Longterm ozone changes in UTLS in the northern hemisphere

1
Long-term ozone changes in UT/LS in the northern
hemisphere

• Johannes Staehelin,
• Christina Schnadt Poberaj,
• and Dominik Brunner
• Institute for Atmospheric and Climate Science,
  ETHZ, Zürich, Switzerland
• Present Address Empa, Dübendorf, Switzerland

2
1. Introduction

• Ozone is strong greenhouse gas in the UT/LS, but
  information about long term changes (UT/LS) is
  very sparse
• In the past information mainly from regular
  ozonesondes
• In this presentation Analysis of regular
  aircraft measurements compared with ozonesondes
• C. Schnadt Poberaj, J. Staehelin, D. Brunner, V.
  Thouret, H. De Backer, and R. Stübi Long-term
  changes in UT/LS ozone between the late 1970s and
  the 1990s deduced from the GASP and MOZAIC
  aircraft programs and from ozonesondes, Atm.
  Chem. Phys., 9, 5343-5369 (2009)

3
Presentation

• 2. Measurements of ozone in upper troposphere/
  lower stratosphere (UT/LS)
• 3. Data treatment
• 4. Ozone changes in UT from regular aircraft
  measurements
• 5. Comparison of UT ozone changes Regular
  aircraft vs. ozonesondes
• 6. LS ozone changes (regular aircraft)
• 7. Conclusions and outlook

4
2. Measurements of Ozone in upper troposphere/
lower stratosphere (UT/LS)a. regular aircraft
5
Measurements from regular passenger aircraft
GASP Global Atmospheric Sampling Program
MOZAIC Measurement of Ozone and Water Vapor by
Airbus In-Service Aircraft ProgramFlight
RoutesGASP (1975-1979) MOZAIC (1994-2001)
6
b. Ozonesonde measurements available for
comparison with GASP/MOZAIC

• BM sonde data have been used with correction
  factors (CF) applied
• Range of allowed CF (BM) 0.9-1.35 (Uccle,
  Payerne), and 0.9-1.2 (Hohenpeissenberg)
• Wallops Island Data normalised using SBUV column
  ozone information
• Sonde data have been corrected for response time
  of the ozone and pressure sensors

7
3. Data treatmenta. regular aircraft measurements

• 9 high bias of GASP data removed (caused by a
  calibration bias (Tiefermann, 1979))
• Spuriously low values (Wozniak, 1979)
• Excluded values lower than 10 ppbv in UT and
  lower than 30 ppbv in LS
• Vertical analysis range
• 330 hPa195 hPa ( 8.5 km11.9 km).
• In UT, 330 hPa235 hPa ( 8.5 km10.8 km)

8
Data treatment of regular aircraft measurements,
cont.

• UT ozone not allowed to exceed seasonally
  variable upper limit in mixing ratio to avoid
  significant biases of individual averages in
  regions with limited sample sizes (recent
  stratospheric intrusions)
• Probability Density Function of MOZAIC UT ozone
  to determine cutoff values
• DJF 80 ppbv
• MAM 120 ppbv
• JJA 120 ppbv
• SON 90 ppbv

9
b. Treatment of aircraft and sonde data

• Discrimination between tropospheric and
  stratospheric ozone At extratropics Interpolate
  ERA40 dynamical tropopause information (2 PVU)
  onto GASP, MOZAIC, and ozonesonde coordinates. In
  tropics use thermal tropopause.
• Horizontal averaging
• 1. Compute daily means over predefined regions
  and 10x10 grid for GASP and MOZAIC UT data
• 2. Calculate multiannual averages of GASP,
  MOZAIC and ozone sondes
• Vertical scaling for comparison with ozonesondes
  Use potential temperature against tropospause,
  binning over vertical bands

10
c. Comparisons of climatologies determination of
changes

• Comparison of long-term changes by aircraft
• and ozonesondes
• - Average GASP/MOZAIC data over Europe
  (42N57N, 5W20E) and USA East
  (30N-50N, 90W-60W) and compute differences
• - Average ozonesonde data over 1975-79 and
  1994-2001 periods and compute differences

11
4. Ozone changes in UT from regular aircraft
measurements
spring
winter
GASP (1975-1979)
fall
summer
winter
MOZAIC (1994-2001)
spring
summer
fall
12
Relative UT ozone concentration changesMOZAIC
(1994-2001)/GASP (1975-1979), in
Spring
Winter
Summer
Fall
Relative difference between GASP and MOZAIC UT
ozone. Data averaged over a 1010o grid.
Troposph. identification2 PVU tropopause in
extratropics, thermal tropopause in tropics (lat.
lt35 N). Grey triangles GASP data biased toward
one year (50 from one year), pink triangles
GASP data available from three years only.
Hatched boxes differences statist. sign. at the
95 level. (Differences only where data from at
least three years are available for the GASP and
number of daily means available for averaging is
ten or more.
13
Relative change in anthropogenic surface NOx
emissions (19751979 vs. 19942000) in from
RETRO (TEAM-Model, Schultz, 2007). Emission
sources power generation, industrial,
residential, and commercial combustion,
transport, and ships.
14
Long-term changes of UT ozone (MOZAIC-GASP/GASP,
in ) Function of season for regions (W USA, NE
USA, ATL, EUR, ME, N IND, N JP, S JP, S CHINA).
Relative differences only if 10 or more daily
regional averages available for GASP and MOZAIC.
Vertical bars 95 confidence intervals of diff..
Bottom rows of numbers numbers of daily means
available for averaging (upper row GASP, lower
row MOZAIC). Numbers are colored in orange (red)
if GASP or MOZAIC data are biased toward one
year 5075 of data from one year (gt75 from one
year). Grey triangles mark regional averages for
which data from three years only are available
for averaging.
15
5. Comparison of UT ozone changes Regular
aircraft vs. ozonesondesAircraft measurements
vs. European Brewer Mast ozonesondes (Uccle,
Hohenpeissenberg, Payerne)
16
Comparison of ozone changes from aircraft vs.
European Brewer Mast sondes
Winter
Spring
Summer
Fall
Relative differences of multi-annual mean UT
ozone profiles between 19751979 and 19942001
() (1990s1970s) by aircraft and sonde data over
Europe at potential temperature distance from the
2 PVU tropopause. GASP and MOZAIC averaged over
EUR SONDE region. Diff. only if number of daily
averages is 10 for all data. Black MOZAIC-GASP,
blue Uccle, orange MOHp, red Payerne.
Horizontal bars 95 confidence intervals of
differences.
17
Comparison of ozone changes from aircraft with
ECC sondes of Wallops Island(legend same as last
picture)
18
Further questions concerning tropospheric ozone
measurements of BM-sondes comparison
Hohenpeissenberg and Payerne with mountain sites
(Jungfraujoch and Zugspitze) PhD thesis Carlos
Ordonez (earlier at PSI)
Differences (1991-2004) Winter Spring Summer Fall
Time series and linear trends for surface ozone ?
and sondes ? Winter Summer
19
6. LS ozone changes (regular aircraft)(equivalent
latitude coordinates)
20
7. Conclusions and Outlook

• Substantial difference in long-term tropospheric
  ozone changes between regular aircraft
  measurements and European Brewer Mast ozonesondes
  - better agreement for ECC station of Wallops
  Island (stratospheric ozone different story)
• UV-sensor believed to be more reliable
• UT ozone changes (based on climatologies) Since
  second part of 1970s grossly consistent with
  continental scale ground-based NOx emission
  changes (RETRO)
• Effect of stratospheric changes ?
• Future ?

21
Ozone production in the troposphere
22
Predicted (ground-based) NOx emissions The Royal
Society, Ground-level ozone in the 21st Century
Future trends, impacts and policy implications,
Science Policy Report 15/08, 2008. UT ozone
decrease ?
23
Stratospheric ozone
Emissions of Ozone Depleting Substances (ODS)
black CFCs grey HCFCs
Chemical Ozone Depletion (by ODS) (EESC
Equivalent Effective Stratos. Chl.
Ozone layer Black Measurements
(60oS-60oN) Grey Numerical simulations
UV-changes
24
Ozone changes from stratosphere (at
midlatitudes) Increase ? (Montreal Protocol and
enhancement of Brewer/Dobson circulation). Ozone
anomalies from model CMAM and measurements Cly
(Shepherd, 2008)
25
Future ? UT/LS ozone up or down ?

• Long-term UT/LS ozone monitoring important
  (Activity B1 Global UTLS monitoring approach of
  IGACO-Ozone and UV radiation Implementation Plan,
  GAW Report. 182, WMO, Geneva, 2009).
• For global coverage for long-term UT/LS ozone
  monitoring, a coherent approach is needed. This
  can be accomplished by designing a network
  integrating current and planned civil aircraft,
  ground-based (ozone sondes and LIDAR) and UT/LS
  satellite measurements. In addition, a concept
  for data quality should be developed making use
  of the different measurement platforms.
• Tasks - Organize a workshop to discuss current
  capabilities and identify gaps.
• - Aim at developing a strategy to solve the
  problems.
• - Prepare assessment report with recommendations.

26
New project (funding from MeteoSwiss, GAW)

• Close comparison between MOZAIC and ozonesondes
  (ECC)
• Note Different EEC ozonesonde types and solute
  concentrations yield significantly different
  results in UT/LS ozone (homogenization required
  !)

27
Evidence for ozone increase over Europe 1950-1990
Arosa (Swiss Alps, 1800 msl.) 1950-1990
28
Emission changes Fossil fuel related
NOx-emissions from continents (TEAM (TNO emission
assessm. Model), Pulles et al.)
29
Acknowledgement

• (Partial) funding from EU-Project RETRO
• We thank Andrew Detwiler for providing us the
  GASP data
• Special thanks to the MOZAIC team for making
  their data available to us. The authors also
  gratefully acknowledge the strong support of the
  MOZAIC program by the European Communities, EADS,
  Airbus and the airlines Lufthansa, Austrian, and
  Air France who have carried the MOZAIC equipment
  free of charge since 1994.

30
(No Transcript)
31
Total ozone series of Arosa
32
Time series of ozone monthly means for the UT
(left panels) and the LS (right panels) from
MOZAIC ozone-measurements. Ozone evolution
similar(i) at many sites in northern
mid-latit.(ii) to ozone at high mountain sites
(strong increase in the second part of the
1990s)Thouret et al., 2006