Validation of AIRS Version 5 Ozone Retrievals AIRS Science Team Meeting - March 28, 2007 Bill Irion - Jet Propulsion Laboratory Michael Newchurch - U. Alabama at Huntsville Sunmi Na - Pusan National University With thanks to Sung-Yung Lee, Annmarie Eldering, Greg Osterman, John Blaisdell, Chris Barnet, Murty Divakarla, Jennifer Wei, Wallace McMillan, Steve Friedman and SHADOZ 1
The problem with Version 4 ozone Relative di fg erences of AIRS & ECMWF vs ozonesondes Correction: matchup is within 100 km and 6 hrs. ECMWF/Sonde AIRS/Sonde Standard deviations shown. Pressures offset for clarity. Like ECMWF, V4 AIRS is too high in troposphere and too low in lower stratosphere; column OK to first order. Irion et al. AIRS Science Team Meeting 2 “Validation of AIRS V5 ozone” 3/28/2007
Simplified Version 4 algorithm for ozone Damping Channel L1B AIRS 3x3 parameter selection observation + (“noise propagation AMSU threshold”) (Constrained) Profiles Cloud Cloud-cleared Physical and radiances and Clearing uncertainties columns Retrieval Cloud-cleared First guess radiances Regression Training by ECMWF Irion et al. AIRS Science Team Meeting 3 “Validation of AIRS V5 ozone” 3/28/2007
Changes in algorithm from V4 to V5 ozone L1B AIRS 3x3 Expanded Decreased observation + channel damping AMSU selection parameter (Constrained) Profiles Cloud Cloud-cleared Physical and radiances and Clearing uncertainties columns Retrieval Cloud-cleared First guess First guess radiances Regression Climatology Training by ECMWF Irion et al. AIRS Science Team Meeting 4 “Validation of AIRS V5 ozone” 3/28/2007
Channel selection changes Irion et al. AIRS Science Team Meeting 5 “Validation of AIRS V5 ozone” 3/28/2007
First O 3 guess from climatology •Monthly zonal-mean climatology from combined ozonesonde, SAGE II and (at high latitudes) MLS measurements and used as a priori for TOMS Version 8 retrievals. McPeters, R. D., J. A. Logan, and G. J. Labow (2003), Ozone Climatological Profiles for Version 8 TOMS and SBUV Retrievals, Eos Trans. AGU, 87 (52), Fall Meet. Suppl., Abstract A21D-0998; McPeters, R.D., G. J. Labow, and J. A. Logan (2007), Ozone climatological profiles for satellite retrieval algorithms, J. Geophys. Res., 112, D05308, doi:10.1029/2005JD006823. Irion et al. AIRS Science Team Meeting 6 “Validation of AIRS V5 ozone” 3/28/2007
AIRS V4 and V5 total O 3 comparison V5 ozone decreases along tropics and high southern latitudes. November 5, 2004 Irion et al. AIRS Science Team Meeting 7 “Validation of AIRS V5 ozone” 3/28/2007
AIRS V4 and V5 comparison vs surface temperature Large decreases in V5 in high latitude cold regions. Irion et al. AIRS Science Team Meeting 8 “Validation of AIRS V5 ozone” 3/28/2007
V3 V4 V5 total ozone comparisons with OMI AIRS column ozone V3–>V4–>V5 show increasing agreement with OMI. Irion et al. AIRS Science Team Meeting 9 “Validation of AIRS V5 ozone” 3/28/2007
Zonal comparisons with OMI over Focus Days Versions 4 and 5 AIRS ozone columns show comparable biases w.r.t OMI, except bias substantially lowered in southern high latitudes. Need to compare to ground-based Dobson/Brewer measurements. Irion et al. AIRS Science Team Meeting 10 “Validation of AIRS V5 ozone” 3/28/2007 Irion et al. AIRS Sci Team Meeting 10 “Validation of AIRS V5 ozone” 3/25/2007
Average zonal bias over all Focus Days AIRS Version 5 - OMI bias symmetric across equator. AIRS positive bias also seen against Brewer/Dobson measurements - see Murty Divakarla’s presentation on Thursday Irion et al. AIRS Science Team Meeting 11 “Validation of AIRS V5 ozone” 3/28/2007
Sonde comparisons with and without regression V5 AIRS-Sonde Average Relative Bias What does a climatology for Regression 1st guess Climatology 1st guess the first guess do to the profile? Retrievals using regression in red Retrievals using climatology in blue Same channels and damping for both Generally improved agreement in mid-to-lower troposphere and lower stratosphere. Mixed results in upper troposphere. Irion et al. AIRS Science Team Meeting 12 “Validation of AIRS V5 ozone” 3/28/2007
Layer comparison with TES At high ozone mixing ratios @ 273 mb 50°-60°N AIRS V4 is high w.r.t. TES while V5 is low. Pre-release TES Version 3 retrievals courtesy of Greg Osterman and Annmarie Eldering Irion et al. AIRS Science Team Meeting 13 “Validation of AIRS V5 ozone” 3/28/2007
AIRS-TES bias at 273 mb coincident with lower O3 @ 273 mb tropopause height TES and AIRS show increased ozone, but AIRS 1st guess at 273mb less suited to AIRS mixing ratios lower lowered tropopause? than TES. Would shifting first guess mixing ratio profile up or down with tropopause Tropopause pressure change this? Pressure Need to carefully look at information content and spectral fits. Irion et al. AIRS Science Team Meeting 14 “Validation of AIRS V5 ozone” 3/28/2007 Irion et al. AIRS Sci Team Meeting 14 “Validation of AIRS V5 ozone” 3/25/2007
Ideas to try for Version 6 • Re-evaluate trapezoid levels? • Shift a priori mapping up or down with retrieved tropopause pressure? • Based on tests deciding new damping factor… Dynamically modify damping factor with skin temperature or tropopause height? Irion et al. AIRS Science Team Meeting 15 “Validation of AIRS V5 ozone” 3/28/2007
Thanks for your time! We’ll be watching like…uh, you know. Irion et al. AIRS Science Team Meeting 16 “Validation of AIRS V5 ozone” 3/28/2007
Recommend
More recommend
