Physik Fachbereich 1 Institut für Umweltphysik Wissam Chehade, Anna Serdyuchenko, Victor Gorshelev, John P. Burrows, Mark Weber 08. 05. 2012 ESA, ESRIN Frascati Harmonisation of GOME, SCIAMACHY,GOME-2 Ozone and Nitrogen Dioxide Cross-sections HARMONICS i Re-analysis of SCIAMACHY and GOME-2 ozone flight model data and retrieval tests: Final Results I. Ozone II. Nitrogen Dioxide
37
Embed
Wissam Chehade , Anna Serdyuchenko , Victor Gorshelev , John P. Burrows, Mark Weber
Har monisation of GO M E, SCIAMACHY, GOME-2 Ozone and N itrogen D i oxide C ross- section s HARMONICS i Re-analysis of SCIAMACHY and GOME-2 ozone flight model data and retrieval tests: Final Results I. Ozone II. Nitrogen Dioxide. Wissam Chehade , Anna Serdyuchenko , Victor Gorshelev , - 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
PhysikFachbereich 1
Institut für Umweltphysik
Wissam Chehade, Anna Serdyuchenko, Victor Gorshelev,
John P. Burrows, Mark Weber08. 05. 2012 ESA, ESRIN Frascati
Harmonisation of GOME, SCIAMACHY,GOME-2 Ozone and Nitrogen Dioxide Cross-sections
HARMONICS
i
Re-analysis of SCIAMACHY and GOME-2 ozone flight model data and retrieval tests:
Final Results I. Ozone
II. Nitrogen Dioxide
PhysikFachbereich 1
Institut für Umweltphysik
Why re-analysis is needed?
Re-analysis of satellite FM cross-sections (WP 100).• Concatenation, Gluing• Absolute Scaling
Comparison to Literature Data.
Ozone Retrieval Tests (WP 300).
I. Re-analysis of SCIAMACHY and GOME-2 ozone flight model data.
PhysikFachbereich 1
Institut für Umweltphysik
Ozone is the most important trace gas.
plays a unique role in the chemistry and dynamics of the stratosphere (UV protection)
and the troposphere (air pollution, climate gas).
Global and long term monitoring is achieved using atmospheric remote sensing.
Accurate absorption cross-sections spectra are needed.
Laboratory absorption cross section spectra (Flight Model FM) for O3 @ 203K, 223K,
243K, 273K, and 293K were measured.
Advantages of satellite FMs: The use of FM cross-section in the trace gas retrievals generally lead to smaller fit residuals in the satellite retrievals.
The a-priori knowledge of the slit function is not even required
PhysikFachbereich 1
Institut für Umweltphysik
Using SCIAMACHY ozone FM cross-sections in the SCIAMACHY WFDOAS retrieval resulted in systematic differences of about 3-5% in the retrieved total ozone column from the 325-335 nm spectral range when compared to collocated GOME. [Eskes et al., 2005]
GOME2 total O3 retrieval (using GOME-2 FM3) resulted in overestimation of 8%. [Weber et al., 2011]
Harmonisation of O3 FM cross-sections from GOME-2 and SCIAMACHY for a consistent retrieval.
PhysikFachbereich 1
Institut für Umweltphysik
Re-analysis of satellite FM cross-
sections (WP 100)
Concatenation, Gluing
Absolute Scaling
PhysikFachbereich 1
Institut für Umweltphysik The basic principle used for the absorption measurements is the Lambert-Beer-
Law.
Iabs
(λ) = I0 (λ) · e−σ(λ)·ρ·l
CATGAS measurements were done using ozonator and a continuous flow of O2/O3 mixture
O3 concentration (ρ) and path length (l) were unkown.
OD (λ) = (I0 (λ) /I
abs (λ) ) = σ(λ).ρ.l
raw data available, Iabs , Io are available
PhysikFachbereich 1
Institut für Umweltphysik
Several measurements were carried out for different O3 concentrations and path lengths („mixtures“) to cover large dynamical range (7 order of magnitude) requires concatenation (scaling and gluing).
PhysikFachbereich 1
Institut für UmweltphysikGOME-2 FM3
A starting spectrum is selected (mixture 3). Optimal optical density 0.1 < O.D. < 1.0 Select the overlap region between the neighboring mixtures. A scaling factor is calculated.
Limitations : No scaling across channel borders.
PhysikFachbereich 1
Institut für Umweltphysik
All temperature data are then normalized to unit area under the curve (preserves relative temperature dependence)
GOME-2 FM3
PhysikFachbereich 1
Institut für Umweltphysik
Reference data : BDM & Bass Paurconvolved with the wavelength dependent GOME-2 instrumental slit function.
The scaling factor is calculated using a least square approach at each of the selected wavelengths between the optical densities and the available reference cross-sections at temperatures close to those used in the CATGAS campaign.
A single scaling factor scales all spectra at once and preserves the temperature dependence.
PhysikFachbereich 1
Institut für Umweltphysik
GOME-2 Revised FM3
PhysikFachbereich 1
Institut für UmweltphysikSCIAMACHY FM
Reference data (Bass Paur) convolved to FM resolution (310-335 nm) band are used to find a single scaling factor for each temperature
PhysikFachbereich 1
Institut für UmweltphysikSCIAMACHY Revised FM
PhysikFachbereich 1
Institut für UmweltphysikImprovements in the re-analysis (WP
100)
What are the changes w.r.t. earlier analysis (Bogumil, Gür
et al.)?
optimising overlap regions with wider wavelength
ranges and different raw data for concatenation
(GOME-2)
use linear fitting (two parameters) rather than ratios
(one parameter) to concatenate/scale OD spectra
(GOME-2)
Different raw data and spectral ranges for
concatenation (SCIAMACHY)
use convolved reference data (BDM, Bass Paur) for
absolute scaling
PhysikFachbereich 1
Institut für Umweltphysik
Comparison to Literature Data
PhysikFachbereich 1
Institut für Umweltphysik
Comparison at Hg and He-Ne laser lines at room temperature Comparison of GOME-2 FM3 and SCIAMACHY FM with literature data at 10 selected wavelengths. The absorption cross sections are given in units of 10-20 cm2.
Comparison of the new SCIAMACHY FM with literature using a non-linear least square fitting program.
The revised data and the high spectral resolution data convoluted with each spectrometer slit function are compared with a non-linear least square fitting program using five parameters:
a linear wavelength shift to correct for the differences in the spectral calibrations (1 parameter). a scaling coefficient to adjust the amplitude of the cross-sections (1 parameter). a third order baseline (cubic) polynomial (3 parameters).
Direct comparisons by non-linear least square fit in DOAS window
PhysikFachbereich 1
Institut für Umweltphysik
Temperature parameterizations and smoothness comparisons The temperature parameterization is used in the total ozone retrieval together with temperature climatology to express the change of ozone absorption with altitude.
PhysikFachbereich 1
Institut für UmweltphysikTemperature smoothness
PhysikFachbereich 1
Institut für Umweltphysik
Ozone Retrieval Tests (WP 300)
PhysikFachbereich 1
Institut für Umweltphysik
Ozone Retrieval Tests (WP 300)
The total ozone column is retrieved using the Weighting Function Differential Optical Absorption Spectroscopy (WFDOAS) method (Coldwey-Egbers et al., 2005, Weber et al., 2005) (325-335 nm).
A wavelength dependent weighting function of ozone and temperature that describes the relative radiance change due to a vertical profile change.
The temperature parameterization is used in the radiation transfer code together with temperature climatology to express the change of ozone absorption with altitude.
PhysikFachbereich 1
Institut für UmweltphysikThe cross-section shifts are tested .
The optimum applied shift improves the fit residual RMS.
The optimum applied shifts are in good agreement to the values found in previous tests.
-0.038 nm
0.018 nm
PhysikFachbereich 1
Institut für Umweltphysik
The total ozone values retrieved are roughly within +0.5% to the values retrieved currently.
The fit residuals (RMS) are improved
PhysikFachbereich 1
Institut für Umweltphysik
The total ozone values retrieved are roughly within +1% to the values retrieved currently.
The fit residuals (RMS) are the same
PhysikFachbereich 1
Institut für Umweltphysik
Retrieval tests using the new ozone cross-section data The new experimental cross-section data are tested in total
ozone retrieval of GOME-2 and SCIAMACHY.
The data have to be convolved to GOME-2 and SCIAMACHY slit
function.
-0.007 nm 0.007 nm
PhysikFachbereich 1
Institut für Umweltphysik
The total ozone columns retrieved from GOME-2 and SCIAMACHY satellites are in good agreement with the amounts retrieved by the current data, +1% and +2% respectively.
fit residuals (RMS) for GOME-2 and SCIA similar to satellite FM.
PhysikFachbereich 1
Institut für Umweltphysik
Is the re-analysis needed?
Comparison to Literature Data.
NO2 Retrieval Tests (WP 300).
II. Re-analysis of SCIAMACHY and GOME-2 NO2 flight model data.
PhysikFachbereich 1
Institut für Umweltphysik Nitrogen dioxide (NO2) is an important trace gas in the Earth’s atmosphere.
In the stratosphere, it is involved in ozone chemistry as a catalyst for ozone destruction and also in the formation of halogen reservoirs such as chlorine nitrate.
In the troposphere, nitrogen oxides (NOx = NO + NO2) are key ingredients for ozone formation.
Long term measurements of NO2 total column, tropospheric column and its stratospheric profile provide unique information about the changes in tropospheric pollution around the globe and the state of the stratosphere.
Nadir measurements from GOME, SCIAMACHY and GOME-2 have been used in many studies on tropospheric NO2 burdens, the importance of different NOx emissions sources (by industry and traffic) and their change over time.
PhysikFachbereich 1
Institut für Umweltphysik
The NO2 absorption cross-section data is not the significant error source for the total vertical column retrieval when the data retrieved from GOME-2 and SCIAMACHY compared with GOME.
The usage of the same settings as applied to data from the predecessor instrument is essential to obtain a high degree of consistency among them. [Richter et al., 2005, 2011]
The setting should provide the best differential NO2 signal.
The least interference of other absorbers in the retrieval
spectral window (DOAS).
Geophysical parameters (common solar spectrum, field of
view, spatial resolution).
Moreover, the instruments have overpass time that differ by about 30 minutes which can make a difference in the stratospheric and tropospheric NO2 due to daytime variations.
PhysikFachbereich 1
Institut für UmweltphysikComparison to Literature Data and
Retrieval Tests Nitrogen dioxide is retrieved in the UV/visible region using the strong differential absorption structures of the NO2 molecule (DOAS fitting technique).
The NO2 absorption cross-section measured at 223K is used in the retrieval in the 425-450 nm DOAS region.
Good consistency between GOME-2 and SCIAMACHY is also achieved by using a larger fitting window (425-497 nm)
[Richter et al. 2011]
PhysikFachbereich 1
Institut für Umweltphysik
The NO2 reference data is compared to Vandaele et al. reference data convolved with the GOME-2 slit function in a non-linear least square fitting program in the two retrieving windows.
Excellent agreement (scaling differences are below 0.5%)
PhysikFachbereich 1
Institut für UmweltphysikBoth data were tested in the vertical NO2 column retrieval using a stratospheric air mass factor on one day of GOME-2 data .
The differences between the columns are hardly noticed.
PhysikFachbereich 1
Institut für Umweltphysik
higher values (1 – 2 %) retrieved in regions of good NO2 fit. slightly higher values (2 – 4 %) over the tropics where the absorption is small and the signal is noisier
PhysikFachbereich 1
Institut für Umweltphysik
The usage of GOME-2 cross-section data produces slightly better fit residuals (up to 5%) mainly in the region where the NO2 absorption is large
PhysikFachbereich 1
Institut für UmweltphysikThe O3 revised data are in good agreement with the literature data.
correct temperature dependence in the DOAS region (better than 3%)
An improvement in the retrieved total ozone column. GOME-2: +0.5% difference to WFDOAS standard SCIAMACHY: +1% difference to WFDOAS standard
Retrieval test for IUP Bremen cross-sections GOME-2: +1% SCIAMACHY : +2%
revised versions of GOME-2 FM3 and SCIAMACHY FM data and IUP experimental ozone cross-sections are ready for use.
Summary
PhysikFachbereich 1
Institut für Umweltphysik The NO2 data are in good agreement with the literature data in the DOAS region (better than 0.5%)
The retrieved vertical column is within few percents (1-
4%) compared to column retrieved with Vandaele data.
The fit residuals (RMS) are up to 5% better when GOME-2 data is used
The NO2 absorption cross-section’s quality is good enough to be used in the retrieval and no re-analysis was required.