Henk Eskes, ERS-ENVISAT symposium 2004 trieval, validation and assimilation of trieval, validation and assimilation of IAMACHY ozone columns IAMACHY ozone columns Henk Eskes, Ronald van der A, Ellen Brinksma, Pepijn Veefkind, Johan de Haan, Pieter Valks Royal Netherlands Meteorological Institute (KNMI) 1) Ozone column retrieval 2) Validation of SCIAMACHY ozone columns 3) Data assimilation and validation
24
Embed
Retrieval, validation and assimilation of SCIAMACHY ozone columns
Retrieval, validation and assimilation of SCIAMACHY ozone columns. Henk Eskes, Ronald van der A, Ellen Brinksma, Pepijn Veefkind, Johan de Haan, Pieter Valks Royal Netherlands Meteorological Institute (KNMI) 1) Ozone column retrieval 2) Validation of SCIAMACHY ozone columns - PowerPoint PPT Presentation
Welcome message from author
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
Henk Eskes, ERS-ENVISAT symposium 2004
Retrieval, validation and assimilation of Retrieval, validation and assimilation of SCIAMACHY ozone columnsSCIAMACHY ozone columns
Henk Eskes, Ronald van der A, Ellen Brinksma,Pepijn Veefkind, Johan de Haan, Pieter Valks Royal Netherlands Meteorological Institute (KNMI)
1) Ozone column retrieval2) Validation of SCIAMACHY ozone columns3) Data assimilation and validation
Henk Eskes, ERS-ENVISAT symposium 2004
Ozone column retrieval: new DOAS algorithm
Heritage:• Based on the OMI operational ozone column algorithm
OMI-DOAS - Pepijn Veefkind, J. De Haan• Implementation for GOME
TOGOMI - P. Valks, R. Van Oss• Implementation for SCIAMACHY
TOSOMI - H. Eskes and R. van der A
ESA ITT - GOME total-ozone algorithm• BIRA• IFE - University Bremen• KNMI
Henk Eskes, ERS-ENVISAT symposium 2004
Innovations compared to e.g. GOME Fast Delivery, GDP vs 3
• New treatment of rotational Raman scattering (J. de Haan)• Empirical air-mass factor approach• TOMS v8 ozone profile data base• FRESCO cloud cover and cloud top height• Radiative transfer improvements• T-dep O3 cross section: ECMWF temperature profiles
Henk Eskes, ERS-ENVISAT symposium 2004
Normalized IncidentSunlight
Ozone Absorption
Single Rayleigh ScatteringCabannes
not scrambled96.2 %
Ramanscrambled
3.8 %
OzoneAbsorption
OzoneAbsorption
Received by sensor
Rotational Raman: impact on retrieval
Henk Eskes, ERS-ENVISAT symposium 2004
New approach to rotational Raman scattering
Difference between old and new treatment of Raman
Henk Eskes, ERS-ENVISAT symposium 2004
Empirical air-mass factor
Henk Eskes, ERS-ENVISAT symposium 2004
DOAS fit example
rms 0.5 %
Henk Eskes, ERS-ENVISAT symposium 2004
SCIAMACHY TOSOMI Example
QuickTime™ en eenTIFF (ongecomprimeerd)-decompressor
zijn vereist om deze afbeelding weer te geven.
Henk Eskes, ERS-ENVISAT symposium 2004
Validation: midlatitude example
BrewerDe Bilt (52.1 N, 5.18E) bias: -1.8% rms: 4.3%
No obviousseasonalbias
Henk Eskes, ERS-ENVISAT symposium 2004
Validation: summary
Av. Bias [%]
RMS Bias number
AllAll -1.3-1.3 5.05.0 9696 Dobson -1.1 4.5 54 Brewer -1.9 5.0 28 SAOZ -3.1 7.6 9 DOAS 4.9 8.1 3 other -1.1 5.4 2Polar Lat >60 -1.7 5.7 11 Lat<-60 1.1 6.5 7Midlat Lat 30 to 60 -1.4 4.9 51 Lat –30 to -60 -4.0 6.3 6(Sub)tropical Lat –30 to 30 -1.0 4.3 21
Henk Eskes, ERS-ENVISAT symposium 2004
Validation: summary vs. latitude
Main conclusions:
• Tosomi 1.5% lower than ground based
• RMS 4.9% increasing with ozone variability (representativity)
• No clear geographical location dependence
• No clear seasonal dependence
Henk Eskes, ERS-ENVISAT symposium 2004
Ozone data assimilation at KNMI
TM3DAM assimilation code:• Driven by 6h meteo from ECMWF (wind, temperature, pres):
analyses and 10-day forecasts• Same vertical levels as ECMWF (subset in lower troposphere)• Second moments advection (Prather)• Sub-optimal Kalman filter, detailed error covariance modelling• Ozone chemistry parametrizations:
• Cariolle gas phase• "Cold tracer" scheme for heterogeneous chemistry