAIC data over
Frankfurt Sept 1994-Aug 1996 at 5 standard
pressure levels. Standard deviations are plotted
as error bars with large cap for MOZAIC and
small cap for Hohenpeissenberg. MOZAIC is 3-13
higher than sondes data Thouret et al., JGR,
1998
51

• CO INSTRUMENT FOR MOZAIC
• OPERATION ON COMMERCIAL AIRCRAFT
• FULLY AUTOMATIC, NO CALIBRATION GAS, NO
  MAINTENANCE
• PRINCIPLE CHOICE EXISTING IR ANALYSER TEI
  MODEL48CTL
• LESS SENSITIVE THAN OTHER TECHNIQUES, BUT STABLE
  IN TIME
• IMPROVEMENTS (30 SEC) 5 , 5 ppbv
• IR DETECTOR MORE SENSITIVE, TEMP. REGULATED
• INTERFERENCE ELIMINATION H2O
• PERIODICAL ZERO MEASUREMENT (20 min.)
• OPERATION TO CABIN PRESSURE VIA EXTERNAL PUMP
• CELL PRESSURISATION TO 2.5 BAR
• ELECTRICAL GROUNDING OPTIMISATION (LOW SIGNALS)
• CALIBRATION WITH NIST CO BOTTLES (?1)
• ..

Nedelec P., J.-P. Cammas, V. Thouret, G. Athier,
J.-M. Cousin, C. Legrand, C. Abonnel, F. Lecoeur,
G. Cayez, and C. Marizy An improved infrared
carbon monoxide analyser for routine measurements
aboard commercial airbus aircraft Technical
validation and first scientific results of the
MOZAIC III programme , Atmos. Chem. Phys., 3,
1551-1564, 2003