GOMOS Overview

GOMOS Applications


The primary scientific objective of GOMOS was the global monitoring of the stratospheric and mesospheric vertical ozone distribution with high accuracy, high vertical resolution and full global coverage. The expected high accuracy of the ozone profile measurements allowed ozone trends to be studied over the lifetime of Envisat in the stratosphere and mesosphere.

GOMOS ozone trend observations addressed:

  • The degree of Arctic (polar) ozone loss under different meteorological conditions section
  • Increased data for determining height-resolved trend analysis section
  • Monitoring, which will indicate the beginning of expected 'recovery' in ozone section

The primary scientific mission objective pointed to the use of an instrument applying the occultation technique, as this was self-calibrating, protecting the measurements from long-term instrument drifts. Stars as light sources had been chosen in order to provide sufficient global coverage of the measurements, which could not be achieved with, for example, a solar occultation instrument.

The occultation technique had one significant advantage, which was that an absolute estimate of the molecule density was obtained from the ratio of two measurements taken with the same instrument within a few seconds. This made the method inherently self-calibrating, since even if the spectral sensitivity of the instrument was changing with time, the ratio will be measured correctly. This protection against long-term drift was ideal for the study of atmospheric trends.

GOMOS measured in the UV and visible wavelength regions where ozone shows strong absorption. The measured wavelength range was 250 nm to 675 nm, covering the ozone Huggins and Chappuis band. The UV wavelengths range was particularly well suited for mesospheric and upper stratospheric measurements. The visible band was well suited for the probing the stratosphere, including the lower stratosphere, where the UV signal was no longer strong enough anymore because of the UV absorption by ozone. Roughly, the UV yielded the best results for the mesosphere and stratosphere above 35 km and the visible band for the stratosphere below 35 km.

A highly desirable follow-on programme with GOMOS or COALA-type instruments provided long-term trend observation.

Secondary Scientific Objectives

NO2 and NO3 play important roles in the nitrogen chemistry relevant to ozone.

Observations of NO2 and NO3 help to address:

  • The degree of denoxification/denitrification in Arctic winter and availability of NO2 to deactivate ClO
  • The abundance of lower stratosphere NO2, which varies with the aerosol loading
  • Future NOx observations relevant to the ozone recovery. As NOx dominates ozone loss on a global scale, NOx observations are important to interpret ozone changes as halogens decrease.

GOMOS therefore observed NO2 and NO3 (enhanced), in the UV-visible band. OClO and BrO are key species in the polar ozone depletion processes. They can be detected with GOMOS if their concentration is enhanced. Measuring the aerosol extinction is an additional secondary objective. Aerosols are important for a number of reasons, including the provision of surfaces for heterogeneous reactions, which can strongly affect the partitioning of chlorine and nitrogen compounds.

Stratospheric water vapour is fundamental to the budget of many trace gases in the stratosphere. It was therefore important to determine its three-dimensional distribution and long-term trends. Water vapour was measured in a near-infrared channel at 926-952 nm.

A further secondary objective was to study stratospheric dynamics. This required measuring the atmospheric temperature and density profiles. Temperature and air density was derived from a near-infrared channel (756-773 nm). Complementary to this, a high-resolution temperature profile could be retrieved from atmospheric scintillation, which was be measured using two fast photometers sensitive in the blue and red, respectively.

The measurement of temperature was also needed for supporting the primary mission objective, ozone monitoring. The ozone cross-sections in the Huggins band (310-350 nm) were strongly dependent on the atmospheric temperature. Therefore, a long-term variation of UV absorption might be due to a temperature variation and could be wrongly attributed to an ozone variation. Temperature measurements were possible with the A-band of O2 at 760 nm. O2 is a perfectly mixed gas and the air can be assumed to be in hydrostatic equilibrium. Therefore, its scale height was directly connected to the atmospheric temperature. The O2 measurements also allow to relate all measurements of ozone density (and other species) to the air density to yield the mixing ratio [O3]/[air], which is a quantity most readily used in models. Furthermore, temperature and air density are essential parameters for atmospheric dynamics, including mixing of gases.

Long-term, campaign and permanent objectives

The scientific objectives were diivided into three groups. The first group consisted of long-term objectives. These objectives were enacted according to the overall GOMOS mission plan during the whole mission lifetime in a more or less regular manner.

The most important of these objectives was the monitoring of stratospheric ozone. The second group consists of campaign type objectives. These had more limited scope than those of the long-term objectives. The activation of a campaign objective was based on an agreement among the partners of the GOMOS mission planning.

The third group consisted of a few permanent objectives which had to be considered when a new mission plan was being prepared. In particular, whenever possible, certain stars (e.g. Sirius) were occulted and predefined locations (e.g. selected validation sites) were prioritised. Therefore, these objectives had an overruling priority.

In more detail, the three groups include:

Long-term objectives

  • Stratospheric ozone monitoring (15-50 km)
  • Stratospheric chemistry: upper stratosphere (35-50 km)
  • Stratospheric chemistry: lower and middle stratosphere (15-35 km)
  • Stratospheric dynamics: large scale dynamics, polar vortex, planetary waves
  • Stratospheric dynamics: small scale processes
  • Mesospheric ozone monitoring
  • Mesospheric noctilucent clouds variation (80-87 km)

Campaign type objectives

Geophysical validation of GOMOS products:

  • Validation of gas concentration retrieval for particular species
  • Co-located measurements e.g. overpass flights over validating ground based instrument or co-incident regions withvalidating satellite instrument
  • Validation of diurnal condition. (Depending on the validating instrument.)

Inter-comparison with MIPAS and SCIAMACHY:

  • At specified times
  • Comparison of ozone and aerosols profiles. (High priority)
  • Comparison of NO2, air, O2 and temperature profiles. (Lower priority)
  • Comparison of H2O, OClO and BrO profiles (Lowest priority) 
  • Interesting diurnal condition: depending on the validating instrument (MIPAS day and night, SCIAMACHY nadir and limb modes day, SCIAMACHY occultation mode twilight) 
  • Co-located simultaneous measurements with uniform global coverage
  • Campaigns: Campaign-specified latitude regions and priorities for gases
  • Special events, for example, volcanic eruption: ozone and aerosol profiles with highest accuracy around the eruption
  • Stellar spectra, with increased occultation starting altitude

Permanent objectives

  • Fixed region measurements
  • Fixed stars measurements
  • Long lasting occultations

The permanent objectives had overruling priority over the other mission objectives. But it must be noted that the permanent and other mission objectives could be fulfilled - or nearly so - at the same time, because of the large number of occultations per orbit. More generally, almost all mission objectives were to some extent fulfilled in any mission plan. The degree of fulfilment, of course, varies. The best results were obtained for the mission objective, which was the target of the optimisation.



GOMOS was a medium resolution spectrometer measuring atmospheric constituents by spectral analysis of the spectral bands between 250 nm to 675 nm, 756 nm to 773 nm, and 926 nm to 952 nm. The primary goal of GOMOS was the accurate detection of stratospheric ozone, allowing long-term monitoring of global trends. Whilst providing ozone profiles from UV-visible occultation spectra in an altitude range of ~15 – 80 km and with a vertical resolution of better than 1.7 km. Additionally, the instrument yielded small-scale turbulence measurements and high-resolution temperature profiles using two fast broadband photometers operated in two spectral channels; between 470 nm to 520 nm and 650 nm to 700 nm, respectively. 

The high sensitivity requirement down to 250 nm had been a significant design driver leading to an all-reflective optical system design for the UVVIS part of the spectrum and to functional pupil separation between the UVVIS and the NIR spectral regions (thus no dichroic separation of UV). Due to the requirement of operating on very faint stars (down to magnitude 4 to 5), the sensitivity requirement to the instrument was very high. Consequently, a large telescope (30 cm × 20 cm aperture) had to be used to collect sufficient signal, and detectors with high quantum efficiency and very low noise had to be developed to achieve the required signal to noise ratio.

In addition, in order to use the entire star signal, a slitless spectrometer design had to be chosen. The price which had to be paid for this "light efficient" design is that a high performance pointing system had to be used to keep the star image fixed at the input of the spectrometers in order not to degrade the spectral resolution and the spectral stability. Achieving a high signal to noise ratio when observing the very weak star signal embedded in strong surrounding atmospheric background and stabilizing the star image, in spite of the satellite disturbances, were major engineering challenges for the GOMOS design.


The main instrument requirements and the resulting design drivers are summarised in the table below:

Requirement description Requirement Design driver for:
Occulting stars characteristics Visual magnitude range: Maximum -1.6 to minimum 2.4 to 4 for stars with 30,000K and 3,000K temperature respectively

High sensitivity and dynamic range requirements for the star tracker

High sensitivity and dynamic range spectrometer detectors (especially in the UV)

Large telescope and high transmission optics needed to collect sufficient signal from the faint stars.

Spectral range of the spectrometer:

250 nm to 675 nm for UV and VIS

756 nm to 773 nm and 926 nm to 952 nm

Wide spectral range, high transmission optics.

Functional pupil separation between UVVIS and NIR.

Very strict contamination control to avoid UV sensitivity degradation.

Broadband sensitive and low noise detectors with high sensitivity in the UV.

High NIR sensitivity.

Spectral sampling

0.3 nm in UVVIS

0.05 nm in NIR

Large sensors (ca. 2,500 used pixels on four sensors each 1,500 pixels wide)

High dispersion, high efficiency gratings

Spectral resolution

1.2 nm in UVVIS

0.2 nm in NIR

High imaging quality optics

Very high pointing system stability requirements

Spectral stability knowledge in dark limb

0.07 nm in UVVIS

0.016 nm in NIR

Star tracker and pointing system

Photometer spectral windows and sampling rate

470 nm to 520 nm and 650 nm to 700 nm

1 kHz sampling rate

Fast (1 kHz), high sensitivity detectors
Short term radiometric stability (over 150 seconds) 1% Spatially uniform detectors and very high pointing stability
Linearity 1% High detector and electronic chain linearity (very challenging for the extremes of the dynamic range)
Pointing stability Better than 40 microradians peak to peak High speed, high accuracy closed-loop pointing system
Number of occultations per orbit 45 on average, i.e., approximately 920,000 occultations during the 4-year mission Challenging requirement for the star pointing mechanism in terms of long term performance and reliability
Angular coverage -10° to +90° with respect to the flight direction. Thus, large instrument angular range observability Large total angular travel range for the mechanism


The overall instrument design

The GOMOS instrument was based on a 30 cm × 20 cm telescope, whose pupil is shared by the UVVIS and NIR spectrometers and by two redundant star trackers.

This function was fulfilled by the optical beam dispatcher. The signal collected by the NIR subpupil was subsequently dichroically split between the NIR spectrometer and two photometers (1 and 2). A two-stage steering front mechanism (SFM) moving a 30 cm×40 cm flat mirror was used to point the line of sight towards the selected star and to track it with very high accuracy as it sets through the atmosphere. The telescope, the optics, all sensors and their associated front-end electronics were mounted on a thermally controlled optical bench. This telescope and optical bench assembly (TOBA) and the SFM were mounted via a GOMOS-interface structure (GIFS) to the spacecraft. The entire spacecraft and external GOMOS instrumentation (optomechanical assembly: OMA) was covered by an optomechanical cover responsible for protecting the instrument from light coming from a different direction other than the defined angular range, and for ensuring a stable, defined thermal environment.

The OMA was connected to the instrument electronics consisting of the sensor detection electronics, the redundant instrument control unit, and the redundant mechanism drive electronics in the payload equipment bay of the satellite via a dedicated harness. An overview of the optical design of GOMOS follows.

The main spacecraft resource requirements of the GOMOS instrument were:

  • Instrument mass: 175 kg
  • Instrument power consumption: 200 W
  • Data rate to satellite: 226 kbit/s

The equipment requirements and design

The spectrometer

A special frame-transfer CCD has been developed to meet the needs of the GOMOS mission.



A special frame transfer CCD has been developed to meet the needs of the GOMOS mission

This is a frame-transfer, UV to NIR sensitive, low-noise and radiation-hardened CCD. The main requirements together with the resulting technological and design choices are summarized below.


technological and design choices are summarised below.


Parameter Requirement (already demonstrated by flight models) Selected technology / design
Spectral range 250 nm to 950 nm Thinned, backside illuminated, anti-reflection coated CCD
Quantum efficiency 20% at 250 nm to 350 nm, 60% at 500 nm to 675 nm , 20% at 950 nm See above
Dark current at 20C < 25 pA/cm2 Inverted mode operation (MPP)
Linearity better than 0.5% Special design of the output stage
Geometrical design of the CCD 2 × 143 lines and 1353 columns with 20 × 27 micrometer pixels  


The steering front mechanism

The steering front mechanism consists of a flat mirror of some 40 cm × 30 cm mounted on a two-stage steering mechanism. A coarse steering mechanism, in azimuth only, steers the mirror coarsely towards the occulting stars within a 100° angular range, while an azimuth and elevation fine-steering mechanism with a range of approximately Ò 4° is performing the acquisition, centering, and tracking of the star as it sets through the atmosphere.




The steering front mechanism

The azimuth coarse-pointing mechanism is using a ball-screw drive, while the fine-steering stages are using voice-coil actuators. The fine-steering stages, together with the mirror, are inertially mounted so that spacecraft microvibrations are not transmitted to the line of sight. Additionally, electrodynamical dampers are used for microvibration damping. The control bandwidth of the mirror is approximately 5 Hz.

The main requirements, together with the resulting technological and design choices, are summarized in the table below.

Parameter Requirement Selected technology/design
Angular range (optical) 100° Two-stage (coarse + fine) design
Open loop pointing accuracy better than +/-0.02° bias and +/-0.01 degrees dynamics High resolution inductosyn angular sensors (1 LSB = 0.0009 degrees)
Torque (acceleration) resolution 5 micro nm Voice-coil actuators together with high linearity electronics
Microvibration rejection 3 microradians residual above 10 Hz Three-axis, frictionless mirror mounting
Number of angular travel cycles
> 1.5 million of 70° average


The science data electronics (SDE)

The science data electronics (SDE) was the CCD sensor control and signal conditioning unit. It controlled a total of 8 CCD's (two for the UVVIS spectrometer, two for the NIR spectrometer, two for the fast photometers, and two for the nominal and for the redundant star tracker). This unit had a high degree of programmability.


During the star occultation measurement, three bands of the spectrometer CCD's were read out. They contained the star, the upper, and the lower background spectrum. The typically seven lines of these bands were binned out in the output register of the CCD before reading. The position, the width, and the separation of these bands were all programmable. This gave a high degree of flexibility in optimizing the detection performance to specific observation objectives. During specific monitoring modes, the CCDs were read out in unbinned mode, and the integration times were programmable between 0.25 and 10 seconds.

The sequencing of the star tracker CCD was adapted dynamically to the different phases of star detection centering and tracking as follows. In the initial phase of star detection, the SDE detected the coordinates of the most-illuminated pixel of the CCD using a programmable integration between 5 ms and 50 ms, depending on the star magnitude. As soon as the star had been detected, a 10 × 10 pixel window centred around the star was read out with programmable integration times between 5 ms and 10 ms. This window follows the star as the tracking system centres the star in the field of view. As soon as the star was centred, the read out window "collapsed" to 8 × 8 pixels and the read out frequency was set to 100 Hz with 5 ms or 10 ms integration times.

The analogue signal conditioning consisted of low noise, programmable gain channels. The analogue to digital conversion was done with 12 bit ADC's.

The telescope

The GOMOS telescope had a rectangular aperture of 30 cm × 20 cm. It had to operate in a 0.6° field of view and had to have very good transmission in the 250 nm to 950 nm range. The small allowable volume (intermirror distance < 250 mm) coupled with high-quality imaging and stability requirement, and coupled with the high stiffness requirement (first eigenmode above 180 Hz) were very challenging. A Cassagrain design based on aspheric primary and secondary mirrors, and based on a CFRP structure had been chosen. The main telescope requirements, together with the resulting technological and design choices, are summarized in the table below.



The GOMOS telescope

Parameter Requirement Selected technology/design
Field of view 0.6° Cassegrain design
Optical transmission

> 82% between 250 nm and 500 nm

> 92% between 500 nm and 952 nm

Al coated UVVIS subpupil and AG coated NIR subpupil
Intermirror distance stability better than 10 mm coefficient of thermal expansion and coefficient of moisture expansion compensated CFRP structure
Imaging quality

UVVIS: 25 mm @ 85% encircled energy

NIR: 30 mm @ 85% encircled energy

Cassegrain design with aspheric primary and secondary mirrors

Sensor Modes

GOMOS Measurement Principle

GOMOS principle

The GOMOS measurement principle was based on stellar occultation. Its main measurement mode was the occultation mode. During this mode the instrument was autonomously acquiring and tracking stars as they set through the atmosphere. The occultation mode had a specific submode called "fictive star" submode. In this submode the instrument was scanning the limb along a programmable trajectory as if a star was present. This submode could be used for limb sounding under bright limb conditions. The sequence of stars used for occultation was generated by a set of CFI software and was provided in form of a timeline of occultation.  Nominal measurements were regularly interrupted by three types of monitoring activities, namely:

  • The uniformity mode (UNI): in this mode the CCD sensors were read out in a non-binned, i.e., pixel by pixel, mode. Depending on whether, the instrument is pointed towards "dark space" or towards uniform limb, the dark current or the photo response uniformity of the CCD sensors is characterised. In this mode the tracking function was disabled. The spectrometer integration time was programmable in the range of 0.25 to 10 seconds.
  • The spatial spread mode (SSM): in this mode the instrument operated as in the uniformity mode, but with the tracking function active. Thus, the optical transfer function of the instrument could be monitored in this mode by observing a star outside the atmosphere
  • The linearity mode (LIN): The linearity mode (LIN): in this mode the sensors were operated in binned mode as in occultation mode but with variable integration times (in the range 0.25 to 10 seconds). This mode was used to monitor the linearity of the detection chains during the instrument lifetime by observing stars outside the atmosphere with variable integration times.



The instrument line of sight could be oriented towards a preselected star and maintained whilst the star was setting behind Earth's atmosphere observed on the horizon. During the star occultation, the ultraviolet, visible, and near-infrared spectra of the star were continuously recorded.

As the star set through the atmosphere, its spectrum became more and more attenuated by the absorption of the various gases in the atmosphere, each of which was characterised by a known, well-defined spectral signature. Back on the ground, these attenuated spectra recorded by GOMOS were compared with the unattenuated stellar spectrum measured a few tens of seconds earlier, outside the atmosphere, so allowing the absorption spectra to be derived very accurately. This radiometrically self-calibrating method was protected from sensitivity drifts and was thus capable of fulfilling the challenging requirement of reliably detecting very small trends in ozone (and other gas) profiles.

During day-side observations, the solar radiation scattered by the atmosphere was superimposed to the star signal as the line of sight started crossing the atmosphere. In order to be able to retrieve the star signal transmitted through the atmosphere without the background component, the (vertically imaging) spectrometers were recording the background spectrum just above and below the star too. These spectra were then used on ground for background removal.

GOMOS used SAGE, as the occultation measurement method which, compared to other instruments, offered the advantage of high measurement accuracy and of very good altitude profiling. However, instead of using the sun as an occulting source, GOMOS used stars to perform the occultation measurements. There were some 100 stars bright enough for GOMOS to observe as they set through the atmosphere. Over one day/one month there are typically 1600/48000 occultations to be chosen from.

Selection criteria like coverage of specific latitudes/longitudes, altitude ranges, etc. could be applied, while still maintaining a good global coverage. Typically, GOMOS performed more than 600/18000 profile measurements per day/month.

Mission Operation

The GOMOS instrument was active from 8 March 2002 to 8 April 2012.

Download the full list of events that occurred during the mission lifetime. These events include maintenance to or anomalies with GOMOS and may impact the availability of data from the instrument.

GOMOS Resources