ESA Earth Home Missions Data Products Resources Applications
EO Data Access
How to Apply
How to Access
  Acknowledgements and contact
List of figures and tables
List of tables
 Additional information on the GOMOS measurements
Periods of data unavailabilities
Evolution of the IPF
List of abbreviations and acronyms
Product types and structure
Other sources of information
GOMOS-related ESA web pages
Publications (peer-reviewed publications and proceedings of conferences)
Secondary products
Scintillation and turbulence
Aerosols and PSC
Product validation
NO2 and NO3 measurements
O3 measurements
CAL/VAL activities
Retrieval and processing issues
Assimilation of GOMOS products
Specific events
GOMOS-related theses
Other technical reports by members of GOMOS SAG, ESL and QWG
Level2 processing
How to
Access tools
GOMOS products toolbox
Data selection
PCD summary in the Level2 products
Star properties
Data availability
Presentation of the GOMOS products
Description of the products
Product content
Product structure
Time availability
Data size
Presentation of the instrument and the measurements
Scientific achievements
Validation results
Measurement characteristics
Occultation obliquity
Geographical and time coverage
Star characteristics
Mission planning
Modified mission scenario since August 2005
Instrument description and measurement principle
Calibration phase and monitoring activities
Measurement technique
GOMOS Product Handbook
Content of the products
Site Map
Frequently asked questions
Terms of use
Contact us


Chapter 2
Presentation of the GOMOS products

2.1 General presentation

2.1.1 Organization and relation of the products Processing levels and main products

The GOMOS products, like all the ENVISAT products, are grouped according to the processing level:

  • Level 0 products:              Reformatted and time-ordered satellite data
  • Level 1b products:           Geolocated calibrated engineering data
  • Level 2 products:              Geolocated geophysical products High level data flow

Figure 2.1 shows the high level data flow of the GOMOS ground processing. The level 0 products are processed by the Level 1b processing chain that creates two products: the level 1b product (mnemonic TRA) and the limb product (mnemonic LIM). Only the first one is used by the Level 2 processing chain to generate two other products: the level 2 product (mnemonic NL) and the residual extinction product (mnemonic EXT).

Besides these products processed off-line, Near Real Time products are generated within 3 hours of ground reception. Their processing uses the predictions of the external atmospheric data, instead of the a posteriori analysis used by the standard off-line processing. The so-called meteo products contain selected profiles extracted from the Near Real Time products at a reduced spatial resolution. They are delivered mainly to the meteorological community.

Figure 2.1: High level data flow of GOMOS processing.

Table 2.1 lists the GOMOS Level 0 products, the Level 1b products, the Level 2 products, including the meteo products, with the naming convention and the general description of their content.

Processing level

Naming convention

File description

Level 0


GOMOS nominal Level 0 product


GOMOS monitoring Level 0 product

Level 1b


GOMOS Geolocated and Calibrated Transmission Spectra Product


GOMOS Geolocated and Calibrated Background Spectra (Limb) Product

Level 2


GOMOS Temperature and Atmospheric Constituent Profiles


GOMOS Residual Extinction


GOMOS NRT Extracted Profiles for Meteo Users

Table 2.1: List of GOMOS Level 0 products, Level 1b products, and Level 2 products. Level 0 processing

The Level 0 processing includes a simple set of operations: the determination of the satellite position and the conversion of satellite binary time to universal time coordinates. There are actually two types of GOMOS Level 0 products: the nominal Level 0 products (the sensor is in nominal occultation measurement mode), and the monitoring Level 0 products (the sensor is in calibration monitoring mode). Level 1 processing

The input data for the Level 1 processing are the Level 0 products and relevant auxiliary data. Level 1 products are divided into two categories: Level 1a and Level 1b. The Level 1a products are the Level 0 products after they have been sorted and filtered by low-level quality checks. They will not be further detailed in the document. The Level 1b products are generated by the Level 1b processing chain from the Level 0 products. The aim of the Level 1b processing is to estimate a set of horizontal transmission functions in the UV-visible-near IR between 250 nm and 952 nm using data measured by the GOMOS spectrometers. There are two types of Level 1b products: the geolocated and calibrated transmission spectra products and the geolocated and calibrated background spectra limb products.

The main steps of the Level 1b processing are illustrated in Figure 2.2 .

Figure 2.2: Simplified architecture of the Level 1b processing.

In a first step, the nominal wavelength assignment, geolocation and datation processing are performed. The nominal wavelength assignment corresponding to a perfect tracking of the star during the measurement is provided by the spectral assignment of one CCD column and by the spectral dispersion law of the spectrometers read in the calibration auxiliary product. Then spectral shifts due to vibrations and imperfect tracking are estimated thanks to the pointing data history produced by the SATU (Star Acquisition Tracking Unit). Each CCD column is then spectrally assigned during each spectrometer measurement. Each measurement of the atmospheric transmission is precisely geo-located, from the satellite location and the known direction of the star. Because of the atmospheric refraction, the light from the star to the instrument does not follow a straight line. A full ray tracing is performed through the atmosphere to compute the exact path of the stellar light, the refraction effects being inferred from the state of the atmosphere given by the ECMWF and MSIS90.

The processing of the spectrometer data is then performed. Anomalies and outliers are first detected and corrected (saturated samples, bad pixels, cosmic ray, modulation correction). Dark charge is removed and a few other instrumental corrections (correction of the SFA mirror reflectivity, internal and external straylight correction, vignetting correction, flat-field correction) are applied. In bright limb occultations, the estimate of the scattered solar light is removed from the central band to get the star signal alone. This background signal is estimated from the signals measured in the upper and lower CCD bands, and it is stored in the geolocated and calibrated background spectra limb product. The full transmission spectra are then computed as the ratio of the estimated star spectrum to the reference spectrum of the current occultation (average of several star spectra measured outside the atmosphere during the occultation). They are stoed in the geolocated and calibrated transmission spectra product.

The processing of the fast photometer data includes some steps similar to the ones applied to the spectrometer samples (detection and correction of anomalies). The estimated central background computed for the spectrometers is substracted from the photometer signals.

The processing steps and the generation of the Level 1b products are presented in more detail in the ATBD reference document (see Chapter 4 for reference). Level 2 processing

The input data needed for the Level 2 processing are the Level 1b products and relevant auxiliary data. The main Level 1b quantities needed are the transmission spectra at different tangent point heights, and the photometric data from the two fast photometers. The aim of the Level 2 processing is to retrieve the vertical profiles of O3, NO2, NO3, O2, H2O and other trace gases profiles, the temperature profile, the aerosol extinction coefficient and wavelength dependency parameters, and information about atmospheric turbulence from the full atmospheric transmission spectra. There are three types of Level 2 products: the products storing the profiles of temperature and atmospheric constituents, the residual extinction products and the products storing selected profiles processed in NRT for meteo users.

The geolocation data and the a priori atmospheric data are also used to deal with the refractive effects and to initialise the inversion. These data are partly replaced by new data from the GOMOS Level 2 processing.


The main steps of the Level 2 processing are illustrated in Figure 2.3 .


Figure 2.3: Simplified architecture of the Level 2 processing.

In a first step, the transmission spectra are corrected for the attenuation and dilution caused by refraction and modulations by scintillations. The measurements from the photometers are used to correct the measured transmissions from the scintillation effects. The spectral inversion of the transmission spectra is then performed to retrieve the constituent line densities. A separate spectral inversion for the IR spectrometers (retrieval of H2O and O2 densities in the 756-773 nm and 926-952 nm spectral ranges) is applied.


o      spectral inversion UV-VIS (O3, NO2, NO3):

The model transmission function is fitted to the refraction-corrected transmissions. The minimization is done by a nonlinear Levenberg-Marquardt method, simultaneously at all wavelengths. For NO2 and NO3, chromatic scintillations caused by isotropic turbulence produce perturbations in the transmission spectra, and subsequent unrealistic oscillations in the vertical profiles of species, mainly NO2 and NO3 below 40 km. The scintillation correction is unable to remove these perturbations in the spectra. A Global DOAS iterative method has been implemented for the retrieval of NO2 and NO3.


o      spectral inversion IR (O2, H2O):

A different algorithm is used for the spectral inversion of O2 and H2O, to take into account the dependence of the apparent cross-sections on the integrated densities. Reference transmission spectra are calculated for different integrated densities of O2 or of H2O. A direct model is used to take into account the dependence of the reference transmissions with the pressure.


After the spectral inversion, the vertical inversion of the line densities is applied to produce the local density values of each constituent. It is performed with the onion-peeling method. Tikhonov regularisation is applied in order to attenuate unphysical oscillations in the profiles, due to noisy data and scintillations. A new temperature profile is also produced from Rayleigh scattering from the UVIS spectrometer and from O2 density from the IR spectrometer. This profile is used to recalculate the effective cross-sections. Then the spectral and the vertical inversions are activated. It has been shown that this iterative process improves the results. The GOMOS temperature profile is also used to update the ray path from the previous computation made during the Level 1b processing and basing on ECMWF/MSIS90 data.

The measurements from the two fast photometers are used to retrieve a high resolution temperature profile of the atmosphere. Due to the variation of the index of air refraction with wavelength, the light beam of an occulted star is more bent in the blue part of the spectrum than in the red part. Thus the time delay between the signals of the two fast photometers provides indications on the density and the temperature profiles in the atmosphere which may be inferred with a high vertical resolution.


The processing steps and the generation of the Level 2 products are presented in more detail in the ATBD reference document (see Chapter 4 for reference). Auxiliary files

The auxiliary product files are used by the GOMOS geophysical processing facility and the calibration processing environment. Figure 2.4 illustrates the data flow of the calibration and the configuration files for GOMOS processing.

Figure 2.4: Data flow of the calibration and the configuration files for GOMOS processing.

Table 2.2 lists the auxiliary files with the naming convention, the general description of their content and the processing level for which they are used.

Naming convention

File description

Processing level


Calibration database

Level 0 to 1b processing


Star catalogue

Observation planning


Stellar Spectra

Level 0 to 1b processing


ECMWF forecast data

Level 0 to 1b processing


Level 1b processing configuration database

Level 0 to 1b processing


Level 2 processing configuration database

Level 1b to 2 processing


Cross section database

Level 1b to 2 processing


Instrument physical characteristics data

Level 0 to 1b processing,

Level 1b to 2 processing

Table 2.2: List of GOMOS auxiliary files.

The calibration file GOM_CAL_AX is used only by the Level 1b processor. The star catalogue GOM_CAT_AX is used by the mission planning software. It contains all possible stars (selected) for use by GOMOS, as well as 7 planets and dark regions (for dark calibration). The star catalogue is used by the Level 1b processor and is also used as a reference for reading the stellar spectra databank file (by the Level 1b processor).

The Cross-section database GOM_CRS_AX contains the specific trace-gas cross sections and is read by the Level 2 processor.

The characterisation file GOM_INS_AX contains the instrument parameters which are assumed to change (due to instrument ageing) during the instrument lifetime: the CCD size, the focal length, … The characterisation file provides inputs to both the Level 1b and the Level 2 processors.

The Level 1b and the Level 2 processor configuration files GOM_PR1_AX and GOM_PR2_AX are used for setting up the geophysical processors.

The stellar spectra database GOM_STS_AX is continuously updated and contains the averaged stellar spectra, as recorded by GOMOS outside the atmosphere.

The file AUX_ECF_AX contains the ECMWF meteorological forecast data needed to compute the atmospheric model used for the processing of the Near Real Time products.