Rutherford Appleton Laborator Remote Sensing Group Tropospheric ozone retrieval from uv/vis spectrometery RAL Space - Remote Sensing Group Richard Siddans, Georgina Miles Brian Kerridge, Barry Latter 21 st May 2014 TEMPO 2 nd Science Meeting, Hampton, VA
12
Embed
Rutherford Appleton Laboratory Remote Sensing Group Tropospheric ozone retrieval from uv/vis spectrometery RAL Space - Remote Sensing Group Richard Siddans,
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Rutherford Appleton Laboratory
Remote Sensing Group
Tropospheric ozone retrieval from uv/vis spectrometery
RAL Space - Remote Sensing Group
Richard Siddans, Georgina MilesBrian Kerridge, Barry Latter
21st May 2014 TEMPO 2nd Science Meeting, Hampton, VA
Remote Sensing Group
RAL Remote Sensing Group
• Development and application of retrieval schemes to retrieve trace gases, aerosol and cloud from satellite observations
• Ozone:– Munro et al., 1998 (Nature), demonstrating first
tropospheric ozone profiles– Role in GOME series (& MAG) since ~1990– Involved ESA’s Climate Change Initiative Essential Climate
Variable programme– Developing NRT initiative for MACC/Copernicus
Atmosphere Service– Involvements in Operational ESA Sentinals 4, 5 and 5
• Designed to fit to high precision in Huggins bands required for tropospheric ozone information
• Currently optimised for GOME, SCIAMACHY and GOME-2A/B• Selected as ESA Ozone CCI prototype nadir profile processor
for sensitivity to tropospheric ozone• Algorithm validation paper about to be submitted to AMTD
(Miles et al, 2014)• A joint IASI+GOME-2 (IR+UV) version is available, currently
developing Chappuis add-on.
January 2008
Surface - 450hPa ozone 2008
J F
M A
M J
J A
S O
N D
Remote Sensing Group
Ozone retrieval from the Chappuis bands
• Why?– Sensitivity to near-surface ozone in visible region– Typically less Rayleigh scattering/brighter over land than in UV
• Why not?– Broadband ozone feature (1 piece of information, not profiles on
its own)– Works in theory, but it’s hard to do in practice. Many other
spectral patterns to interfere with ozone fit:– Surface/scattering/other gases/water vapour… – Instrumental ‘features’/spectroscopic errors…
Remote Sensing Group
Ozone retrieval using the Chappuis Bands for near-surface information
UV+Vis retrieval simulations indicate a signal to noise ratio of ≥300 is needed to add information to the UV-only profile retrieval
This implies a Chappuis-only total column fit precision of 2-4DU is required for Chappuis bands to add useful information on near-surface ozone to UV profile retrieval
Dark surface in Chappuis region
Over ‘grass’ surface type
(Miles et al., 2012)
Simulated AKs
AK0-2km0-6km6-12km
Remote Sensing Group
Chappuis-only DOAS slant column fit • Spectrally averaged GOME-2A B3/4 measurements (450-550nm)• WL range after Richter et al, 2012• Fit 7 leading principle components from land spectral databases
(ASTER/USGS)• Further spectral patterns to fit:
– O2, O3, O4, H2O(2), NO2, Ring(2), liquid water, registration/shifts parameters(2), , air molecule Ring effect(2) and polynomials to represent Rayleigh, cloud and aerosol scattering
• All parameters are fit in a straightforward least squares manner. A third order polynomial is subtracted from all patterns (other than the fitted polynomial. patterns).