1.

INTRODUCTION

Gravity is a well-established tool of today’s Earth Observation from space. Measurement of the gravity field reveals Earth’s state of mass balance and its dynamics; provides the geoid as reference of sea level, global ocean circulation and height systems; and the variations of gravity and of the geoid measure mass exchange processes in the Earth system¹.

GOCE² orbiting from 2009 to 2013, delivered a gravity map of Earth with mean global accuracy of 2.4 cm in terms of geoid heights and 0.7 mGal for gravity anomalies, at 100 km spatial resolution³, and applications of its data continue to be developed⁴. GOCE’s instruments measured full-tensor gravity gradients and GNSS positions while orbiting at mean orbit altitudes of 255 km (nominal mission) and 225 km (extended mission), in drag-free mode. The low controlled altitude, the full tensor measurements, the drag-free control and the accurate angular accelerations measured as a byproduct of the gradiometer payload were all instrumental in GOCE’s outstanding result.

GRACE, in orbit from 2002 to 2017, provided monthly estimates of Earth’s global gravity field at scales of a few hundred kilometers and larger. The time variations of the gravity field were used to determine changes in Earth’s mass distribution, with applications ranging from measurement of continental water content (season dependent changes in large river basins, groundwater extraction), to ice and snow accumulation and depletion in the polar regions and large glaciers, to monitoring of global mean eustatic sea-level variations⁵. GRACE consisted of two identical satellites in nearcircular, polar (89° inclination) orbits, initially at 500 km altitude, at mutual along-track distance of 220 km. The instantaneous distance measured by a dual-band microwave ranging instrument (24 GHz, 32 GHz) was the main datum, supplemented by GPS positions and non-gravitational acceleration data measured by high precision accelerometers. The satellite altitude decayed naturally under air drag down to about 320 km at end of life, and the ground track did not have a fixed repeat pattern.

GRACE Follow-On⁶, launched on 22 May 2018, is meant to continue the GRACE time series for at least five years, with largely unchanged on board systems, plus technology demonstration of a more precise, laser based ranging interferometer.

Acceleration measurement errors (temperature drifts, angular rates), the relatively high and variable altitude, and the onedimensional North-South sampling are known to affect the GRACE gravity models. Improvements of the spacecraft design (thermal control, attitude measurement and control) can reduce the systematic errors. Beyond that, however, aliasing due to poor modelling of high frequency ocean and atmospheric mass variations dominates and even a substantially improved instrument such as the laser ranging interferometer on board GRACE-FO cannot deploy its full potential⁷. A single pair of satellites – as would be the case of a Mass Change Designated Observable (MC DO) mission flying alone - will not meet operational needs, since providing only partial information and being unable to support key applications, e.g. ground water and aquifer monitoring and management, at the required spatio-temporal resolution. Needs exist also in support of science advancements, especially for climate, since gravity data are required to derive Essential Climate Variables¹³ products, as identified by GCOS 2016 Implementation plan. From a (pre-) operational standpoint, among current services, those for land, climate, ocean, and emergency management would especially benefit from improved mass change data as available only from a double pairs constellation. Thus, a Next Generation Gravity Mission¹⁸ dedicated to mass transport in the Earth system will require not only improvements in the instrument, the spacecraft (disturbing accelerations) and the mission design (sampling), but also a constellation of two pairs of satellite in an optimal orbit configuration and a strategy for reducing the aliasing errors.

From a programmatic point of view, ESA is considering the NGGM as a candidate Mission of Opportunity in cooperation with NASA: interactions - at scientific and technology level - are ongoing between ESA and NASA in order to prepare the next steps to arrive to a joint constellation. Over the past decade, system studies and technology development for the NGGM mission have advanced the maturity of the system concept and the readiness level of the key technologies (attitude and drag control, proportional thrusters^15,16, laser optics and electronics) for the mission to be proposed for development after 2022 and launch around 2028. Different candidate implementation scenarios exist, to be traded in the Phase A under preparation and planned to start in early 2021.

The paper focusses on the status of the ESA NGGM studies, concerning both the platform (featuring drag control with an optimized thruster layout, refined over the last part of the Phase 0) and the Laser Tracking Instrument^17,19 (featuring a new “off-axis” Retro-Reflector laser interferometer design currently under test). A discussion of alternative drag compensation scenarios will be presented, in order to address the entwined impact of different levels of drag-compensation designs and related orbit altitudes on the mission performance.

2.

THE NEXT GRAVITY MISSION CONCEPT

Soon after the selection of GOCE as its first Earth Explorer (1999), ESA started preparatory studies towards a Next Generation Gravity Mission (NGGM). Since the mid-2000’s, these studies have focused on a mission concept devoted to sustained observation and quantification of mass transport processes in the Earth system (encompassing AOHIS: Atmosphere – Ocean – Hydrology – Ice – Solid Earth) over a decade-long time span. The objective is achieved by monitoring the time variation of the Earth gravity field by Satellite-to-Satellite Tracking (SST) at high spatial and temporal resolution, using a heterodyne laser interferometer measuring the variation of the distance between two satellites flying in formation at low altitude. Such measurements of mass transport in the Earth system are unique to this type of mission and they complete science’s picture of Global Change with otherwise unavailable data.

The NGGM mission and system design will build on the experience of GOCE¹⁷, in particular for the design of the attitude and orbit controls, GRACE, for the concept of an SST mission with a direct inter-satellite metrological link between two spacecraft flying in low-Earth orbit, and GRACE-FO for the laser interferometer instrument.

The science case and design concept of the new mission have been built up in parallel, linked studies of the user science institutes and the engineering teams. The tiny gravity signals associated to mass transport in the Earth system may be detected using “virtual gradiometers” formed by satellites orbiting in formation, with separation among members of tens to hundreds of km. Early studies addressed mission design options such as type of satellite formation, altitude, intersatellite distance, number and accommodation of sensing instruments (accelerometers) per satellite, ranging technology. Design requirements and performance of formations such as “pearl-string”, “pendulum” and “cartwheel” were studied. A notable outcome of this phase was the selection of the “Bender constellation”⁸, consisting of two pairs of satellites flying in pearl-string formation in near-polar and mid-inclination orbits, as the best compromise between mission performance and implementation cost.

The science studies^9,10 produced a detailed assessment of the impact on the gravity field solutions of ocean tide aliasing, which was found to cause larger errors than non-tidal signals, and developed ocean tide mitigation strategies to be applied in post-processing.

The NGGM mission will be devoted to measuring the time-variations of the gravity field with order of 100 km spatial resolution and weekly or better time resolution. The NGGM objectives will be achieved by employing an inter-satellite laser interferometer (282 THz, corresponding to 1064 nm wavelength), and designing the flight system to exploit at best its expected performance in path length measurement (<10 nm/√Hz, NGGM goal specification). This implies in particular to endow the satellites with devices for the accurate measurement (and suppression as necessary) of the disturbing accelerations from the external and internal environment, and for the precision pointing of the laser beam.

3.

MISSION AND SYSTEM REQUIREMENTS

The mission requirements, consolidated through the series of system studies, are summarized as follows.

The instrument set on board each satellite of the NGGM constellation shall comprise a laser interferometer, two accelerometers, a GNSS receiver and retro-reflectors for laser ranging from the ground.

The lifetime requirement shall be at least 7 years.

A time series of measurements taken at constant altitude is preferred to a variable-altitude profile. The minimum design altitude will be selected around 340 km, a value compatible with the detailed measurement of the gravity field and with the resources needed for orbit maintenance and drag compensation over the complete lifetime.

The altitude will be maintained, GOCE-like, within a band of about 100 m around a specified value which will be selected to realize a controlled longitude shift (and to control aliasing as mentioned below).

The ESA satellite pair shall be conceived as one-half of a Bender pair, providing global coverage (via the polar pair) and extended coverage of the mid-latitudes (via the inclined pair). Current orbit design values are in Table 1.

Table 1.

Orbit definition

The target spatial resolution in the gravity field mapping shall be ~100 km (half-wavelength) at mission completion.

The mission performance assumed for the Phase 0 studies, stated in terms of geoid accuracy, shall reach 1 mm accuracy, at 3 days intervals with 500 km spatial resolution and at 10 days intervals with 150 km spatial resolution. At the moment of writing, the user requirements are under review by a joint Ad-hoc Science Study Team composed by US and European representatives of the scientific community, for a univocal prioritization of the threshold and target mission requirements for observation systems flying at different altitude ranges.

The temporal resolution shall be such that atmospheric and ocean tide (AO + OT) error signals can be decoupled from signals from other Earth system constituents (ice, hydrology, oceans and solid Earth), taking into account aliasing periods. Indeed, the double satellite pair mission concept have the intrinsic potential to retrieve the full AOHIS signal in contrast to the single-pair mission, where even a tailored post-processing is not able to achieve the same performances as a double-pair mission without post-processing.

For intersatellite distances in the range 70–100 km, the NGGM performance is relatively constant and lengths >100 km do not provide any benefit in terms of gravity field recovery. The current baseline is therefore set to 100 km.

Figure 1 shows the measurement requirements (threshold and goal) of the fundamental observables of the mission: intersatellite distance variation and projection of the differential non-gravitational acceleration along the satellite-to-satellite direction. Ultra-precise accelerometers such as those that made up the gradiometer of GOCE¹¹ can provide the required non-gravitational acceleration measurement performance.

Figure 1.

System measurement performance requirements. Left: sat-sat distance variation. Right: non-gravitational relative acceleration, taking in account the in-flight lesson learnt from GOCE and accelerometers of the GOCE GRADIO-class.

The non-gravitation relative acceleration requirements in Figure 1 are derived taking in account the in-flight lesson learnt from GOCE, where the performances of the GOCE GRADIO-class²⁰ accelerometers have been modelled considering the colored noise of the Analogic-to-Digital-Converter of the capacitive detector (of the proof-mass motion) at high frequency, the estimated thermal drift below the science MBW, and the noise floor of 9.81 • 10^-12 m/s²/Hz^1/2 on the basis of the in-flight measured noise plus margin. The noise floor of the non-gravitational relative acceleration can be then lowered down to some extent, allowing a better exploitation of the laser ranging instrument (measuring precisely the inter-satellite distance variations) in the milli-Hz region, enabling for new science.

Two accelerometers symmetrically arranged around the centre of mass will be embarked on each NGGM satellite, marking a conceptual difference with respect to GOCE (where three couples of accelerometers were orthogonally mounted to form a tri-axis gradiometer) or flying mission like GRACE FO (where only one accelerometer is installed in the centre of mass). The primary objective of the accelerometer sensor suite on NGGM is to measure the satellite nongravitational acceleration in the satellite-to-satellite direction, with a low-frequency noise (below 1 mHz, where it becomes the dominant error source) possibly better than in GOCE. Moreover, the accelerometers shall provide the measurements utilized on board for the orbit and formation maintenance, the drag compensation, the control of the satellite angular accelerations and rates, the high stability pointing of the laser beam.

A new generation of electrostatic accelerometers, of the MicroSTAR class, under development at ONERA²¹ is a promising candidate for NGGM; in fact its performance can be adapted to the mission needs by playing with the following parameters:

All these improvements will be analyzed and tested (where possible) during the pre-development activities in support of the NGGM Phase A.

The driving spacecraft system requirements concern the orbit, drag-free and attitude control subsystem (DFACS) and the on-board propulsion, addressed in the next chapter. A stringent temperature stability requirement applies in the compartment enveloping the optical bench with the temperature-sensitive items, including parts of the laser equipment and the accelerometers:

4.

ORBIT AND SPACECRAFT DESIGN

As noted, the “Bender formation” emerged as the best compromise between gravity field coverage and satellite implementation. This consists of two pearl-string pairs, one in near-polar orbit (inclination between 88° and 90°) and another in medium inclination orbit (between 65° and 75°). The polar pair ensures complete geographic coverage while the combination of the two pairs provides sufficiently isotropic sampling of the gravity signal, East-West as well as North-South.

Later studies¹⁰ addressed inter-satellite distance and the orbit definition of the two satellite pairs: altitude / repeat period, inclination, nodal spacing. The basic requirements of time-variable gravity measurements from space are: (a) orbit altitude as low as possible to maximize signal strength, (b) retrieval periods as short as possible to maximize the time resolution of the gravity field solutions, and (c) ground track coverage as dense as possible within the retrieval period for maximum spatial resolution. The orbit altitude is driven by satellite engineering constraints (configuration, drag cross section, propulsion type and allowed propellant load) and a suitable range was found around 340 km circular altitude; detailed solutions to realize specific repeat patterns, which depend on the semi-major axis, were sought around that value. The study identified a group of constellations with comparable performance; besides, it was found that the quality of the retrieved time-variable gravity field model was modulated over time due to the relative drift of the nodes of the two pairs. Eventually, the orbit parameters in Table 1 were selected. Specific repeat patterns are no longer required, but the longitude shift of 1.3° per 7-day cycle (the gravity retrieval period) guarantees that, despite the differential nodal drift, the relative pattern of the two sets of ground tracks remains constant.

The attitude and the environmental disturbing accelerations will be controlled within a measurement bandwidth between 1 mHz and 100 mHz according to the requirements in Table 2. The formation of the two satellites will tend to drift under the action of differential air drag and differential accelerometer bias: dedicated controls, acting below the measurement bandwidth, will ensure that the mean semi-major axis remains within 100 m of nominal and the relative distance between 10% of the nominal intersatellite distance.

Table 2.

Orbit, drag-compensation and attitude control requirements (ASD = amplitude spectral density)

The spacecraft propulsion enacting the DFACS and orbit control functions is the main challenge of the spacecraft design. The thrust range and modulation capability imposed by the mission, coupled with the lifetime requirement, can only be satisfied by electric propulsion. Electric propulsion trades propellant mass for electric power, and ~1 kW-level power generation is a demanding task in a mission which needs to keep the drag cross section small (below 1m²) with high seasonal and orbital variation of the solar aspect angle. Moreover, high specific impulse is only available over a limited thrust range and thrust demand varies by a factor ~3 to 5 both at orbit frequency and with epoch over the 11-yr solar flux cycle. This leads to using two different thruster types for different tasks, dubbed Drag Compensation Thruster (DCT) and Fine Control Thruster (FCT), discussed in Chapter 6. Requirements have been defined for each type of thruster, in terms of thrust throttling range, specific impulse, total impulse, response time, noise and beam divergence, under ceiling requirements on total propellant mass and total power demand. Thrusters meeting these requirements have already been demonstrated ^2,15,16.

The spacecraft will be built on the heritage of platforms used in the ESA Copernicus program. The main modifications are in the solar array - to install the large panels, including deployed wings, as needed by the electric propulsion - and in the internal layout - to make room for the interferometer core near the center of mass and for the laser beam baffles which run through the spacecraft body.

Figure 2 shows the launch configuration of an NGGM pair under the VEGA-C faring at the end of Phase 0, with the two spacecraft aligned vertically on opposite sides of the launch vehicle dispenser, and the orbit configuration with deployed solar array. Four star tracker heads view through the top solar panel, in a position minimizing the distance to the accelerometers and optical bench inside the spacecraft. Four heads are needed to compensate the variation of the occultation of the sky by the Earth, as the spacecraft is rotated seasonally around roll for the solar array to follow the sun; at least two trackers are active at any given time, as in GOCE.

Figure 2.

NGGM dual launch configuration under the VEGA-C fairing, and configuration in orbit.

The internal layout is dictated by the requirement that the optical reference for the inter-satellite distance measurement by the laser interferometer shall be placed in center of mass and the two accelerometers close to and symmetric with respect to the center of mass. The temperature-sensitive parts of the payload including the accelerometer sensor heads, the optical bench assembly and the retro-reflector are accommodated in a dedicated temperature-controlled enclosure occupying the central volume of the spacecraft. The payload and service equipment boxes are accommodated on either side of the central bay, grouped according to function.

5.

LASER METROLOGY INSTRUMENT (LMI)

5.1

Laser Interferometer

In NGGM for the first time the laser interferometer will be the primary payload on a satellite specifically designed to match the laser metrology performance (i.e., the performance shall be limited by the instrument and not by a non-optimal accommodation, environmental effects such as temperature fluctuations and the dynamic effects of residual air drag).

The laser interferometer will implement the Michelson scheme in the heterodyne version (particularly suitable for measurements over very long distances) and operate with continuous wave sources at 1064 nm wavelength. For NGGM, two interferometer schemes are under evaluation.

The first scheme is inherited from the GRACE Follow-On LRI. In such an interferometer (see Figure 3) the laser beam transmitted by the follower satellite (Satellite 2) is received by the leader satellite (Satellite 1) where it is “regenerated” by a second laser source, phase-locked with a frequency offset (heterodyne frequency) to the incoming beam, and retransmitted to Satellite 2. In the “optical transponder” scheme, a source with limited optical power output (~25 mW, provided directly by the “master oscillator”) is sufficient to achieve the required signal-to-noise ratio on the photo receiver. On the other hand, two laser sources must be active simultaneously, one on each of the satellites.

Figure 3.

Functional scheme of the “optical transponder” interferometer concept (LH = Laser Head; CAV = optical reference cavity for laser frequency stabilization; ICU = Instrument Control Unit (including phasemeter); M1,2 = Mirror 1, 2; BS = Beam Splitter; P = Photo receiver; RR = retro-reflector; CoM = Center of Mass).

In the second interferometer concept, the optical transponder is replaced by a passive retro-reflector on Satellite 1 which intercepts the laser beam transmitted by Satellite 2 and back-reflects it. Here, the heterodyne frequency is generated by an acousto-optic modulator on Satellite 2, where also the two beat signals produced by the interference of the laser beams are detected by the photo receivers (the combination of the photo receiver outputs produces a sinusoidal signal whose phase is proportional to the inter-satellite distance variation).

The “retro-reflector” scheme requires a source with larger optical power output (~500 mW, provided by a fiber amplifier stage after the “master oscillator” of the same power and quality of the “transponder” case). The active part of the interferometer is located on Satellite 2 only, while a simple passive retro-reflector is sufficient on Satellite 1. With this scheme the acquisition of the optical link between the satellites is significantly simplified: it is sufficient to illuminate Satellite 1 with the laser to get the return beam and no laser frequency scan is necessary to bring the beat signal within the photo receiver bandwidth. Moreover, by replicating all the interferometer elements on both satellites, they can be made identical thus realizing a functionally fully redundant system: in case of failure of the active part on Satellite 2, the position of the two satellites along the orbit can be swapped, keeping the same orientation, and the measurement can continue with the interferometer active on Satellite 1. All these features reduce the system complexity and increase its robustness, key aspects for an operational gravity mission, and motivate the trade-off with the flight-proven “optical transponder” scheme¹⁹.

Figure 4.

Functional scheme of the “retro-reflector” interferometer concept (same nomenclature as in Figure 3).

The very low laser signal levels received after retro-reflection may cause phase errors in the presence of straylight generated by multiple reflections of the very intense laser beam traveling on the optical bench before launching to the other spacecraft. Thus, a beat signal can be originated by the interference of this straylight with the local oscillator laser signal: the amplitude is comparable to the distance measurement signal but the phase is different, not related to the distance variation between the spacecraft. Additionally, cycle slips in the phase measurement may occur due to the limited signal-to-noise ratio and limit the achievable performance. These potential problems are being addressed by a breadboard verification with representative optical setup and the expected optical power levels. The results of the ongoing tests in the setup of Figure 5 will be the subject of a later publication.

Figure 5.

Breadboard of the “retro-reflector” interferometer concept arranged for the performance tests (courtesy STI/TASI).

The objective is to achieve an intrinsic interferometer noise level of 2 nm/√Hz on the distance variation measurement, with 500 mW laser power delivered on the optical bench and 20 pW (BOL) / 4 pW (EOL) received on the photodiode after the round trip path. This objective defines the attenuation of the intensity of the laser beam required on the breadboard, a factor >10¹⁰, for the analysis of the measurement noise and of the occurrence of cycle slips when tracking a moving target (the expected relative spacecraft velocity is up to ±0.1 m/s). Having realized such conditions, the test will verify the level of straylight suppression and measure the performance of a differential wavefront sensor based on a quadrant photodiode in detecting the incoming direction of the retro-reflected beam.

The ultimate limiting factor of the performance of both interferometer concepts is the stability of the laser frequency ν: a frequency variation δν induces a distance variation measurement error δd = d·(δν/ν), where d is the distance between the satellites (nominally 100 km). Consequently, to achieve the inter-satellite distance measurement error of Figure 1, the frequency stability of the “master oscillator” shall be kept under the limits shown in Figure 6. The required stability can be achieved by locking the frequency of the “master oscillator” to the resonance of an optical cavity made from low thermal expansion material and thermally insulated. Such a frequency stabilization system is now flying on GRACE-FO.

Figure 6.

Upper limits of the frequency variation of the “master oscillator”.

5.2

Link Acquisition Metrology

Initial laser beam acquisition, its maintenance and re-acquisition of the link for the heterodyne phase measurement are critical for implementing autonomous operations of the NGGM constellations. In GRACE FO a fast steering mirror on both satellites is used to scan a relatively large field of view (about 3 mrad) and in parallel to scan the laser frequency on one of the two satellites, until the laser offset frequency is < 15 MHz with similar satellite pointing accuracies. The 5-dimensional scan (pitch and yaw angles for each spacecraft and the laser frequency sweep) is potentially rather lengthy (i.e. up to 8 hours) and complex: after that the scan patterns (a slow hexagonal pattern on the master spacecraft and a fast Lissajous pattern on the transponder satellite) are performed, the downlinked data are processed and the angular and frequency offsets are uploaded again, in order to command the instrument into re-acquisition mode. A similar 5-D scan is then performed and the laser link is finally acquired in a few minutes.

Since NGGM design offers a better satellite pointing control due to the use of proportional electric thrusters, the LMI DWS (Differential Wavefront Sensor) signal can be used as sensor, complemented by a dedicated acquisition light system: this new design reduces the number of elements in the optical path, simplifies the optical bench, reduces the electrical power demand and improve the system reliability (with two acquisition light system in cold redundancy).

In both NGGM LMI schemes, the laser beams of the laser interferometer have a far-field divergence of 100 µrad (~20 arcsec), semi-cone angle. To establish the optical link between the satellites necessary to operate the interferometer, Satellite 2 must aim the emitted laser beam in the direction of Satellite 1 with a precision <100 µrad.

Once this condition is achieved, the beam is “passively” back reflected towards the direction of Satellite 2 in the “retro-reflector” concept, whereas in the “optical transponder” concept the pointing of the return beam to the Satellite 2 must be actively performed by Satellite 1, again with a precision <100 µrad. The initial orientation of the satellites towards each other shall be within about ±3 mrad, achievable from the information provided by the attitude sensors of the two satellites and from the knowledge of their relative position calculated by an on-board orbit propagator, fed with tracking measurements from ground.

An auxiliary optical metrology, named the Acquisition and Pointing Metrology System (APMS), has been devised to achieve the required laser beam pointing starting from the ±3 mrad initial condition and avoiding long and critical searching procedures, another important aspect for an operational mission. Its task is to identify the direction of the satellite (Satellite 1 in Figure 7) towards which the laser interferometer beam emitted by Satellite 2 has to be directed, and to drive the pointing of the beam until the interferometer signal is acquired.

Figure 7.

Operating concept of the APMS.

The APMS consists of an independent laser source emitting a beam with larger divergence (>3 mrad semi-cone angle) from the satellite to be intercepted (only Satellite 1 in the “retro-reflector” concept) towards the satellite which must point the interferometer beam (Satellite 2 in the “retro-reflector” concept). The emitted beam is detected by a receiver located on Satellite 2 which drives the attitude control to orient the interferometer towards the direction of this “artificial guide star”. Figure 7 illustrates the operating concept, which can be briefly summarized as follows:

• • Laser beam NOT acquired: both spacecraft are flying in their nominal orbit, with absolute pointing uncertainty at mrad level;

• • APMS switched ON: ALS & ALD are turned on, and the ALD on Satellite 2 detects the artificial “guidestar” emitted by the other satellite. Pointing accuracy on Satellite 2 is brought down to <100 µrad, sufficient to intercept Satellite 1 with the laser beam emitted by Satellite 2;

• • LMI switched ON (Satellite 2): the DWS signal is used to fine pointing down to µrad accuracy;

• • APMS switched OFF: science mode is initiated, while Satellite 2 pointing control is performed based on the DWS signal, and satellite 1 pointing control is nominally performed by the on-board AOCS.

Therefore, the acquisition procedure is by far simplified, since 4 degrees of freedom of the initial acquisition process are eliminated with respect to GRACE FO, ensuring full on-board autonomy and fast (re-)acquisition times¹⁹.

The APMS receiver designed for detecting the faint light received from the emitter is characterized by a 0.86° field of view (semi-cone angle) and 20 mm clear aperture. It is endowed with a 2048x2048 pixel CMOS (Complementary Metal-Oxide Semiconductor) imaging sensor. 633 nm wavelength is currently assumed for the emitter, to match the maximum spectral response of the image sensor.

A flight-representative breadboard of the APMS receiver has been implemented and its performance characterized on a test set-up reproducing the illumination conditions produced by an emitter located at 100 km virtual distance (Figure 8). The test results confirm that the receiver can measure the angular direction of the emitter with accuracy <10 µrad by collecting an optical power < 1 pW while taking images with an integration time of 10 ms. Under the same conditions, the measurement error spectral density is <1 µrad/√Hz, a performance that makes this metrology device suitable not only for the initial acquisition of the optical interferometric link between the satellites, but also for the fine control of the laser beam pointing during the measurement phase of the mission. The test results confirm also that the required illumination of the receiver can be achieved using an emitter with optical output power of merely 10 mW, and far field divergence of the beam of 10 mrad (half-cone angle).

Figure 8.

Left: Breadboard of the APMS receiver on the test set-up. Right: Measurement accuracy of the beacon angular direction vs. optical power received by the detector.

A detailed qualification plan is under preparation, in order to bring both the emitter and the receiver units of the Acquisition and Pointing Metrology System to TRL 6, in parallel to the NGGM Phase A study.

6.

DRAG COMPENSATION SYSTEM

The drag compensation is implemented by a system of ion thrusters of two types, DCT (Drag Control Thruster) and FCT (Fine Control Thruster), operating in different thrust ranges.

Two 20mN-class DCT thrusters, in cold redundancy, provide the main force components for in-track (X-axis), in-band (1mHz to 100 mHz) drag compensation and orbit maintenance. Two candidate implementations compliant with the requirements exist, both of them with flight heritage, one of them being GOCE, which carried out the same tasks as envisaged in NGGM. The main difference is the thrust level: GOCE demonstrated operation in a range between 70µN and 20 mN; in NGGM, the altitude is considerably higher (340 km vs. 260 km) and a thrust range extended at the lower limit would be advantageous, even if it means reducing the upper limit. Two operating regimes have been specified for the DCT:

Ten mN-class FCT microthrusters provide the cross-track (Y) and radial (Z) drag compensation force components, and 3-axis torques for angular drag control and attitude control. The minimum operating set comprises 8 thrusters; configurations with 1 or 2 extra cold-redundant thrusters have been studied, and other options are possible including 10 operating thrusters with compliant operation in case of 1 failure and gracefully degraded operation in case of 2 failures. The specified minimum operating thrust range is between 50µN and 1 mN; reasonably efficient operation below 50µN and above 1 mN is an important asset. A number of candidate technologies exist for the FCT ^15,16; the technology readiness of the entire subsystem (as opposed to the thruster alone) will be a crucial selection factor.

The optimal configuration and operation of the FCT system has been the subject of intensive study in the last few years, continuing today. Stringent system constraints apply: propellant mass (DCT+FCT) not exceeding 100 kg; subsystem peak power (DCT+FCT) not exceeding 350W; identical Leader and Follower spacecraft with identical thruster layouts (the forces acting on the leading and trailing spacecraft are not identical); 1 (or 2) redundant thrusters must cope with the failure of any one of the nominal set of 8; total impulse per thruster not exceeding the demonstrated lifetime.

The redundancy problem is solved by means of a complex multi-stage optimization algorithm which tailors the thruster layout to the expected envelope of thrust demand, as determined by detailed simulation of the satellite operation in the drag environment. The solution is therefore dependent on the fidelity of the models (atmosphere model driven by the epoch in the solar cycle; satellite-to-atmosphere interaction driven by accommodation coefficients; surface areas; achievable thruster moment arms). The best solution achieved so far is tailored to low-to-average solar activity (as befits a mission beginning in the late 2020, a time of solar minimum) and has 4 thrusters (plus 1 redundant) on the aft side of the spacecraft and another 4 +1 on the forward side; all the aft thrusters have large components in the +X direction (counter-drag) whereas the forward thrusters mostly act perpendicular to the X-axis, for efficient torques as well as Y, Z forces. The system is tightly coupled and sensitive to the thrust range available from the FCT; a wider range improves the robustness. Because of that, the completion of the design must go hand-in-hand with the advancement of thruster technology characterization in the lab; thruster parameters such as maximum and minimum achievable thrust, specific impulse and power-to-thrust ratio as function of thrust, as well as system overhead (equipment mass, idle power of the electronics) will decide the design that will fly.

The system as studied so far is designed to be compliant with the full set of requirements of Table 2; Figure 9 and Figure 10 illustrate an example. One of the current subjects of study is the impact on the system performance of Figure 1 when one or more of the requirements of Table 2 are relaxed. The errors in the Y- and Z-axis acceleration components, for example, only enter the main measurement as projections on the acceleration along the line of sight between the satellites; the corresponding specifications could therefore be relaxed. Some further relaxation of the thrust requirements will come from the concurrent actuation of the magnetic torquers, which are included in the design but not yet in the simulations; mag torquers were already implemented in the GOCE control. Yet another area of investigation is the attitude control requirements. All of these aspects will drive the final solution, which will anyway retain, as distinguishing features, at least the in-track DCT drag compensation and the laser beam steering by orienting the satellites without continuously moving mirrors, which degrade the performance of the measurement of both the satellite-satellite distance (changing the measured optical path) and the non-gravitational accelerations (producing a microvibration background noise).

Figure 9.

Three months of simulated control data; commanded force and torque components and individual thruster outputs.

Figure 10.

Performance analysis of the example scenario simulated in Figure 9.

7.

CONCLUSIONS

ESA’s Next Generation Gravity Mission is a project to improve our knowledge and monitoring of geophysical phenomena revealed by Earth’s gravity field, in the wake of the GOCE, GRACE and GRACE-FO missions. Extensive preparation activities over more than 10 years have advanced the maturity of the system and the readiness level of the key technologies (attitude and drag control, proportional thrusters, laser optics and electronics) to high enough level for the mission to be proposed for adoption in 2022 and launch in the 2026-2028 time frame. Concurrently, in the USA, the “Decadal Strategy for Earth Observation from Space” for 2017-2027 takes continuity of mass change and gravity measurement data, begun with GRACE and continued with GRACE-FO, as one of its priorities, and encourages NASA to seek international partnership opportunities to implement the mission. In this context, a NGGM mission in cooperation with NASA is considered as the most mature candidate for a Mission of Opportunity in the current decade.

The key instrument of NGGM is a laser interferometer which shall measure the distance variation of two spacecraft separated by 100 km with the resolution of few nanometers. Two interferometer schemes suitable to meet the NGGM measurement requirements have been defined, together with the auxiliary metrology system (APMS) for optical link acquisition, being at TRL 4 and already tested with results meeting the performance requirements. The breadboard of the “off-axis retro-reflector” scheme is almost complete and is going to be tested at the time of writing: the results will provide the Agency with the necessary information to carry out the technology predevelopments in parallel to the foreseen NGGM Phase A.

A discussion on the improvements to be carried on state-of-art electrostatic accelerometers has been presented, together with the current status of the drag compensation system and its performance. Alternative drag compensation scenarios are under investigation, where lateral and angular drag compensation requirements can be relaxed (together with a progressive relaxation of the thruster requirements), while retaining the inline drag compensation and orienting the satellites – without moving mirrors - for maintaining the inter-satellite laser link: these scenarios will be further analysed and in-depth assessed in the on-going pre-Phase A activities.