Mission objectives and scientific Requirements Document (MRD)

for the

Gravity Field and Steady State Ocean Circulation Explorer

(GOCE) Mission

Draft Version 1.5 - 15 November 1997

1. Introduction

For the post-2000 time frame two general classes of Earth Observation missions have been identified to address user requirements (see e.g. ESA 1995b), namely:

Earth Explorer Missions - these are research/demonstration missions

Earth Watch Missions - these are pre-operational missions

Nine Earth Explorer missions have been identified as potential candidates for Phase A study (Reports for Assessment: The Nine Candidate Earth explorer Missions, ESA SP-1196 (1-9), 1996). After a selection process four missions were recommended for further study including among others the Gravity Field and Steady State Ocean Circulation Explorer (hereafter GOCE) Mission.

The purpose of this document is to define the mission objectives and scientific requirements of the GOCE and to provide guidelines for the technical implementation of the mission. The document is divided into 7 chapters addressing the scientific background of the mission (chapter 2), the objectives of GOCE (chapter 3), the observation requirements (chapter 4), the mission elements (chapter 5), the mission products (chapter 6), and the concluding remarks (chapter 7). The instrument specifications and critical items are included in Annex 1 and Annex 2..

2. Scientific background

The Earth is a dynamic system constantly undergoing changes. The Earth's gravity field reflects the Earth's geological history, formed by processes such as post-glacial rebound and tectonics. The unique contribution of a dedicated gravity field mission is the potential for providing a fuller description of the dynamic processes in the solid Earth.

The geoid is the surface of equal gravitational potential which on average correspond to the ocean surface. It is the referense surface to which the sea surface may be compared, by means of which the steady state ocean circulation may be inferred. Knowledge of this mean circulation is required by oceanographers and by the builders of models of the Earth's climate system. The geoid is also required to be precisely known by geodesists and surveyors.

For this reason the Earth's gravity field and geoid surface, determined uniformly and globally with significant improved spatial resolution and accuracy, are of fundamental importance. Only by the use of a satellite mission can these requirements be met.

Marine Geoid and its Impact on Ocean Circulation - the absolute value of the ocean dynamic topography require the determination of the 'hypothetic' ocean at rest, i.e. the marine geoid. This is the basic requirement for modelling ocean circulation and interpreting satellite altimeter data. Unfortunately, current insight into geoid uncertainties and their impact on the absolute dynamic topography is limited, particularly at shorter wavelengths less than 2000 km.

Given the continuing need to study and predict climate variation and climate change by the combined use of altimetry, global ocean circulation models, and high-quality global in-situ data, it is thus essential to significantly reduce errors in our knowledge of the geoid, in particular, at shorter wavelength. The same also applies to the growing field of operational oceanography for which radar altimetry are an important data source. This could be achieved once and for all through a single dedicated gravity field mission. The need for this has been strongly articulated in international scientific programmes including the World Ocean Circulation experiment (WOCE), the Climate Variability and Prediction program (CLIVAR) and the Global Ocean Observing System (GOOS).

Gravity Field and Solid Earth Processes - specific issues to be addressed by an accurate and detailed determination of the gravity field includes discrimination between active and passive models of rifting; identification of anomalous mass which may drive basin subsidence; and determination of the deep density structure beneath the continents and of the mechanical strength of the continental lithosphere.

Understanding of mantle processes, in particular convection patterns, and of post-glacial mass readjustment, will greatly benefit from improved and more detailed knowledge of the Earth's gravity field. Moreover, the accurate and detailed determination of the anomalous gravity field plays a key role in advancing understanding of the dynamics of the continental lithosphere. It is also necessary to identify the contribution of post-glacial rebound to sea level change, and the impact of Solid Earth processes on the global ocean. Further progress in the understanding of these processes and improvement in geopotential models require global determination of the gravity field, at a spatial resolution down to half wavelengths of between 50 and 400 km.

Geodesy- at present, the different orthometric height systems (national datum) differ by the order of decimeters between islands and between islands and continents, over distances of a few 100 km. Between continents these differences may be as large as a metre. Sea level in one part of our planet can therefore not be properly compared with sea level in other parts, nor can changes be precisely separated into sea level rise and vertical land uplift or subsidence respectively. One of the key objectives in geodesy is therefore to improve and unify the different orthometric height systems using high quality gravity field data.

In recent years several concepts have emerged for dedicated gravity field missions. The GPS receiver and a micro-accelerometer will make CHAMP (the German mission planned for launch in mid-1999) a good candidate for gravity field mapping at degree and order below 40-50 (wavelength of about 1000 km). Currently, NASA are considering to fly a dedicated gravity field mission named GRACE with the aim to retrieve very accurate observations of the time variations of the gravity field at long and medium wavelengths (to degree and order 60). Beyond this degree and order we have to rely mostly on current knowledge which cause the degree error rms to become too large. For example, at long wavelengths (degree less than 14) the ocean dynamic topography signal is separable from the geoid. However, the geoid error becomes comparable to the dynamic topography signal around degree 14, that is at wavelength of about 3000 km whereas at scales of 1000 km and less geoid error dominates.

3. GOCE Mission Objectives

The aim of the Gravity Field and Steady-State Ocean Circulation Mission (GOCE) Mission is to provide global and regional models of the Earth's gravity field and of the geoid, its reference equipotential surface, with high spatial resolution and accuracy. Such models would be used in a wide range of research and application areas, including global ocean circulation, physics of the interior of the Earth and unification of height systems. In oceanography for instance, it would contribute to the recomputation and correction of a major deficiency which is the lack of an accurate and precise reference surface for the interpretation of the mean sea level as mapped by satellite altimetry, allowing longer term studies of the variations of the sea surface. It will moreover benefit to all mission which requires highly accurate orbits. It will also make it feasible, should it become necessary, to lower the altitude of future altimeter satellites, since the orbital error due to gravity field knowledge will be considerably smaller. Furthermore, it will impact on the reliability of topography studies from SAR interferometry, where the orbit errors are presently posing a limitation.

Geodesy (responsibility C. Tscherning, H. Sünkel)

The gravity potential provides the foundation for all practical height determinations, since the surface of zero height - the geoid- is a surface where the potential is equal to a constant. The knowledge of the height of the geoid above the ellipsoid is extremely important, because heights provided by GPS has to be converted to height above the geoid by the orthometric height (or by the normal height). The practical determination of orthometric height is conducted using the long wavelength information provided e.g. through a spherical harmonic expansion (SHE) combined with local gravity or topographic information. This determination is currently disturbed by errors in the longer wavelengths, often of the order of 0.5 - 2.0 m. The goal is to be able to compute geoid height differences to better than 0.01 m for distances of 100 km. This can be achieved if local gravity data is combined with an improved SHE.

Geodetic height systems have currently their zero-values fixed through a mean sea level as calculated at or near tide-gauges. Due to changes in the sea surface topography this means that the height systems at present differs with values of the order of decimeters between two islands and between an island and a continent having distances up to a few hundred km. Between continents the differences may be up to a meter. These height systems (national datum) should be unified (by determining the differences relative to an accurate geoid) based on data from a gravity field mission enhanced with local data. Such a geoid surface would in turn serve as a stationary reference for the study of all topographic processes, including dynamic ocean topography (and therefore ocean circulation), the evolution of the ice sheets and land surface topography.

Geodynamics (responsibility R. Sabadini, R. Rummel)

The open questions that in order to be solved require a high resolution gravity mission are related to the structure and composition of the continental lithosphere, where mineral resources of strategic importance are located, and to the dynamics of the oceanic lithosphere at ocean ridges and subduction zones where the interaction between plates is responsible for dynamic processes that control the evolution of important tectonic structures. New achievements in the comprehension of the tectonics of plate interiors and plate boundaries will allow to understand the impact of tectonic forces on the stress field in active seismic regions and to improve the knowledge on the seismic risks in areas where earthquakes cause human and economical losses.

This mission will also allow a substantial improvement of the knowledge of the rheology of the mantle that controls the dynamics of the interior and of the whole Earth; mantle rheology impacts not only the long-time scale convective pattern of our planet but its comprehension also enables the interpretation of post-glacial rebound data and of present-day sea-level changes.

Marine Geoid/Ocean circulation (responsibility P. Woodworth, C. Le Provost)

The main objective of the GOCE mission for oceanography is a better understanding of the mean (or 'absolute') ocean circulation via an accurate determination of the geoid. At the present time, the mean sea surface (MSS) is known to several centimeters accuracy, owing to the high precision of recent altimeter missions, specifically TOPEX/POSEIDON. However, the configuration of the ocean dynamic topography or sea surface topography ( = MSS - Geoid height) is limited because of the multi-decimeter errors in the geoid. If, through GOCE, the dynamic topography can be determined at the centimetric level, then the mean strengths of the main ocean current can be inferred with major benefits to several areas of oceanographic research.

Two examples may be given of the impact of improvement in knowledge of the ocean transports at large (basin) and small spatial scales. For the former, an error of 1 cm in sea surface slope integrated across the North Atlanctic at 30_ N, which might be aimed at achieving by a gravity field mission, corresponds to 7 Sv in volume transport (assuming barotropic flow). This is of order 10% of the North Atlanctic transport. Present knowledge of the geoid is more than an order of magnitude worse than this. For the latter, most current systems, from strong (e.g. Gulf Stream) to weaker (e.g. Scottish coastal current) are of the order 100 - 300 km in width and will have sea surface topography signatures of order 100 - 30 cm respectively. In addition, many important flows through straits are of this spatial scale (e.g. Drake Passage). It is clear that the geoid will be required to be known at the centimeter level at this spatial scale if the geostrophic currents associated with such systems are to be resolved at the 10-20% level.

As regards climate-oriented research, a better knowledge of the ocean circulation will lead to significantly improved estimates of transports of the huge amounts of heat, fresh water and salt, and of dissolved quantities including pollutants, and to improved knowledge of the carbon cycle essential to climate studies.

Atmospheric density modelling (Responsibility K. Wakker, G. Balmino)

A drag free system will be installed on board the GOCE satellite which will be in orbit at about 260 km of altitude (see chapter 5 Mission Elements). If the drag on the satellite can be determined to a precision of a few percents from the thruster activity and from the modelling of the radiation pressures (from the Sun and Earth), estimation of total air density can be obtained at satellite level. This, in turn, can contribute to better understanding and prediction of the behaviour of the thermosphere.

4. Observation requirements

The results of a gravity field mission may be made available to the users in different forms. The most important are :

filtered measurements of gravity gradients at satellite altitude,

coefficients of a spherical harmonic expansion of the gravity potential,

derived surface gravity anomalies and

derived geoid heights

The mission performance requirement ranges for the different scientific aims; it is expressed in terms of geoid heigths and gravity anomalies (Table 1). Those values are supplemented by the corresponding spatial resolution they apply to. The values shall be understood as 'most likely' values i.e. as equal to the error standard deviation assuming a normal. distribution.

The mission shall last at least 8 months. There is no need for a repeat orbit. Actually a constantly drifting ground track that ensures a uniform coverage is desirable. A polar orbit would certainly be ideal but Sun-Synchronous orbits are acceptable because the effects of polar gaps can be minimized or surpressed by, for instance, including long wavelength information from present global model data and recent release of gravity field data from Russia.

The gravity field and geoid height recovery requirements shall be achieved on a global scale but Ocean Circulation derived requirements do not apply to the polar gaps.

Accuracy Spatial Resolution
Geoid Gravity (half wavelength)
Ocean Circulation

- Mesoscale

- Basin scale

2 cm

0.2 cm

<0.1 cm

100 km

200 km

1000 km


- Continental lithosphere (thermal structure,

post-glacial rebound)

- Mantle composition, rheology

- Ocean lithosphere and interaction with

asthenosphere (subduction processes)

1-2 mgals

1-2 mgals

5-10 mgals

50-400 km

100-5000 km

100-200 km


- Ice and land vertical movements

- Rock basement under polar ice sheets

- World-wide height system

2 cm

<5 cm

1-5 mgals

100-200 km

50-100 km

50-100 km

Table 1. GOCE observation requirements

5. Mission Elements

Orbit : The reference orbit is a dawn-dusk sun-synchronous orbit at a mean altitude of 250 km. The baseline mission duration is 8 months.

Satellite: The satellite that will be the space segment of this mission is called GOCE. The requirements of this mission are special and the satellite is built around the payload. It is a slender satellite configured to minimise aerodynamic drag. Some of its subsystems are conventional but others have uncommon designs due to the special needs of the mission.

Instruments: The payload of GOCE includes two instruments, which provide complementary measurements of the gravity field:

a gravity gradiometer that measures the gravity gradients, from which the medium and short wavelength terms of the field can be derived.

a GPS/GLONASS receiver called GRAS from which the precise satellite position can be obtained and from which the long wavelength terms of the field can be derived.

The instruments are decribed in detail in Annex 1 of this document.

Ground Segment: It includes a dedicated single ground station at a northern latitude. ESOC would be in charge of mission and satellite control and would perform a precise orbit determination for the satellite. The corrected and calibrated ouput will be provided to a science data centre or a consortium (for example as done in the Hipparcos project) which will be responsible for verification and qualification of the data as well as the generation of the final (global) geophysical data products.

The whole data reduction process is not time critical. Data and certain products will be archived for 10 years and will be retrieved and provided to users on request.

Synergistic Observations: Some scientific studies running in parallel to the definition of GOCE are investigating high accuracy ground knowledge of the geoid to help in the calibration of the instruments. Although the fulfilment of the scientific requirements may not need such a 'ground truth' calibration, if implemented it could simplify the in-orbit calibration.

GOCE does not fly over the poles. The final processing of data after the execution of the mission will make use of any existing gravity data over the poles. The availability of data produced by, e.g. a campaign using aircraft, will improve the performance of the mission. The mission performance figures quoted in this document have assumed that polar data are not available.

Ground support requirements for tracking will be satisfied with the availability of the IGS network of GPS receivers

6. Products

The level of data to be produced will be the following:

level 0, raw payload data as sent by GOCE. The satellite and instrument data will

be provided at a frequency (data rate) of 1 Hz.

level 1a, data 'depacketised' and sorted in files with calibration data attached but not applied.

level 1b, data calibrated and corrected. Level 1b sets of data to be produced include: gravity gradients and orbit data of GOCE as derived from GRAS, accelerometer outputs, thruster actrivity parameter and attitude control.

level 2, final geophysical outputs. Several main types of outputs: global gravity potential, modelled as harmonic coefficients, and global gridded values of geoid heights and gravity anomalies. Also regional models as appropriate.

After validation of the satellite data, the scientific data will be pre-processed and stored as level 1a data on appropriate media as they are generated during the mission. The pre-processing, essentially consisting of channel decommutation and reformatting, will not contain any scientific evaluation. The raw data will contain the readouts from the instruments together with calibration, time attitude and other housekeeping information like temperatures as required. Data will be calibrated and corrected to provide the level 1b product of the gradiometer. The output will be gravity gradients at the satellite orbit for maximum precision but already referred to an Earth reference frame if all gradients can be derived with sufficient reliability. This data shall also include the attitude, angular velocity and angular acceleration history of the satellite. These tasks will require to develop special algorithms which will be part of the general industrial development of the satellite and which are needed to verify the gradiometer.

GPS/GLONASS data will be calibrated and corrected to generate the relevant part of the level 1b product. The output will be a precise orbit determination of GOCE. To do so it will be necessary to use the International GPS Geodynamics Service (IGS) of which ESOC forms part. This service routinely computes high accuracy ephemeris of the GPS satellites.

Once processed, level 1a and 1b data will be transferred to a science data centre that will perform the final data reduction. The delivery will be done by any suitable media. A delivery rate of once a week can be used as a guideline. The algorithms required to convert the data from level 1 to 2 are, thanks to the work of the CIGAR consortium, quite mature. The level 2 final mission products as provided by a data processing centre or consortium will include:

a global gravity potential model represented by its harmonic coefficients, for most general use, together with error estimates,

a global geoid height grid for oceanography,

and a global gravity anomaly grid for geodesy and geophysics.

Regional models will be provided as requested. Further processing of GOCE data will also provide better models of atmospheric density.

Level 1a data will be archived by ESA as optical disks for a period of 10 years. No other service apart from physical storage and duplication on request will be provided. Level 2 products will also be archived and distributed on request

7. Concluding Remarks

Annex 1: Instrument description

Capacitive Gradiometer

This gradiometer is based on ambient temperature, closed loop, capacitive accelerometers. The principle of operation is based on the measurement of the forces needed to maintain a proof mass at the centre of a cage. A six degrees of freedom servocontrolled electrostatic suspension provides control of the proof mass on translation and rotation. Any movement of the proof mass will produce differences in capacitance between the branches of a capacitance bridge. This difference will be sensed amplified and corrected. The correction is done by adjusting the electrical potential of electrodes that act upon the proof mass until the difference in capacitance is reduced to zero.

The present concept of the gradiometer has six accelerometers with one pair located along each main direction: X (velocity) Y (perpendicular to orbit) and Z (Earth). This configuration is able to recover the three diagonal terms of the gravity gradient matrix. It will also provide -but with degraded performance- the non diagonal terms and the angular velocities and accelerations of the satellite.

The separation between each pair of accelerometers, which defines the gradiometric base, will be around 0.5 m.

The basic instrument characteristics of the capacitive gradiometer are specified in Table A1.
Mass (kg) Power (w) Data Rate




Calibration Bandwidth (Hz)
7050 1.9*.9*.9 1 in 10000 0.005-0.1

Table A1. Instrument charateristics

The expected final in-orbit gradiometer accuracy will be Å 2-5 mE (1 E = 10-9 s-2)

The gradiometer is also used for the determination of the non gravitational accelerations acting upon the satellite. The instrument performance requirements are given in Table A2.
Orbital Frequency

(0.00018 Hz)

0.005 Hz 0.1 Hz


Not applicable 4 10-9 m/s-2ÃHz


10-8 ms-2 2 10-9 m/s-2ÃHz 2 10-9 m/s-2ÃHz

Table A2. Instrument performance requirements

Inductive Gradiometer

This gradiometer is based on the sensing of displacements by detectors working at the temperature of liquid He. The instrument is inside a reservoir that contains the fluid. . Two proof masses are maintained levitating by inductances. Any relative displacement of the masses caused by external accelerations provokes changes in the currents going through them.. The changes are sensed by a SQUID detector. The baseline design for a pair of accelerometers includes: six Squids, two proof masses, seventeen coils interacting with the proof masses, four independent current loops for seven different circuits and heat switches to enable the independent setting of currents in any one of these loops. This allows the detection of all accelerations (linear and angular) and the in-orbit correction of angular and linear misalignment of the proof masses. It also provides a system that is much more sensitive to differential movement of the proof masses, (that we want to measure), than to the common movement of them, (that we want to reject).

It has also three pairs of differential accelerometers along the axes X, Y and Z of the satellite. It provides the full gravity tensor plus information on satellite angular velocities and accelerations. The structure incorporates six identical accelerometer sets placed on the six faces of a cube. Most elements are in niobium.

The cryostat will contain the gradiometer and the liquid helium for cooling. Preliminary calculation shows that 100 litres are needed for 8 months of mission life. This keeps the temperature of the accelerometers and cryogenic apparatus at about 2 K

The separation between each pair of accelerometers, which defines the gradiometric base, will be around 0.15 m.

The basic instrument characteristics of the inductive gradiometer are listed in Table A3.
Mass (kg) Power (w) Data Rate


Volume (m) Calibration Bandwidth (Hz)
130 1002 1.5*0.9*0.9 1 to 100000 0.005 - 0.1

Table A3. Instrument characteristics

Of the 130 kg of mass, 30 correspond to the gradiometer, 80 to the cryostat and 20 to electronics. The expected final in-orbit gradiometer accuracy will be Å 1 mE.

The gradiometer is also used for the determination of the non gravitational accelerations acting upon the satellite. The instrument performance requirements are given in Table A4.
Orbital Frequency

(0.00018 Hz)

0.005 Hz 0.1 Hz
Angular Acceleration
Linear Acceleration 10-8 ms-2 2 10-9 m/s-2ÃHz 2 10-9 m/s-2ÃHz

Table A4. Instrument performance requirements


GRAS is a geodetic quality GPS/GLONASS receiver, i.e. it provides measurements at two frequencies for ionospheric corrections to be applied. More information can be seen in the Earth Explorer Atmospheric Profile Mission Report for Assessment (ESA-SP-1196(7)). The main interface characteristics of GRAS are provided in Table A5.
Mass (kg) Power (w) Data Rate (Kbps) Volume (m) Effective Carrier phase

noise (mm)

315 1 to 10 0.3*0.06*0.2 1

Table A5. GRAS characteristics

For satellite to satellite tracking, a antenna looking to the zenith is sufficient. The mass of this antenna is 3 kg and its volume 0.1*0.1* 0.3 m

The gradiometer is unable to provide good performance on the determination of the accelerations outside the measurement bandwidth. GRAS is used to complement the SGG measurements and to provide calibration of the gradiometer outside the measurement bandwidth, e.g. to provide absolute common scale factor errors.

Annex 2: Critical Items

The values of this attachement are not a list of specifications. They are the industrial assumptions taken in the elaboration of the error budget of Goce and they are provided for information only.

Most of the times, it is necessary to specify the performances on two frequency domains:

Taking into account that the phase A has not yet started, it has to be understood that all the values provided below are pending of consollidation by the industrial contractor.

Mission Duration and Launching Time: The minimum duration of the mission is 9 months with a nominal launching date is spring equinox 2004.

Orbit (Density times Area): The orbit shall be sun synchronous dawn-dusk with an orbit altitude at a minimum compatible with the maximum drag forces. It is expected that it will be around 250 km of altitude.

Density parameters

The satellite attitude and drag control devices will be dimensioned against a 2 sigma solar activity probability. Scientific performance are going to be calculated with respect to an altitude equivalent to 1 sigma solar activity probability.

The corresponding solar activity values are:

Gravity Tensor: The error analysis has been done assuming the following values for the gravity tensor.

Gravity gradient tensor (integrated value) in the range 'o'

Gravity gradient tensor in the range 'w'

Resulting external accelerations: The following accelerations are expected.


In the 'o' domain: 4 10-6 rad/s2 for each of the three axis.

In the 'w' domain:
At 0.005 Hz: At 0.1 Hz:

The figures above should be ensured by adequate configuration and centering of the satellite. In principle no active control is needed


In the 'o' domain:

In the 'w' domain:
At 0.005 Hz: At 0.1 Hz:

To ensure the figures above it is necessary to provide active control on axes X (along velocity) and Y (perpendicular to orbit) of the satellite. The low value along Z will be ensured by the naturally small perturbations along that axis.

Attitude Control

In the 'o' domain: 0.1 _ = 1.75 mrad for yaw, pitch and roll

In the 'w' domain:
At 0.005 Hz: At 0.1 Hz:

To ensure the 'w' figures above it is not necessary to provide active control of the angular stability, velocities or accelerations. The pointing accuracy on 'o' needs active controlling.


Capacitive Inductive

Gradiometric baseline: 0.5 m Gradiometric baseline: 0.15 m


Linear accelerations rejection ratio: Linear accelerations rejection ratio

In the 'o' domain 1.5 10-5 In the 'o' domain 3 10-7

Angular accelerations rejection ratio: Angular accelerations rejection ratio

In the 'o' domain 1.5 10-5 In the 'o' domain 3 10-6

Intrinsic error

Total: 1.16 10-12 m/s2ÃHz Total: 1.22 10-13 m/s2ÃHz

Instrument dimensional stability

Relative acceleration between Not applicable

proof masses 1.12 10-12 m/s2ÃHz

Resulting Gradiometric Error Budget (all values in mE).


All the values below are after post processing and for each axis:

Intrinsic noise 3 mm

It includes anti spoofing and selective availability

Multipath 3 mm

CoM to satellite possitioning 3 mm

Non gravitational forces 5 mm

Contribution to the final satellite possition budget:

The derived requirement on recovery of non

gravitational accelerations is TBD

GPS&Glonass Ephemeris 5 mm

Contribution to the final satellite possition budget: