Euclid — Mapping the geometry of the dark Universe
Euclid is a medium-class ("M-class") mission and is part of ESA's "Cosmic Vision" (2015–2025) scientific program. Euclid was chosen in October 2011 together with Solar Orbiter, out of several competing missions. The mission will investigate the distance-redshift relationship and the evolution of cosmic structures. It achieves this by measuring shapes and redshifts of galaxies and clusters of galaxies out to redshifts ~2, or equivalently to a look-back time of 10 billion years. It will therefore cover the entire period over which dark energy played a significant role in accelerating the expansion. 1) 2)
The Greek mathematician Euclid (born ~300 BCE, Alexandria, Egypt), is the most prominent mathematician of Greco-Roman antiquity, best known for his treatise on geometry, the Elements. His Elements is one of the most influential works in the history of mathematics.
Until about thirty years ago astronomers thought that the Universe was composed almost entirely of ordinary matter: protons, neutrons, electrons and atoms. In the intervening years the emerging picture of the composition of the Universe has changed dramatically. It is now assumed that ordinary matter makes up only about 4% of the Universe, and that the mass-energy budget of the Universe is actually dominated by two mysterious components: dark energy and dark matter.
Dark energy, which accounts for the vast majority (76%) of the energy density of the Universe, is causing the expansion of the Universe to accelerate. The existence and energy scale of dark energy cannot be explained with our current knowledge of fundamental physics.
Figure 1: The Universe evolves from a homogeneous state after the big bang through cooling and expansion. The small initial inhomogeneities grow through gravity to produce the large-scale structures that we see today (image credit: Euclid Assessment Study Report) 3)
The remaining 20% of the energy density is in the form of dark matter, which, like ordinary matter, exerts a gravitational attraction, but unlike normal matter does not emit light. The nature of dark matter is unknown, although several candidates are predicted by supersymmetric extensions of the standard model of particle physics. Plausible candidates for the cold dark matter are the axion and the lightest supersymmetric particle; massive neutrinos can account for the hot dark matter. One possibility to explain one or both of these puzzling components is that Einstein's Theory of General Relativity, and thus our understanding of gravity, needs to be revised on cosmological scales. Together, dark energy and dark matter pose some of the most important questions in fundamental physics today.
Euclid's cosmological probes: Euclid will map the large-scale structure of the Universe over the entire extragalactic sky - or half of the full sky excluding the regions dominated by the stars in our Milky Way. It will measure galaxies out to redshifts of ~2, which corresponds to a look-back time of about 10 billion years, thus covering the period over which dark energy accelerated the expansion of the Universe.
Euclid is optimized for two primary cosmological probes:
1) Weak gravitational Lensing (WL): Weak lensing is a method to map the dark matter and measure dark energy by measuring the distortions of galaxy images by mass inhomogeneities along the line-of-sight.
2) Baryonic Acoustic Oscillations (BAO): BAOs are wiggle patterns, imprinted in the clustering of galaxies, which provide a standard ruler to measure dark energy and the expansion in the Universe.
Weak gravitational lensing requires extremely high image quality because possible image distortions by the optical system must be suppressed or calibrated-out to be able to measure the true distortions by gravity.
Figure 2: Left: Illustrations of the effect of a lensing mass on a circularly symmetric image. Weak lensing elliptically distorts the image, flexion provides an arc-ness and strong lensing creates large arcs and multiple images. Right: Galaxy cluster Abell 2218, strongly lensed arcs can be seen in around the cluster. Every background galaxy is weakly lensed [image credit: Credit for Abell 2218: NASA, ESA, and Johan Richard (Caltech, USA)] 4)
The Euclid baryonic acoustic oscillations experiment involves the determination of the redshifts of galaxies to better than 0.1%, this can only be accomplished through spectroscopy.
Euclid is optimized to tackle some of the most important questions in modern cosmology: How did the Universe originate and why is it expanding at an accelerating rate, rather than slowing down due to the gravitational attraction of all the matter in it?
The discovery of this cosmic acceleration in 1998 was rewarded with the Nobel Prize for Physics in 2011 and yet we still do not know what causes it. The term 'dark energy' is often used to signify this mysterious force, but by using Euclid to study its effects on the shapes and locations of galaxies across the Universe, astronomers hope to come much closer to understanding its true nature and influence.
The Euclid Consortium, Partnerships and Euclid Flagship mockup galaxy catalog:
The Euclid Consortium is an organization that brings together teams of researchers in theoretical physics, particle physics, astrophysics and space astronomy, and also engineers, technicians, and management and administrative staffs working in public research laboratories and contributing to the Euclid mission. Together with ESA (European Space Agency) and aerospace industry, they are part of the Euclid Collaboration. 5)
The Euclid Consortium has been selected by ESA in June 2012 as the single official scientific consortium having the responsibility of the scientific instruments, the production of the data and of leading the scientific exploitation of the mission until completion. It is funded by national space agencies and national research organizations and led by the Euclid Consortium Lead (ECL) and a Euclid Consortium Board (ECB). The ECL and ECB are the primary contact points with ESA and with the Euclid Consortium members.
The Euclid Consortium is responsible of the definitions of the scientific goals, the mission concept, the science requirements and the survey. It is also in charge of the design, construction, tests, integration and delivery to ESA of the imaging and spectroscopic instruments VIS and NISP. The Euclid Consortium Science Ground Segment (EC SGS) is responsible of the design, development tests, integration and operation of the data processing tools, pipelines and data centers. It shares with the Euclid Consortium Science Working Groups (SWG) the responsibility of the scientific production and delivery of Euclid data releases, and their scientific analysis and interpretation. The Euclid Consortium COMS group organises and coordinates all Euclid Consortium communication activities.
Partnerships: In total, about 1500 scientists are registered in the EC, of which more than 900 are researchers in astrophysics, cosmology, theoretical physics and particle physics. More than 200 laboratories covering all fields in astrophysics, cosmology, theoretical physics, high energy, particle physics and space science that are relevant for the Euclid missions are contributing to Euclid. The Euclid Consortium comprises scientists from 14 European countries: Austria, Belgium, Denmark, France, Finland, Germany, Italy, Netherlands, Norway, Spain, Switzerland, Portugal, Romania and the UK. Canada and the USA through NASA and a few US laboratories are also contributing and members of the Euclid Consortium. — The Euclid Consortium meets once a year for its annual Euclid Consortium Meeting.
Figure 3: Members of the Euclid consortium at the Euclid Meeting in London, June 05-08, 2017, (Photo credit: LOC of the London Euclid Consortium Annual Meeting)
A key ingredient in order to prepare the scientific exploitation of this "golden" dataset is the development of synthetic observations of the real survey: a cosmological simulation that matches the expected volume and complexity of the real data. In a massive coordinated effort, a team of scientists of the Euclid project have worked together over the last year to develop the largest simulated galaxy catalogue ever produced, the so-called Euclid Flagship mockup galaxy catalog. 6)
The Euclid Flagship mockup galaxy catalog is based on the record-setting 2 trillion (2 x 1012) dark-matter particle simulation performed on the supercomputer Piz Daint, hosted by the Swiss National Supercomputing Center (CSCS). The simulation code was developed by a team of scientists at the University of Zurich, led by Joachim Staled from the Institute for Computational Science. This unique dataset reproduces with exquisite precision the emergence of the large scale structure of the Universe, with hundred of billions of dark matter halos hosting the galaxies we see in the night sky today.
Using this dark-matter cosmic web from the Flagship simulation, a group of scientists of the Euclid Consortium at Institut de Ciències de L'Espai (ICE, IEEC-CSIC) and Port d'Informació Científica (PIC) in Barcelona, in collaboration with the Cosmological Simulations Working Group, led by Pablo Fosalba (ICE, IEEC-CSIC) and Romain Teyssier (Institute for Computational Science at the University of Zurich), have built a synthetic galaxy catalog using state-of-the-art scientific pipelines that implement the Halo Occupation Distribution technique, a sophisticated recipe to relate dark and luminous matter in the universe.
The Euclid Flagship mockup galaxy catalog contains more than 2 thousand million galaxies distributed over the 3D cosmological volume that Euclid will survey. Synthetic galaxies in this simulation mimic with great detail the complex properties that real sources display: ranging from their shapes, colors, luminosities, and emission lines in their spectra, to the gravitational lensing distortions that affect the light emitted by distant galaxies as it travels to us, the observers. A dedicated web portal, CosmoHub hosted by PIC, the Euclid Spanish Science Data Center, will distribute the Flagship mockup data to the 1000+ members of the Euclid Consortium.
Armed with this new virtual universe, scientists will be able to assess the performance of the Euclid mission as a whole, the so-called Science Performance Verification. The Science Performance Verification exercise uses a full end to end simulation of the Euclid mission developed by ESA and the Euclid Consortium and represents a critical milestone of the project. Moreover the Euclid Flagship mockup will be an essential tool to develop the data processing and the science analysis pipelines developed by the Euclid Science ground Segment and the Science Working Groups and will set the science for the exciting discoveries that await the Euclid mission when the real data shall come.
Figure 4: Euclid Flagship mockup galaxy catalog: False color images showing a small portion (0.3%) of the full light-cone simulation of mockup galaxies in the Euclid survey. Light-cone stripes extend 500 Mpc/h (vertical) x 3800 Mpc/h (horizontal axis). The 2D "pencil beam" images result from a slice of the 3D light-cone, projected from a 40 Mpc/h width (in the direction orthogonal to the image plane). From top to bottom, panels display the full sample of galaxies in the mockup, and the sub-samples expected from observations in the VIS and NISP-Halpha channels. The galaxy mockup has been produced using a Halo Occupation Distribution pipeline developed by the Institut de Ciències de l'Espai (ICE) and Port d'Informació Científica (PIC) in Barcelona, it is based on the 2 trillion dark-matter particle Flagship run produced by U. Zurich (image credit: J. Carretero/P. Tallada/S. Serrano for ICE/PIC/U.Zurich and the Euclid Consortium Cosmological Simulations SWG)
Figure 5: Euclid Flagship mockup galaxy catalog: Galaxy Types in the Flagship Mock Catalog — Top panel: False color images showing a small portion (0.3%) of the full light-cone simulation (similar to Image 1), but showing different galaxy types with different colors. Central galaxies are colored in green, and satellites in red. Bottom panel: zoom in of the top panel image that displays the local universe with greater detail. Central galaxies populate all dark-matter halos of the cosmic web, whereas satellite galaxies tend to reside in the most massive halos, that is, in the highest density peaks of the underlying dark-matter distribution (image credit: J. Carretero/P. Tallada/S. Serrano for ICE/PIC/U.Zurich and the Euclid Consortium Cosmological Simulations SWG)
Euclid Science Case (Ref. 7):
Euclid is ESA's next cosmology mission after Planck. While Planck mapped the structure and the properties of the early Universe at a redshift of z~1100, Euclid will map the evolution of cosmic structures from z~2 until now. Euclid uses the clustering of matter, which includes both dark and luminous matter, to measure the accelerated expansion of the Universe at different cosmological times. The accelerated expansion is attributed to a substance of unknown nature, dubbed Dark Energy (DE). Euclid aims at measuring the equation of state parameter w(z), relating the DE ‘fluid' pressure with its density, to an accuracy of 2% for the constant component, and 10% for a possible variation as a function of redshift. If w(z)= constant =-1, then the universe can be described by General Relativity with an additional cosmological constant providing the accelerated expansion and Cold Dark Matter for the mass (ΛCDM). However at the moment such a description cannot be reconciled with the standard model of particle physics. A significant non-zero variation would mean that either a new "dark energy" component with exotic physical properties exists in the Universe or our understanding of General Relativity needs to be revisited. It is the accuracy of the Euclid measurements that will put strong constraints on any fundamental physics theory.
Euclid will use a number of cosmological probes to measure the clustering properties but is optimised for two methods: (1) Galaxy Clustering (GC): measurement of the redshift distribution of galaxies from their Hα emission line survey using near-infrared slitless spectroscopy and (2) Weak Lensing (WL): measurement of the distortion of the galaxy shapes due to the gravitational lensing caused by the - predominantly dark - matter distribution between distant galaxies and the observer. The resulting galaxy shear field can be transformed into the matter distribution. This is done in a number of redshift bins to derive the expansion of the dark matter as a function of redshift. In addition, GC provides also direct information of the validity of General Relativity because we can monitor the evolution of structures subject to the combined effects of gravity, which forces clumping of matter, and the opposing force caused by the accelerated expansion. GC maps the distribution of the luminous, baryonic matter whereas WL measures the properties of the combination of both luminous and dark matter. The complementarity of the two probes will provide important additional information on possible systematics, which limit the accuracy of each of the probes.
The mission will address the following items and associated key questions:
• Is the Dark Energy simply a cosmological constant, or is it a field that evolves dynamically with the expansion of the Universe?
• Alternatively, is the apparent acceleration instead a manifestation of a breakdown of General Relativity on the largest scales, or a failure of the cosmological assumptions of homogeneity and isotropy?
• What is Dark Matter? What is the absolute neutrino mass scale and what is the number of relativistic species in the Universe as opposed to the structures predicted by the cold (non-relativistic) dark matter component?
• What are the initial conditions after the Big Bang, the power spectrum of primordial density fluctuations, which seeded large-scale structure, and are they described by a Gaussian probability distribution?
In order to accomplish the above science objectives, Euclid must survey a large fraction of the sky: 15,000 deg2, or 36% of the celestial sphere and image billions of galaxies out to z~2 up to an AB magnitude of 24.5 for 10σ extended objects in the visible band and of 24 for 5σ point sources in the NIR bands. The mission lifetime has been set to 6 years (6 months margin included), hence a telescope with a stable, large field of view is required.
Euclid is a spaceborne optical/near-infrared survey mission of ESA to investigate the nature of dark energy, dark matter and gravity by observing the geometry of the Universe and on the formation of structures over cosmological timescales. Euclid will use two probes of the signature of dark matter and energy: Weak gravitational Lensing, which requires the measurement of the shape and photometric redshifts of distant galaxies, and Galaxy Clustering, based on the measurement of the 3D distribution of galaxies through their spectroscopic redshifts. The mission is scheduled for launch in 2020 and is designed for 6 years of nominal survey operations. 7)
The Euclid Spacecraft is composed of a Service Module (SVM) and a Payload Module (PLM). The Service Module comprises all the conventional spacecraft subsystems, the instruments warm electronics units, the sun shield and the solar arrays. In particular the Service Module provides the extremely challenging pointing accuracy required by the scientific objectives. The Payload Module consists of a 1.2 m three-mirror Korsch type telescope and of two instruments, the visible imager and the near-infrared spectro-photometer, both covering a large common field-of-view enabling to survey more than 35% of the entire sky. All sensor data are downlinked using K-band transmission and processed by a dedicated ground segment for science data processing. The Euclid data and catalogs will be made available to the public at the ESA Science Data Center.
The Sunshield, part of the SVM, protects the PLM from illumination by the sun and supports the photovoltaic assembly supplying electrical power to the spacecraft. The overall spacecraft envelope, compatible with the Soyuz ST fairing, fits within a diameter of 3.74 m and a height of 4.8 m (Figure 6).
Figure 6: Euclid spacecraft overview (image credit: Euclid Consortium)
Mechanical and Thermal Architecture: The SVM (Figure 7) is an irregular hexagonal base built around a central cone that provides the interfaces with the launcher and with the PLM and encloses the Hydrazine and Cold Gas propellant tanks. External equipment attached to the SVM include a high-gain antenna, three low-gain antennas, hydrazine and cold-gas thrusters, on dedicated pods to enhance thrust efficiency, and sun sensors.
Figure 7: SVM overview (the central cone aperture) is sealed by MLI belonging to the SVM, not shown here (Euclid Consortium)
The SVM accommodates the equipment grouped on six side panels (Figure 5) according to functions: (Telemetry and Telecommand, (TT&C), Attitude and Orbit Control (AOCS), Central Data Management (CDMS) and Electric Power (EPS), payload and Fine Guidance Sensor (FGS) warm electronics). Each panel can be individually dismounted to ease the integration of equipment and their access. The SVM footprint and the sunshield are designed to keep the PLM in the shadow within the allowed range of variation of the sun direction in the spacecraft reference frame. The PLM is structurally connected to the SVM via three glass-fiber bipods attached to the SVM in six points along the central cone upper ring of 2.25 m diameter and in three points to the baseplate of the PLM. This connection forms an isostatic mount preventing the SVM induced distortion to affect the PLM structure. The lower ring of the central cone is connected to the launcher vehicle adapter interface, a standard 1666 mm adapter, and locked at launch with a low shock separation clamp band.
The Sun Shield (SSH) consists of a carbon fiber reinforced plastic (CFRP) frame made of 2 vertical poles with diagonal stiffeners and 2 struts slanting toward the SVM. The front side carries the photovoltaic assembly in 3 identical panels, while the rear side is covered by Kapton MLI. On the top edge, an optical baffle consisting of 3 blades with decreasing height has the function of attenuating the sunlight diffracted towards the PLM down to negligible levels. A dedicated feature is embedded on a corner of the SSH to provide extra radiation shielding to the VIS Instrument focal plane.
The thermal control is based on a passive design using radiators, multilayer insulation (MLI) and heaters operated in Pulse Width Modulation. The design drivers are the short-term temperature stability of the PLM conductive and radiative interface under the maximum commanded Solar Aspect Angle (SAA) change, and minimal (<25 mW) heat flux into the coldest NISP radiator. High performance Kapton MLI is installed on the on SVM top floor, PLM bottom and Sun Shield rear side to minimise the heat flux and thermal disturbances onto the PLM.
Figure 8: SVM equipment accommodation (image credit: Euclid Consortium)
Electrical and Data Handling Architecture: The spacecraft provides 28V regulated power to equipment and instruments electronics through protected lines individually commandable provided by the on-board PCDU (Power Conditioning and Distribution Unit). The PCDU also provides power to the heaters, to the pyro actuators and controls the charge and discharge of the battery. The battery is used only during the launch phase and is design to provide up to 419 W of power and a total energy of 300 Whr. In the other phases of the mission the sun shield three panels provide a power between 2430 and 1780 W depending from the spacecraft orientation and ageing of the panels.
One centralised on-board computer CDMU (Command and Data Management Unit) provides spacecraft and AOCS command, control and data processing. The CDMU is a modular unit including standard core boards plus dedicated I/O boards to interface AOCS and spacecraft units and devices. The Processor Module is based on a general-purpose space qualified microprocessor (LEON-FT) with minimum computational power of 40 MIPS and 5 MFLOPS. Two processor modules are comprised in a single-failure tolerant unit.
The number of scientific exposures and high-resolution images generate a high science data volume and require large on-board memory capable of hosting the 850 Gbit of daily generated data. The on-board MMU (Mass Memory Unit) has a capacity of 4 Tbit EoL sufficient to store 72 hours of scientific data and 20 days of spacecraft housekeeping data. The MMU stores instruments data and housekeeping and other ancillary data in named files organized in a two level folders' structure.
The commands and telemetries are distributed and collected mostly via two standard Mil-Std-1553 buses, one dedicated to the spacecraft equipment and another to the instruments and mass memory, although some spacecraft equipment have dedicated connections. The instruments deliver high volume scientific data via high speed SpaceWire links directly into the mass memory. The platform bus handles non-packet remote terminals (RT) and the FGS, and is characterised by cyclic communication frames at 10 Hz, linked to the AOCS control cycle. The transfer layer protocol of the science bus is based on a cyclical communication frame at 60 Hz, maximising the efficiency of data transfer per communication frame.
Files stored in the mass memory are downloaded using the standard CCSDS File Delivery Protocol (CFDP) using the reliable transfer with acknowledges for the downlink and the simple unreliable transfer for uplink. Both the X- and K-band communication link can be used for the file transfer. The baseline configuration expect the directives of the CFDP to be transmitted via X-band to ensure visibility of the file downlink in progress by Ground also in case of adverse weather conditions, while data are downloaded via K-band to maximize the data rate. Any other combination is however possible.
Figure 9: CDMS interfaces (image credit: Euclid Consortium)
RF communications: The telecommunications architecture includes two independent sections: an X-band section used for telecommands,
The K-band section supports downlink of recorded science and housekeeping telemetry at two different data rates: nominal at 73.85 Mbit/s and reduced at 36.92 Mbit/s to allow 3 dB extra link budget margin in adverse weather. The K-band section features two K-band modulators operated in cold redundancy, directly connected with the mass memory unit for the download of files. A 70 cm diameter K-band HGA (High Gain Antenna) is installed on a two-degree-of-freedom pointing mechanism with angular range of ±55° in azimuth and -70° to + 40° in elevation.
AOCS (Attitude and Orbit Control): The Euclid image quality requirements demand very precise pointing and small jitter, while the survey requirements call for fast and accurate slews. The AOCS provides 75 mas (milli-arcseconds) Relative Pointing Error (RPE) over 700 sec and 7.5 arcsec of Absolute Pointing Error both with 99.7% Confidence Level. A FGS (Fine Guidance Sensor) with 4 CCD sensors collocated within the VIS imager focal plane provides the fine attitude measurement. Cold gas thrusters with micro-newton resolution provide the forces used to actuate the fine pointing. The Gyro and FGS-based attitude control rejects the low frequency noise (less than 0.1 Hz) ensuring that the RPE requirement is met. Three star trackers (STR) provide the inertial attitude accuracy to comply with the APE requirement. The STRs are mounted on the SVM and thus subject to thermoelastic deformation when large slews are executed. To solve this problem, the FGS will be endowed with absolute pointing capabilities (based on a reference star catalogue), allowing cross-calibration of STR and FGS reference frames. A high performance gyroscope is included to reduce high frequency attitude estimation noise, to manage FGS delays and recovery from temporary outages. Four reaction wheels execute all the slews (field slews, 50-100 arcsec dithers, and large slews between different sky zones). After each slew maneuver the wheels are controlled to slow down until friction stops them. Keeping the reaction wheels at rest during operation ensures noise-free science exposures by eliminating the micro-vibration associated to reaction wheel actuation.
The micro-propulsion employed for fine attitude control is based on cold-gas Nitrogen thrusters in a configuration of two branches with six thrusters each. Four high-pressure tanks provide storage of 70 kg Nitrogen, sufficient for 7 years operation with nearly 100% margin.
Orbit control and attitude control in non-science modes are actuated by two redundant branches of ten 20 N hydrazine thrusters. In each branch, two thrusters, one on either side of the spacecraft, provide torque-free thrust for the Trajectory Control Maneuvers on the way to SEL2, monthly Station Keeping Maneuvers at SEL2, and disposal at end of life. The other eight thrusters provide force-free torques for angular momentum and attitude control in non-science modes. Hydrazine storage is provided by one central tank with 137.5 kg propellant mass capacity with 10% volume margin over and above the prescribed ΔV margins.
PLM (Payload Module)
The Euclid PLM is designed around a three mirrors anastigmatic Korsch Silicon Carbide (SiC) telescope feeding the two instruments, VIS and NISP. The light separation between the two instruments is performed by a dichroic plate located at exit pupil of the telescope. The PLM is in charge of providing mechanical and thermal interfaces to the instruments (radiating areas and heating lines). Whereas NISP is a stand-alone instrument with interface bipods, VIS is delivered in several separate parts: a FPA (Focal Plane Assembly) connected to proximity electronics, readout shutter unit and calibration unit, with dedicated mechanical and thermal interfaces with PLM.
The secondary mirror (M2) is mounted on a mechanism (M2M) for 3-DOF adjustment to compensate for launch and cool-down effects. In addition, the PLM hosts the FGS, used as pointing reference by the AOCS. All these detectors are mounted on the structure carrying the VIS focal plane, in order to ensure precise co-alignment.
Except for proximity electronics of the focal planes and FGS, all electronics are placed on the SVM to minimise thermal disturbances to the PLM.
Figure 10: Schematic functional view of the Euclid PLM (image credit: Euclid Consortium)
Overall PLM architecture
The PLM is divided in two cavities, separated by the base plate:
• The front cavity including the telescope primary and secondary mirrors as well as the M2 refocusing mechanism and the associated support structure. This cavity is thermally insulated by a baffle.
• The instrument cavity including the telescope folding mirrors, the tertiary mirror, the dichroic splitter, the FGS, the two instruments (VIS and NISP), the shutter and calibration source for the VIS channel.
The PLM mechanical architecture is based on a common SiC baseplate which supports on one side telescope M1 and M2 mirrors and on other side the other optics and the two instruments. This architecture is well adapted to the selected thermal architecture with telescope and instrument cavities both passively controlled at neighbor temperatures.
The optical accommodation on the baseplate consists in implementing two folding mirrors (FoM1 and FoM2) at the entrance of the instrument cavity (between M2 and M3) to fold the optical beam in the plane of the baseplate. A third folding mirror (FoM3) allows having the VIS instrument close to an efficient radiative area.
Figure 11: External front view of the PLM (image credit: Euclid Consortium)
Figure 12: External front view of the PLM without the external baffle (image credit: Euclid Consortium)
Figure 13: Rear view of the PLM showing the instrument cavity (image credit: Euclid Consortium)
PLM Optical design: The optical combination is a Korsch telescope design with a FoV offset of 0.47°. The useful pupil diameter is 1.2 m and the focal length is 24.5 m, i.e. plate scale of 8.3 mas/µm. The central obscuration has been minimized by designing a thin spider and careful design of M1 and M2 baffles. The resulting collecting area is larger than 1 m2.
The challenging spectral transmission and the VIS out-of-band rejection requirements are met by complementing the dichroic plate with highpass filter on FoM1 and FoM2 and bandpass filter on FoM3.
Table 1: Specification of the telescope
PLM Thermal design: The passive thermal concept is adapted from the one successfully designed for Gaia PLM, requiring minimum heating power and providing best thermal stability. The telescope is cooled-down to its equilibrium temperature around 130 K. In nominal operations, only local heating capacity at constant power is needed to adjust instruments interface temperatures to the prescribed value. The required heating power in operational mode is therefore very low, ~140 W. Additional heating lines are installed for optics decontamination during commissioning phase and for survival mode, to keep instruments in their non-operational temperature range.
This cold telescope offers high thermo-elastic stability (SiC CTE is reduced to 0.4 µm/m/K) and cold environment (135 K) to the instruments. The heat leaks to cold instruments units are minimised (actually the PLM becomes a thermal sink for all units except NISP detector and VIS-FPA electronics which are connected to out-looking radiators). Specific harness design and highly decoupled conductive and radiative thermal interfaces allow minimising the heat leaks from the SVM and the sunshield.
Image quality performances: Thanks to the high thermal and dimensional stability of the telescope, and to the capacity for accurate in-orbit alignment, the Euclid PLM design offers excellent image quality performances, with large margins on the visible PSF (Point Spread Function) ellipticity, Full Width at Half Maximum (FWHM) and near Infrared Encircled Energy (EE) radius as shown in Table 2 and Table 3.This allows improved mission performances and offers flexibility for relaxed image motion requirements, i.e. reduced pointing stability and larger tolerance to micro vibrations.
Table 2: VIS image quality performance
Table 3: NISP image quality performance
Euclid mission development status:
• 23 October 2017: Technical problems discovered during ground testing of U.S.-built detectors for the European Space Agency's Euclid astronomy mission will delay the completion of the telescope's scientific payload, jeopardizing the observatory's 2020 launch target, the head of NASA's astrophysics division said last week. 8)
- Officials expect the problem, traced to an electronics package, will delay assembly of the detectors with Euclid's NISP (Near-Infrared Spectrometer and Photometer) instrument at least 12 months.
• 26 April 2017: ESA's Euclid mission has passed another important milestone with the delivery of the first three HgCdTe state-of-the art detectors for the Near-Infrared Spectrometer and Photometer instrument. When completed, the NISP instrument will include 16 of these detectors. Each of them is composed of 2040 x 2040 pixels, 18 µm in size. 9)
Figure 14: Photo of an Euclid NISP detector (image credit: CPPM (Centre de Physique des Particules Marseille))
• 07 February 2017: The first flight hardware, in the form of four detectors known as CCDs (Charge Coupled Devices), has been delivered to MSSL (Mullard Space Science Laboratory) by the UK company e2v. Each large area CCD comprises 4096 x 4132 pixels. The VIS instrument is being built by a consortium of nationally funded institutes led by MSSL. 10)
- The delivered CCDs will soon undergo extensive calibration with the flight electronics at MSSL in Surrey before being integrated into the VIS focal plane at CEA in Paris. Meanwhile, the CDR (Critical Design Review) for the VIS instrument is underway and this will be followed by the CDR for the telescope later in the year. The VIS instrument will be integrated into the telescope at Airbus facilities in Toulouse in 2018. 11)
Figure 15: Euclid VIS CCD device (image credit: e2v)
• 17 December 2015: Euclid, ESA's dark Universe mission, has passed its preliminary design review, providing confidence that the spacecraft and its payload can be built. It's time to start 'cutting metal'. 12)
- "This is really a big step for the mission," says Giuseppe Racca, Euclid's project manager. "All the elements have been put together and evaluated. We now know that the mission is feasible and we can do the science."
- First proposed to ESA in 2007, Euclid was selected as the second medium-class mission in the Cosmic Vision program in October 2011. Italy's Thales Alenia Space was chosen as the prime contractor in 2013.
• 29 November 2013: Multilateral agreements for Euclid and Gaia have been signed at the Science Program Committee meeting, held on 28 November 2013 at ESA Headquarters in Paris, France. 13)
- This agreement is between the ESA and the funding agencies of the eleven European countries participating in the Euclid Mission Consortium: the Agenzia Spaziale Italiana (Italy); the Centre National d'Etudes Spatiales (France); the Deutsches Zentrum für Luft- und Raumfahrt e.V. (Germany); the Danish Space Research Institute (Denmark); the Fundação para a Ciência e a Tecnologia (FCT), Space Office (Portugal); the Ministerio de Economía y Competividad (Spain); the Nederlandse Onderzoekschool Voor Astronomie (The Netherlands); the Norwegian Space Centre (Norway); the Romanian Space Agency (Romania); the United Kingdom Space Agency (UK); and the University of Helsinki (Finland). Switzerland participates via PRODEX.
- The infrared sensors for the Near-Infrared Spectroscopy and Photometry Instrument (NISP) will be provided by NASA in line with a Memorandum of Understanding signed in January 2013.
• 24 January 2013: NASA has joined the European Space Agency's (ESA's) Euclid mission, a space telescope designed to investigate the cosmological mysteries of dark matter and dark energy. 14)
- "NASA is very proud to contribute to ESA's mission to understand one of the greatest science mysteries of our time," said John Grunsfeld, associate administrator for NASA's Science Mission Directorate at the agency's Headquarters in Washington.
- NASA and ESA recently signed an agreement outlining NASA's role in the project. NASA will contribute 16 state-of-the-art infrared detectors and four spare detectors for one of two science instruments planned for Euclid.
- In addition, NASA has nominated three U.S. science teams totaling 40 new members for the Euclid Consortium. This is in addition to 14 U.S. scientists already supporting the mission. The Euclid Consortium is an international body of 1,000 members who will oversee development of the instruments, manage science operations and analyze data.
• First proposed to ESA in 2007, Euclid was selected as the second medium-class mission in the Cosmic Vision program in October 2011. Italy's Thales Alenia Space was chosen as the prime contractor in 2013.
Launch: A launch of the Euclid spacecraft is planned for 2021on a Soyuz-ST-B Fregat-MT vehicle from Kourou, French Guiana.
Orbit: The Euclid spacecraft will be launched into a halo orbit about L2 (Lagrangian Point 2).
Table 4: Mission characteristics of Euclid 15)
Scientific instruments (VIS, NISP)
The VIS (visible imager) instrument is optimized to register resolved images of galaxies in the 550-900 nm passband. 16) The VIS nominal survey images have a 10 σ extended source detection limit of AB 24.5 magnitude and are used to determine the shapes of at least 30 galaxies per arcmin2 over the survey area.
The VIS instrument consists of a CCD based FPA (Focal Plane Array) employing one wide visible passband, a shutter mechanism to close the optical path for read out and dark calibration, and a calibration unit for flat field measurements. 17) The FPA supports 6 x 6 CCDs (4 k x 4 k pixels each) with 0.10 arcsec pixel plate scale, giving a geometric field of greater than 0.5 deg2 including the gaps between the detectors. The VIS data processing and instrument control units are mounted in the spacecraft service module. The VIS central data processing unit collects data from 144 CCD quadrants, arranges all the pixels in the correct order and compresses this very large image (24 k x 24 k), in approximately 250 seconds. The VIS power and mechanisms control unit activates the shutter and the calibration unit. To have full control over the sources of systematic errors no additional VIS image processing will be done on board, all CCD data is transferred to ground.
Figure 16: VIS and weak lensing channel characteristics
Science aims: The core task of VIS is to enable Weak Lensing measurements. The dark matter (and ordinary matter) aggregates under the influence of gravity as the Universe expands. These overdensities deflect light differentially so that the light from background objects is distorted, and they appear to have a measurable change in ellipticity. In general there is only mild distortion, corresponding to weak gravitational lensing. The mass distribution can be inferred from statistical averages of the shapes of background galaxies distorted by this effect, enabling Euclid to map dark matter. By using galaxies further and further away, the rate at which the aggregation has occurred can be characterized. This is directly affected by the expansion history of the Universe, which appears at more recent times to be increasingly driven by dark energy, so the characteristics of the dark energy can consequently also be constrained.
Figure 17: The placement of the VIS instrument within the Payload Module (VI-FPA, VI-RSU, VI-CU – see Figure 19 for details). The large radiator for the VIS detection chain electronics (the ROE) is on the far side. The yellow instrument is the NISP. The PLM is shown "upside down" with the telescope below it. The light path from the hole in the primary mirror is shown in red before the dichroic, and orange after it. FMs are fold mirrors, and M3 is the third mirror in the Korsch telescope (image credit: Airbus Defence & Space)
FPA (Focal Plane Array): Photons within the broad red bandpass (550–900 nm) are detected by the array of 36 CCDs, each of which is a CCD273-8415,16 specifically designed for VIS and manufactured by e2v Technologies. The CCDs are located in a 6 x 6 matrix on the front of the FPA. Close butting of the CCDs provides a >90% filling fraction of active Si.
The FPA has dimensions of ~0.45m on each side. The 36 CCDs are held in a Silicon Carbide (SiC) structure. The detector array is maintained relative to the optical focal plane with tight tolerances to ensure image quality. There is a separate block of electronics which digitize the signals from the CCDs, and is supported separately, with only the CCD electrical flexible interconnects between them (Figure 21). The electronics block holds twelve Readout Electronics (ROE – one for three CCDs), and their closely coupled power supply units (R-PSU, one per ROE). Within the support structure for the ROEs are thermal shields to isolate the cold CCDs from the warm ROEs. A radiator to space maintains the ROE thermal environment. The detector block with the 36 CCDs is maintained at a temperature of 150–155 K by the PLM environment.
The detection chain consists of the CCDs, ROEs and their associated R-PSUs. The architecture of the chain has been set up to ensure sufficient redundancy, while minimizing the use of system resources. Manufacturability, testability and ease of access were important considerations. The tradeoff identified the optimal configuration as one ROE supporting three CCDs, each with four video chains (Figure 18), each with its own power supply unit. In order to minimize system noise levels, the power supply for each ROE is accommodated close to it on the outside of the Electronics Block. A single clocking and bias generation circuit is implemented for each CCD half. Loss of one ROE results in loss of <10% of the focal plane, and the consequent thermal perturbations are manageable. Within each ROE it is envisaged that individual CCDs may be lost without loss of the entire unit, providing further redundancy.
The CCDs are read out through four readout nodes at a rate of 70 kpix s–1. Analog signals are converted to a 16 bit resolution. In order to maintain thermal stability, all circuitry remains powered during exposures. Care is taken in the design of mixed signal circuitry handling very low level analogue inputs, to prevent cross-talk and other noise pick-up: multi-layer printed circuit boards are used with separate ground planes for analogue and digital functions. System grounding and decoupling is carefully planned to prevent circulating currents in ground lines from introducing noise sources. All of the CCDs are read out synchronously, removing electrical co-interference that could result from a slew of non-synchronized clocks; an LVDS (Low-Voltage Differential Signaling) interface (Figure 20) is used for the synchronization.
Data are transmitted through a SpaceWire port to the payload data handling unit (Figure 20), with SpaceWire communication and command decoding, together with other digital functions such as clock sequence generation carried out in a single space-qualified field-programmable gate array (FPGA) per ROE. The packaging of the ROE and PSU electronics is designed to optimize conductive thermal paths while minimizing the parasitic heat conduction to the CCDs.
Figure 18: Left: The block diagram for an ROE, interfacing to 3 CCDs. Right: The mechanical packaging of two ROEs each with 3 CCDs to make a slice for 6 CCDs (image credit: Euclid Consortium)
Figure 19: The five units comprising VIS. The two units at the bottom are in the Service Module and the other three in the Payload Module (image credit: Euclid Consortium)
Figure 20: VIS electrical architecture. The three units to the right (MMU, CDMU and PCDU) are spacecraft units (image credit: Euclid Consortium)
Shutter: A shutter is located in front of the FPA to block the light to the detector array while the detectors are not making exposures or reading them out. This is a momentum-compensated (linear and angular) mechanism in order to minimize any disturbances to the spacecraft and the NISP instrument during actuation. It opens and closes within 10 seconds. The mechanism is electrically cold-redundant. The previously envisaged launch lock is now considered unnecessary and, pending positive results from testing, will be removed. The shutter does not seal against any Payload Module structure, and therefore scattered light paths require careful analysis. Surface treatments for the shutter minimize scattered light when it is opening and closing.
Figure 21: An expanded view of the focal plane array. The detector matrix is on the left, supported on six cold SiC beams each holding six CCDs within a SiC frame. The top row is offset to the rear, to show its relationship of each triplet of detectors to their readout electronics (ROE) which digitize the signals. The cold frame holding the CCDs is mechanically decoupled from the framework holding the ROEs. Each ROE has its own power supply unit (R-PSU). Items labelled TS are the thermal shrouds to isolate the cold detector plate from the warm ROEs (image credit: Euclid Consortium)
Calibration Unit: A calibration unit is provided to flood the FPA with light at different wavelengths within the VIS passband. CCD flat fields taken with this unit allow pixel-pixel variation to be measured as a function of wavelength. It is not required to provide high levels of stability or of large-scale spatial uniformity, as the photometric throughput is calibrated from stellar sources during routine observations, so a simple lens projection within the baffle provides sufficient uniformity and control of out-of-field light. The smoothest flat fields are at ~750 nm. While the wavelength dependence of the CCD flat field is slow, the number of independent illumination sources has been increased from 3 to 6 to ensure its adequate characterization.
There is no shutter for the calibration unit. Flat field calibrations therefore have the sky signal included, but because they are short exposures, and because many flat fields are combined to produce a master flat, this simple solution is acceptable. The calibration unit is electrically cold-redundant.
CDPU (Command and Data Processing Unit): The CDPU in the Service Module controls the instrument, transitioning between the different instrument modes and sequencing the operations within exposures, and monitoring instrument status to generate the housekeeping data. It also takes the science data from the ROEs, losslessly-compressing it and buffering it before transfer to the spacecraft bulk memory (MMU). It consists of a Maxwell SCS750 PowerPC triple-voting processing unit which performs the compression and science packet generation, a data routing board to handle the instrument status and the science data from the ROEs, power routing hardware to the PMCU and a multiplexer board to interface the 12 ROEs to both halves of the unit. The CDPU is dual-redundant except for the multiplexer board (Figure 20).
Power and Mechanism Control Unit: The PMCU provides power to the Calibration Unit and the Shutter, and controls its opening and closing. It also accepts housekeeping data from the shutter and calibration unit, and heater power to the FPA to ensure an appropriate thermal environment. The PMCU is also dual-redundant.
VIS operations and calibration:
VIS is designed to be simple to operate because this promotes the stability needed for weak lensing measurements. Almost all of the observations are carried out with a 565 second exposure time, with the entire focal plane active. The CCDs are then read out, the data are digitized, buffered and compressed, and the process repeated. The Euclid Reference Survey covers 15000 deg2 for the Wide Survey and ~40 deg2 for the Deep Survey and four separate exposures are made for each field, with displacements of 100 arcsec between them (with an additional lateral 50 arcsec for the fourth exposure). This is in order to cover gaps in the detector matrix, to permit some recovery of the spatial resolution, to minimize radiation damage impact on the data and to allow cosmic rays to be identified. This results in 50% of the sky being observed in four exposures, and another 47% with three, because of the gaps. Optimizations of the Reference Survey are currently being considered, and this may lead to a different total survey area, more exposures per field and different size displacements between them. However, any changes are required to be compatible with the current operating capabilities of the instruments and satellite.
The array takes some 80 seconds to read out in total. The VIS exposures take place during the spectroscopic exposures of NISP which have the same duration. After this, the time during the NISP photometric exposures is limited for science observations, because of pointing disturbances caused by the NISP filter wheel. However, shorter exposures will be taken during a small subset of NISP photometric exposures. These add dynamic range, improve sampling and assist in cosmic ray elimination: one or two of these short exposures can be included within the telemetry and shutter lifetime budgets. The time occupied by the NISP photometric exposures will mainly be used for VIS calibrations, such as flat and bias fields and dark exposures. Dark exposures will be of three types: in the first the dark current in the detectors (which is extremely low) can be measured, together with hot pixels (which will be relatively few given the cold operational temperature of the CCD); in the second a few lines of charge will be injected electrically into the CCD, and in the third charge will be injected over the whole region of the CCD and the CCD clocked backwards and forwards in a "pocket-pumping" mode.
Flat field calibrations are made with the Calibration Unit. Because of possible cross-contamination with NISP, these will be done at the end of the sequence of four exposures just before a spacecraft slew. The calibration exposures are short (10 seconds), incurring no time overhead.
Linearity calibrations require a slightly different operational sequence, with different durations for each of the four exposures. In order not to inordinately constrain the shutter repeatability, the shutter will open while the CCD is being read out continuously, after which the readout will stop for a precise duration set by the synchronization timing in VIS, before the readout begins again and the shutter closes. This will generate stellar images superimposed on illuminated strips in the image, caused by reading out when the shutter is open. These can be subtracted to produce accurate estimates of the flux in the image, and hence to determine the linearity.
NISP (Near Infrared Spectro Photometer)
NISP is one of the two Euclid instruments and is operating in the near-IR spectral region (900-2000 nm) as a photometer and spectrometer. The instrument is composed of:
- a cold (135 K) optomechanical subsystem consisting of a Silicon carbide structure, an optical assembly (corrector and camera lens), a filter wheel mechanism, a grism wheel mechanism, a calibration unit and a thermal control system
- a detection subsystem based on a mosaic of 16 HAWAII2RG cooled to 95 K with their front-end readout electronic cooled to 140 K, integrated on a mechanical focal plane structure made with molybdenum and aluminum. The detection subsystem is mounted on the optomechanical subsystem structure
- a warm electronic subsystem (280 K) composed of a data processing / detector control unit and of an instrument control unit that interfaces with the spacecraft via a 1553 bus for command and control and via Spacewire links for science data.
Euclid will carry out an imaging and spectroscopic wide survey of the entire extra-galactic sky (15000 deg2) along with a deep survey covering at least 40 deg2. To achieve these science objectives, the current Euclid reference design consists of a wide field telescope to be placed in L2 orbit by a Soyuz launch with a 6 years' mission lifetime. The payload consists of a 1.2 m diameter 3-mirror telescope with two channels: a VISible imaging channel (VIS) and a NISP (Near Infrared Spectrometer and Photometer) channel. Both instruments observe simultaneously the same Field of View (FOV) on the sky and the system design is optimized for a sky survey in a step-and-stare tiling mode. 18) 19)
The NISP Instrument is operating in the 920-2000 nm range at a temperature lower than 140K, except for detectors, which are cooled down to ~95 K or below. The warm electronics will be located in the service module, at room temperature (around 20°C). The NISP instrument has two main observing modes: the photometric mode, for the acquisition of images with broad band filters, and the spectroscopic mode, for the acquisition of slitless dispersed images on the detectors.
In the photometer mode the NISP instrument images the telescope light in the wavelength range from 920 - 2000 nm (Y, J, H bands). The spatial sampling is required to be 0.3 arcsec per pixel. The FOV of the instrument is 0.55deg2 having a rectangular shape of 0.763º x 0.722º. In the spectrometer mode the light of the observed target is dispersed by means of grisms covering the wavelength range of 950 – 1850 nm. In order to provide a flat resolution over the specified wavelength range, four grisms are mounted in a wheel. These four grisms yield three dispersion directions tilted against each other by 90° in order to reduce confusion from overlapping (due to slitless observing mode). The field and waveband definitions used in the individual configurations for spectroscopy and photometry are:
• Three photometric bands: Y-bBand: 950 - 1192 nm, J-band: 1192 - 1544 nm, H-band: 1544 – 2000 nm.
• Four slitless spectroscopic bands:
- Red 0°; 90° and 180° dispersion: 1250 - 1850 nm
- Blue 0° dispersion: 920 -1300 nm.
The spectral resolution shall be higher than 250 for a one arcsec homogenous illumination object size. For such an object, the flux limit in spectroscopy shall be lower than 2 x 10-16 erg·cm-2·s-1 at 1600 nm wavelength. As with all slitless spectrographs, the real resolution varies with the object size (the smaller the size is, higher the resolution is). The image quality of the instrument in flight shall deliver a 50% radius encircled energy better than 0.3 arcsec and a 80% one better than 0.7 arcsec. There is a variation due to diffraction with the wavelength.
The NISP budgets are presently the following:
European Contributor countries for NISP are: France, Italy, Germany, Spain, Denmark and Norway, ESA for the engineering detectors and USA (NASA) for the flight detectors.
The NISP instrument consists of three main Assemblies:
• The NI-OMA (Opto-Mechanical Assembly), composed of the Mechanical Support Structure (NI-SA) and its thermal control (NI-TC), the Optical elements (NI-OA), the Filter Wheel Assembly (NI-FWA), the Grism Wheel Assembly (NI-GWA), the Calibration Unit (NI-CU). The NI-OMA structure supports the Optical elements, the calibration unit, the Filter and Grism Wheel Units and the detection system. It provides the thermo-mechanical interface towards the Euclid PLM.
• The NI-DS (Detector System Assembly) is composed by the Focal Plane Assembly (NI-FPA; the mechanical part of NI-DS) and by the Sensor Chip System (NI-SCS) compose). The NI-DS comprises the 16 H2RG detectors and associated 16 ASICS (Sidecars), passively cooled at operating temperature (<100K for the detectors; 140K for the ASICS Sidecar). Thermal stabilization of the detector is "naturally" obtained thanks to the very good thermal stability provided by the Euclid PLM at the NISP interfaces.
• The Warm Electronics Assembly (NI-WE), composed of the Instrument Data Processing Unit and Control Unit (NI-DPU/DCU), and the Instrument Control Unit (NI-ICU). The NI-ICU is managing the commanding and the control of the instrument. It is interfaced with the satellite via a 1553 bus. The NI-DPU/DCU controls the Sensor Chip System and basic image processing such as co-adding (DCU function) and the science onboard data processing, the compression and transfer of scientific data to the S/C Mass Memory using Spacewire links (DPU function). The NI-DPU/DCU functions are regrouped in a single mechanical box for controlling eight detectors. There are two NIDPU/DCU boxes.
The NI-DS is screwed on the NI-OMA (SiC panel to SiC panel). The NI-OMA+NI-DS is located in the Euclid spacecraft Payload module in a cold environment (130 K). The Warm electronic are located in the Euclid spacecraft Service Module at room temperature. A dedicated harness interconnects the NI-OMA, the NI-DS, the NI-WE and different spacecraft electronics boxes.
An overview of NISP subsystems is shown in the Figures 22 and 23. The overall throughput of the NISP instrument will enable scientists to measure photometric and spectroscopic redshifts of galaxies with sufficient accuracy. The NISP photometer will be able to get a signal-to-noise ratio of at least 5 down to magnitude 24.0 in about 100 seconds per filter. The NISP spectrograph will provide redshift of about 30 million emission line galaxies over the redshift range 0.7 to 2.0 in about 4000 seconds.
NISP successfully passed the Critical Design Review in November 2016 and is now en route for building the flight model.
Figure 22: Overview of the subsystems composing the NISP instrument (the warm electronic subsystem is not shown on this figure). The top panel shows the elements of the NISP Opto-Mechanical Assembly and Detector Assembly: NISP Calibration Unit (NI-CU), NISP Camera Lens Assembly (NI-CaLA), NISP Structure Assembly (NI-SA-ST and NI-SA-HP as for Structure and HexaPodes, respectively), NISP Corrector Lens Assembly (NI-CoLA) and the NISP Detector System (NI-DS). NI-FWA and NI-GWA are the NISP Filter Wheel Assembly and the Grism Wheel Assembly, respectively (image credit: Euclid Consortium/NISP team, Ref. 19)
Figure 23: The panel on top shows the NISP focal plane and the elements of the NISP Detector System (brown color): NISP Sidecar Support Structure (NI-SSS), NISP Sensor Chip System (NI-SCS) and NISP Cold Support Structure (NI-CSS).- The bottom panel shows the filter positions (left), grism positions (right) and the transmission curves of the Y, J and H filters and the blue and red grims (image credit: Euclid Consortium/NISP team, Ref. 19)
The NISP Calibration Unit allows in-flight calibration of the infrared detector array. The unit provides stable illumination of the image plane at five different infrared wavelengths, allowing for small-scale flat field calibration and measurements of the detector linearity. The design is relatively simple with 2 x 5 LEDs (one nominal and redundant per wavelength) inside the calibration unit pointing to a small patch of Spectralon PTFE material. The Lambertian scattered light is directly pointed towards the detector through a set of baffles without going through any of the optics. Control of the LED brightness and thus received flux is performed by current and duty cycle regulation of the drive signal in the instrument control unit (ICU). As the unit operates under cryogenic conditions, finding and qualifying suitable LEDs has become a major challenge, especially for longer wavelengths >1.6um.
Previous work has shown that commercially available off-the-shelf devices are not usable in our case. We have therefore procured the raw LED dies for the required wavelengths and submitted them to a uniform and controlled assembly and packaging process. The full formal space qualification of the LEDs is expected to take place well into the 2017. Nevertheless, initial tests of proofing devices, especially under cryogenic conditions, have confirmed that the LEDs are highly durable. The structural model of the calibration unit has passed vibration testing to confirm the validity of the mechanical design. Current development steps include long-term cryogenic storage and lifetime cycle-testing of the LEDs as well as fine-tuning of the interior optical layout of the calibration unit.
Figure 24: Cross-cut of the NISP calibration unit / View into one of the custom produced LEDs for the NISP calibration unit / The structural model of the NISP calibration unit, as used in the vibration campaign of the NISP STM. This model does not include LEDs, harness and optical elements ((image credit: Euclid Consortium/NISP team)
The NISP Warm Electronics:
The NISP warm electronics are composed of two Data Processing Units (NI-DPU) and one Instrument Control Unit (NI-ICU).
Data Processing Unit (NI-DPU): The full system is shown in the electrical drawing reported in figure "NISP functional electrical scheme", each DPU unit is mounted around a shared Compact PCI bus structure with the exception of the main power supply system and the 8 x DCU boards each one managing one SCE/SCA detection pair. The two Data Processing Units (NI-DPU) are both including:
- 8 x Detector Control Units (DCU) that provide clock and power to the readout electronic. In addition, these units will preprocess the data using FPGA boards
- Central Processor Unit that finalize the on board data processing, compress and format the data sending them via SpaceWire link to the central spacecraft memory.
Each DPU is hosting the following boards:
- CPCI Data Processors based with two Maxwell SCS750
- CPCI Data Routers
- CPCI Data Buffer
- Power Supply.
Except for the DCUs, all the boards in the DPU are cold redundant. Each DCU receives the data of one 2 k x 2 k detector from one SIDECAR and performs the low level pre-processing foreseen in HW consisting of:
- Group of frames averaging
- Telemetry extraction
- Extraction of sub-sets of programmed raw detector lines to be used on ground for monitoring purposes
- Co-added Frame data buffering and Spacewire transmission to Data Buffer Boards.
At this interface level redundancy is supported by the full duplication of DPU hardware. Averaged data groups can be configured to be transmitted to one of the Data Buffer Boards available in each DPU, this is accomplished by duplication of the 8 x Spacewire (SpW) links. The same redundant configuration is available at each DCU TMTC interface bus: the control link based on the RS485 standard can be configured to be driven by one of the two available CPCI data router boards.
The Data Buffer board allows the storage of up to 46+46 averaged frames with telemetry and ancillary data from the 8 x handled detection channels in double-buffering mode to ease the further data processing.
A functional demonstrator model of the DPU (with one Maxwell board, one DCU board and one Data Buffer board) has been manufactured and tested with a first version of the application SW.
Figure 25: DPU (design / Demonstrator model), image credit: Euclid Consortium/NISP team
Instrument Control Unit Hardware (NI-ICU):
• One Instrument Control Unit (NI-ICU) in charge of:
- Interface with the spacecraft via a 1553 bus for the commanding of the NISP
- Housekeeping management
- General power supply
- Command signal to the cryo-mechanism, to the 5 LED's calibration source and to the NI-OMA and NI-DS heaters (heater constant power is applied in open loop with power setpoint determined by ground operators).
The ICU has two sections (nominal and redundant), which are identical and operate in cold redundancy. Each NI-ICU (N or R) is divided in three boards, all of them interconnected by means of a backplane motherboard:
• LCPS (Low Voltage Power Supply): provides DC/DC converters to generate all the necessary secondary power supplies, as well as the 1553 transceivers for the NI-DPU link (1553 controller logic is actually located in the CPDU board.
• CDPU (Central Data Processing Unit): contains a LEON2-FT CPU embedded in a MDPA ASIC, which manages all the functions of the NI-ICU. This module also includes a RTAX FPGA that extends the functionalities of the MPDA, with the main aim of interfacing with the DAS module. The 1553 transceivers for the S/C link and the test connector are also located in this board.
• DAS (Data Acquisition System): this board features all the analogue acquisition and driving electronics that are used to interface with the
Figure 26: Left: NI-ICU mechanical design, Right: EBB of the CDPU board (image credit: Euclid Consortium/NISP team)
The warm electronic will be placed in the service module of the spacecraft at ambient temperature. A harness under Prime contractor responsibility provides the link with the NI-OMA and NI-DS. This cable will carry LVDS signal for scientific data, housekeeping signals, control command and power supply for equipment.
The main challenge of the warm electronics is to process the amount of data delivered by the detector during the integration of the following frame. The onboard data processing is complexified by the fact that the amount of downlink accepted to ground is very limited. Only final frames can be sent to ground, but as described later HgCdTe detectors deliver lots of frame to achieve final science performances.
The ICU Application SW (ASW) is devoted to manage the satellite/platform interface, the ICU/DPU interface and all the functionalities related to instrument commanding. It is in charge of the following functions:
- TM/TC exchange with S/C CDMU on Nom/Red 1553 link
- TC decoding and distribution to NISP instrument subunits: NI-FWA electronics, NI-GWA electronics, NI-CU electronics, NI-TC electronics, NI-DPU/DCU/SCE
- Global instrument monitoring and HK packet generation
- Time management, propagation of OBT to the DPUs, TM time tagging and high level instrument internal synchronizations
- NISP operating mode management
- Execution of autonomous functions and FDIR algorithms and processes
- Control of the calibration unit (ON/OFF, intensity level and current absorption handling)
- Control of filter wheels (reference position, position switch)
- Thermal control (open loop) of the NI-FPA detector cold-plate through temperature sensors and heaters
- High level handling of macro-commands submission to detector system
- Thermal control of the NI-OMA through temperature sensors and heaters
- Management of software maintenance, memory patch and dump (EEPROM patching is performed by the Boot SW)
Instrument Control Unit Hardware Application Software (ICU ASW): The ICU ASW is based on RTEMS real-time operative system, in the space-qualified version by EDISOFT. Telecommand and Telemetry packets will be based on the Packet Utilization Services (PUS) standard, with the implementation of services tailored to the specific needs of the Euclid project.
A coordinated effort is in place with the Prime of the Spacecraft and with the VIS CDPU ASW team in order to ensure a common approach and, as far as possible, implementation of services between NISP and VIS, so that the SW interfaces with the Spacecraft can be simplified and standardized. The interface with the DPU is based on a second MIL-STD-1553 bus, similar to the one used between the Spacecraft and the Euclid instruments, in which the DPUs are configured as Remote Terminals and the ICU as the Bus Controller. The SW interface and communication protocol is an internally defined one, with the aim of reducing as much as possible the load of management tasks on the DPU processor, since this resource is needed for the demanding data processing tasks. The ICU ASW will decode the PUS formatted high level TCs and implement the low level sequences towards the two active DPUs.
Onboard Data Processing:
The routine science NISP operations foresee 20 fields of observation per day, each one composed by four dithers where four exposures each are taken, for a total maximum assigned science data telemetry of 290 Gbit/day. A dark exposure is taken during the spacecraft slew. This limited amount of allowed telemetry, together with the huge number of frames typically produced by IR detectors operated in multi-accumulation mode, have as a consequence the need to perform part of the processing pipeline directly on-board and to transfer to ground only the final products for each exposure. Moreover, final data must be also compressed to fit with the assigned telemetry throughput. A number of readout modes have been envisioned for the NISP instrument in the various development stages. Multi-accumulation (MACC) is at the moment the preferred modes for both spectrograph and photometer readout. MACC readout is a peculiar Up the Ramp process (UTR) where detector readouts are grouped in contiguous sets of readouts uniformly placed along the accumulated charge ramp. The data processing can be split into two main stages: stage 1 is implemented in the NI-DCU, directly interfaced to the SCS, where the first static basic pre-processing steps are performed, while stage 2, performed in NI-DPU, is devoted to the processing and compression of the final data frames.
Figure 27: Pre-processing HW structure connected to 1x SCS single pair (H2RG SCA + SCE) from a total of 16 x located inside the SCS system (image credit: Euclid Consortium/NISP team)
The software architecture is dictated by the science requirements and depends on the hardware organization, in terms of DPU power, internal memory, available links with both DCU and SVM. During the previous different phases of the project various processing possibilities were analyzed, in terms of computational complexity, DPU internal memory needs, amount of final data and quality of results. As a result, the foreseen on-board pre-processing pipeline7 will be as depicted in Figure 6-2 where the violet blocks represent the operations performed inside the DPUs. This operational flow is sequentially repeated to cover the 17 exposures (4 spectro + 12 photo + 1 dark) to be performed during each single cycle.
At the end of the pipeline described in Figure 6-2 final generated data, with their associated header and metadata to properly re-construct images on ground, are transmitted to the spacecraft Mass Memory Unit, to be downlinked to ground. The most crucial constraint for the on-board processing is given by the need to keep up with the on-going observations, so the previous work was mostly concentrated to verify the algorithm performances, especially in terms of time spent. Current development steps include the integration of the data processing with the overall DPU Application Software structure.
Figure 28: On-board data processing pipe-line for the Euclid NIS/NIP instrumental modes. The pipe-line is subdivided in three different sections on the base of the involved hardware, in the order: SCE analog hardware, FPGA hardware and sequential processing hardware (image credit: Euclid Consortium/NISP team)
Optical performances: The following figure shows the evaluation of the Encircled Energy compared to required values (the maximum radius of the Encircled Energy (EE) at 80% and at 50% shall be lower than the specified ones. The Figure 29 shows that NISP complies with its optical requirement.
Figure 29: EE (Encircled Energy) performance evaluation (image credit: Euclid Consortium/NISP team)
IR detector QE and noise performances: The detector QE and the detector noise are a major concern for the future NISP performance. One important goal is to ensure not only mean specified values but also that 95% of pixels meet these requirements to ensure the efficiency of coverage in the full survey.
Figure 30: QE measurements on the first flight detectors between 920nm and 2300 nm (left). Flat Field images of flight detectors compared with NRE phase detectors produced shows an improvement in the spatial homogeneity of the pixel responses (image credit: Euclid Consortium/NISP team)
The homogeneity of the pixel response is excellent. The CDS readout noise shows the same performances demonstrated during the NRE phase. The mean dark has been shown to be very low around 0,005 e–/pix/s at 100 K with a sharp distribution and well inside NISP specifications. From these first tests on Flight parts, the project can expect a very high quality NIR Focal Plane.
During the 6 years of the survey, Euclid will collect more more than 500,000 visible (VIS) and near infrared images (NISP imaging and NISP spectroscopy) that will be transferred to Earth on a daily basis cadence. 20)
The Euclid space images will be complemented by visible imaging data from ground-based telescopes, covering the same sky as the Euclid mission, in order to derive photometric redshifts (distance measurements) of lensed sources. Euclid space and ground data will provide photometry of each galaxy in 7 different filters ranging from 460 nm to 2000 nm (g,r,i and z bands from the ground, Y, J and H from Euclid) down to 24.0 to 24.5 AB magnitude. Overall, the Euclid input images represent several million of images and about 30 Petabytes of image data.
In addition to the data volume one challenge of the science ground segment is to process and analyze heterogeneous data sets that mix together space and ground based images. They will be obtained with several telescopes providing visible or near infrared data with different depth and resolution in order to derive astrometric, photometric, morphometric and spectroscopic parameters of sources and possibly time domain information.
About 10 billion sources will be observed by Euclid, out of which more than 1 billion will be used for weak lensing, and about 30 million galaxy redshifts will be obtained. The science ground segment is a most challenging part of this mission and represents about 50% of the resources provided by the Euclid Consortium.
The ground segment activities are shared by ESA and the Euclid Consortium.
The Science Ground Segment (SGS) of the Euclid mission is composed of the combination of two independent sections: the ESA Science Operations Center (SOC), and the Euclid Consortium Science Ground Segment (EC SGS). SOC will be the single interface to the ESA Mission Operations Center (MOC), and hence to the Euclid Operations Ground Segment (OGS). The SOC is also in charge of the survey planning, managing the downlinked data (i.e. performing quick look, daily quality reports and telemetry editing) and providing the EC with the data necessary to perform further processing of the science data.
The scientific processing of the data, down to the production of the scientific results, is the responsibility of the Science Ground Segment managed by the Euclid Consortium. The processing activity will occur in national Science Data Centers (SDCs, Figure 32). Two SDCs will host the Instrument Operations Teams (IOTs), one for each of the instruments, which will perform the monitoring of instrument performance. Other SDCs will be in charge of providing large external data sets obtained from ground based surveys complementing the Euclid data.
The SGS will be designed consistently with the principle of maximizing the impact of Euclid in both primary and legacy sciences. For this purpose, one of the drivers in the design will be making data available publicly as soon as possible. The SGS will therefore be responsible for the archive and the production of the official Euclid data releases.
Figure 31: Overview of the 2 elements of the Ground Segment and its main components, the Operation Ground Segment (OGS) and the Science Ground Segment (SGS). The OGS is under ESA responsibility. The SGS is split into an ESA component and a Euclid Consortium component (ECSGS). VIS, NIS, SIR, EXT, MER, SPE, SHE, PHZ, LE3 and SIM are Euclid Consortium Organization Units in charge of defining and prototyping the algorithms (image credit: Euclid Consortium/ESA/SGS Team)
Figure 32: Organization of the SDCs (Science Data Centers) under responsibility of the Euclid Consortium. The SDCs are computing centers in charge of implementing and running the data processing pipelines (image credit: Euclid Consortium/ESA/SGS Team)
1) "Euclid Science Goals," ESA, URL: http://sci.esa.int/euclid/42267-science/
2) "Euclid Mapping the geometry of the dark Universe,"Definition study report, ESA, ESA/SRE(2011)12, September 2011 (Revision 1), URL: http://sci.esa.int/science-e/www/object/doc.cfm?fobjectid=48980
3) "Expansion history of the Universe," ESA, 5 March 2010, URL: http://sci.esa.int/euclid/46673-expansion-history-of-the-universe/
4) "Week lensing effects," ESA, 23 July 2014, URL: http://sci.esa.int/euclid/46675-weak-lensing-effects/
6) "Scientists of the Euclid Consortium announce the release of the largest simulated galaxy catalog ever built," Joint EC, ICE (IEEC-CSIC), PIC, CSCS and University of Zurich Press Release, 07 June 2017, URL: http://www.euclid-ec.org/?page_id=4133
7) Giuseppe D. Racca, René Laureijs, Luca Stagnaro, Jean-Christophe Salvignol, José Lorenzo Alvarez, Gonzalo Saavedra Criado, Luis Gaspar Venancio, Alex Short, Paolo Strada, Tobias Bönke, Cyril Colombo, Adriano Calvi, Elena Maiorano, Osvaldo Piersanti, Sylvain Prezelus, Pierluigi Rosato, Jacques Pinel, Hans Rozemeijer, Valentina Lesna, Paolo Musi, Marco Sias, Alberto Anselmi, Vincent Cazaubiel, Ludovic Vaillon, Yannick Mellier, Jérôme Amiaux, Michel Berthé, Marc Sauvage, Ruyman Azzollini, Mark Cropper, Sabrina Pottinger, Knud Jahnke, Anne Ealet, Thierry Maciaszek, Fabio Pasian, Andrea Zacchei, Roberto Scaramella, John Hoar, Ralf Kohley, Roland Vavrek, Andreas Rudolph, Micha Schmidt, "The Euclid mission design," Presented at the SPIE Astronomical Telescopes and Instrumentation conference in Edinburgh, Scotland, United Kingdom, 6 June- 1 July 2016, DOI: 10.1117/12.2230762, URL: https://arxiv.org/ftp/arxiv/papers/1610/1610.05508.pdf
8) Stephen Clark, "Detector trouble expected to delay ESA's Euclid dark energy mission," Spaceflight Now, 23 October 2017, URL: https://spaceflightnow.com/2017/10/23/detector
9) "Detector delivery marks another Euclid milestone," Euclid, 26 April 2017, URL: http://sci.esa.int/euclid/59050-detector-delivery-marks-another-euclid-milestone/
10) "First Euclid flight hardware delivered," Euclid, 07 Feb. 2017, URL: http://sci.esa.int/euclid/58777-first-euclid-flight-hardware-delivered/
11) "First Euclid flight hardware delivered - VIS detectors," The Euclid Consortium Newsletter - Spring 2017, URL: https://www.euclid-ec.org/Documents/Newsletter/EC-Newsletter_issue07.pdf
12) "Euclid dark Universe mission ready to take shape," Euclid, 17 Dec. 2015, URL: http://sci.esa.int/euclid/57039-euclid-dark-universe-mission-ready-to-take-shape/
13) "Multilateral agreements signed at SPC (Science Program Committee) meeting," 29, Nov. 2013, URL: http://sci.esa.int/cosmic-vision/53274
14) "NASA Officially Joins ESA's 'Dark Universe' Mission," NASA, 24 Jan. 2013, URL: https://www.nasa.gov/mission_pages/euclid/news/euclid20130124.html
16) Mark Cropper, S. Pottinger, S. Niemi, R. Azzollini, J. Denniston, M. Szafraniec, S. Awan, Y. Mellier, M. Berthe, J. Martignac, C. Cara, A.-M. di Giorgio, A. Sciortino, E. Bozzo, L. Genolet, R. Cole, A. Philippon, M. Hailey, T. Hunt, I. Swindells, A. Holland, J. Gow, N. Murray, D. Hall, J. Skottfelt, J. Amiaux, R. Laureijs, G. Racca, J.-C. Salvignol, A. Short, J. Lorenzo Alvarez, T. Kitching, H. Hoekstra, R. Massey, H. Israel, the Euclid collaboration, "VIS: the visible imager for Euclid," Proceedings of SPIE, Vol. 9904, 'Space Telescopes and Instrumentation 2016: Optical, Infrared, and Millimeter Wave,' 99040Q , July 19, 2016, DOI: 10.1117/12.2234739, URL: https://arxiv.org/ftp/arxiv/papers/1608/1608.08603.pdf
17) A. D. Short, D. Barry, M. Berthe, N. Boudin, O. Boulade, R. Cole, M. Cropper, L. Duvet, J. Endicott, L. Gaspar Venancio, J. Gow, P. Guttridge, D. Hall, A. Holland, H. Israel, R. Kohley, R. Laureijs, J. Lorenzo Alvarez, J. Martignac, J. Maskell, R. Massey, N. Murray, S.-M. Niemi, P. Pool, S. Pottinger, T. Prod'homme, G. Racca, J-C Salvignol, W. Suske, M. Szafraniec, P. Verhoeve, D. Walton, R. Wheeler, "The Euclid VIS CCD detector design, development, and programme status," Proceedings of SPIE, Volume 9154, 'High Energy, Optical, and Infrared Detectors for Astronomy VI', 91540R, 23 July 2014, doi: 10.1117/12.2057587, URL of abstract: https://doi.org/10.1117/12.2057587
18) Thierry Maciaszek, Anne Ealet, Knud Jahnke, Eric Prieto, Rémi Barbier, Bogna Kubik, Yannick Mellier, FlorentBeaumont, William Bon, Anne Bonefoi, Michael Carle, Amandine Caillat, Anne Costille, Doriane Dormoy, Franck Ducret, Christophe Fabron, Aurélien Febvre, Benjamin Foulon, Jose Garcia, Jean-Luc Gimenez, Emmanuel Grassi, Philippe Laurent, David Le Mignant, Laurent Martin, Christelle Rossin, Tony Pamplona, Patrice Sanchez, Sebastien Vives, Jean Claude Clémens, William Gillard, Mathieu Niclas, Aurélia Secroun, Benoit Serra, Jérome Amiaux, Jean Christophe Barrière, Michel Berthe, Cyrille Rosset, Juan Francisco Macias-Perez, Natalia Auricchio, Adriano De Rosa, Enrico Franceschi, Gian Paolo Guizzo, Gianluca Morgante, Francesca Sortino, Massimo Trifoglio, Luca Valenziano, Laura Patrizii, T. Chiarusi, F. Fornari, F. Giacomini, A. Margiotta, N. Mauri, L. Pasqualini, G. Sirri, M. Spurio, M. Tenti, R. Travaglini, Stefano Dusini, F. Dal Corso, F. Laudisio, C. Sirignano, L.Stanco, S.Ventura, Enrico Borsato, Carlotta Bonoli, Favio Bortoletto, Andrea Balestra, Maurizio D'Alessandro, Eduardo MedinaCeli, Ruben Farinelli, Leonardo Corcione, Sebastiano Ligori, Frank Grupp, Carolin Wimmer, Felix Hormuth, Gregor Seidel, Stefanie Wachter, Cristobal Padilla, Mikel Lamensans, Ricard Casas, Ivan Lloro, Rafael Toledo-Moreo, Jaime Gomez, Carlos Colodro-Conde, David Lizán, Jose Javier Diaz, Per B. Lilje, Corinne Toulouse-Aastrup, Michael I. Andersen, Anton N. Sørensen, Peter Jakobsen, Allan Hornstrup, Niels-Christian Jessen, Cédric Thizy, Warren Holmes, Ulf Israelsson, Michael Seiffert, Augustyn Waczynski, René J. Laureijs, Giuseppe Racca, Jean-Christophe Salvignol, Tobias Boenke, Paolo Strada, "Euclid Near Infrared Spectrometer and Photometer instrument concept and first test results obtained for different breadboards models at the end of phase C," Proceedings of SPIE, June 2016, URL: https://www.duo.uio.no/bitstream/handle/10852/52728
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).