Swarm (Geomagnetic LEO Constellation)
Swarm is a minisatellite constellation mission within the Earth Explorer Opportunity Program of ESA, proposed under the lead of DNSC (Danish National Space Center) of Copenhagen, Denmark (formerly DSRI). In January 2007, DNSC became DTU Space, an institute at the Technical University of Denmark. The Swarm mission will be the 4th mission in ESA's Earth Explorer Program, following GOCE, SMOS, and CryoSat-2.
The first mission to ever map the Earth's magnetic field vector at LEO was the NASA MagSat spacecraft (launch Oct. 30 1979). Due to the low perigee (perigee=350 km, apogee=551 km), MagSat remained in orbit for only seven and a half months until June 11, 1980. About 20 years later, the Danish Ørsted micro satellite (1999-), the German CHAMP (2000-), the Argentine SAC-C (2000-) have been designed specifically for mapping the LEO magnetic field. Common to these recent missions is the magnetometry package, which utilizes a vector field magnetometer co-mounted with a star tracker (2 in the case of CHAMP) on an optical bench. As the accuracy of the instrument package has constantly increased, as well as the modelling methods have been improved towards optimized signal decomposition, it has been realized that simultaneous data from several points in space is needed, if the ultimate modelling barrier, the spatial-temporal ambiguity, has to be broken.
The overall objective of the Swarm mission is to build on the Ørsted and CHAMP mission experiences and to provide the best ever survey of the geomagnetic field (multi-point measurements) and its temporal evolution, to gain new insights into the Earth system by improving our understanding of the Earth's interior and climate. 1) 2) 3) 4) 5) 6) 7) 8) 9) 10) 11) 12) 13) 14) 15) 16) 17) 18) 19)
This will be done by a constellation of three satellites, two will fly at a lower altitude, measuring the East-West gradient of the magnetic field, and one satellite will fly at a higher altitude in a different local time sector. Other measurements will also be made to complement the magnetic field measurements. Together these multipoint measurements will allow the deduction of information on a series of solid-Earth processes responsible for the creation of the fields measured.
Background on the discovery of electromagnetism:
The history of magnetic discovery goes back to about 110 B.C., when the earliest magnetic compass was invented by the Chinese. They noticed hat if a "lodestone" (natural magnets of iron-rich ore) was suspended so it could turn freely, it would always point in the same direction, toward the magnetic poles. This directional pointing property of magnetic material was eventually introduced into the making of an early compass and used for maritime navigation . By the 13th century, the directive property of magnetism was widely recognized and used in navigation. The mariner's magnetic compass is the first technological application of magnetism and, one of the oldest scientific instruments.
Until 1820, the only magnetism known was that of iron magnets and of lodestones. It was the Danish physicist Hans Christian Ørsted, professor at the University of Copenhagen, who, in 1820, was first to discover the relationship between the hitherto separate fields of electricity and magnetism. Ørsted showed that a compass needle was deflected when an electric current passed through a wire, before Faraday had formulated the physical law that carries his name: the magnetic field produced is proportional to the intensity of the current. Magnetostatics is the study of static magnetic fields, i.e. fields which do not vary with time. 20) 21)
Magnetic and electric fields together form the two components of electromagnetism. Electromagnetic waves can move freely through space, and also through most materials at pretty much every frequency band (radio waves, microwaves, infrared, visible light, ultraviolet light, X-rays and gamma rays). Electromagnetic fields therefore combine electric and magnetic force fields that may be natural (the Earth's magnetic field) or man-made (low frequencies such as electric power transmission lines and cables, or higher frequencies such as radio waves (including cell phones) or television (Ref. 22).
Figure 1: The early history of electromagnetic discovery made by scientists throughout the centuries (image credit: ESA, Ref. 16)
Background on the Earth's magnetic field:
The Earth has its own magnetic field, which acts like a giant magnet. Geomagnetism is the name given to the study of this field, which can be roughly described as a centered dipole whose axis is offset from the Earth's axis of rotation by an angle of about 11.5º. This angle varies over time in response to movements in the Earth's core. The angle between the direction of the magnetic and geographic north poles, called the magnetic declination, varies at different points on the Earth's surface. The angle that the magnetic field vector makes with the horizontal plane at any point on the Earth's surface is called the magnetic inclination.
This centered dipole exhibits magnetic field lines that run between the north and south poles. These field lines convergent and lie vertical to the Earth's surface at two points known as the magnetic poles, which are currently located in Canada and Adélie Land. Compass needles align themselves with the magnetic north pole (which corresponds to the south pole of the 'magnet' at the Earth's core).
The Earth's magnetic field is a result of the dynamo effect generated by movements in the planet's core, and is fairly weak at around 0.5 gauss, i.e. 5 x 10-5 tesla (this is the value in Paris, for example). The magnetic north pole actually 'wanders' over the surface of the Earth, changing its location by up to tens of km every year. Despite its weakness, the Earth's dipolar field nevertheless screen the Earth from charged particles and protect all life on the planet from the harmful effects of cosmic radiation. In common with other planets in our solar system, the Earth is surrounded by a magnetosphere that shields its surface from solar wind, although this solar wind does manage to distort the Earth's magnetic field lines.
The Earth's magnetic field shows deviations, called anomalies, from the idealized field of a centered bar magnet. These anomalies can be quite large, affecting areas on a regional scale. One example is the SAA (South Atlantic Anomaly), which affects the amount of cosmic radiation reaching the passengers and crew of any plane and spacecraft led to cross it (Ref. 22).
Figure 2: Artist's view of solar wind interacting with Earth's magnetic field (image credit: DTU Space)
Figure 3: Schematic view of the geomagnetic field, produced mainly by a self-sustaining dynamo in the outer fluid core (image credit: GFZ)
Figure 4: Map of the geomagnetic field strength at the surface of the Earth derived from the model produced using data from the Oersted satellite (image credit: LETI) 22)
Figure 5: Magnetic field contributions (image credit: ESA) 23)
The primary research topics to be addressed by the Swarm mission include: 24)
• Core dynamics, geodynamo processes, and coremantle interaction. - The goal is to improve the models of the core field dynamics by ensuring long-term space observations with an even better spatial and temporal resolution. Combining existing Ørsted, CHAMP and future Swarm observations will also more generally allow the investigation of all magnetohydrodynamic phenomena potentially affecting the core on sub-annual to decadal scales, down to wavelengths of about 2000 km. Of particular interest are those phenomena responsible for field changes that cannot be accounted for by core surface flow models. 25)
• Lithospheric magnetization and its geological interpretation. - The increased resolution of the Swarm satellite constellation will allow, for the first time, the identification from satellite altitude of the oceanic magnetic stripes corresponding to periods of reversing magnetic polarity. Such a global mapping is important because the sparse data coverage in the southern oceans has been a severe limitation regarding our understanding of plate tectonics in the oceanic lithosphere. Another important implication of improved resolution of the lithospheric magnetic field is the possibility to derive global maps of heat flux.26) 27)
• 3-D electrical conductivity of the mantle. - Our knowledge of the physical and chemical properties of the mantle can be significantly improved if we know its electrical conductivity. Due to the sparse and inhomogeneous distribution of geomagnetic observatories, with only few in oceanic regions, a true global picture of mantle conductivity can only be obtained from space.
• Currents flowing in the magnetosphere and ionosphere. - Simultaneous measurements at different altitudes and local times, as foreseen with the Swarm mission, will allow better separation of internal and external sources, thereby improving geomagnetic field models. In addition to the benefit of internal field research, a better description of the external magnetic field contributions is of direct interest to the science community, in particular for space weather research and applications. The local time distribution of simultaneous data will foster the development of new methods of co-estimating the internal and external contributions.
The secondary research objectives include:
• Identification of the ocean circulation by its magnetic signature. - Moving sea-water produces a magnetic field, the signature of which contributes to the magnetic field at satellite altitude. Based on state-of-the-art ocean circulation and conductivity models it has been demonstrated that the expected field amplitudes are well within the resolution of the Swarm satellites. 28)
• Quantification of the magnetic forcing of the upper atmosphere. - The geomagnetic field exerts a direct control on the dynamics of the ionized and neutral particles in the upper atmosphere, which may even have some influence on the lower atmosphere. With the dedicated set of instruments, each of the Swarm satellites will be able to acquire high-resolution and simultaneous in-situ measurements of the interacting fields and particles, which are the key to understanding the system.
Historic background of Swarm: Ref. 14)
• The first Swarm proposal was made in 1998, prior to launch of the Ørsted mission.
• In early 2002, the Swarm mission was proposed to ESA by Eigil Friis-Christensen of DNSC (Copenhagen, Denmark), Hermann Lühr of GFZ (GeoForschungszentrum, Potsdam, Germany), and Gauthier Hulot of IPG (Institut de Physique du Globe, Paris, France) with support from scientists in seven European countries and the USA. In the meantime, the Swarm team comprises participation of 27 institutes on a global scale. The mission was selected for feasibility studies in 2002. The initial mission proposal considered a Swarm constellation of 4 spacecraft. 29)
• In May 2002 there were three mission candidates: ACE+, EGPM and Swarm; they were chosen for a feasibility study.
• At the end of two parallel feasibility studies, the Swarm mission was selected as the 5th mission in ESA's Earth Explorer Program in May 2004. Phase A was completed in Nov. 2005, resulting in a constellation of 3 spacecraft.
• New Concept – Constellation to characterize external sources:
- The external contributions are highly influenced by solar activity and local time
- Simultaneous satellites in different orbital planes are necessary in order to overcome the time-space ambiguity in the measurements. The optimum constellation depends on the scientific objectives.
- But, measurements of high accuracy are not sufficient! A better understanding of the various sources is equally important, in particular when doing measurements with unprecedented precision, where new phenomena appear in the data. For this, additional and independent key information is needed: a) electric field, b) ionospheric conductivity.
• In 2006, the Swarm project was in Phase B, ending with the PDR (Preliminary Design Review) in the summer 2007.
The construction of the Swarm constellation commenced in November 2007 with the Phase C/D kick-off meeting. The Swarm project CDR (Critical Design Review) took place on Oct. 14, 2008 at ESA/ESTEC. 30)
Legend to Figure 6: The magnetic field and electric currents near Earth generate complex forces that have immeasurable impact on our everyday lives. Although we know that the magnetic field originates from several sources, exactly how it is generated and why it changes is not yet fully understood. ESA's Swarm mission will help untangle the complexities of the field.
Space segment concept:
The Swarm mission architecture is driven by the requirement for separation of the various sources contributing to the Earth's magnetic field. Hence, the space segment concept employs a three-minisatellite constellation with the following characteristics:
- Three spacecraft in two different orbital planes, with two satellites in a plane of 84.7º inclination and with one satellite in a plane of 88º inclination
- The two satellites in the 87.4º inclination orbit will fly at a mean altitude of 450 km, their east-west separation will be 1-1.5º, and the maximum differential delay in orbit will be about 10 s.
- The satellite in the higher inclination orbit (88º) will fly at a mean altitude of 530 km.
In November 2005, ESA selected EADS Astrium GmbH, Friedrichshafen, Germany as the prime contractor for the Swarm spacecrafts. The Swarm consortium (main subcontractors) consists of: 34)
- EADS Astrium Ltd., UK (mechanical, thermal, AIV)
- GFZ Potsdam, Germany (end-to-end system simulator, calibration & validation)
- DTU Space, Copenhagen, Denmark [level 1b processor and instruments (VFM magnetometer and STR star tracker)]
The spacecraft design is governed by the following requirements:
1) Magnetic cleanliness: magnetometers on deployable boom, non-magnetic materials and caution during handling
2) Magnetic field vector attitude knowledge: ultra-stable connection between VFM (Vector Field Magnetometer) and STR (Star Tracker) assembly on the optical bench
3) Ballistic coefficient: small ram surface in flight direction to minimize air drag
4) Accelerometer proof-mass vs satellite CoG (Center of Gravity) location.
Figure 7: Artist's rendition of the Swarm constellation (image credit: ESA)
An important design measure is the accommodation of the magnetometer package at a distance from the main body/platform sufficient to minimize any magnetic disturbance. A boom ensures a magnetically 'clean' environment and provides very stable accommodation for the magnetometer package. Due to envelope constraints of the launcher fairing, the boom must be deployable. 35)
Optical bench: The vector magnetometer is mounted on an ultra-stable silicon carbide-carbon fiber compound structure (the SWARM optical bench). Both optical bench and scalar magnetometer are installed on a deployable conical tube of square cross section. The position tolerance of the optical bench to its tube interface has to be fixed within 0.2 mm. 36) 37)
The design driver of the composite tube assembly of Swarm is thermal stability. The main cause for observed thermal distortion is the non-uniformity of the cross-sections arising from the different adaptations of the filament winding process in order to manufacture the carbon fiber reinforced structure. The manufacture of the structure required use of thermally controlled high precision bonding jigs to join the composite tubes to the metallic fittings.
The scalar magnetometer and optical bench are fixed to a deployable large beam of square cross section, the SWARM (Carbon-fiber Tube Assembly (CTA) which fulfils the following main functions (Figure 8):
• Separate the sensitive instruments from the spacecraft to comply with the very high magnetic cleanliness requirements
• Provide a suitable stable structure for the fixation of instruments.
The chosen manufacturing technology for the SWARM tube was filament winding. The SWARM tube has a conical taper. Since the amount of fibers in a cross section is constant the tube had two main characteristics: the wall thickness increased linearly from the root to the tip and due to nature of the winding process the fiber angle became steeper at the tip than at the root. The overall effect is a variation of properties along the length of the tube.
The Swarm carbon-fiber tube assembly was subjected to various tests: Thermal distortion was measured by establishing a 65ºC gradient between the tip and the hinge and a 10ºC gradient between opposite sides of the CTA. The hole pattern of the optical bench was accurate to within 0.2 mm (Ref. 36).
Figure 8: The Swarm optical bench, carbon-fiber tube assembly (image credit: RUAG)
Figure 9: Configuration and performance requirements of a Swarm spacecraft (image credit: EADS Astrium)
Figure 10: Configuration details of a Swarm spacecraft (image credit: EADS Astrium)
The three identical Swarm minisatellites consist of the payload and the platform elements. The platform comprises the following subsystems: structure/mechanisms, power, RF communications, AOCS (Attitude and Orbit Control Subsystem), thermal control, and onboard data handling.
The AOCS design is based to a maximum extent on the CryoSat AOCS design of EADS Astrium. The gyro-less AOCS provides 3-axis stabilization with an Earth pointing attitude control in all modes. The requirements call for: 38)
- An attitude pointing control within a band of < 5º about all axis (roll, pitch, and yaw), the pointing stability is < 0.1º/s
- Provision of a sufficient torque capability for launcher tip-off rate damping and attitude acquisition
- Minimize acceleration and magnetic stray field disturbances to scientific instruments
- Provision of a high ΔV capability for orbit & attitude control maneuvers.
Figure 11: Functional architecture of AOCS (image credit: EADS Astrium)
The AOCS is tightly coupled with the propulsion subsystem. Actuation is provided by a cold gas propulsion subsystem, referred to as OCS (Orbit Control Subsystem), and magnetic torquers (used for ΔV maneuvers and to complement the magnetic torquers). The cold gas propulsion system is provided by AMPAC-ISP, UK. - Attitude sensing is provided by a star tracker assembly (3 star tracker heads), 3 magnetometers, and a CESS (Coarse Earth and Sun Sensor) assembly used in safe mode situations and in initial acquisition sequences, respectively (CESS is of CHAMP, GRACE, and TerraSAR-X heritage). A dual frequency GPS receiver (GPSR) is used to provide PPS (Precise Positioning Service) to the OBC and instruments for on-board datation.
Note: the star tracker (STR) assembly and optical bench are described below under a separate heading.
The nominal attitude has a nadir orientation. Rotation maneuvers of S/C about roll, pitch and yaw are used for instrument calibration and orbit Control. The safe mode is Earth-oriented. Pointing requirements are 2º about all axes, with limitations on use of actuators.
The Swarm rate damping design, in support of the critical spacecraft deployment phase, employs magnetic rate damping - magnetometers in combination with magnetic torquers and thrusters - to provide a significantly cheaper implementation than with the use of gyroscopes. From a control theory point-of-view, rate damping with magnetometers using 2-axis measurement is as "safe" as with gyroscopes using 3-axis measurement: Global asymptotical stability is achieved except for the case when the magnetic field does not change. This is only in near-equator orbits possible with perfect field symmetry which is in practice not realistic. The result is confirmed by the evaluation of the observability criterion where no loss of this property could be detected except for the mentioned case. Since SWARM is in a polar inclination orbit, the control concept is considered "clean". 39)
Rate damping design: The RDM controller is a simple proportional controller on the S/C rate with reference rate zero. The S/C rate is computed by processing and derivation of the FGM measurements. The controller outputs the torque commands for the torquer and the thruster. A dead band for the thruster inhibits the thruster activation for low rates which can be covered by the torque rod.
Figure 12: Schematic of the Swarm RDM controller (image credit: EADS Astrium)
Each spacecraft features 2 propellant tanks, each with a capacity of 30 kg of N2. The thrusters provide thrust levels of 20 and 40 mN. The cold gas thruster system was developed and space qualified by Ampec-ISP, Cheltenham, UK consisting of 24 OCT (Orbit Control Trusters) and 48 ACT (Attitude Control Thrusters) for the Swarm constellation. The assembly and test of Ampac's SVT01 series of cold gas thrusters has included design modifications, full qualification and verification of suitability to operate with a new propellant. In 2010, a set of 72 units has been supplied and integrated into the constellation of three Swarm spacecraft. 40)
A GPS receiver provides the functions of timing and position determination. The spacecraft dry mass is about 370 kg.
Figure 13: Software architecture of AOCS (image credit: EADS Astrium)
EPS (Electrical Power Subsystem): The two body-mounted solar arrays and the varying orbits of the satellites require a MPPT (Maximum Power Point Tracking) system. Important requirements are related to the magnetic cleanliness of the satellites and result in following specific PCDU (Power Conditioning and Distribution Unit) design requirements: 41)
- Minimization of magnetic moment i.e. minimizing of magnetic materials and current loops
- Selection of switching frequencies outside the ‘forbidden' frequency ranges
- Minimizing spacecraft surface charging by use of negative bus voltage concept (battery + is connected to spacecraft structure).
The PCU part of the PCDU covers all tasks to control the power flow in the unit from the different sources and performs the communication with the OBC (On Board Computer).
The main power requirements for the PCDU are defined as follows:
- Solar array input: 0 to -125 V, max. 21 A (each of 2 panels)
- Maximum power per panel: 750 W
- Main bus voltage range -22 V to -34 V
- Maximum battery charge current 24 A
- Continuous discharge current 0 to 14 A.
Maximum discharge current/power up to 0.5 h: 20 A / 440 W.
Figure 14: Architecture of the PCDU (image credit: EADS Astrium)
Negative bus voltage concept: The Swarm satellite requires the positive line of the power system connected to structure. This implies that all bus protection functions have to be allocated in the ‘hot' negative line. As all essential functions, ( i.e. bus voltage control) need to be independent from the auxiliary supplies, they have to be supplied by the negative bus voltage. Figure 15 shows a principle grounding/power supply diagram of the main functional blocks in the PCDU.
Power control concept: The PCDU uses a simple concept for control of the battery state of charge and the bus voltage:
- Whenever the bus voltage and the charge current are below the limits, the MPPTs are active
- When the either the bus voltage attains the ‘battery end-of-charge voltage or the battery attains the charge current limit, the MEA (Main Error Amplifier) supersedes the tracker operation.
A bus overvoltage detection logic has been implemented in the PCDU, which performs a rapid ramp-down of the solar regulator current by using hysteresis control.
The MEA is composed of 3 identical separated control stages and a majority voter. Each control stage has a dedicated set of sensors and receives the relevant set commands for the bus voltage via redundant internal control busses. The charge current limitation is implemented in a ‘cascade configuration', using the output of the current error amplifier as a set signal for the voltage amplifier. This assures a low and constant bus impedance during all MEA control modes. The implementation of the regulation concept is given in Figure 16.
Figure 15: Grounding scheme of the Swarm PCDU (image credit: EADS Astrium)
Figure 16: Schematic of the bus control concept (image credit: EADS Astrium)
Table 1: Overview of spacecraft parameters
RF communications: S-band for TT&C spacecraft monitoring services and for science data transmission.
Figure 17: Artist's rendition of the Swarm constellation in orbit (image credit: EADS Astrium)
Figure 18: Photo of the three Swarm satellites at the launch site (image credit: ESA)
Figure 19: The Swarm satellites separated by a few centimeters (image credit: ESA, M. Shafiq)
Legend to Figure 19: Attached to the tailor-made launch adapter, the three Swarm satellites sit just centimeters apart. This novel part of the rocket keeps the satellites upright within the fairing during launch and allows them to be injected simultaneously into orbit. 42)
DBA (Deployable Boom Assembly): The Swarm DBA, consisting of a 4.3m long CFRP tube and a hinge assembly, is designed to perform this function by deploying the CFRP tube plus the instruments mounted on it. mounted on a 4.3m long deployable boom. Deployment is initiated by releasing 3 HDRMs (Hold Down Release Mechanisms) , once released the boom oscillates back and forth on a pair of pivots, similar to a restaurant kitchen door hinge, for around 120 seconds before coming to rest on 3 kinematic mounts which are used to provide an accurate reference location in the deployed position. The motion of the boom is damped through a combination of friction, spring hysteresis and flexing of the 120+ cables crossing the hinge. Considerable development work and accurate numerical modelling of the hinge motion was required to predict performance across a wide temperature range and ensure that during the 1st overshoot the boom did not damage itself, the harness or the spacecraft. - Due to the magnetic cleanliness requirements of the spacecraft, no magnetic materials could be used in the design of the hardware. 43)
Launch: The Swarm constellation was launched on Nov. 22, 2013 (12:02:29 UTC) on a Rockot vehicle from the Plesetsk Cosmodrome, Russia. The launch was provided by Eurockot Launch Services. Some 91 minutes after liftoff, the Breeze-KM upper stage released the three satellites into a near-polar circular orbit at an altitude of 490 km. 44) 45) 46) 47) 48)
The launch was planned for the fall of 2012, but due to the recent Breeze-M (Briz-M) failure the launch was postponed to permit proper investigations of the cause. In Nov. 2012, ESA is still expecting, from the Russian Ministry of Defence, the launch manifest for the year 2012/13 for Rockot launchers indicating the launch date for Swarm. 49) 50)
Note: Rockot, a converted SS-19 ballistic missile, has been grounded since February 1, 2011 when the Rockot vehicle with the Breeze-KM upper stage failed to place the Russian government's GEO-IK2 geodesy satellite of 1400 kg (Kosmos 2470) into its intended orbit of 1000 km. However, in the meantime, the Rockot/Breeze-KM vehicle demonstrated its reliability by lifting 4 Russian spacecraft (Gonets-M No.3, Gonets-M No.4, Strela-3/Rodnik, and Yubileiny-2/MiR) successfully into orbit on July 28, 2012.
On April 9, 2010, ESA awarded a contract to Eurockot, for the launch of two of its Earth observation missions. The contract covers the launch of ESA's Swarm magnetic-field mission and a 'ticket' for one other mission, yet to be decided. Both will take place from the Plesetsk Cosmodrome in northern Russia using a Rockot/Breeze-KM launcher. Eurockot is based in Bremen, Germany and is a joint venture between Astrium and the Khrunichev Space Center, Moscow. 51) 52) 53) 54)
After release from a single launcher, a side-by-side flying lower pair of satellites at an initial altitude of 460 km and a single higher satellite at 530 km will form the Swarm constellation. The constellation deployment and maintenance require a total ΔV effort of about 100 m/s.
In LEOP (Launch and Early Orbit Phase), at least three ground stations will be involved. LEOP is expected to last 3 days for the full activation of the satellites, followed by an orbit acquisition phase of up to three months. In parallel with the orbit acquisition phase, the commissioning phase will start in order to check out all satellite subsystems and the payload. The commissioning phase is currently expected to last three months. After the commissioning phase the nominal mission phase of 4 year starts.
Orbits of the constellation:
Accurate determination and separation of the large-scale magnetospheric field, which is essential for better separation of core and lithospheric fields, and for induction studies, requires that the orbital planes of the spacecraft are separated by 3 to 9 hours in local time. For improving the resolution of lithospheric magnetization mapping, the satellites should fly at low altitudes - thus experiencing some drag, but commensurate with the goals of a multi-year mission lifetime. The three satellites are being flown in 3 orbital planes with 2 different near-polar inclinations to provide a mutual orbital drift over time (Figure 20 and 21).
• Two satellites (Swarm A+B) are in a similar plane of 87.4º inclination. The satellite pair of 87.4º inclination will fly at a mean altitude of 450 km, their east-west separation shall be 1-1.4º, and the maximal differential delay in orbit shall be about 10 s. The formation-flying aspects concern the satellite pair, a side-by-side formation, requiring some formation maintenance.
• One higher orbit satellite (Swarm C) in a circular orbit with 88º inclination at an initial altitude of 530 km. The right ascension of the ascending node is drifting somewhat slower than the two other satellites, thus building up a difference of 9 hours in local time after 4 years.
Note: Due to ASM instrument problems on Swarm-C (Charlie), it was decided prior to launch to place Charlie with Alpha on the lower orbit, and Bravo on the higher orbit.
Table 2: Overview of the Swarm orbit configuration
Figure 20: Orbit altitude projection over mission time (left); Local time evolution of the S/C in two orbital planes (right), image credit: DTU Space
Figure 22: Polar projection of the Swarm orbit configuration over time (image credit: ESA)
Figure 23: The Swarm tandem pair provides a stereo view (image credit: DTU Space)
• December 19, 2016: We would normally associate jet streams with the weather but, thanks to ESA's magnetic field mission, scientists have discovered a jet stream deep below Earth's surface – and it's speeding up. Launched in 2013, the trio of Swarm satellites are measuring and untangling the different magnetic fields that stem from Earth's core, mantle, crust, oceans, ionosphere and magnetosphere. Together, these signals form the magnetic field that protects us from cosmic radiation and charged particles that stream towards Earth in solar winds. 57)
- Measuring the magnetic field is one of the few ways we can look deep inside our planet. As Chris Finlay from the Technical University of Denmark noted, "We know more about the Sun than Earth's core because the Sun is not hidden from us by 3000 km of rock." The field exists because of an ocean of superheated, swirling liquid iron that makes up the outer core. Like a spinning conductor in a bicycle dynamo, this moving iron creates electrical currents, which in turn generate our continuously changing magnetic field. Tracking changes in the magnetic field can, therefore, tell researchers how the iron in the core moves.
Figure 24: ESA's Swarm satellites have led the discovery of a jet stream in the liquid iron part of Earth's core 3000 km beneath the surface. In addition, Swarm satellite data show that this jet stream is speeding up (image credit: ESA)
- The accurate measurements by the unique constellation of Swarm satellites allow the different sources of magnetism to be separated, making the contribution from the core much clearer. A paper published today in Nature Geoscience describes how Swarm's measurements have led to the discovery of a jet stream in the core. 58)
- Phil Livermore from the University of Leeds in the UK and lead author of the paper said, "Thanks to the mission we have gained new insights into the dynamics of Earth's core and it's the first time this jet stream has been seen, and not only that – we also understand why it's there." One feature is a pattern of ‘flux patches' in the northern hemisphere, mostly under Alaska and Siberia. "These high-latitude flux patches are like bright spots in the magnetic field and they make it easy to see changes in the field," explained Dr Livermore.
- Swarm reveals that these changes are actually a jet stream moving at more than 40 km a year – three times faster than typical outer-core speeds and hundreds of thousands of times faster than Earth's tectonic plates move. "We can explain it as acceleration in a band of core fluid circling the pole, like the jet stream in the atmosphere," said Dr Livermore.
- So, what is causing the jet stream and why is it speeding up so quickly? The jet flows along a boundary between two different regions in the core. When material in the liquid core moves towards this boundary from both sides, the converging liquid is squeezed out sideways, forming the jet. "Of course, you need a force to move the fluid towards the boundary," says Prof. Rainer Hollerbach, also from the University of Leeds. "This could be provided by buoyancy, or perhaps more likely from changes in the magnetic field within the core." As for what happens next, the Swarm team is watching and waiting.
- Rune Floberghagen, ESA's Swarm mission manager, added, "Further surprises are likely. The magnetic field is forever changing, and this could even make the jet stream switch direction.This feature is one of the first deep-Earth discoveries made possible by Swarm. With the unprecedented resolution now possible, it's a very exciting time – we simply don't know what we'll discover next about our planet."
Figure 25: Earth's magnetic field is thought to be generated largely by an ocean of superheated, swirling liquid iron that makes up Earth's outer core 3000 km under our feet. Acting like the spinning conductor in a bicycle dynamo, it generates electric currents and thus the continuously changing electromagnetic field (image credit: ESA/AOES Medialab)
• October 28, 2016: Satellite engineers have been puzzling over why GPS navigation systems on low-orbiting satellites like ESA's Swarm constellation, sometimes black out when they fly over the equator between Africa and South America. Thanks to Swarm, it appears ‘thunderstorms' in the ionosphere are to blame. 59) 60)
- Launched in 2013, the Swarm trio is measuring and untangling the different magnetic fields that stem from Earth's core, mantle, crust, oceans, ionosphere and magnetosphere – an undertaking of at least four years.
- As with many satellites, ESA's three Swarm satellites carry GPS receivers as part of their positioning system so that operators keep them in the correct orbits. In addition, GPS pinpoints where the satellites are making their scientific measurements. — However, sometimes the satellites lose their GPS connection. In fact, during their first two years in orbit, the link was broken 166 times.
- The new study (Ref. 60) connects these losses of GPS signal to EPIs (Equatorial Plasma Irregularities), which occur when the electron density in the ionosphere's F region undergoes large, rapid changes. EPIs mostly occur close to Earth's magnetic equator, which might explain why the loss of GPS signal to Swarm satellites also occurred most frequently when the satellites were at low latitudes.
- Nearly all of the 166 loss of GPS signal events adhered to this pattern, and the researchers confirmed that the 161 low-latitude events coincided with EPIs that caused significant depletion of the ionospheric electron density. The other five events at high latitudes corresponded to polar patches or increased geomagnetic activity. In addition, when a Swarm satellite flew through an area of strong EPIs, it experienced loss of GPS signal for at least one channel about 95% of the time. These results indicate that EPIs play a critical role in causing loss of GPS signal for satellites in low Earth orbit.
- The authors noted other distinct patterns. Swarm B, for example, underwent fewer GPS signal loss events than Swarm A or C: Swarm B's orbit keeps the satellite 50 km higher above Earth than Swarm A and C, which means it inhabits an area with lower electron density. The authors indicate that the lower background density limits the magnitude of local density depletions, protecting the satellite from the most intense EPIs.
- This finding suggests that satellites flying at higher altitudes could be at lower risk for loss of GPS signal. Also, the bandwidth of the Swarm satellites' GPS receivers was updated in 2015, which may help decrease issues with GPS signal in the future. With further study and more data over time, researchers may soon determine how best to limit loss of GPS signal, whether that be through further changes to the GPS receivers or adjusted orbits.
Figure 26: The red dots in the map show where the Swarm-C satellite lost its GPS connection between launch in November 2013 and March 2015. These losses in tracking signal were down to equatorial plasma bubbles. The green line denotes the geomagnetic equator (image credit: NASA blue marble/GFZ Potsdam/ESA)
Claudia Stolle from the GFZ research center in Potsdam, Germany said, "Ionospheric thunderstorms are well known, but now we have been able to show a direct link between these storms and the loss of connection to GPS. "This is thanks to Swarm because it is the first time that high-resolution GPS and ionospheric patterns can be detected from the same satellite."
These thunderstorms occur when the number of electrons in the ionosphere undergoes large and rapid changes. This tends to happen close to Earth's magnetic equator and typically just for a couple of hours between sunset and midnight. As its name suggests, the ionosphere is where atoms are broken up by sunlight, which leads to free electrons. A thunderstorm scatters these free electrons, creating small bubbles with little or no ionized material. These bubbles disturb the GPS signals so that the Swarm GPS receivers can lose track.
Resolving the mystery of blackouts is not only good news for Swarm, but also for other low-orbiting satellites experiencing the same problem. It means that engineers can use this new knowledge to improve future GPS systems to limit signal losses.
Christian Siemes, who works at ESA on the mission, said, "In light of this new knowledge, we have been able to tune the Swarm GPS receivers so they are more robust, resulting in fewer blackouts. Importantly, we are able to measure variations in the GPS signal which is not only interesting for engineers developing GPS instruments, but also interesting to advance our scientific understanding of upper-atmosphere dynamics."
ESA's Swarm mission manager, Rune Floberghagen, added, "What we see here is a striking example of a technical challenge being turned into exciting science, a true essence of an Earth Explorer mission such as Swarm. These new findings demonstrate that GPS can be used as a tool for understanding dynamics in the ionosphere related to solar activity. Perhaps one day we will also be able to link these ionospheric thunderstorms with the lightning we see from the ground" (Ref. 59).
• October 3, 2016: Oceans might not be thought of as magnetic, but they make a tiny contribution to our planet's protective magnetic shield. Remarkably, ESA's Swarm satellites have not only measured this extremely faint field, but have also led to new discoveries about the electrical nature of inner Earth (Figure 27). 61) 62)
- The magnetic field shields us from cosmic radiation and charged particles that bombard Earth from the Sun. Without it, the atmosphere as we know it would not exist, rendering life virtually impossible. Scientists need to learn more about our protective field to understand many natural processes, from those occurring deep inside the planet, to weather in space caused by solar activity. This information will then yield a better understanding of why Earth's magnetic field is weakening. Although we know that the magnetic field originates in different parts of Earth and that each source generates magnetism of different strengths, exactly how it is generated and why it changes is not fully understood.
- While the mission is already shedding new light on how the field is changing, this latest result focuses on the most elusive source of magnetism: ocean tides. When salty ocean water flows through the magnetic field, an electric current is generated and this, in turn, induces a magnetic response in the deep region below Earth's crust – the mantle. Because this response is such a small portion of the overall field, it was always going to be a challenge to measure it from space.
- Last year, scientists from the Swiss Federal Institute of Technology, ETH Zurich, showed that if it could be measured from space – never done before – it should also tell us something about Earth's interior. However, this all remained a theory – until now. 63)
- Thanks to Swarm's precise measurements along with those from Champ – a mission that ended in 2010 after measuring Earth's gravity and magnetic fields for more than 10 years – scientists have not only been able to find the magnetic field generated by ocean tides but, remarkably, they have used this new information to image the electrical nature of Earth's upper mantle 250 km below the ocean floor.
- Alexander Grayver, from ETH Zurich, said, "The Swarm and Champ satellites have allowed us to distinguish between the rigid ocean ‘lithosphere' and the more pliable ‘asthenosphere' underneath." The lithosphere is the rigid outer part of the earth, consisting of the crust and upper mantle, while the asthenosphere lies just below the lithosphere and is hotter and more fluid than the lithosphere. "Effectively, ‘geo-electric sounding from space', this result is a first for space exploration," he continues. "These new results are important for understanding plate tectonics, the theory of which argues that Earth's lithosphere consists of rigid plates that glide on the hotter and less rigid asthenosphere that serves as a lubricant, enabling plate motion."
- Roger Haagmans, ESA's Swarm mission scientist, explained, "It's astonishing that the team has been able to use just two years' worth of measurements from Swarm to determine the magnetic tidal effect from the ocean and to see how conductivity changes in the lithosphere and upper mantle. Their work shows that down to about 350 km below the surface, the degree to which material conducts electric currents is related to composition. In addition, their analysis shows a clear dependence on the tectonic setting of the ocean plate. These new results also indicate that, in the future, we could get a full 3D view of conductivity below the ocean."
- Rune Floberghagen, ESA's Swarm mission manager, added, "We have very few ways of probing deep into the structure of our planet, but Swarm is making extremely valuable contributions to understanding Earth's interior, which then adds to our knowledge of how Earth works as a whole system."
Figure 27: Magnetic field sources: The different sources that contribute to the magnetic field measured by Swarm. The coupling currents or field-aligned currents flow along magnetic field lines between the magnetosphere and ionosphere (image credit: ESA, DTU Space)
• August 2016: The proceedings of the ‘Living Planet Symposium 2016', held in May 2017 in Prague, were only available in mid-August 2016. The current mission status of the Swarm ground segment and a summary of the major operational challenges since launch are covered mostly in the Ground segment chapter (at the end of the file, Ref. 113). Only the platform operations are stated here.
- After a challenging commissioning and orbit acquisition phase to achieve the target constellation, the three satellites are working remarkably well and the complete flight operations segment is running smoothly and without major interruptions despite the additional complexity of a multi-spacecraft mission. The Swarm platform has proven to be very robust and is behaving remarkably well. All units, nominal and redundant, are working as expected on all three spacecraft, and there have been few anomalies since launch.
- The main platform related activity has been the redefinition of the strategy to handle the different errors in the on-board MMU (Mass Memory Unit). The new approach takes into account the error type and the error area to better decide when to ignore and when to re-initialize the memory. It is particularly important to differentiate between SEFI (Single Event Functional Interruptions) and SSB (Single Stuck Bits); as both disturbances are regularly observed in Swarm MMUs. Table 3 summarizes the non-correctable errors reported by the three spacecraft.
Table 3: Summary of SEFI/SSB since commissioning phase. For Swarm-B 40 of the non-correctable errors (stuck bits) belong to the same address
- With the old strategy, all non-correctable errors (single/multi) detected by the memory scrubber were recovered from ground by performing an MMU re-initialization, which resulted in about 30 minutes of data loss each time. From launch until early 2015, about 5 hours (in total for the three satellites) of data were lost due to unnecessary MMU resets.
- In case of a SEFI, the decision is based on the impacted area; the MMU re-initialization is not performed if the affected area is storing housekeeping and science data and not MMU meta-data, since the anomaly condition disappears after the affected memory is rewritten.
- Finally, in order to improve the performance of the GPSR (GPS Receivers) – as another platform subsystem - for better scientific exploitation and to minimize the failures due to loss of synchronization, a number of GPSR setting changes were commanded by patching the GPSR on-board software. The updates include a stepwise increase of the GPS field-of-view up to 88º in order improve performance of the GPS with a reduced number of satellites in tracking.
• August 2016: Since completion of the orbit acquisition phase in April 2014 one satellite (Swarm-B) is flying in a higher orbit with an inclination of 87.8º and an altitude decaying from 520 km. The other two satellites are Swarm-A (trailing) and Swarm-C (leading). They form the lower pair with an initial altitude of 473 km, an inclination of 87.4º and an ascending node difference of 1.4º. 64)
- The original mission analysis foresaw a decay of the lower pair down to 300 km altitude within 4 years after launch. The target altitude of the launcher injection orbit was selected accordingly with some margin due to uncertainties in the solar activity prediction. However the final altitude selection had to be provided more than half a year before launch. Following several launch delays, the major part of the mission falls now beyond the maximum of the current solar cycle. Because of the lower radio flux and geomagnetic activity, the air drag forces are now much lower and the actual decay takes longer.
- As a first countermeasure the target for the inclination difference between Swarm-B and Swarm-A/C was reduced to 0.4º shortly before the start of the orbit acquisition maneuver sequence early 2014 such that the LTAN drift between the orbit planes of Swarm-B and A/C has been reduced to 1.5 h per year to avoid a too large difference towards the end of the mission.
• May 10, 2016: With more than two years of measurements by ESA's Swarm satellite trio, changes in the strength of Earth's magnetic field are being mapped in detail. Launched at the end of 2013, Swarm is measuring and untangling the different magnetic signals from Earth's core, mantle, crust, oceans, ionosphere and magnetosphere – an undertaking that will take several years to complete. 65)
- Although invisible, the magnetic field and electric currents in and around Earth generate complex forces that have immeasurable effects on our everyday lives. The field can be thought of as a huge bubble, protecting us from cosmic radiation and electrically charged atomic particles that bombard Earth in solar winds. However, it is in a permanent state of flux.
- New results from the constellation of Swarm satellites show where our protective field is weakening and strengthening, and importantly how fast these changes are taking place. It shows clearly that the field has weakened by about 3.5% at high latitudes over North America, while it has strengthened about 2% over Asia. The region where the field is at its weakest – the South Atlantic Anomaly – has moved steadily westward and weakened further by about 2%. — In addition, the magnetic north pole is wandering east, towards Asia.
- Chris Finlay, senior scientist at DTU Space in Denmark, said, "Swarm data are now enabling us to map detailed changes in Earth's magnetic field, not just at Earth's surface but also down at the edge of its source region in the core. "Unexpectedly, we are finding rapid localized field changes that seem to be a result of accelerations of liquid metal flowing within the core."
- Rune Floberghagen, ESA's Swarm mission manager, added, "Two and a half years after the mission was launched it is great to see that Swarm is mapping the magnetic field and its variations with phenomenal precision. "The quality of the data is truly excellent, and this paves the way for a profusion of scientific applications as the data continue to be exploited."
- It is clear that ESA's innovative Swarm mission is providing new insights into our changing magnetic field. Further results are expected to lead to new information on many natural processes, from those occurring deep inside the planet to weather in space caused by solar activity.
Legend to Figure 28: The magnetic field is thought to be produced largely by an ocean of molten, swirling liquid iron that makes up our planet's outer core, 3000 km under our feet. Acting like the spinning conductor in a bicycle dynamo, it generates electrical currents and thus the continuously changing electromagnetic field. Other sources of magnetism come from minerals in Earth's mantle and crust, while the ionosphere, magnetosphere and oceans also play a role. ESA's constellation of three Swarm satellites is designed to identify and measure precisely these different magnetic signals. This will lead to new insight into many natural processes, from those occurring deep inside the planet, to weather in space caused by solar activity. - It is thought that accelerations in field strength are related to changes in how this liquid iron flows and oscillates in the outer core.
• December 22, 2015: The Swarm team is pleased to inform the users that an ASM-VFM Residual dataset produced for the investigations of the scalar residuals between the readings of the scalar magnetometer (ASM) and the vector magnetometer (VFM) is now available in the "Advanced" folder of the ESA FTP server covering the period from the beginning of mission until 17 July 2015 for all Swarm spacecraft. 67)
• July 2015: With the excellence of the science data from the three-satellite constellation, mission exploitation continues at a rapid pace. Scientific productivity is extremely high in all mission areas, ranging from the deep interior outer core via the mantle, lithosphere and crust, through to the thermosphere and magnetosphere. Swarm has meanwhile produced its initial field models in areas while also demonstrating beyond all doubt that its constellation approach is instrumental in the disentanglement and separation of the complex and dynamic contributors to the total magnetic and electric field measurements. 68)
- Constellation maintenance activities are proceeding very well, in particular keeping the lower pair at optimum operation for measurement of magnetic field gradients. This is especially relevant and important for achieving the best possible estimate of all contributors to the total measured magnetic field.
- Spring saw the release of the first official geophysical models following a period of validation activities. Likewise, in terms of external fields and geospace measurements, Swarm instruments continue to demonstrate their feasibility to detect current systems and ionospheric features, additionally underlining the high quality of mission data. This holds for both elements of the electric field instrument – the Langmuir probe and the thermal ion imager.
- The mission's calibration/validation effort continues, with particular emphasis on detailed assessment of instrument data quality. As a direct result, significant improvement to all mission products have been achieved through lessons learned from onboard tests and/or processing algorithm evolution.
• June 22, 2015: After about 1 1/2 a years on orbit, the three Swarm satellites have provided a first glimpse inside Earth and started to shed new light on the dynamics of the upper atmosphere – all the way from the ionosphere about 100 km above, through to the outer reaches of our protective magnetic shield. 69)
- Rune Floberghagen, ESA's Swarm Mission Manager, said, "These results show that all the meticulous effort that went into making Swarm the best-ever spaceborne magnetometry mission is certainly paying off."
- "For the time being, the priority is to make sure that the science community can take advantage of the first results of the mission," remarks Nils Olsen, who leads the Swarm Satellite Constellation Application and Research Facility, a consortium of European, US and Canadian institutes in charge of producing advanced models of the various sources of the geomagnetic field from Swarm data.
• May 2015: Swarm continues to acquire excellent science data. Satellite constellation maintenance operations are proceeding, this is particularly relevant and important to achieving the best possible estimate of all contributors to the total magnetic field. Last year, early mission data were used to derive candidate solutions for the 2015 International Geomagnetic Reference Field (IGRF) model. The IGRF is a main field model that (by convention) is updated every five years, and which is used by practically all applications communities and services in need of geomagnetic data. IGRF-12, as the final 2015 model is called, is based on a combination of Swarm, historical satellite data and ground-based observatory data. In addition, a Swarm Initial Field Model, which includes also the computation of the crustal magnetic field at high spatial resolution, has been produced and made available to the community. 74) 75)
Figure 29: Total number of Swarm satellite data (stacked histogram) as a function of (left) time and (right) latitude, respectively. For vector (red), scalar (purple), and N-S gradient (blue and green) data, the dark/normal/light color represents SW-A/SW-B/SW-C. E-W gradient data are shown in black and grey (image credit: Swarm Science Team, Ref. 75)
• On November 22, 2014, the three Swarm spacecraft were 1 year on orbit. 76)
- The three-satellite constellation continues to acquire excellent quality science data from the seven identical instruments on board. Over the summer, early mission data were reprocessed in order to support the development of candidate solutions for the 2015 IGRF (International Geomagnetic Reference Field) model, for which the delivery date was 1 October. Such candidate models have been produced by various members of the Swarm science team, using data from both classes of magnetometers. The IGRF is a main field model that (by convention) is updated every five years, and which is used by practically all applications communities and services in need of geomagnetic data. 77)
• October 2014: Early results of the Swarm ASM (Absolute Scalar Magnetometer). 78)
- The ASM instruments have been operating on the three Swarm satellites, since November 26, 2013. The first results confirm that our initial goals are fully met.
- For the last nine months, the ASM instruments have been providing simultaneous scalar and vector data at a 1 Hz sampling rate, continuously except some days where specific operations were carried out. — The new burst mode confirmed to be a powerful tool to analyze the instruments' functioning during the commissioning phase, and allowed to demonstrate the very low noise of the magnetometers and the cleanliness of the electromagnetic environment in the ASM band. Further analysis of the signals collected during the commissioning phase is currently being carried out by IPGP to look for potential geophysical signals of interest.
- The ASM scalar data offers the best performance ever attained in space and the consistency of the readings delivered by the different instruments has been demonstrated. This should allow the scientists to study very tiny signals, which were not detectable before Swarm. Moreover, despite the challenge it represented, a direct in-orbit verification of the Swarm mission's requirement for the accuracy of the magnetic field's magnitude has been carried out, by taking full advantage of the constellation configuration.
- The vector mode experiment has also proven itself very successful and the ASM ability to function simultaneously as an absolute scalar magnetometer and an autonomous vector field magnetometer is now validated in orbit, which is a world first ! It provides thus a unique opportunity to cross calibrate in orbit two different types of vector instruments (the Vector Fluxgate Magnetometer and the ASM in vector mode). - The vector mode data have demonstrated to have their own scientific merit, and may become an official product of the Swarm mission at some point. The ASM data is presently being used to build a geomagnetic field model, independently of the VFM.
- The primary role of the ASM is to provide absolute measurements of the magnetic field's strength at 1 Hz, for the in-flight calibration of the VFM (Vector Field Magnetometer). As the Swarm magnetic reference, the ASM scalar performance is crucial for the mission's success. Thanks to its innovative design, the ASM offers the best precision, resolution and absolute accuracy ever attained in space, with similar performance all along the orbit. In addition, thanks to an original architecture, the ASM implements on an experimental basis a capacity for providing simultaneously vector measurements at 1 Hz. This new feature makes it the first instrument capable of delivering both scalar and vector measurements simultaneously at the same point. Swarm offers a unique opportunity to validate the ASM vector data in orbit by comparison with the VFM's. Furthermore, the ASM can provide scalar data at a much higher sampling rate, when run in "burst" mode at 250 Hz, with a 100 Hz measurement bandwidth. An analysis of the spectral content of the magnetic field above 1 Hz becomes thus possible. These different ASM new capabilities have been operated on the three Swarm satellites since the beginning of the mission, nine months ago. The calibration and validation activities have been carried out until the end of June 2014.
- The ASM instruments early health check occurred after the first switch on, 4 days after launch on Nov. 26, 2013 and tested in their various operating modes. First, the good health was assessed, thanks to the different diagnosis modes which are available on board. Then the different measurement modes were exercised: scalar mode (scalar measurements only at 1 Hz), burst mode (scalar measurements only at 250 Hz), vector mode (simultaneous scalar and vector measurements at 1 Hz). All the functional verifications were successful on the 3 satellites. A quick look at the magnetometers' intrinsic noise confirmed their excellent resolution, around 1 pT/√Hz, in accordance with the ground measurements.
The same verifications were then carried out on the redundant models. The same successful results were obtained on Swarm Alpha and Bravo, whereas the redundant ASM on Swarm Charlie showed a major malfunction, with no resonance signal. This model had been tested on ground during the launch campaign at Plesetsk and proved to be fully operational. Unfortunately, it did not survive the launch and is now considered as definitively lost. Since the ASM nominal model is fully operational on Charlie and designed for the specified four-year nominal lifetime, this failure should not impact the mission. This was detected before the beginning of the orbit change maneuvers. As the mission's lifetime of the upper satellite is likely to be extended, it was considered preferable to have a redundant model available on this satellite. Therefore it was decided, among other considerations, to place Charlie with Alpha on the lower orbit, and Bravo on the higher orbit. — After these early functional verifications, the detailed assessment of the ASM performance started, on the nominal models only, on each satellite.
Launched in November 2013, Swarm is providing unprecedented insights into the complex workings of Earth's magnetic field, which safeguards us from the bombarding cosmic radiation and charged particles. Measurements made over the past six months confirm the general trend of the field's weakening, with the most dramatic declines over the Western Hemisphere. But in other areas, such as the southern Indian Ocean, the magnetic field has strengthened since January. The latest measurements also confirm the movement of magnetic North towards Siberia.
These changes are based on the magnetic signals stemming from Earth's core. Over the coming months, scientists will analyze the data to unravel the magnetic contributions from other sources, namely the mantle, crust, oceans, ionosphere and magnetosphere. This will provide new insight into many natural processes, from those occurring deep inside our planet to space weather triggered by solar activity. In turn, this information will yield a better understanding of why the magnetic field is weakening.
The field is particularly weak over the South Atlantic Ocean – known as the SAA (South Atlantic Anomaly). This weak field has indirectly caused many temporary satellite ‘hiccups' (called Single Event Upsets) as the satellites are exposed to strong radiation over this area.
Legend to Figure 30: The image is a 'snapshot' of the main magnetic field at Earth's surface as of June 2014 based on Swarm data. The measurements are dominated by the magnetic contribution from Earth's core (about 95%) while the contributions from other sources (the mantle, crust, oceans, ionosphere and magnetosphere) make up the rest. Red represents areas where the magnetic field is stronger, while blues show areas where it is weaker.
Legend to Figure 31: These changes are based on the magnetic signals that stem from Earth's core. Shades of red represent areas of strengthening, while blues show areas of weakening over the 6-month period.
- Engineers have spent the last five months commissioning the identical satellites and carefully guiding them into their orbits to provide the crucial measurements that will unravel the mysteries of Earth's magnetic field.
- The commissioning phase was completed and operations Phase-E2 began on April 9,2014. The final orbital constellation was reached on April 17, where the lower pair is formed by Swarm ‘Alpha' and ‘Charlie', and Swarm ‘Bravo' is in the upper orbit.
- Swarm has a challenging task ahead. Together, the satellites will measure and untangle the different magnetic readings that stem from Earth's core, mantle, crust, oceans, ionosphere and magnetosphere. In addition, information will also be provided to calculate the electric field near each satellite – an important counterpart to the magnetic field for studying the upper atmosphere.
- Two satellites are now orbiting almost side by side and have started their operational life at 462 km altitude. The third is higher, at 510 km. The readings made at different locations will be used to distinguish between the changes in the magnetic field caused by the Sun's activity and those signals that originate from inside Earth.
- Swarm is now in its fine-tuning phase but it has already produced enough information to build models of the magnetic field for comparison with existing models. This proves that only a few months of Swarm data agree very well with a decade or more of predecessor missions.
- Scientists will start to have access to the mission's magnetic field data in a couple of weeks.
Legend to Figure 32: Data from Swarm were used to generate a model of the magnetic field from Earth's lithosphere. The image compares the Swarm model with the Chaos-4 model and shows good agreement, especially considering Swarm is still only in the calibration and validation phase of the mission. The colors in the image show differences between the two models.
• February 2014: Since the Swarm constellation was launched last November, engineers have been busy putting the satellites through their paces to make sure that the craft and instruments are working correctly. This commissioning phase is an essential part of the mission before it starts providing data to further our understanding of the complex and constantly changing magnetic field. 83)
- Some tricky maneuvers are now under way to steer the trio of Swarm satellites into their respective orbits so that they can start delivering the best-ever survey of Earth's magnetic field.
- Since the intensity of solar activity is currently lower than anticipated, the original plan of where to place the satellites at the beginning of science operations has been reviewed recently by the scientific community and experts in ESA. Low solar activity means the satellites experience lower atmospheric drag, as clearly demonstrated by ESA's GOCE mission.
• On Nov. 26, 2013, a significant milestone for ESA's magnetic field mission was reached, the Swarm satellites have completed the critical first phase of their new mission. 84)
• Immediately following separation, the three satellites started transmitting their first signals to Earth, marking the start of the critical ‘launch and early orbit phase', known as LEOP (Launch and Early Orbit Phase). After separation, the project acquired signals from the first two Swarm satellites 91 minutes into the mission, followed by the third at the 95-minute mark — starting the LEOP phase. Around 23 hours GMT on Nov. 22, each of the three satellites deployed their 4 m-long boom carrying instruments essential to the mission's scientific success. LEOP was formally declared as completed by Flight Operations Director Pier Paolo Emanuelli on Nov. 24, 2013 at 19:30 GMT (Ref. 84).
• First contact was established with the trio a few minutes after deployment through the Kiruna station in Sweden and the Svalbard station in Spitsbergen, Norway. All three satellites are controlled by ESA teams at ESOC (European Space Operation Center) in Darmstadt, Germany (Ref. 44). The ESA control team gave nicknames to the trio of spacecraft, namely: Alpha, Bravo and Charlie.
Sensor complement: (VFM, ASM, CEFI, MAC-04 accelerometer, LRR)
High-precision and high-resolution measurements of the strength, direction and variation of the magnetic field, complemented by precise navigation, accelerometer and electric field measurements, will provide the necessary observations that are required to separate and model various sources of the geomagnetic field. 85) 86)
The observation concept is mainly determined by the following drivers set by the scientific payload: Each of the satellites carries an identical payload:
• High magnetic cleanliness required by the magnetometers (sub nT-range for VFM and ASM)
• EFI (Electrical Field Instrument) requires in-flight pointing with control accuracy of 5º
• ACC (Accelerometer) requires precise and stable accommodation in CoG (Center of Gravity).
Figure 33: Schematic overview of the Swarm sensor complement (image credit: GFZ Potsdam)
Table 4: Key performance requirements
VFM (Vector Field Magnetometer):
VFM is the prime instrument of the Swarm mission developed at DTU Space. The objective is to measure the magnetic field vector, on the boom, together with the star tracker for precise attitude measurement. The boom mounted Swarm vector magnetometer instrument consist of a triple star sensor block and a CSC (Compact Spherical Coil) vector magnetometer sensor, mounted on a stable optical bench (Figure 8). Each satellite contains the optical bench with one CSC and three CHU (Camera Head Unit). 87) 88) 89)
The three star sensor units are arranged with the boresights 90º from each other so as to ensure that only one CHU may be affected by Sun or Moon intrusion at any given time. Hereby an attitude solution accurate in all three degrees of freedom can be delivered to the CSC throughout the entire mission. The CSC sensor and the triple star sensor block are mounted on either end of a highly stable mechanical structure.
The CSC vector sensor is supported by a zero CTE (Coefficient of Thermal Expansion) CFRP (Carbon Fiber Reinforced Polymer) adapter that on the one end matches the zero CTE CFRP tube, used to displace the CSC sensor from the star sensor heads (CHU), and on the other end matches the 32 ppm CTE CSC sensor, by means of a finger section. The rotational symmetry of this design ensures an excellent angular stability.
The other end of the CFRP tube is attached to a CSiC bracket holding the three CHUs. The CSiC exhibit a heat distribution capacity second to none, minimizing thermal biases of this section, from the inevitable thermal gradient induced when the sun happens to illuminate any of the three CHUs. Because the CSiC is weakly magnetic, this material can only be used at distances larger than 20 cm from the CSC sensor.
Each CHU is fitted with a straylight suppression system that is thermally decoupled from the optical bench. This separation minimizes thermal excursions from the time varying sun impingement over an orbit to less than a few degrees C. The straylight suppression system is mechanically mounted on an external thermal CFRP shroud, which also provides for thermal control of the entire optical bench. The material selection for all thermal protection has been performed to suppress soft or hard magnetic parts as well as parts that can generate magnetic fields under thermal gradients.
VFM instrument: The VFM (fluxgate type) is based on the fluxgate transducer using a ringcore with amorphous magnetic material, which has a very low noise (10-20 pT rms). It has an extremely high stability < 0.05 nT/year. VFM consists of a CSC (Compact Spherical Coil) sensor, non redundant, mounted on the deployable boom, an internally redundant data processing unit (DPU) and the connecting harness. The spherical coils that create a homogeneous vector field inside the sphere are mounted on an isotropic and extremely stable mechanical support. In feedback conditions the sensor is used as a nulling device and the coils define uniquely the magnetic axes of the sensor. The VFM exhibits high linearity (< 1ppm), a component accuracy of 0.5 nT and precision of 50 pT rms.
The operation of the fluxgate sensor is based on the extreme symmetry of the positive and negative magnetic saturation levels of the ferromagnetic sensor core material. Continuous probing of the core saturation levels by a high frequency excitation magnetization current enables the sensor to detect deviations from the zero field with only tens of pT noise and sub-nT long term stability.
The mounting of the VFM sensor is using a sliced adaptor ring. The optical bench ensures mechanical stability of the system. Three star trackers provide full accuracy attitude.
Table 5: Specification of the VFM instrument
Figure 34: The VFM flight model with redundant electronics unit or DPU (left) and CSC sensor (right), image credit: DTU Space
The µASC (micro Advanced Stellar Compass) of DTU Space provides the high accuracy, inertial attitude determination for the Swarm vector magnetometer. The microASC is a fully autonomous, internally hot/cold redundant star tracker, featuring up to four cameras. The microASC features a split DPU (Data Processing Unit) and CHU (Camera Head Unit) enabling the low power dissipation and very low magnetic disturbance CHU, to be placed close to many types of science instruments, including the CSC sensor (see µASC description below).
Inter-calibration: determining the internal angels between the CSC and the three CHUs: The optical bench provides a mechanically stable platform for the CSC and the three CHUs and will ideally fixate the internal angles between these. The prime objective of the inter-calibration is to establish the internal angles with the highest possible accuracy. The measurement frame of the CSC sensor is defined by the orientation of the compensation coils on the outside of the CSC sensor sphere. These coils form a nearly orthogonal triad, which has been thoroughly calibrated and orthogonalized prior to the mounting on the bench structure. Similarly the measurement frame of any of the CHUs is defined by the mechanical arrangement of the optics relative to the CCD sensor of the unit. Also this measurement frame has been established prior mounting the unit on the bench.
Despite the effort to minimize thermo-elastic deformations and the effort to make the platform as stiff and stable as possible, small residual variations exists. A secondary objective for the inter-calibration is therefore to assess the size of these residual errors, e.g. gravity release effects.
Finally, the mounting of the sensor units to the stiff optical bench will cause stresses to be built into the mounting interfaces. These stresses may cause minute changes to the internal calibration of the sensors. A third objective of the inter-calibration is therefore to verify the pre mounting calibration of the sensor units.
ASM (Absolute Scalar Magnetometer):
ASM is provided by CNES (French Space Agency) and CEA-LETI (French Atomic Energy Commission - Laboratoire d'Electronique de Technologie et d'Instrumentation), Grenoble, France. The objective is to measure magnetic field strength and to calibrate the VFM device to maintain the absolute accuracy during the multi-year mission. ASM is positioned at the very tip of the boom. The required main performance characteristics of the ASM are: absolute accuracy of < 0.3 nT (2σ), resolution < 0.1 nT within its full-scale range of 15000-65000 nT. 90) 91) 92) 93) 94) 95) 96) 97) 98)
Measurement concept: To overcome the limitations of the OVM (Overhauser Magnetometers) identified during the Oersted and Champ programs, a new magnetometer has been designed for the Swarm mission. The ASM pumped helium magnetometer relies on a low pressure helium vapor as the sensing medium (Figure 35), with the optical pumping process the counterpart of the dynamic nuclear polarization. 99) 100)
One important difference is however due to the fact that the optical pumping is a much more efficient polarization method, leading to an almost complete polarization. As a consequence, the signal amplitude does no longer depend on the magnetic field strength and a resolution of 1 pT/ (Hz)1/2 is now obtained over the complete measurement range.
Figure 35: Relevant helium energy levels involved in the ASM magnetometer (image credit: LETI)
As compared to most optically pumped magnetometers, the ASM operates with linearly polarized pumping light instead of circularly polarized light. The main reasons for that choice are the following:
- the strong interaction between the laser pumping beam and the helium atoms can in general affect their energy level and result in so-called light shifts whenever the pumping light wavelength is detuned from the helium transition center wavelength. Now using linearly polarized light suppresses this effect, thus significantly increasing the instrument's accuracy.
- the key parameter governing the optical pumping angular dependence is then the direction of the laser polarization, whereas it is the propagation direction of the pumping beam that matters in circularly polarized light. Now when trying to design an isotropic instrument, i.e an instrument whose performances are independent of the sensor attitude, it is obviously easier to control the direction of the linear polarization than to rotate the whole sensor in order to align it properly with respect to the magnetic field direction. In our case the isotropy is thus simply achieved thanks to the use of an amagnetic piezoelectric motor which permanently controls the laser polarization and the RF magnetic field directions so that they are both perpendicular to the static magnetic field. The resulting magnetometer architecture is illustrated in Figure 36.
Figure 36: Isotropic helium magnetometer architecture (image credit: LRTI)
Contrary to the Overhauser solution based on a design trade-off between instrument's resolution and omnidirectionality, the helium magnetometer is always operated in the optimal operational conditions thanks to this servo loop, but this is achieved at the expense of the use of a dedicated mechanism. As for the sensor anisotropy, resulting from a combination of induced and remanent contributions, a typical signature corresponding to the flight configuration is presented in Figure 37.
As for the environment susceptibility, the ASM significantly broader resonance line (close to 70 nT as compared to less than 7 nT for the OVM) reduces the impact of inhomogeneous magnetic fields on the magnetometer performances, while the principles of operation and architecture of the helium device makes it robust to low frequency radiated magnetic fields, thus making the EMC specifications much easier to meet in that respect.
Last but not least, the short metastable helium relaxation time (of the order of one millisecond) results in a much higher bandwidth for the helium magnetometer than was the case for the NMR sensors. While this feature is of no direct interest for the calibration of the vector instruments (the scalar data are sampled for that purpose at a fixed 1 Hz frequency), it opens new opportunities for the exploitation of the scalar instrument.
Instrument: The ASM magnetometer is based on the ESR (Electron Spin Resonance) principle and makes use of the Zeeman effect which splits the emission and absorption lines of atoms in an ambient magnetic field. The pattern and amount of splitting is a signature of the magnetic field strength. The optically pumped helium magnetometer uses a High Frequency (HF) discharge within a gas cell to excite 4He atoms from the ground state to the metastable state. This metastable level is split by the Earth magnetic field into 3 Zeeman sublevels. The separation of those sublevels is directly proportional to the ambient field strength and equals half the gyro frequency (eB/2m - where m is the electron mass).
Given the role of the laser in the ASM instrument, the following specifications have to be met:
• Wavelength stability with piezoelectric modulation piezoelectric modulation around the D0 transition : λ=1082.908 nm in air standard (std)
• RIN (Relative Intensity Noise) lower than -135 dB (Hz)1/2 at 1 kHz and 2 mW of output optical power
• Spectral linewidth lower than the D0 absorption width of 1.7 GHz
• Specific space environment and space design requirements.
The fiber laser consists of a pump laser diode, a WDM (Wavelength Division Multiplexer), an Yb doped FBG (Fiber Bragg Grating), an optical isolator and a splitter to allow a feedback control. In this architecture, the noise reduction loop acts on the pump diode current by detecting the low frequency fluctuations of the output optical power from the dedicated photodiode located in the 20 % splitter output. The corresponding block diagram is presented in Figure 38.
Figure 38: Block diagram of the laser architecture (image credit: CEA-LETI, CNES)
Athermal design of the LFA (Laser Fiber Assembly): The ASM laser has to be able to pump the 4He at 1082.908 nm (in air std) within the temperature range of [-5ºC; +50ºC] which is the specified qualification operating temperature range of the ASM electronics on the Swarm satellites. One of the main challenges to take up for the LFA conception was to reduce the thermal wavelength drift as much as possible in a passive way. We could have used, for example, a 1083 nm laser diode with a thermo-electric Peltier device but the laser consumption would have been highly increased.
The role of the piezoelectric actuator (made by the CEDRAT Group) is to modulate and to allow a fine tuning of the laser wavelength around the 4He D0 transition. A picture of the final LFA design is given in Figure 39.
Figure 39: Photo of the LFA (image credit: CEA-LETI, CNES)
Legend to Figure 39: The pre-stressed piezoelectric actuator on the right is glued on the Zerodur®; the LFA is then fixed in the ASM electronics box with the titanium bridge put upon the Zerodur®.
Another challenge to take up was the design of a suitable fixation system of the LFA allowing it to comply with the shocks and vibrations specifications. A solution has been developed using a titanium bridge, coned-disc springs (also known as Belleville washers) and elastomers (with suitable thickness and hardness, and a low outgassing characteristic). - The pieces of elastomer are located under the LFA and between the titanium bridge and the LFA in order to dampen any tri-axis vibrations or shocks. This fixation system has successfully passed the vibrations and shocks qualification tests. The final titanium fixation bridge is shown on the LFA of Figure 39.
Figure 40: RIN measured at 1 kHz with a SWARM flight model laser (image credit: CEA-LETI, CNES)
RIN performance: For both consumption and simplicity reasons, in the ASM instrument it has been chosen to detect the magnetic resonance signals coming from the 4He cell around 1 kHz, which corresponds to the 1 kHz modulation of the continuous part of the LA0 magnetic resonance signals between the Zeeman sub-levels. This signal is high enough to allow the measurement of the magnetic field with a low noise laser: the corresponding RIN measurement of the ASM laser is shown in Figure 40 (obtained from an electronic spectrum analyzer and an InGaAs photodiode). As shown, the low frequency feedback control loop presents the opportunity to reduce the RIN from about -105 dB (Hz)1/2 at 2 mW of output power around 1 kHz down to -140 dB (Hz)1/2. The low noise laser specifications are thus met.
Table 6: Specification of the ASM instrument
The instrument assembly consists of a DPU (Digital Processing Unit) and a separately installed sensor connected to the electronic box by a bundle of optical fibers and electrical cables (harness). A specific sensor bracket is designed to mechanically interface 2 identical sensors with the satellite boom (a cold redundancy has been chosen for Swarm, each sensor being connected to a dedicated DPU located within the satellite main body).
Figure 41: Photo of two ASM sensors and their sensor mount (image credit: CNES)
Figure 42: ASM DPU and ASM sensor functional architecture (image credit: CEA, LETI, CNES)
Figure 43: Photo of the ASM DPU and sensors (image credit: CNES, CEA-LETI, Gérard Cottet)
The ASM characteristics make the instrument ideally suited not only for the traditional role of scalar magnetometers as absolute references for the calibration of the on-board vector instruments, but also for extended operational capacities, such as higher frequency scalar measurements (of potential interest for magnetosphere studies for the low frequency part of the spectrum) or autonomous scalar / vector operations. Last but not least, the helium magnetometer can be operated in a zero field configuration with only very minor evolutions in the sensor overall design, thus extending its initial capabilities to new missions in planetary exploration (Ref. 99).
Electromagnetic cleanliness: Differential scalar measurements have been performed throughout the ASM development phases up to the satellite final ground magnetic tests. This allowed first to select the sensors parts in order to minimize their residual magnetic signature and second to evaluate the accuracy of the magnetic data that will be delivered by the ASMs for the three satellites of the Swarm constellation both at instrument and satellite level. This differential method has been adapted to the various rather challenging tests configurations met within the Swarm program. It allowed to demonstrate that the ASM performances meet the mission requirements, with measurement uncertainties below 25 pT. It finally contributed to improve the quality of the magnetic measurements that were carried out at satellite level and check the perturbation models that had been established. 101)
Characterization and removal of thermomagnetic perturbations onto ASM instrument. 102)
A step-by-step perturbation identification and removal process has then been set up in order to characterize the global perturbation and has resulted in several hardware upgrades. Thanks to dedicated scalar differential measurements performed in the CEA-Leti Magnetic Characterization Facility of Herbeys, all elements which have been characterized as sources of thermomagnetic perturbations have been progressively removed or replaced.
The test process has led to several hardware upgrades (3 in total) of both the mechanical interface of the ASM with the satellite boom and later even some ASM components. The original titanium interface bracket including the heaters for the ASM thermal control has been replaced by CFRP (Carbon Fiber Reinforced Polymer) components and the ASM internal harness connectors have been removed. These successive evolutions can be simply summarized as follows: the less metallic material submitted to thermal gradient in the vicinity of the ASM, the smaller
In the end, this exhaustive test campaign has demonstrated the very efficient reduction of thermomagnetic perturbations generated in the close environment of the ASM: the remaining maximum effects have been characterized in the 20-30 pT +/- 5 pT peak-to-peak range (Figure 4) which is well below the accuracy specification of 300 pT (2σ) allowed for ASM for the Swarm mission
Figure 44: Amplitude of the heater-generated thermomagnetic perturbation on the ASM sensor w.r.t assembly configuration (measurement accuracy of ± 5 pT), image credit: Université Grenoble Alpes, CEA, LETI, CNES
Thanks to the removal and/or replacement of metallic components in the close environment of the ASM sensor, it has been possible to get rid of these perturbations before the launch of the Swarm satellites. The study puts further emphasis on one of the basic design rules for the conception and integration of magnetic sensors on-board space vehicles: avoid as much as possible metallic elements in the vicinity of these instruments, even if they do not show any static magnetic signature. They can still give rise to magnetic perturbations if they are submitted to thermal gradients, which turn to be very difficult to compensate for by modelling as these elements are never simple in shape and as it also requires a very precise 3D knowledge of their internal temperature distribution.
EFI (Electrical Field Instrument):
EFI, also referred to as CEFI (Canadian Electric Field Instrument), is provided by Canada (CSA funding, design by the University of Calgary, with ComDev Ltd. of Cambridge, Ontario as instrument manufacturer). The CEFI sensor is based on SII (Suprathermal Ion Imager)-a Canadian particle detector design that has already proven its capability - to gather precise measurements of ion winds. The goal of the CEFI instrument is to characterize the electric field about the Earth by measuring the plasma density, drift, and acceleration at high resolution; also for plasma density mapping in conjunction with GPS.. CEFI derives its heritage from the CPA (Cold Plasma Analyzer) instrument on Freja, the Nozomi TPA instrument and the CUSP, JOULE and GEODESIC sounding rocket missions. 103) 104) 105) 106) 107)
The plasma ion measurements are derived from energy-angle distributions that are generated by two orthogonal 2D electrostatic analyzers on each satellite. The ion bulk flow velocity and temperature are related to the distribution moments by transfer functions whose form are determined from simulations of the analyzers. The electric field is determined from measurements of the ion velocity and the magnetic field. 108)
The main sources of error come from uncertainties in the instrument transfer functions, the sensor-to-plasma potential difference, particle Poission noise, galactic cosmic ray event, and detector gain variations.
The CEFI instrument is comprised of three main parts: the SII (Suprathermal Ion ImagerI) sensors, the LP (Langmuir Probe) sensors, and the Electronics Assembly. The electronics assembly contains all of the electronics necessary to support power supply, sensor data acquisition, instrument control and communications with the spacecraft bus. The electronics assembly and SII sensors will be positioned on the ram face of each Swarm spacecraft along with the Langmuir probes positioned preferably on the ram and nadir faces of each spacecraft and connected to the electronics assembly with wire harnesses.
Figure 45: The EFI instrument measurement concept (image credit: University of Calgary)
Figure 46: Schematic of the CEFI device (image credit: University of Calgary)
The SII sensors are of CPA heritage; they are using a unique particle focusing scheme developed at the University of Calgary. Ions enter a narrow aperture slit and are then deflected by a pair of hemispherical grids that create a region having electric fields directed radially inward. Incoming low-energy positive ions are accelerated toward the center of the spherical system, whereas ions with larger kinetic energies travel farther toward the edge of the detector, creating an energy spectrum as a function of detector radius (Figure 47).
Particles arriving from out of the plane of Figure 47 land at different azimuths on the image plane. The resulting image from each SII sensor is a 2D cut through the ion distribution function, from which one can calculate ion density, drift velocity (2D), temperature, and higher-order moments. The two SII sensor head assemblies are oriented such that the aperture slits are oriented perpendicularly to each other, enabling 3D characterization of the ion distribution.
Range of operational conditions:
• Natural variability
- Ion densities (108-10 m-3)
- Ion and electron temperatures (0.1-0.5eV)
- Ionospheric plasma flow (~200 m/s)
• Active biasing of face plate
• Passive biasing of material components associated with different work functions and contact potentials.
Figure 47: Schematic view of the particle focusing system (image credit: University of Calgary)
Figure 48: Illustration of the CEFI sensor head assembly (image credit: University of Calgary)
When the charged particles strike the MCP Microchannel Plate) detector, the signal is amplified through secondary emission processes. The voltage applied across the MCP controls the gain of the device. In parts of the orbit where ion flux is high, the voltage applied to the front surface is reduced to limit the gain of the device and preserve its life. This gain adjustment is part of an automatic gain control realized through the use of a feedback loop using the CEFI instrument faceplate current as a control input. Where sufficient gain control cannot be achieved via the MCP voltage alone, an electrostatic shutter will gate the incoming ions with duty cycles ranging from 100% to well below 1%.
A Langmuir Probe assembly is part of the CEFI device to provide measurement of electron density, electron temperature and spacecraft potential. The LP design is based on hardware flown on the Cluster and Rosetta missions of ESA and was developed by the Swedish Institute of Space Physics. The overall height of the LP sensor is 10 mm.
The instrument electronics assembly includes the following electrical subsystems:
• Instrument controller
• Detector readout electronics
• HVPS (High Voltage Power Supply)
• LVPS (Low Voltage Power Supply). The primary function of the LVPS is to generate low voltages for other electronics within the Swarm CEFI instrument.
• Langmuir probe assemblies.
Figure 49: CEFI electrical block diagram (image credit: University of Calgary)
Figure 50: Engineering model of the EFI instrument with the two orthogonal sensor heads (image credit: COM DEV)
MAC-04 (Micro Accelerometer-04):
MAC-04, or simply ACC, is an electrostatic accelerometer instrument assembly designed and developed at VZLU (Výzkumný a zkušební letecký ústav, a.s. - Aeronautical Research and Test Institute), Prague, Czech Republic. EADS Astrium (as prime contractor to ESA for the Swarm mission) awarded the MAC-04 contract to VZLU in March 2008. VZLU is the lead of a MAC-04 consortium, involving 14 Czech institutions and companies. 109)
The overall objective is to measure non-gravitational perturbing accelerations (time and spatial variability), such as air drag, winds, Earth albedo (reflected solar radiation from the cloud/snow coverage and the thermal radiation of the Earth), and solar radiation pressure on the spacecraft. In-situ air density measurements together with magnetic data can be used to obtain new insights on the geomagnetic forcing of the upper atmosphere.
Typical magnitudes of non-gravitational forces are presented in Table 7. For spacecraft altitudes below 800 km the atmospheric drag is the dominant acceleration; at higher altitudes, the direct solar radiation pressure and other radiative forces are increasing in dominance.
Table 7: Expected values of non-gravitational accelerations in LEO
All these radiation effects have a common feature – namely a slow change of magnitude with time and position in orbit. However, for the solar radiative effects, the illumination of the spacecraft is a precondition for any acceleration detection in its orbital path.
The solar activity has a direct impact on atmospheric drag. The thermosphere usually expands and contracts in line with the sun's 11 year solar cycle. During solar maximum when solar activity increases, it causes the thermosphere to heat up — reaching temperatures of 1100ºC — and expand. The opposite happens during solar minimum. As solar activity increases, EUV (Extreme Ultraviolet) radiation heats our planet's gaseous envelope, causing it to swell and reach farther into space than normal. However, despite of the sun's possible rapid fluctuations, Earth's thermosphere response is a fairly slow process. These considerations led to the design performances of the MAC-04 accelerometer as shown in Table 8.
Table 8: Performance specification of the 3-axis MAC-04 accelerometer
The MAC-04 assembly is composed of a cubic proof mass which is free-floating in the cubic cavity. The center of the sensor will be accommodated in the spacecraft's center of gravity. The proof mass is separated from external influences by the satellite structure and the construction of the micro-accelerometer.
Free motion of the proof mass is realized by virtue of the gravitational law. The cavity is rigidly connected to the satellite body. The gravitational as well as all perturbing forces acting on a satellite produce their acceleration contributions which are identical to the one of cavity. The difference between the acceleration of the cavity and the acceleration of the proof mass is the sum of all accelerations produced by the non-gravitational forces acting on a satellite.
A precise measurement of the proof mass position enables to properly detect its small relative displacements with respect to the satellite-fixed cavity. Applying a known electrostatic force, the instrument can compensate and measure the action of the non-gravitational forces. This proof mass position control is performed by a feedback loop of the servo control electronics.
A block diagram of the sensor control circuit is shown in Figure 51. Part of the block diagram placed in the dashed border represents the dynamics of the proof mass. The constants A1, A2 and A3 represent the shape and geometry of the electrodes and their distance to the proof mass.
Figure 51: Block diagram of proof mass position regulation system (image credit: VZLU)
Figure 52: Major elements of the MAC-04 (image credit: ESA)
Figure 53: The MAC-04 electronics stack (left) and the instrument (right), image credit: ESA
Figure 54: Photo of the mechanical part of the MAC04 accelerometer (image credit: VZLU)
MAC-04 science data output 1Hz rate:
- Linear acceleration: 3 vector components
- Angular acceleration: 3 vector components
- Linear positions: 3 vector components
- Angular positions: 3 vector components
- Temperatures: 8 sensor values.
Some background of Czech accelerometer instrument projects in spaceflight:
During the last two decades, the Czech Republic has developed several accelerometer instrument models within the framework of the non-gravitational accelerometer program. There is great interest among the theoreticians in satellite dynamics since our understanding of the non-gravitational effects is rather limited. In many cases, only rough phenomenological models are available. In situ measurements of the non-gravitational effects have thus great importance for checking the theoretical concepts. The MAC (Micro Accelerometer) program (formerly referred to as MACEK) is devoted to such measurements with an expected threshold up to 5 x 10-11 ms-2 .
Table 9 provides an overview of the various spaceborne missions whose payload included a MAC instrument. The obtained measurements are of different quality. The first two instrument models (flown on Resurs-F1 and on the STS-79 flight of the Space Shuttle) were mainly of in-flight technology verification nature. The MIMOSA project was the first small satellite dedicated to the research of the thermosphere. Unfortunately the failure of unlocking in one axis spoiled the mission goal.
The first MAC-04 instrument of VZLU, developed under in the frame of the project TEASER (Technological Experiment And Space Environment Resistance), was launched with the Russian Tatiana-2 mission in Sept. 2009. The project was funded Ministry of Industry and Trade of the Czech Republic. - However, after spacecraft separation from the upper stage of the launch vehicle, the satellite stabilization system did not work correctly (infrared Earth sensor failed). When all trouble shooting didn't bring any tangible results, the communication with the spacecraft was terminated after one month of "operations" with the spacecraft.
As of mid-2010, the development of the MAC-04 instrument for the Swarm mission is completed and instrument qualification for the mission is underway. Although the measurement principle is quite simple, past experiments showed high demands on the precise adjustment of the accelerometer in ground conditions. Moreover the complexity and the high precision adjustment of the sensor mechanical parts in combination with electronics properties (e.g. temperature dependence) represent a challenging task in the MAC-04 accelerometer verification.
Table 9: Space missions with the MAC micro-accelerometer
Table 10: Basic as designed/build performances of different accelerometer models
The Czech Republic formally became ESA's 18th Member State on 12 November 2008. Already in 1996, Czech Republic signed the formal Framework Cooperation Agreement. In November 2003, however, the Czech Republic became an ESA European Cooperating State by signing the Plan for European Cooperating States Agreement (PECS Agreement) and entering the ESA PECS Program. 110)
LRR (Laser Retro Reflector):
The LRR instrument is being provided by GFZ Potsdam. The objective is spacecraft POD (Precise Orbit Determination) to cm-level accuracy by SLR (Satellite Laser Ranging).
Figure 55: Photo of the LRR unit (image credit: ESA)
Figure 56: Placement of instruments on the spacecraft (image credit: ESA) 111)
Table 11: Anticipated signals at 400 km (reference 7)
Star tracker assembly and OB (Optical Bench):
The STR (Star Tracker) assembly provides the attitude of the VFM. Both instruments are co-mounted in a common optical bench to ensure proper alignment for the determination of the highly accurate magnetic field components. The µASC (Micro Advanced Stellar Compass) of DTU Space is being used in the star tracker assembly. It features two fully cold/hot redundant DPU's. With full cross-strapping, each DPU can control one to four CHU's (Camera Head Unit). Mission specific baffles can be designed for optimum performance. The attitude is autonomously calculated based on all brighter stars in the FOV of the CHUs; µASC can provide 22 true solutions per second. The absolute accuracy is < 1 arcsec. The instrument mass is < 1.4 kg (3 x CHU, BFL's & DPU). The power is < 5.7 W (3 x CHU+DPU). The instrument can also support asteroid science - Near Earth Object (NEO) detection and planets triangulation.
Figure 57: Configuration of the µASC instrument for the Swarm mission (image credit: DTU Space)
OB (Optical Bench): The purpose of the OB is the transference of the attitude from the extremely precise star trackers to the magnetometer field components. The OB ensures a highly mechanical stable platform for the magnetometer and the star trackers. A exhaustive thermo mechanical design and analysis is carried out to determine and minimize any thermal gradient that could cause a shift in the relative attitude between the two systems.
The star trackers are very magnetically clean; the separation between the two instruments (STR and VFM) is about 40 cm to reduce magnetic perturbation from the STRs. The exploitation of the symmetric system has resulted in a cylindrical tube holding the VFM sensor, minimizing transversal thermal gradients. Emphasis has been on the matching of material parameters, the use of iso-static support interfaces, and a detailed analysis of the loads. The instruments are calibrated as stand alone, and once integrated in the OB, an intercalibration and system verification is carried out to determine the relative orientation between the VFM and STR and to verify that the stability is as required.
Figure 58: The Optical Bench with three star tracker cameras (yellow, only two of them are shown) and the magnetometer sensor (green), image credit: DTU Space
Figure 59: Alternate view of a Swarm spacecraft and instrument locations (image credit: ESA)
The ground segment consists of the following elements: 112)
1) CDAE (Command and Data Acquisition Element), located at the Kiruna ground station, Kiruna, Sweden.
2) FOS (Flight Operations Segment), located at ESA/ESOC, Darmstadt, Germany. ESOC is in charge of monitoring and planning of satellite operations.
3) PDGS (Payload Data Ground Segment), located at ESA/ESRIN, Frascati, Italy. The main tasks of PDGS are the generation of data products from the science data, data archiving, and data distribution to the user community.
Figure 60: Overall architecture of the Swarm mission elements (image credit: ESA)
Figure 61: Ground segment architecture (image credit: ESA)
Figure 62: Swarm payload data ground segment (image credit: ESA)
A ground segment for multisatellite operations:
The concept for control of the SWARM mission during the routine phase is presented in Figure 63. It is based on the use of a single control center at ESOC, in conjunction with a prime ground station at Kiruna, augmented by external stations (Svalbard and Esrange) when required, and interconnected by a general purpose, highly available ground network. This is collectively called the FOS (Flight Operations Segment). 113)
Figure 63: Elements of SWARM Flight Operations Segment (image credit: Flight Control Team)
The control center, otherwise referred to as the FOCC (Flight Operations Control Center) in ESA parlance, is comprised of the following systems:
- The SWARM MCS (Mission Control System) to support, with both hardware and software, the data archiving and processing tasks essential for controlling the mission. Together with the MCS, the EDDS (EGOS Data Dissemination System) takes care of providing the Swarm data to all external interfaces.
- The SWARM MPS (Mission Planning System), supporting command request handling and the scheduling of spacecraft/payload operations.
- The FD (Flight Dynamics) System, supporting all activities related to attitude and orbit determination and prediction, preparation of slew and orbit maneuvers, spacecraft dynamics evaluation and navigation in general.
- The Spacecraft Simulator, to support procedure validation, operator training and the simulation campaign before each major phase of the mission.
Data acquired by the FOS is retrieved directly from the ground stations by the PGDS (Payload Data Ground Segment) located at ESRIN, which is in charge of processing the raw data to generate scientific products and making them available to the scientific community. Additionally, satellite housekeeping data is provided through the EDDS server to the PLSO (Post-Launch Support Office) located at ESTEC and to the spacecraft prime contractor providing industrial support.
Mission control system and mission planning
The main challenge of the FOS was the multi-satellite system of Swarm, which necessitated the development of a multi-domain MCS distributed across several servers, and organized in a nominal and a backup chain. This was driven by the need for flexibility for constellation operations and parallel activities.
The Swarm MCS is based on ESOC's generic SCOS-2000 (Spacecraft Control & Operation System-2000) infrastructure software and implements three separate domains in hardware and software, one per spacecraft, while a fourth domain handles all processes common to the three main domains. The system allows the control of the three spacecraft at the same time when required, as demonstrated during the LEOP phase, when two satellites were commanded in parallel and data was received, processed and archived for the three of them: real-time telemetry (VC0 and VC1 data streams) and recorded TM from the on-board mass memories (VC2 and VC4 data streams). In routine, only one satellite is commanded at a time, but processing of the recorded telemetry dumped during the passes may occur in parallel for the three satellites depending on the separation between spacecraft passes.
Commanding is supported by the MPS, which generates two command schedules, the SIGR (Schedule Increment Ground Schedule) and the SIOS (Schedule Increment On-Board Schedule).
The SIGR contains commands to be sent in real time and is typically related to automated pass operations by setting up the link configurations to the respective ground stations and management of the on-board mass memory (start and suspend transmission of stored data and deletion of old data).
The SIOS provides time-tagged telecommands to be loaded into the satellite MTL (Mission Timeline) on-board and mainly controls the critical data downlink strategy (transponder switch on/off, on-board statistics housekeeping, instrument mode transitions, etc.).
The default planning interval is based on seven days corresponding to a calendar week from Monday 00:00:00 UTC until Sunday 23:59:59 UTC. It is nominally prepared on Thursday of week N-1 with an execution time starting on Monday of week N and is uplinked on Friday.
All for less: A data downlink strategy
The data downlink strategy is based on just two ground station passes per satellite per day during working hours; with each pass allowing 4 to 9 minutes of commanding.
During the first pass a long data dump after the overnight out-of-coverage is performed while the second pass is a few hours after the first contact and allows to dump any remaining data that could not be retrieved during the first pass. The second pass makes the schedule more robust against outages and guarantees that the backlog is quickly recovered in case a pass is lost. It also allows recalling any data from previous passes and offers a second commanding window, which is crucial for complex payload and platform special operations that cannot be performed in a single pass.
After each pass, old data is partially deleted. This strategy ensures that always up to three days of science data (fill status of packet store up to 70%) and up to two days of housekeeping data (fill status of packet store up to 50%) are available on-board. Any data gap larger than 5 minutes detected on ground is systematically recovered by re-dumping the data. An example of the evolution of the mass memory fill level for one satellite and one week is presented in Figure 64.
Figure 64: Swarm-A mass memory fill level (science: green; housekeeping TM: blue) on week 15, 2016 (image credit: Flight Control Team)
Aspects on maneuvering the Swarm satellite constellation
The three months of commissioning in 2014 were characterized by a very tight and dynamically changing schedule of activities. All operational issues could be solved during that time, including the challenging orbit acquisition phase to achieve the final constellation.
The near polar orbit for good global coverage and regular 24 hours local time coverage about every 9 months was achieved straight away due to the nominal injection into the common separation orbit. It remained to establish the relative differences between the lower pair formed by Swarm-A/Swarm-C and the upper satellite Swarm-B. The targets in altitude and inclination difference were selected to be 50 km and 0.4º. Mainly the latter determines the relative drift rate of the LTAN (Local Time of Ascending Node) of the orbital planes between the lower pair and Swarm-B. This drift is still on-going and the LTAN difference will reach 4 hours in October 2016 and should then ideally remain within 6h ± 2h during the continuous natural decay down to 300 km until 2022 or later. To achieve this, another maneuver campaign will be needed to slow down the relative drift.
In addition, after launch the lower pair had to be separated by 5-6 minutes in LTAN and maneuvered to fly side by side with less than 10 seconds along track difference. The LTAN separation of the lower pair was achieved indirectly by performing the inclination changes of Swarm-A and Swarm-C at different times leaving, on average, six weeks in between with different nodal precession. The firing direction of the maneuvers was selected such that the semi-major axis was lowered at the same time. To achieve the required delta-v of 32m/s with the low thrust cold gas system capable of 2x0.05N, each satellite had to perform more than 130 maneuvers. The finalization of all maneuvers was completed during only 12 weeks, whereby each week was dedicated to a batch of maneuvers with a single satellite only. For each satellite a small test batch was performed to obtain a first thruster calibration. The subsequent batches consisted of 22-34 consecutive orbits with a 20 minutes maneuver around each ascending and descending node and slews in between. Here the main challenge was to come up with a robust strategy, which could be adjusted easily in case of maneuver failures and excluded any collision risk when Swarm-A approached the side-by-side configuration after completion of four additional revolutions during the six weeks drift phase. The original pre-launch maneuver plan is shown in Figure 65.
This paid off when the first Swarm-A maneuver sequence did, indeed, abort after 69% of the planned delta-v due to an on-board software problem with a shared memory address. 114)
About 36% of the loaded fuel was consumed for the orbit acquisition. This leaves enough fuel for a later maneuver campaign to slow down the relative node precession in the near future in order to stay within a LTAN difference of 6 h ±2 h. The decision about when and how far to slow down needs to be taken by the end of 2016 at the latest and requires also the best possible prediction and planning of the further decay from 440 km down to about 300 km altitude. 115)
Figure 65: Planned initial orbit acquisition after commissioning phase (image credit: Flight Control Team)
Routine and non-routine maneuvers:
The initial orbit for the lower pair, Swarm-A and Swarm-C, was targeted at an altitude of 468 km and an inclination of 87.35 º. It is now maintained in a side-by-side constellation separated along track such that their ascending node crossing time differences are between 4 and 10 seconds. The lower limit guarantees that both spacecraft will not accidentally collide even in the case of anomalies on-board.
In addition, their eccentricity vectors are to be kept close enough to ensure an altitude difference of always less than 5 km. Also for mitigating collision risks, the eccentricities are kept as close as possible. Since the end of commissioning, 9 maneuvers on Swarm-A have been performed using a total of approximately 80 g and none on Swarm-C until now.
The other spacecraft, Swarm-B, was placed in a higher orbit (altitude circa 516 km), with an inclination higher than the lower pair by 0.4º at 87.75º. Its orbit is not controlled after the orbit acquisition phase except for necessary collision avoidance maneuvers.
Two collision avoidance maneuvers had to be executed so far, one on Swarm-B and one on Swarm-A each using around 30 g of fuel to avoid debris that endangered the safety of the satellite.
Several attitude slews were also performed to calibrate various instruments on board. Slew maneuvers were for instance performed in May 2014 to characterize the observed residuals of the two magnetometers VFM and ASM. They required careful preparation in order to ensure that they would not affect the power or thermal budget of the satellite and were compatible with the flight domain of the AOCS. Furthermore, during the routine phase the on-board fuel is spent for normal attitude maintenance (12 g per week on average).
Table 12: Fuel consumption from launch until April 2016. All values in kg
Challenges of Swarm operations
Although the actual spacecraft commissioning phase was concluded in spring 2014, the investigations for some payload instruments are still on-going.
One of the payloads of the Swarm constellation is the EFI (Electrical Field Instrument) in cooperation with the University of Calgary (UoC) in Canada and devoted to the measurement of spacecraft potential, electron temperature, ion properties and ultimately the electric field. Two Langmuir probes (LP) are used to measure the electron properties and spacecraft potential, while a Thermal Ion Imager (TII) is used to capture the plasma particles and produce 2D maps with two CCD sensors.
The intended concept of the TII was full-time operations of the imager, but, in practice, some indications of image degradation arose after a period of continuous operation depending on the satellite. Therefore, in order to maximize the scientific return, it was decided to operate the TII for just a limited and fixed number of orbits per day, in order to ensure good quality data in regions where the physical phenomena, especially at high latitudes, are of higher interest. The LPs are not affected and are always producing good quality data.
This modified concept required the need of a formal coordination between the FOS and the scientific community in order to define the times of activation: in particular a new data interface called OPF (Operations Planning File) has been defined to automatically process the inputs created by UoC and integrate them in the mission planning process at ESOC.
In parallel with the scientific measurements, several tests were carried out, mainly for Swarm-C, in order to raise the TII voltages and the temperatures of the inner instrument and to scrub possible contaminants from inside the chamber, suspected as the possible source for the image degradation. After these tests, several parameter setting updates were applied and, during the last year, the number of orbits used for science operations has successively increased, a sign of the improvement in the continuous and step-by-step fine tuning of the instrument operations. Moreover, a periodic calibration of the CCD gain maps was performed for the two sensors of each TII separately, in order to provide the conditions for good data exploitation.
Table 13 summarizes the different kind of tests performed in order to characterize the TII image anomaly. All these unforeseen activities, far beyond the original assumptions, resulted in a significant extra workload for the FCT. Large efforts were necessary to prepare, schedule and execute all the tests in a setup closer to a "extended commissioning phase" than routine operations.
Table 13: Summary of EFI tests performed since commissioning phase
The health status of the VFM (Vector Field Magnetometers) and the ASM (Absolute Scalar Magnetometers) is excellent, with the exception of the failure of both ASMs on Swarm-C just after launch (this was one of the reasons for the selection of Swarm-B as the upper satellite). The ASMs on Swarm-A and Swarm-B are routinely operated in Vector Mode and the three active VFMs are producing data at 50 Hz on the three spacecraft. Careful monitoring of temperatures, voltages, currents and other payload parameters is performed by the FCT (Flight Control Team) on a regular basis to detect any anomalous behavior.
A residual bias was identified between the measurement of the VFM and the ASM, which was presumably related to a thermal effect of the instruments. In order to characterize this behavior, Swarm-B was slewed four times by 90º. The spacecraft remained in this special attitude for 5 orbits. This operation was performed together with coordinated slews of Swarm-A and Swarm-C, four times by 90º in the reverse direction and offset by 3 orbits. The two satellites remained in each attitude position for 6 orbits.
Additional tests became also necessary for the ACC (Accelerometer) instruments on-board the Swarm satellites. The dependency of the ACC performance with respect to temperature variations appeared to be more complex than anticipated during the design phase. Tests were created in order to reduce thermal variations by using different On/Off heater strategies, in some cases with the nominal and the redundant heaters used in parallel. Those tests required careful preparation to schedule and execute more than 6000 commands synchronized with Sun eclipses.
Figure 66 shows an example of heater profile activation for one of the tests performed. The heater profiles are designed to analyze the impact of various delay activations versus start and end of eclipses in order to achieve stabilization of the ACC temperature. The tmax and tmin times closely match the entry and exit of Sun eclipses and are used as reference for the heater activation.
Test campaigns with attitude thruster activations were conducted to deduce ACC scale factors needed to adjust accelerometer deviations for all satellites. All three axes were calibrated using dedicated thruster activations designed to minimize the impact on the attitude and the orbit.
Figure 66: Example of heater profile activation during ACC thermal tests (image credit: Flight Control Team)
Figure 67 shows an example of the along track scale factor calibration scenario used for all Swarm satellites. It shows the impact of the thruster activations selected on the orbit and rotation of the satellite. It consisted in four separated firings designed to test along-track acceleration in both positive and negative direction and to bring the satellite back to its original position and rotation at the end of the operation. Those tests required a de-activation of the AOCS and had to be geolocalized for best performance. They, therefore, required a very careful preparation and analysis from the FCT before their execution. Currently, a six months calibration campaign is on-going to verify the stability of the scale factor over time.
Figure 67: ACC scale factor calibration maneuvers (image credit: Flight Control Team)
1) E. Friis-Christensen, H. Lühr, G. Hulot, "Swarm: A constellation to study the Earth's magnetic field," Earth Planets Space (special issue on Swarm), Vol. 58, No 4, April 2006, pp. 351-358, URL: http://www.terrapub.co.jp/journals/EPS/pdf/2006/5804/58040351.pdf
2) N. Olsen, R. Haagmans, T. J. Sabaka, A. Kuvshinov, S. Maus, M. E. Purucker, M. Rother, V. Lesur, M. Mandea, "The Swarm End-to-End mission simulator study: A demonstration of separating the various contributions to Earth's magnetic field using synthetic data," Earth Planets Space, Vol. 58, 2006, pp. 359-370
3) "ESA's Magnetic Field Mission Swarm," URL: http://www.esa.int/esaLP/ESA3QZJE43D_LPswarm_1.html
4) M. Thomsen, J. M. G. Merayo, "Data Handling System for Magnetic Nanoprobe SWARM (MAGNAS)," Proceedings of IAC 2004, Vancouver, Canada, Oct. 4-8, 2004, IAC-04-U.4.01
5) A. Schönenberg, R. Haagmans, A. Regan, A. Ginati, Y. Menard, "Swarm - Three Explorers of the Earth Magnetic Field and its Environment," 5th IAA Symposium on Small Satellites for Earth Observation, Berlin, Germany, April 4-8, 2005
7) Y. Menard, R. Bock, E. Neri, R. Haagmans, "Swarm Mission Concept," Proceedings of the First Swarm International Science Meeting, Nantes, France, May 3-5, 2006, WPP-261
9) E. Friis-Christensen, H. Lühr, D. Knudsen, R. Haagmans, "Swarm – An Earth Observation Mission investigating Geospace," Advances in Space Research, Volume 41, Issue 1, 2008, pp. 210-216
10) Roger Haagmans, "ESA's supporting activities related to Mass Transport in the Earth System," Joint GSTM and DFG SPP Symposium, GFZ Potsdam, Oct. 15-17, 2007, URL: http://www.massentransporte.de/fileadmin/20071015-17-Potsdam/mi_1410_07_haagmans.pdf
11) Max Pignède , Mario Merri, Vemund Reggestad, Peter Fritzen, Michael Irvine, "The Swarm constellation simulator - A brand new, but still operationally responsive development," Proceedings of the Ground System Architectures Workshop (GSAW 2009), Torrance, CA, USA, March 23-26, 2009, URL: http://sunset.usc.edu/gsaw/gsaw2009/s4/pignede.pdf
12) Nils Olsen, Juliane Maries, "From Ørsted to Swarm – Challenges for exploring the Earth's magnetic field from Space," Nordic Space Volume 15, No 2, 2007, pp.9-12, URL: http://www.spacecenter.dk/~nio/papers/NSA188-Olsen-2007.pdf
14) Eigil Friis-Christensen, "Scientific goals of Swarm," Proceedings of ESA's Second Swarm International Science Meeting, June 24-26, 2009, Potsdam, Germany, WPP-303, URL: http://www.congrex.nl/09c24/S1/S1_00_Friis-Christensen_presentation.pdf
15) Yvon Menard, "Swarm Mission Status," Proceedings of ESA's Second Swarm International Science Meeting, June 24-26, 2009, Potsdam, Germany, WPP-303, URL: http://www.congrex.nl/09c24/S1/S1_02_Menard_presentation.pdf
16) R. Haagmans, R. Bock, H. Rider, K. Fletcher, "Swarm - ESA's Magnetic Field Mission," ESA, BR-302, October 2012, URL: http://esamultimedia.esa.int/multimedia/publications/BR-302/offline/download.pdf
17) "Swarm : Why magnetism matters," Astrium, Oct. 4, 2013, URL: http://www.astrium.eads.net/en/news2/why-magnetism-matters.html
18) Richard Holme, "Swarm-Near-Earth Multi-Satellite Probing of the Geomagnetic Field," 2012 NCEO / CEOI (National Centre for Earth Observation/Centre for EO Instrumentation)Joint Science Conference, Nottingham Jubilee Campus, UK, Sept. 17-20, 2012, URL: http://www.nceo.ac.uk/assets/presentations/2012_conference_NOTTINGHAM-NCEOCONF2012_0918_Holme.pdf
19) "Swarm - The Earth's Magnetic Field and Environment Explorers," ESA, Earth Explorers Users Consultation Meeting, 19-20 April 2004, URL: http://esamultimedia.esa.int/docs/EEUCM/Swarm_handout.pdf
20) "Magnetic Fields -- History," NASA, URL: http://www-spof.gsfc.nasa.gov/Education/whmfield.html
21) Yvon Menard, Roger Haagmans and the Swarm Project Team, Rune Floberghagen, Juan Piñeiro, "Swarm: ESA's magnetic field mission - A Journey to Earth's Core," ESA Bulletin No 151, August 2012, pp.2-15, URL: https://earth.esa.int/c/document_library/get_file?folderId=131556&name-DLFE-3101.pdf
22) Jean-Michel Léger, "Spaceborne scalar magnetometers to map the Earth's magnetic field," CLEFS CEA, No 56 , Winter 2007-2008, URL: http://www.cea.fr/var/cea/storage/static/gb/library/Clefs56/pdf-gb/Clefs%2056_p100_102_LegerGB.pdf
24) E. F. Christensen, "Scientific Goals of Swarm," Proceedings of the First Swarm International Science Meeting, Nantes, France, May 3-5, 2006, WPP-261
25) N. Olsen, R. Haagmans, T. J. Sabaka, A. Kuvshinov, S. Maus, M. E. Purucker, M. Rother, V. Lesur, M. Mandea, "The Swarm End-to-End mission simulator study: A demonstration of separating the various contributions to Earth's magnetic field using synthetic data," Earth Planets Space (special issue on Swarm), Vol. 58, No 4, April 2006, pp. 359-370, URL: http://www.dsri.dk/~nio/papers/EPS_58040359.pdf
26) S. Maus., H. Lühr, M. Purucker, "Simulation of the high-degree lithospheric field recovery for the Swarm constellation of satellites," Earth Planets Space (special issue on Swarm), Vol. 58, No 4, April 2006, pp. 397-407, URL: http://www.geomag.us/info/Smaus/Doc/swarm_e2e_Maus.pdf
27) E. Thébault, "Global lithospheric magnetic field modelling by successive regional analysis," Earth Planets Space (special issue on Swarm), Vol. 58, No 4, April 2006, pp. 485-495, URL: http://www.terrapub.co.jp/journals/EPS/pdf/2006/5804/58040485.pdf
28) C. Manoj, A. Kuvshinov, S. Maus, H. Lühr, "Ocean circulation generated magnetic signals," Earth Planets Space (special issue on Swarm), Vol. 58, No 4, April 2006, pp. 429-437, URL: http://www.geomag.us/info/Smaus/Doc/Manoj_EPS_swarm.pdf
30) "Construction of Swarm satellites to begin," Nov. 8, 2007, URL: http://www.esa.int/esaLP/SEMQXVUOM8F_LPswarm_0.html
31) ESA's Swarm trio on its way to watch over our Planet's magnetic shiled," ESA, Nov. 22, 2013, URL: http://www.esa.int/Our_Activities/Observing_the_Earth/The_Living_Planet-Programme/Earth_Explorers/Swarm/ESA_s_Swarm_trio_on_its_way_to-watch_over_our_planet_s_magnetic_shield
32) W. Xiang, J. L. Joergensen, "Formation Flying: A Subject Being Fast Unfolding in Space," Proceedings of the 5th IAA Symposium on Small Satellites for Earth Observation, April 4-8, 2005, Berlin, Germany
33) Nils Olsen, Terence J. Sabaka, Luis R. Gaya-Pique, "Study of an Improved Comprehensive Magnetic Field Inversion Analysis for Swarm," Danish National Space Center, Scientific Report 1/2007, URL: http://esamultimedia.esa.int/docs/DNSC-Scientific-Report_1_2007.pdf
34) "EADS Astrium selected for Swarm Satellites," Nov. 17, 2005, URL: http://www.comspacewatch.com/news/viewpr.html?pid=18338
35) Albert Zaglauer, "Swarm Satellite Status," Proceedings of ESA's Second Swarm International Science Meeting, June 24-26, 2009, Potsdam, Germany, WPP-303, URL: http://www.congrex.nl/09c24/S1/S1_01_Zaglauer_presentation.pdf
36) Carlos Pereira, Christian Oram, Cornel Boesch, Sascha Svilar, Jacques Viertl, Andreas Fix, "Comparative Development of Dimensionally Stable Structures for the Deployable Sunshield Assembly of GAIA and Composite Tube Assembly of SWARM," Proceedings of IAC 2011 (62nd International Astronautical Congress), Cape Town, South Africa, Oct. 3-7, 2011, paper: IAC-11.C2.2.2
37) Carlos Pereira , Ian Rathband, "The Composite Tube Assembly of Swarm," Proceedings of the 12th European Conference on Spacecraft Structures, Materials & Environmental Testing, ESA/ESTEC, Noordwijk, The Netherlands, March 20-23, 2012, SP-691
38) Thomas Usbeck, Stefan v. d. Nüll, Domenico Reggio, "The SWARM AOCS - Tailored from the CryoSat AOCS to the Specifics of a Magnetic Mission," Proceedings of the 7th International ESA Conference on Guidance, Navigation & Control Systems (GNC 2008), June 2-5, 2008, Tralee, County Kerry, Ireland
39) Peter Zentgraf, Domenico Reggio, "Magnetic Rate Damping for Satellites in LEO," Proceedings of the 32nd AAS Guidance and Control Conference, Breckenridge, CO, USA, Jan. 31.- Feb. 4, 2009, AAS 09-074
40) Simon Edwards, Nicholas Solway, Peter Smith, "Development of a cold gas thruster for Swarm," Proceedings of Space Propulsion 2010, San Sebastian, Spain, May 3-6, 2010
41) Nikolaus Breier, Bernhard Kiewe, Olivier Mourra, "The Power Control and Distribution Unit for the Swarm Satellites," Proceedings of the 8th European Space Power Conference, Constance, Germany, Sept. 14-19, 2008, ESA SP-661, Sept. 2008
42) "Satellites packed like sardines," ESA, Nov. 8, 2013, URL: http://www.esa.int/Our_Activities/Observing_the_Earth/Swarm/Satellites_packed-like_sardines
43) Paul McMahon, Hans-Juergen Jung, Jeff Edwards, "Swarm Deployable Boom Assembly (DBA) Development of a Deployable Magnetometer Boom for the swarm Spacecraft," Proceedings of the 15th ESMATS (European Space Mechanisms and Tribology Symposium) 2013, Noordwijk, The Netherlands, Sept. 25-27, 2013, ESA, SP-718, URL: http://www.esmats.eu/esmatspapers/pastpapers/pdfs/2013/mcmahon.pdf
44) "ESA's Swarm trio on its way to watch over our planet's magnetic shield," ESA, Nov. 22, 2013, URL: http://www.esa.int/Our_Activities/Observing_the_Earth/Swarm/ESA_s_Swarm-trio_on_its_way_to_watch_over_our_planet_s_magnetic_shield
45) "Swarm Satellites Launched Successfully," Eurockot, Nov. 22, 2013, URL: http://www.eurockot.com/2013/11/swarm-satellites-launched-successfully/
46) "SWARM – Earth's magnetic field to be accurately measured by three European satellites flying in formation," DLR, Nov. 22, 2013, URL: http://www.dlr.de/dlr/en/desktopdefault.aspx/tabid-10081/151_read-8817/year-all/#gallery/13060
47) "Satellite trio to explore the Earth's magnetic field," Space Daily, Nov. 27, 2013, URL: http://www.spacedaily.com/reports/Satellite_trio_to_explore_the_Earths_magnetic-field_999.html
48) Marie-Helene Thibeault, "University of Calgary technology part of European Space Agency mission launched into space," University of Calgary, Nov. 22, 2013, URL: http://www.ucalgary.ca/utoday/issue/2013-11-22/university-calgary-technology-part-european-space-agency-mission-launched-space
49) "Swarm Mission Control ready for Tripple Launch," ESA, Nov. 19, 2013, URL: http://www.esa.int/Our_Activities/Operations/Swarm_mission_control_ready-for_triple_launch
50) "Preparing to launch Swarm," ESA, Sept. 13, 2013, URL: http://www.esa.int/Our_Activities/Observing_the_Earth/The_Living_Planet-Programme/Earth_Explorers/Swarm/Preparing_to_launch_Swarm
52) Peter Freeborn, York Viertel, Anna Zorina, "Meeting the Future Launch Demand for Earth Observation Missions," 8th IAA (International Academy of Astronautics) Symposium on Small Satellites for Earth Observation, Berlin, Germany, April 4-8, 2011; URL of presentation, IAA-B8-0501, http://media.dlr.de:8080/erez4/erez?cmd=get&src=os/IAA/archiv8/Presentations/IAA-B8-0501.pdf
54) Peter Freeborn, York Viertel, "Eurockot launch services for ESA Earth observation swarm satellites," Proceedings of the 63rd IAC (International Astronautical Congress), Naples, Italy, Oct. 1-5, 2012, paper: IAC-12-D2.2.10
55) Nils Olsen and the Swarm Mission Simulator Team, "Swarm Magnetic Field Models and Other Derived Data Products," IAGA 2005, Toulouse. France, July 19, 2005, URL: http://www.geos.ed.ac.uk/research/geospace/presentations/pub_store/serial110.pdf
57) "There's a jet stream in our core," ESA, Dec. 19, 2016, URL: http://m.esa.int/Our_Activities/Observing_the_Earth/Swarm/There_s_a-jet_stream_in_our_core
58) Philip W. Livermore, Rainer Hollerbach, Christopher C. Finlay, "An accelerating high-latitude jet in Earth's core," Nature Geoscience, Dec. 19, 2016, doi:10.1038/ngeo2859
59) "Swarm reveals why satellites lose track," ESA, Oct. 28, 2016, URL: http://m.esa.int/Our_Activities/Observing_the_Earth/Swarm/Swarm-reveals_why_satellites_lose_track
60) Chao Xiong, Claudia Stolle, Hermann Lühr, "The Swarm satellite loss of GPS signal and its relation to ionospheric plasma irregularities," Space Weather, Volume 14, Issue 8, August 13, 2016, pp: 563–577, DOI: 10.1002/2016SW001439, URL of abstract: http://onlinelibrary.wiley.com/doi/10.1002/2016SW001439/abstract
61) "Magnetic oceans and electric Earth," ESA, Oct. 3, 2016, URL: http://m.esa.int/Our_Activities/Observing_the_Earth/Swarm/Magnetic-oceans_and_electric_Earth
62) Alexander V. Grayver, Neesha R. Schnepf, Alexey V. Kuvshinov, Terence J. Sabaka, Chandrasekharan Manoj, Nils Olsen, "Satellite tidal magnetic signals constrain oceanic lithosphere-asthenosphere boundary," Science Advances, 30 Sept. 2016: Vol. 2, no. 9, e1600798, DOI: 10.1126/sciadv.1600798, URL: http://advances.sciencemag.org/content/advances/2/9/e1600798.full.pdf
63) N. R. Schnepf, A. Kuvshinov, T. Sabaka, "Can we probe the conductivity of the lithosphere and upper mantle using satellite tidalmagnetic signals?," Geophysical Research Letters, Vol. 42, article online 8 APR 2015, doi:10.1002/2015GL063540, URL: https://www.ethz.ch/content/dam/ethz/special-interest/erdw/geophysics/epm-dam/images/Tides/Schnepf_Kuvshinov_Sabaka_2015_GRL.pdf
64) Detlef Sieg, Frank Diekmann, "Options for Further Orbit Evolution of the Swarm Mission," Proceedings of the Living Planet Symposium 2016, Prague, Czech Republic, May 9-13, 2016 (ESA SP-740, August 2016)
65) Earth's magnetic heartbeat," ESA, May 10, 2016, URL: http://www.esa.int/Our_Activities/Observing_the_Earth/Swarm/Earth_s-magnetic_heartbeat
66) "Swarm constellation," ESA, >Oct. 15, 2013, URL: http://www.esa.int/spaceinimages/Images/2013/10/Swarm_constellation
67) "Swarm ASM-VFM Residual dataset," ESA, Dec. 22, 2015, URL: https://earth.esa.int/web/guest/missions/esa-operational-eo-missions/swarm/news
68) "Programs in Progress - Status at end of July 2015 -Swarm," ESA Bulletin No 163, 3rd quarter 2015, pp: 57-58, URL: http://esamultimedia.esa.int/multimedia/publications/ESA-Bulletin-163/offline/download.pdf
69) "Magnetic complexity begins to untangle," ESA, June 22, 2015, URL: http://www.esa.int/Our_Activities/Observing_the_Earth/Swarm/Magnetic-complexity_begins_to_untangle
70) P. Alken, S. Maus, A. Chulliat, P. Vigneron, O. Sirol,G. Hulot, "Swarm equatorial electric field chain: First results," Geophysical Research Letters, Volume 42, Issue 3,16 February 2015, pp: 673–680, doi: 10.1002/2014GL062658
71) Hermann Lühr, Jaeheung Park, Jesper W. Gjerloev, Jan Rauberg, Ingo Michaelis, Jose M. G. Merayo, Peter Brauer, "Field‐aligned currents' scale analysis performed with the Swarm constellation," Geophysical Research Letters, Volume 42, Issue 1, 16 January 2015, pp: 1–8, doi: 10.1002/2014GL062453
72) A. Spicher, T. Cameron, E. M. Grono, K. N. Yakymenko, S. C. Buchert, L. B. N. Clausen, D. J. Knudsen, K. A. McWilliams, J. I. Moen, "Observation of polar cap patches and calculation of gradient drift instability growth times: A Swarm case study," Geophysical Research Letters, Volume 42, Issue 2, 28 January 2015, pp: 201–206, doi: 10.1002/2014GL062590
73) Gauthier Hulot, Pierre Vigneron, Jean‐Michel Léger, Isabelle Fratter, Nils Olsen, Thomas Jager, François Bertrand, Laura Brocco, Olivier Sirol, Xavier Lalanne,Axel Boness, Viviane Cattin, "Swarm's absolute magnetometer experimental vector mode, an innovative capability for space magnetometry," Geophysical Research Letters, Volume 42, Issue 5, 16 March 2015, pp: 1352–1359, doi: 10.1002/2014GL062700
74) ESA Bulletin No 161, May 11, 2015, p. 74, URL: http://esamultimedia.esa.int/multimedia/publications/ESA-Bulletin-161/offline/download.pdf
75) Nils Olsen, Gauthier Hulot, Vincent Lesur, Christopher C. Finlay, Ciaran Beggan, Arnaud Chulliat, Terence J. Sabaka, Rune Floberghagen, Eigil Friis-Christensen, Roger Haagmans, Stavros Kotsiaros, Hermann Lühr, Lars Tøffner-Clausen, Pierre Vigneron, "The Swarm Initial Field Model for the 2014 geomagnetic field," Geophysical Research Letters, Vol. 42, doi:10.1002/2014GL062659.URL: ftp://ftp.space.dtu.dk/pub/cfinl/publications/2015_Olsen_Hulot_Lesur_Finlay_etal.pdf
76) "Swarm celebrates one year in orbit," ESA, Nov. 24, 2014, URL: https://earth.esa.int/web/guest/missions/esa-operational-eo-missions/swarm/news/-/article/swarm-celebrates-one-year-in-orbit
77) "Swarm," ESA Bulletin, No 160, November 2014, p. 73
78) Isabelle Fratter, Jean-Michel Leger, Francois Bertrand, Thomas Jager, Gauthier Hulot, Laura Brocco, Pierre Vigneron, "Swarm Absolute Scalar Magnetometers first in-orbit results," Proceedings of the 65th International Astronautical Congress (IAC 2014), Toronto, Canada, Sept. 29-Oct. 3, 2014, paper: IAC-14,B1,3.1
79) "Swarm reveals Earth's changing magnetism," ESA, June 19, 2014, URL: http://www.esa.int/Our_Activities/Observing_the_Earth/Swarm/Swarm_reveals-Earth_s_changing_magnetism
80) "Earth's ever-changing magnetic field," ESA, June 19,2014, URL: http://www.esa.int/spaceinvideos/Videos/2014/06/Earth_s_ever-changing_magnetic_field
81) "Swarm's precise sense of magnetism," ESA, May 7, 2014, URL: http://www.esa.int/Our_Activities/Observing_the_Earth/Swarm/Swarm_s_precise-sense_of_magnetism
82) R. Mackenzie, R. Bock, D. Kuijper, P. Ramos-Bosch, D. Sieg, G. Ziegler, "A Review of Swarm Flight Dynamics Operations from launch to routine phase," Proceedings of 24th International Symposium on Space Flight Dynamics, Laurel, Maryland, USA, 5 - 9 May, 2014, URL: http://issfd.org/ISSFD_2014/ISSFD24_Paper_S14-4_Mackenzie.pdf
83) "Swarm heads for new heights," ESA, Feb. 6, 2014, URL: http://www.esa.int/Our_Activities/Observing_the_Earth/Swarm/Swarm_heads-for_new_heights
84) "End of the beginning for Swarm Trio," ESA, Nov. 26, 2013, URL: http://www.esa.int/Our_Activities/Operations/End_of_the_beginning_for_Swarm_trio
85) J. M. G. Merayo, J. L. Jørgensen, E. Friis-Christensen, P. Brauer, F. Primdahl, P. S. Jørgensen, T. H. Allin, T. Denver, "The Swarm Magnetometry Package," Proceedings of the 6th IAA Symposium on Small Satellites for Earth Observation, Berlin, Germany, April 23 - 26, 2007
86) "Swarm - The Earth's Magnetic Field and Environment Explorers Technical and Programmatic Annex," ESA, Annex to ESA SP-1279(6) April 2004, URL: http://esamultimedia.esa.int/docs/EEUCM/SWARM_TPA.pdf
87) Peter S. Jørgensen, José Merayo, John L. Jørgensen, "Pre-Flight Calibration of the Swarm Vector Magnetometer Package," Proceedings of ESA's Second Swarm International Science Meeting, June 24-26, 2009, Potsdam, Germany, WPP-303, URL: http://www.congrex.nl/09c24/S1_Posters/S1_P01_Jorgensen_paper.pdf
88) Jose M. G. Merayo, "The Vector Field Magnetometer (VFM) on Swarm," Proceedings of ESA's Second Swarm International Science Meeting, June 24-26, 2009, Potsdam, Germany, WPP-303
89) Jos é M. G. Merayo, John L. Jørgensen, Eigil Friis-Christensen, Peter Brauer, Fritz Primdahl, Peter S. Jørgensen, Thomas H. Allin, Troelz Denver, "The Swarm Magnetometry Package," Springer Netherlands, April 2008, ISBN: 978-1-4020-6942-0
93) Jean-Michel Leger , François Bertrand, Thomas Jager, Matthieu Le Prado, Isabelle Fratter, Jean-Claude Lalaurie, "Swarm absolute scalar and vector magnetometer based on helium 4 optical pumping," Proceedings of ESA's Second Swarm International Science Meeting, June 24-26, 2009, Potsdam, Germany, WPP-303, URL: http://www.congrex.nl/09c24/S1/S1_05_Leger_paper.pdf
94) W. Fourcault, J-M. Léger, V. Costes, I. Fratter, L. Mondin, "Athermal Fiber Laser for the Swarm Absolute Scalar Magnetometer," ICSO 2010 (International Conference on Space Optics), Rhodes Island, Greece, Oct. 4-8, 2010, URL: http://smsc.cnes.fr/SWARM/PDF/LEGER_ICSO_PAPER.pdf
95) Isabelle Fratter, Jean-Michel Léger, François Bertrand, Thomas Jager, Matthieu Le Prado, William Fourcault, "Development and Space Qualification of the Swarm Absolute Scalar Magnetometer," IEEE sensors 2010, Waikoloa, HI, USA, Nov. 1-4, 2010, paper No 1291, URL: http://smsc.cnes.fr/SWARM/PDF/Sensors2010_paper_1291_FRATTER_2_IEEE.pdf
96) D. Ramecourt, T. Guettler Sergeant, T. Buret, C. Rougane, I. Fratter, D. Veyrié, O. Gilard, V. Costes, J. M. Léger, B. Charrat, "SWARM laser: qualification and integration of the optical fiber components," ICSO 2010 (International Conference on Space Optics), Rhodes Island, Greece, Oct. 4-8, 2010, URL: http://smsc.cnes.fr/SWARM/PDF/ICSO-DRA.pdf
97) I. Fratter, J.-C. Lalaurie, M Venet , J.-M. Léger, T. Jager, F. Bertrand, S. Moralès, "The Swarm Absolute Scalar Magnetometer : new features, capabilities and performances," 25th IUGG (International Union of Geodesy and Geophysics) General Assembly, Melbourne, Australia, June 28 to July 7, 2011, URL: http://smsc.cnes.fr/SWARM/PDF/IAGA_13.2_presentation.pdf
99) Jean-Michel Leger, Francois Bertrand, Thomas Jager, Isabelle Fratter, "Spaceborne scalar magnetometers for Earth's field studies," Proceedings of IAC 2011 (62nd International Astronautical Congress), Cape Town, South Africa, Oct. 3-7, 2011, paper: IAC-11-B1.3.9
100) T. Jager, J-M. Léger, F. Bertrand, I. Fratter, J-C. Lalaurie, "Swarm Absolute Scalar Magnetometer accuracy: analyses and measurement results," IEEE sensors 2010, Waikoloa, HI, USA, Nov. 1-4, 2010, paper No 1564, URL: http://smsc.cnes.fr/SWARM/PDF/Sensors2010_Paper_3.pdf
101) F. Alcouffe, F. Bertrand, T. Jager, M. Le Prado, J-M. Léger, I. Fratter, "The Swarm Absolute Scalar Magnetometer Magnetic Cleanliness Program," Proceedings of 2012 ESA Workshop on Aerospace EMC, Venice, Italy, May 21-23, 2012, SP-702
102) T. Jager, J. M. Léger, I. Fratter, P. Lier, P. Pacholczyk, "Magnetic Cleanliness and thermomagnetic effect: Case staudy of the Absolute Scalar Magnetometer and its environment on Swarm satellites," Proceedings of the 2016 ESA Workshop on Aerospace EMC(Electromagnetic Compatibility), Valencia, Spain, May 23-25, 2016, (ESA SP-738, May 2016)
103) Brian Moffat, John Hackett, David Knudsen, Jan-Erik Wahlund, Lennart Åhlén, Nico Stricker, "Overview of the Canadian Electric Field Instrument (CEFI) for Swarm," Proceedings of the 13th Canadian Astronautics Conference, ASTRO 2006, Montreal, QC, Canada, organized by CASI (Canadian Astronautics and Space Institute), April 25-27, 2006
105) David Knudsen, Johnathan Burchill, Laureline Sangalli, "Electric Fields from Ion Distribution Images on Swarm," IPELS (International Workshop on the Interrelationship between Plasma Experiments in Laboratory and Space), Aug. 6-10,, 2007, Cairns, Australia, URL: http://wwwrsphysse.anu.edu.au/~web112/ipels2007/knudsen-ipels2007_invited.pdf
106) David Knudsen, Johnathan Burchill, Anders Eriksson, "Swarm Electric Field Instrument and Langmuir Probe," Proceedings of ESA's Second Swarm International Science Meeting, June 24-26, 2009, Potsdam, Germany, WPP-303
107) R. Marchand, J. Burchill, D. Knudsen, "Modelling sheath effects and particle distributions in the Swarm EFI," Proceedings of ESA's Second Swarm International Science Meeting, June 24-26, 2009, Potsdam, Germany, WPP-303, URL: http://www.congrex.nl/09c24/S1_Posters/S1_P03_Marchand_presentation.pdf
108) J. K. Burchill, D. J. Knudsen, "Error Sources for the EFI Ion Velocity, Ion Temperature, and Electric Field L1B Products," Proceedings of ESA's Second Swarm International Science Meeting, June 24-26, 2009, Potsdam, Germany, WPP-303, URL: http://www.congrex.nl/09c24/S1_Posters/S1_P02_Burchill_paper.pdf
109) Radek Peresty, Milan Chvojka, Viktor Fedosov, "Use of the highly sensitive electrostatic accelerometer for orbit perturbation effects investigation on board of LEO spacecraft," Proceedings of the 61st IAC (International Astronautical Congress), Prague, Czech Republic, Sept. 27-Oct. 1, 2010, IAC-10.B1.3.2
112) Rune Floberghagen, "Ground Segment and Data Products," Proceedings of ESA's Second Swarm International Science Meeting, June 24-26, 2009, Potsdam, Germany, WPP-303, URL: http://www.congrex.nl/09c24/S1/S1_07_Floberghagen_presentation.pdf
113) Frank-Jürgen Diekmann, Ignacio Clerigo, Giuseppe Albini, Laurent Maleville, Alessandro Neto, David Patterson, Ana Piris Niño, Detlef Sieg, "A Challenging Trio in Space 'Routine' Operations of the Swarm Satellite Constellation," Proceedings of the Living Planet Symposium 2016, Prague, Czech Republic, May 9-13, 2016 (ESA SP-740, August 2016)
114) R. Mackenzie, D. Kuijper, P. Ramos-Bosch, D. Sieg, G. Ziegler, "A review of Swarm flight dynamics operations from launch to routine phase", 24th International Symposium on Space Flight Dynamics, May 9, 2014
115) D. Sieg, F. J. Diekmann, "Options for the further orbit evolution of the Swarm mission", Proceedings of the Living Planet Symposium 2016, Prague, Czech Republic, May 9-13, 2016 (ESA SP-740, August 2016)
The information compiled and edited in this article was provided by Herbert J. Kramer from his documentation of: "Observation of the Earth and Its Environment: Survey of Missions and Sensors" (Springer Verlag) as well as many other sources after the publication of the 4th edition in 2002. - Comments and corrections to this article are always welcome for further updates (firstname.lastname@example.org).