Minimize SCIAMACHY Product Handbook

Total Column: Nadir Retrieval Schemes of DOAS-type

Table of Contents


5.2 Total Column: Nadir Retrieval Schemes of DOAS-type

Many molecules of atmospheric relevance have structured absorption spectra in the UV-VIS spectral range (fig. 5-3). These can be used to determine the total atmospheric amount of the species from remote sensing measurements of scattered sunlight using the DOAS method. This powerful retrieval technique was originally developed for ground-based measurements using artificial light sources or scattered sunlight (Solomon et al. 1987, Platt 1994) but has successfully been adapted to nadir measurements from GOME (Burrows et al. 1999 and references therein). Two main ideas form the basis of the DOAS approach (see also fig. 5-4).


the isolation of the high frequency structures of molecular absorbers from broad band scattering features (Rayleigh, Mie) by a high pass filter,


the separation of spectroscopic retrievals and radiative transfer calculations.

The first step of the data analysis consists of determining the total amount of absorption and scattering by dividing the Earthshine radiance by the direct solar irradiance. The latter provides the absorption free background. The molecular absorption cross section together with a polynomial is then fitted to the logarithm of this ratio, yielding the trace gas concentration along the light path (slant column concentration). Fig. 5-5 depicts a typical fit for NO2.

Finally, the average light path through the atmosphere is calculated using a radiative transfer model. The light path is often expressed as airmass factor which is the light path enhancement factor relative to a vertical transection of the atmosphere. Based on the AMF the vertical column concentration of the absorber may finally be calculated. Obviously clouds drastically modify the light path through the atmosphere and need to be properly taken into account when calculating the AMF (see chapter 5.3). The most important cloud parameter is the cloud fraction, as fractional cloudiness blocks the light path to atmospheric layers below the cloud. Using the DOAS algorithm, atmospheric columns of a number of species can be determined, including O3, NO2, SO2, HCHO, BrO, and OClO (Burrows et al. 1999, Borell et al. 2003).


click to enlarge

fig. 5-4:

The main steps of the DOAS retrieval. For further details see the text. (graphics: IUP-IFE, University of Bremen)



click to enlarge

fig. 5-5:

Typical SCIAMACHY NO2 fit results from a measurement over a polluted area in China on January 15th, 2006. The red line is the scaled NO2 laboratory cross-section, the dashed blue line the result of the fit after subtraction of all contributions with the exception of NO2. (graphics: IUP-IFE, University of Bremen)

One limitation of the classical DOAS technique is the assumption that the atmosphere is optically thin in the wavelength region of interest. In addition, ‘line-absorbers’ such as H2O, O2, CO, CO2 and CH4 usually cannot be retrieved precisely by standard DOAS algorithms because their strong absorption also depends on pressure, temperature and wavelength and in addition their spectra are often not fully spectrally resolved by SCIAMACHY. To overcome these drawbacks several DOAS-type techniques were developed to account for such effects and to permit successful retrievals of the trace gas species. They are for example the WFM-DOAS (Buchwitz et al. 2000, de Beek et al. 2006), AMC-DOAS (Noël et al. 2004), TOSOMI (Eskes et al. 2005), SDOAS (Van Roozendael et al. 2006), IMAP (Frankenberg et al. 2005a, 2005b) and IMLM (Houweling et al. 2005). Figure 5-6 illustrates an example of one day of ozone columns derived from nadir observations with one of these schemes, the TOSOMI algorithm. Table 5-1 summarises the DOAS-type retrieval algorithms as applied to SCIAMACHY data and references to it. (fig. 5-6), (table 5-1)


click to enlarge

fig. 5-6:

One day of total ozone densities obtained with the TOSOMI algorithm. (image: KNMI/ESA)
Parameter Spectral Window (nm) Layer Quantity (column) Retrieval Algorithm Retrieval Algorithm Reference
O3  325 335  troposphere, stratosphere  total  operationalWFM-DOASTOSOMISDOAS  Spurr 2000 Weber et al. 2005 Eskes et al. 2005 Van Roozendael et al. 2006
NO2  425-450  troposphere, stratosphere  total, tropospheric   operational DOAS SDOAS   Spurr 2000 Richter et al. 2005 Van Roozendael et al. 2006
BrO  335-347(55)  troposphere, stratosphere  total  DOAS   
SO2  315-327  troposphere  total  DOAS   
HCHO  335-347(55)  troposphere  total  DOAS   
OClO  365-389  stratosphere  total  DOAS   
H2O  688-700  troposphere  total  AMC-DOAS  Noel et al. 2004
CO  2321(24)-2335   troposphere  total  WFM-DOAS IMLMIMAP  de Beek et al. 2006 de Laat et al. 2006 Frankenberg et al. 2005b
CH4  1627(30)-1671  troposphere  total  WFM-DOAS IMLMIMAP   de Beek et al. 2006 Gloudemans et al. 2005 Frankenberg et al. 2005a
CO2  1558(63)-1594(85)  troposphere  total  WFM-DOAS IMLMIMAP  de Beek et al. 2006 Houweling et al. 2005 Frankenberg et al. 2005a  

Table 5-1: Atmospheric geophysical parameters and retrieval algorithms – nadir trace gases.



Table of Contents


0 Attachments
Average (0 Votes)
The average rating is 0.0 stars out of 5.