Validation of Tropospheric Emission Spectrometer (TES) nadir stare ozone profiles using ozonesonde measurements during Arctic Research on the Composition of the Troposphere from Aircraft and Satellites (ARCTAS)

1
Validation of Tropospheric Emission Spectrometer
(TES) nadir stare ozone profiles using ozonesonde
measurements during Arctic Research on the
Composition of the Troposphere from Aircraft and
Satellites (ARCTAS)
Boxe et al., in preparation (JGR)
2
Acknowledgements

• NASA-JPL
• TES Team
• ARC-IONS Network

3
Objectives

• Validate TES V003 and V004 stare ozone retrievals
  with ozonesonde measurements.
• To Show that TES ozone stare retrievals
  theoretical and empirical errors, and biases are
  small and consistent with previous validation
  investigations (Worden et al., 2007 Nassar et
  al., 2008).
• To show that TES is sensitive to lower
  tropospheric ozone.

4
TES Stare Mode

• Stare In nadir mode, point at specific locations
  over sites for validation and other scientific
  interest for up to about 4 minutes. Such
  observations are made for as long as the target
  is in within 45 of the nadir direction (up to
  210 seconds).

5
TES Stare Mode Bratts Lake, April 18th, 2008
6
Ozonesondes are Vital for Validating TES
Retrieval
In Situ Vertical Profiling of Ozone Coincident
with TES retrievals . Precision of (35)
and an accuracy of (510) up to 30 km
altitude (Smit et al., 2007).
7
How do sondes work?

• Electrochemical Concentration Cell (ECC) sondes
  (platinum-platinum, silver-platinum, and
  platinum-carbon electrodes).

8
Error Analysis
random instrument spectral errors ? f.s vector
errors
forward model errors ? f.s vector errors
uncertainty due to unresolved fine structure
9
TES operator applied to sonde data
a priori constraint vector
TES averaging kernel
Sonde Data
accounts for TES sensitivity and vertical
resolution
10
Arctic Intensive Ozonesonde Network
Study(ARCIONS)
11
Spring and Summer TES ozone and ozonesonde
comparisons
Bratts Lake (April 2nd, 2008), 50o N, 105o
W Barrow (April 10th, 2008), 71o N, 157o W Barrow
(April 12th, 2008), 71o N, 157o W Barrow (April
14th, 2008), 71o N, 157o W Bratts Lake (April
18th, 2008), 50o N, 105o W Egbert (July 5th,
2008), 44o N, 80o W Yellowknife (July 5th,2008),
62o N, 114o W Egbert (July 7th, 2008), 44o N, 80o
W Yellowknife (July 7th,2008), 62o N, 114o W
12
Select Spring and Summer TES ozone and
ozonesonde comparisonsApril 10th, 2008, Barrow
13
Select Spring and Summer TES ozone and
ozonesonde comparisonsApril 10th, 2008, Barrow
14
Select Spring and Summer TES ozone and
ozonesonde comparisonsApril 10th, 2008, Barrow
15
Select Spring and Summer TES ozone and
ozonesonde comparisonsApril 10t\h, 2008, Barrow
16
Select Spring and Summer TES ozone and
ozonesonde comparisonsApril 10th, 2008, Barrow
17
Select Spring and Summer TES ozone and
ozonesonde comparisonsJuly 5th, 2008, Yellowknife
18
Select Spring and Summer TES ozone and
ozonesonde comparisonsJuly 5th, 2008, Yellowknife
19
Select Spring and Summer TES ozone and
ozonesonde comparisonsJuly 5th, 2008, Yellowknife
20
Select Spring and Summer TES ozone and
ozonesonde comparisonsJuly 5th, 2008, Yellowknife
21
Select Spring and Summer TES ozone and
ozonesonde comparisonsJuly 5th, 2008, Yellowknife
22
Conclusions

• TES ozone profiles are usually positively biased
  (i.e., by no more than 15) in the
  troposphere-to-mid-stratosphere and negatively
  biased in the upper stratosphere (i.e., by no
  more than 20) when compared to ozondesonde data.
  Bias is consistent with prior sonde comparisons
  (H. Worden et al. 2007, 2006, 2004, and Nassar et
  al. 2008).
• Theoretical and empirical errors are generally
  consistent, both showing analogous vertical
  distribution ( 5 to 20 for V003 data and 5 to
  15 for V004 data).
• TES is sensitive to lower tropospheric ozone,
  even very close to the surface.