Sonde data for TES validation

1
Sonde data for TES validation

• Jennifer Logan
• Harvard University

2
Ozonesonde Data

• Data from WOUDC and SHADOZ analyzed at
  Harvard and provided to TES team for early
  validation as ascii files in a common format.
• Obviously bad data omitted
• April, 2005
• Data for Sept. 2004 onwards, for 21 stations
• No regular sonde data at AVDC
• July, 2005
• Data for Sept. 2004 onwards, for 10 stations
• Status of sonde data at AVDC
• April, 2005 AVE sonde data at AVDC
• June, 2005 SHADOZ data at AVDC
• September CMDL, Canadian data at AVDC

3

• Use sonde data to examine
• Are the validation profiles typical
• Spatial variability of tropospheric ozone on a
  given day, using European stations
• Temporal variability using European data
• Use TOMS data to examine
• Temporal variability at sonde locations - mainly
  reflects the stratosphere
• Contrast mid-latitude and tropical profiles
• MOZAIC data in Europe would be best for looking
  at tropospheric variability vs. time, but not yet
  available
• MOZAIC data for August 2004 just released.

4
Most stations have weekly data
5
Are the sonde data from fall 2004 typical?

• Variability is very high in UT/LS
• Variability is higher in spring than in fall.
• TES averaging kernels are broad, so need good
  match in space/time

Black mean one sd., for 1985-2000.
STRAT.
TROP.
Color - profiles on Nov. 5, 10, 12
6
Spatial variability on the same day - Oct. 13,
2004
Hohen., Payerne 330 km apart Legion., Linden
470 km Hoh/Pay. to Lind./Leg 540-1190 km
Payerne and Hohen. very similar up to 400 hPa.
TROP.
STRAT.
Oct. 13
7
Another example - November 10
8
Nov. 5th 15 ppb difference, almost 50. Can
TES differentiate such a gradient?
9
Median difference between two stations 330 km
apartand 25th and 75th percentiles
Median difference is mostly lt10 in the trop.,
but differences often exceed 30 more than 25 of
the time in the UT/LS
10
Temporal differences large in UT/LSNo daily
data available - 2 day differences
Compare blue and turquoise, brown and purple,
green and red
All profiles within a week of central (blue)
profile
11
Another example - Hohenpeissenberg, Nov.
Soundings on Wed/Fri in fall, Mon., Wed., Fri. in
winter/spring
12
Daily ozone columns from TOMS Columns indicate
when temporal variability is an issue
Dashed lines month start Red line TES match
13
STRAT
TROP.
Comparison of mid-latitude and tropical
variability All profiles in two months
Payerne Oct-Nov
Ascension Sept-Oct
14
Issues The criteria used so far for TES give
relatively few matches, even with data from 20-30
locations. Stricter criteria gt fewer
matches Analysis of MOZAIC data will provide
daily profiles, and show trop. variability for
Europe, as 5 flights in one day - but only for
0-11 km.
15
MOZAIC flights offer the chance to examine
variability on short time scales - but note that
profiles are not vertical. They are within 250
km of each airport
Profiles measured within 24 h
10 h
24h
17 h
Summer
Winter