ESPRESSO, the ultimate rocky exoplanets hunter for the VLT Denis Mégevand*a, Filippo M. Zerbib, Alexandre Cabralc, Paolo Di Marcantoniod, Manuel Amatee,

Francesco Pepea, Stefano Cristianid, Rafael Reboloe, Nuno C. Santosf, Hans Dekkerg, Manuel Abreuh, Michael Affolteri, Gerardo Avilag, Veronica Baldinid, Paul Bristowg, Christopher Broegi, Pedro Carvask, Roberto Ciramid, João Coelhoc, Maurizio Comarid, Paolo Conconib, Igor Corettid, Guido Cupanid, Valentina D'Odoricod, Vincenzo De Capriob, Bernard Delabreg, Pedro Figueiraf,

Michel Fleurya, Ana Fragosoe, Ludovic Genoleta, Ricardo Gomesc, Jonay Gonzalez Hernandeze, Ian Hughesa, Olaf Iwertg, Florian Kerberg, Marco Landonib, Jorge Limak, Jean-Louis Lizong, Christophe Lovisa, Charles Mairea, Marco Mannettad, Carlos Martinsf, André Moitinhok, Paolo Molarod, Manuel

Monteirof, José Luis Rasillae, Marco Rivab, Samuel Santana Tschudie, Paolo Santind, Danuta Sosnowskaa, Sergio Sousaf, Paolo Spanòb, Fabio Tenegie, Giorgio Tosob, Eros Vanzellad, Matteo

Vield, Maria Rosa Zapatero Osoriol

aObservatoire de l’Université de Genève, Ch. des Maillettes 51, CH-1290 Versoix, Switzerland

bINAF – Osservatorio Astronomico di Brera, Via Bianchi 46, I-23807 Merate, Italy cLOLS, Faculty of Sciences, Univ. of Lisbon, Estrada do Paço do Lumiar 22, 1649-038 Lisbon,

Portugal dINAF – Osservatorio Astronomico di Trieste, Via Tiepolo 11, I-34143 Trieste, Italy

eInstituto de Astrofísica de Canarias, C/. Vía Láctea, E-38200 La Laguna, Tenerife, Spain fCentro de Astrofísica da Universidade do Porto, Rua das Estrelas, 4150-762 Porto, Portugal

gESO, European Southern Observatory, Karl-Schwarzschild-Strasse 2, 85748 Garching, Germany

hCAAUL, Faculty of Sciences, Univ. of Lisbon, Tapada da Ajuda, Edificio Leste, 1349-018 Lisbon, Portugal

iPhysics Institute of University of Bern, Siedlerstrasse 5, CH-3012 Bern, Switzerland kSIM, Faculty of Sciences, Univ. of Lisbon, Campo Grande, Edificio C8, 1749-016 Lisbon, Portugal lCentro de Astrobiología, INTA, Carrettera Ajalvir km 4, 28850, Torrejòn de Ardoz, Madrid, Spain

ABSTRACT

ESPRESSO, the VLT rocky exoplanets hunter, will combine the efficiency of modern echelle spectrograph with extreme radial-velocity precision. It will be installed at Paranal on ESO's VLT in order to achieve a gain of two magnitudes with respect to its predecessor HARPS, and the instrumental radial-velocity precision will be improved to reach 10 cm/s level. We have constituted a Consortium of astronomical research institutes to fund, design and build ESPRESSO on behalf of and in collaboration with ESO, the European Southern Observatory. The project has passed the preliminary design review in November 2011. The spectrograph will be installed at the so-called "Combined Coudé Laboratory" of the VLT, it will be linked to the four 8.2 meters Unit Telescopes (UT) through four optical "Coudé trains" and will be operated either with a single telescope or with up to four UTs. In exchange of the major financial and human effort the building Consortium will be awarded with guaranteed observing time (GTO), which will be invested in a common scientific program. Thanks to its characteristics and the ability of combining incoherently the light of 4 large telescopes, ESPRESSO will offer new possibilities in many fields of astronomy. Our main scientific objectives are, however, the

*denis.megevand@unige.ch; phone 0041 22 3792353; fax 0041 22 3792205; www.unige.ch/sciences/astro

search and characterization of rocky exoplanets in the habitable zone of quiet, near-by G to M-dwarfs, and the analysis of the variability of fundamental physical constants. In this paper, we present the ambitious scientific objectives, the capabilities of ESPRESSO, the technical solutions for the system and its subsystems, enlightening the main differences between ESPRESSO and its predecessors. The project aspects of this facility are also described, from the consortium and partnership structure to the planning phases and milestones.

Keywords: Spectrograph, high resolution, high stability, VLT, radial velocity, exoplanets

1. INTRODUCTION A high-resolution, ultra stable, fiber-fed, cross-dispersed echelle spectrograph, called ESPRESSO, is the successor of a line of echelle spectrometers (CORAVEL, Elodie, Coralie, HARPS) designed to give high accuracy radial-velocity (RV) measurements. ESPRESSO stands for Echelle SPectrograph for Rocky Exoplanets and Stable Spectroscopic Observations. It combines the best properties of two leading instruments HARPS and UVES, enhancing the stability of HARPS with an unprecedented resolution. This instrument is intended as an additional second-generation instrument for the VLT, as recommended by the ESO STC during its 67th meeting, in October 2007 and the ESO council at its 11th meeting in December 2007. It will be installed in the underground Combined Coudé Laboratory (CCL) of the VLT, and will be linked to the four UT telescopes through four optical paths called Coudé Trains. Thanks to its special Front-End (F/E) subsystem, it will be able to collect the light of any UT independently or together, allowing several different configurations amongst which a 4-UT mode which will make it the first instrument mounted on a 16-m equivalent telescope, mode originally foreseen for the VLT. This linking scheme present several advantages, but also added complexity. These scientific and technical aspects are detailed later at the opto-mechanical level, as well as from the control or operational points of view.

2. SCIENCE WITH ESPRESSO 2.1 Context

Spectroscopy is a baseline tool of modern astrophysics. During the last two decades, the high precision radial velocity technique (RV) has been the most successful way to discover and characterize exoplanets. In that respect, highly stable and accurate spectrographs are amongst the cornerstones of the astrophysical instrumentation. High-resolution spectrographs have always been in high demand at major observatories (see e.g. HARPS@3.6, UVES@VLT or HIRES@Keck). As telescopes get larger, the capabilities of high-resolution spectrographs extend to fainter objects, but also the precision of the spectroscopy may be enhanced. The HARPS spectrograph at ESO-3.6m telescope [1] has been a pioneering instrument in this respect. It clearly appears today that high-precision spectroscopy has a great potential and many exciting discoveries are just around the corner.

The need for a ground based follow-up facility capable of high RV accuracy has been recently stressed in the ESO-ESA working group report on extrasolar planets. In fact, terrestrial planets in habitable zones are one of the main scientific topics of the next decades in Astronomy, and one of the main science drivers for the new generation of ELT. This ESO-ESA report [2], p. 63, states: a) high precision radial velocity instrumentation for the follow-up of astrometric and transit detections, to ensure the detection of a planet by a second independent method, and to determine its true mass. For Jupiter-mass planets, existing instrumentation may be technically adequate but observing time inadequate; for Earth-mass candidates, special purpose instrumentation (like HARPS) on a large telescope would be required. The same concept is reiterated in the first recommendation: b) Support experiments to improve radial velocity mass detection limits, e.g. based on experience from HARPS, down to those imposed by stellar surface phenomena” [2], p. 72.

Exoplanet research is just one trigger among many others for high-precision spectroscopy. Do the fundamental constants vary? This is one of the six big open questions in cosmology as listed in the ESA-ESO report for fundamental cosmology [3]. In the executive summary the document states: “Quasar spectroscopy also offers the possibility of better constraints on any time variation of dimensionless atomic parameters such as the fine-structure constant α and the proton-to-electron mass ratio. Presently there exist controversial claims of evidence for variations in α, which potentially relate to the dynamics of dark energy. It is essential to validate these claims with a wider range of targets and atomic tracers." This can be done only with improved spectroscopic capabilities.

ESPRESSO combines an unprecedented accuracy in the radial-velocities determination together with the largest photon collecting area available today. It will certainly provide breakthroughs in many areas of astronomical research, many of which we cannot anticipate.

2.2 Hunt for rocky exoplanets in the habitable zone

Since 2004, research teams using the RV technique have been making the headlines several times for their discoveries of low-mass extrasolar planets, some of which having only a few times the mass of the Earth. Of the few tens of exoplanets known today with a mass below 25 Earth Masses, the majority has been discovered with HARPS. The rate of these discoveries increases steadily. In the present high-precision program of HARPS it is estimated that more than 30% of the stars harbor extra-solar planets with masses below the mass of Neptune [4]. HARPS is unveiling an exoplanet population which was not known before. These discoveries were made possible thanks to the sub-m s-1 precision reached by this instrument. Given the faint magnitude of the star and/or their tiny radial-velocity signal induced by the planet, most of the observed objects would have remained out of reach of existing facilities that were limited to ~3m s-1. On the other hand, this emerging population is supposed to be only the tip of the iceberg. Considering the observational bias towards small masses, on the one hand, and the prediction made by planet-formation models [5], on the other hand, one should expect a huge amount of still undiscovered planets, even in already observed stellar samples. This has been confirmed recently by a statistical analysis of the HARPS survey [6] and a large number of planet candidates found by the Kepler satellite [7].

ESPRESSO is designed to explore this new domain and to enter into unknown territory. This goal can only be obtained by combining high efficiency with high instrumental precision. ESPRESSO will be optimized to obtain best radial velocities on quiet solar-type stars. A careful selection of these stars will allow focusing the observations on the best-suited candidates: non-active, non-rotating, quiet G to M dwarfs. With a precision of 10 cm s-1 or better it will be possible to detect rocky planets of few Earth masses in the habitable zone of solar type stars. By extending the sample towards the lighter M-stars the task becomes even ‘easier’. Given its efficiency, spectral resolution and spectral domain, ESPRESSO will be most efficient up to M4 dwarfs.

The discovery and the characterization of this new population of very light planets will open the door to a better understanding of planet formation and deliver new candidates for follow-up studies by transit, astrometry, transit spectroscopy, Rossiter-McLaugling effect, etc. ESPRESSO will also be the ideal (and maybe the only) machine to make spectroscopic follow-up of Earth-sized planets discovered by the transit technique.

2.3 Do fundamental physical constants vary?

Observations of metal lines of intervening galaxies along the line of sights of distant QSOs gave the first hints that the fine-structure constant might change its value over time, being lower in the past by about 6 parts per million [8],[9] and also µ = mp/me has been found to vary [10]. However, other recent results suggest no variation and even a spatial change of the fine-structure constant across the sky has been suggested [11]. The debate is very open and calls for new, more precise measurements. A high-resolution spectrograph such as ESPRESSO, which combines a large collecting area with extreme precision, can provide decisive advances. Thanks to its extreme stability it will provide an increase in the accuracy of the measurement of these fundamental constants by at least 1 order-of-magnitude. A confirmation of variability with high statistical significance is of crucial importance. It will open a window to new physics beyond the Standard Model and possibly shed new light on the nature of Dark Energy. On the other hand, even non-detections are important to further constrain various cosmological models [12]. Few QSOs are sufficiently bright to be observed in the 1-UT mode, but many more are observable in the 4-UT mode, making ESPRESSO an ideal instrument to explore new limits in the variation of fundamental physical constants.

2.4 A Pandora-box of other science

Given its efficiency, the high spectral resolution - and in particular the R = 200’000 mode -, the extreme high radial-velocity precision, and the possibility of combining up to 4 UTs, we expect that ESPRESSO will not only be able to fulfill the main scientific objectives, but also open a new parameter space in observational astronomy with hopefully many new and unexpected results. A preliminary and certainly incomplete list of additional unique and exciting sciences cases, which can be addressed with ESPRESSO by the astronomical community, will be given here:

• Chemical composition of stars in local galaxies • Investigation of metal-poor stars

• Asteroseismology and stellar physics • Diffuse stellar bands in the interstellar medium • Chemical enrichment of IGM • Galactic winds and tomography of the IGM • Chemical properties of protogalaxies • Cosmology • Further extra-solar planet science

o The Rossiter-McLaughlin effect o Transmission spectroscopy o Planets in nearby open clusters and galaxies

3. INSTRUMENT DESIGN OVERVIEW ESPRESSO has been designed following these major guidelines: a) to measure the Radial Velocity of stars for searching rocky planets, b) to produce stellar spectra of astronomical object able to tackle the variation of physical constants and c) to analyze the chemical composition of stars in nearby galaxies. Trade-offs considered during the phase A indicated as the best option the location of ESPRESSO in the VLT Combined Coudé Laboratory (CCL), fed by any or all of the VLT UTs, via optical relay systems deployed in the Coudé path of each UT and through the incoherent combined focus ducts, called Coudé trains, illustrated in Figure 1 and described in [13].

Figure 1 ESPRESSO Coudé train optical relay system

When operated for ESPRESSO, the selected VLT telescope acquires the object through the normal means and delivers its image to the Nasmyth Focus. The first element of the Coudé Train subsystem (The P4 prism), purposely inserted in the optical path when ESPRESSO is in use, transfers the image of the object and of at least 17 arcsec diameter area around it into the Coudé relay optical system. The Relay transports the image through the UT telescope fork into the specific UT Coudé room and through the tunnels to the Coudé Combined Laboratory (CCL).

Four similar but not identical Coudé trains are built, one per UT, in order to cope with each UT peculiarity, e.g. the length of the tunnel to the CCL, maintaining a similar interfaces with a Front-End Subsystem located in the CCL. The last element of the Coudé Train is the ADC. These devices, one per UT, are physically located on the Front-End bench but conceptually belong to the Coudé Train since they concur to materialize the Optical Interface with the Front-End.

The Front-End reimages the field received from the Coudé onto the focal plane on which the fiber-to- the-spectrograph feeding is located. In performing such a re-imaging the Front-End applies pupil stabilization (via a dedicated tip-tilt stage), field stabilization (via another dedicated tip-tilt), focusing (dedicated stage) and provide the mean to inject calibration light (Flat, ThAr, Comb) into the fiber-to- spectrograph if and when needed. A sketch of the Front-End is given in Figure 2; its mechanical configuration is described in [14], while the guiding algorithms applied are detailed in [15].

Figure 2 ESPRESSO Front-End subsystem

A Calibration Unit (CU) has the task to produce various calibration lights (balanced white, spectral reference) and to distribute it in the proper way to the Front-End unit. The various light sources include a Laser Driven Light Source (LDLS), a Thorium Argon lamp (ThAr), A Fabry-Perot Interferometer (FP) and a Laser Frequency Comb (LFC), which is formally another subsystem feeding via optical fibers its light to the CU where it is processed as another lamp. The CU includes two calibration selector modules (CSM), as two fibers for each Front-End have to be fed. One CSM is illustrated in Figure 3.

LDLS

ThAr 2

ThAr 1

LFC

FPCS

Spare

Variable ND Filter wheel

ND Filter in a wheel6pos.

A 1

A 3

A 2

A 4

Linearstage A1

Calibration Selector A

Linearstage A2(on A1)

Lens fixed to Fiber connection (not moving)

Lens fixed to exit Fiber connection (on stage 2)

Beamblock Figure 3 Calibration selector module for the fiber A

The science case for ESPRESSO calls for high precision spectroscopy and a relative uncertainty of about 10 cm/s. The whole instrument design is optimized to support such a performance. In order to deliver this kind of precision a laser frequency comb (LFC) is envisaged for wavelength calibration across the full spectral range. The biggest advantage of such a calibration source is that it is directly tied to the frequency standard. The use of LFCs is new in astronomy and is a quickly developing field. The first operational system deployed at an observatory will be at ESO's 3.6 m telescope with HARPS. The schematic of the HARPS set-up is shown in Figure 4 as an illustration of what can be expected for ESPRESSO. Currently, the ESPRESSO project is investigating several options for the LFC in a small but global market.

Figure 4 Laser Frequency Comb: a) schematics - b) spectra obtained on HARPS

The Fiber-link subsystem relays the light from the Front-End to the spectrograph and forms the spectrograph pseudo-slit inside the vacuum vessel. The 1-UT HR and UHR modes use a bundle of 2 octagonal fibers each, one for the object and one for the sky/calibration. In the HR mode the fiber has a core of 100µm, equivalent to 1 arcsecond on the sky; in the UHR mode the core is 50µm. In the 4-UT mode 4 bundles of 2 (object, sky/calibration) fibers each are simultaneously fed and combined into a single bundle of 2 (object, sky/calibration) square 200 µm fibers each.

An essential task performed by the Fiber-link subsystem is the Scrambling of the Near Field (NF) and Far Field (FF). A high scrambling gain, both in FF and NF, is a crucial issue to achieve the required RV accuracy. Studies and pre-development activities are currently been carried out to maximize the scrambling gain and minimize the consequent loss in transmission. The selection of octagonal fibers is an outcome of these studies. This subsystem is described in [16].

The main processing of the light in ESPRESSO is carried out by the Spectrograph Subsystem. The Spectrograph of ESPRESSO is a pupil-sliced, two-arm, bench-mounted, cross-dispersed Echelle spectrograph. The layout is shown in Figure 5 and described in detail in [17].

Figure 5 Optical Layout of the ESPRESSO Spectrograph

The major novelty of the spectrograph design with respect to existing cross-dispersed Echelle spectrographs is the adoption of an anamorphic pupil slicer. It is known that the resolving power of the spectrograph depends on the size of the pupil at the location of the grating. However only the size along the dispersion direction is relevant. By using an anamorphic collimator, it is possible to form an “ovoidal” pupil that preserves the required size only in the dispersion direction. In addition the anamorphic pupil can be “sliced” and overlapped allowing a reduction of the effective beam with no loss on resolution. In ESPRESSO this is used twice: a first time for the main Echelle grating and a second time for the Cross-Disperser which is a Slanted VPHG, a second important novelty in the design. The main design characteristics of the instrument are described in Table 1.

Table 1 Summary of the characteristics of ESPRESSO in its main modes

Parameter singleUHR singleHR multiMR

Wavelengths Blue arm: 380 – 520 nm Red arm: 520 – 780 nm

Spectral coverage Full Spectra format Up to 4 spectra per order (1 object + 1 reference fibers, 2 spectra / fiber) Resolving power 200’000 130’000 55’000 Aperture on sky 0.5 arcsec 1.0 arcsec 4x1.0 arcsec Spectral sampling >2 pixels >3.5 pixels >4 pixels (binned x 2) Spatial sampling >4 pixels >8 pixels >5 pixels (binned x 4) Sky/ Simultaneous reference Yes (mutually exclusive) Instrumental RV precision <10 cm/s <10 cm/s <1 m/s

The optical elements are assembled on a Nickel-plated Structural Steel optical bench, whose characteristics are given in Table 2.

Table 2 Optical bench characteristics

Material Ni-plated structural Steel St-52 Overall dimensions 4100 x 1500 x 1510 mm Weight ~ 3950 kg Maximum deformation 46.4 µm First Eigen-frequency 107.14 Hz

To stabilize the spectrograph environment, the optical bench is placed in a vacuum vessel where the temperature is stabilized to 0.001°K, under a pressure lower than 5x10-3 mbar (see Figure 6), and where no moving or dissipating parts are located.

Figure 6 The Vacuum vessel and the optical bench

Such a fine control is obtained with a multiple shells’ thermal enclosure, the first one permitting a stabilization of the temperature to the tenth of Kelvin, the second one to the hundredth, and the vacuum vessel being the ultimate shell.

Table 3 Requirements on the scientific detectors

Detector type Monolithic backside illuminated CCD detectors with 16 output amplifiers Light sensitive area ~ 90 mm x 90 mm Pixel size 9 or 10 µm Optical coating Individually optimized AR coatings for the specific spectral range Slow read mode < 60 sec full frame reading with RON < 3 e- RMS Fast read mode < 20 sec full frame reading with RON < 6 e- RMS CTE ≥ 0.999995, from few e- to full well Full well > 50’000 e- Dark current ≤ 1 e-/pixel/hour; target 0.5 e-/pixel/hour

The science detectors of ESPRESSO are two large format CCD detectors. ESPRESSO has a “red” and a “blue” arm optimized for the shorter and longer wavelengths in the useful spectral range. The optimization includes a specific detector per arm. The detectors are controlled by an NGC controller and accommodated in a cryostat to be operated in vacuum and at the given temperature. The requirements of the Call for Tender prepared for these detectors are presented in Table 3.

In Figure 7, we can see the layout of the various component of the ESPRESSO spectrograph in the CCL. One can remark that the Front-End is in a moderately stable temperature environment, as no regulation is applied, but the CCL being located underground, it is stable to ± 3°K.

Figure 7: Location of ESPRESSO in the CCL

4. INSTRUMENT CONTROL, OPERATIONS AND SOFTWARE The Control of the instrument is more complex than any traditional VLT instrument mainly because of the distribution of functions along a path of several tens of meters. The classic centralized LCU scheme does not apply to ESPRESSO. Delocalized Programmable Logic Controller (PLC) units talking with motorized functions and sensors via dedicated industrial buses form the core of the control hardware. These units are linked to one single Instrument Work Station (IWS) via TCP/IP protocol through Paranal LAN.

The Control Architecture is described in [18] and its general scheme is reported in Figure 8. Three PLCs, located in the Control Electronic Cabinets inside the CCL, control the F/E devices, the Vacuum & Thermal System, the Calibration Unit, two exposure meters and various sensors (temperature, etc.).

Two Next Generation Controllers (NGCs) control the scientific detectors, one dedicated to the blue arm and the other to the red arm. Six exposure shutters are present, one shutter per fibers pair. The devices of the four Coudé Trains, with the relative control hardware and software, are situated in the telescope area of each UT. In the final ESPRESSO configuration, the Coudé Train control part will be integrated in the VLT Telescope Control Software (TCS).

Figure 8 Control hardware

ESPRESSO has three main instrument modes, which correspond each to the choice of a particular fiber to feed the spectrograph. These modes are the following:

• singleHR (single High Resolution) mode. In this mode, ESPRESSO is used as a standard VLT instrument. Any, but only one, UT is used and its light brought to the instrument. Only one fiber of 1.0” feeds the spectrograph, resulting in a spectral resolution of about 130,000.

• singleUHR (single Ultra High Resolution) mode. In this mode also, ESPRESSO is used as a standard VLT instrument. Any, but only one, UT is used and its light brought to the instrument. Only one fiber of 0.5” feeds the spectrograph, thus resulting in a very high spectral resolution of about 200,000, at the cost of slit losses.

• multiMR (multi Mid Resolution) mode. In this mode, 2, 3 or 4 UTs are simultaneously used and connected to ESPRESSO. All the telescopes point the same source and are used as an up to 16-m equivalent telescope. The fibers from the used telescopes are brought together and their light coupled into a single larger fiber, resulting in a lower spectral resolution of about 55,000.

The VLT control room, where the users operate, hosts the User Stations for the instrument operations for the 4 UTs, and it will be possible to select ESPRESSO as the working instrument from any console, when in single(U)HR mode. The

VLTI user station will be used for ESPRESSO operations in multiMR mode. The Instrument Workstation (IWS), where the ESPRESSO control software runs, is located in the VLT computer room.

The control SW (CS) is developed conforming to the latest ESO standards, based as far as possible on the ESO generic standard instrument packages (ICB – ICS Base and ic0) provided with the ESO VLT common software kit. The control SW architecture is presented in Figure 9.

Figure 9 Control software

The complete Data Flow System (DFS) including the CS and the science software is shown in Figure 10. The DFS is described in [19]. As stated there, following the very positive experience gained with HARPS, ESPRESSO has been conceived since the beginning not as a “simple” standalone instrument, but actually as a “science-generating machine”. The final goal is to provide the user with scientific data as complete and precise as possible immediately after the end of an observation, increasing in this way the overall efficiency and the ESPRESSO scientific output. An integrated view is

thus mandatory and has been adopted already in the preliminary design phase of the project: operations consider all the aspects, from the observation preparation through instrument operations and control, to the data reduction and analysis. Since ESPRESSO will be installed at Paranal Observatory, all this will be integrated into ESO Paranal Data Flow environment ensuring compatibility, easiness of operations and maintenance.

Figure 10 Data Flow System

The success of the search for rocky planets relies not only on the precise measurements of stellar radial velocities (RV), but also and above all on well-balanced time series of these RVs. To schedule correctly the observations, a specific tool helping the observer to spread correctly the points over the whole stellar orbit is thus really important. A specific Observation Preparation SW (EOPS) tool is developed for the visitor mode observation, while in service mode, the support astronomer will use the standard ESO tools while the PI will be able to modify target priorities as he gets results from the preceding observations.

To allow reactivity, a complete Data Reduction Software (DRS) will enable the immediate removal of the instrumentation signature, as well as the computation of the radial velocity. The immediately reduced data will have scientific quality and will be the basic products sufficiently generic to be used as inputs for further analysis of all main ESPRESSO science cases.

The Data Analysis Software (DAS) is the part of the science software that is non-common to the various scientific domains. Due to the very high stability and resolution foreseen for ESPRESSO, we will supply the users with robust software modules giving reliable results and not just ‘quick look’ numbers, for most of the ESPRESSO science cases. When possible we will introduce automatic procedures and flexible software modules that can do the same operation for different kind of targets (e.g., quasars and stars).

5. PROJECT AND MANAGEMENT The Consumer is the Astronomical Community of the member states of ESO. Astronomers of these countries will be allowed to ask ESO for observing time with this instrument following exactly the same procedure as for any other ESO instrument. ESO ensures the necessary technical and organizational support. No exceptions or restrictions apply.

The Customer is ESO. On behalf of the Astronomical Community, ESO sub-contracts the realization of this instrument to the Contractor. ESO specifies the Technical Requirements and establishes a Statement of Work applicable to the Contractor, and verifies through Reviews held at various stages of the project that the project is led in fulfillment with

the defined technical and managerial frame. As customer, ESO supervises the integration of the instrument and the software within the VLT environment.

The Science Team is distinguished from the Consortium. The Science Team is only linked to the instrument by the GTO agreement with ESO. The Science Team composition ensures the scientific return for each of the partners. Its composition will thus be directly linked to the investment each partner makes for the construction project. An internal Memorandum of Understanding states the rules for the science team constitution, competences and rights.

The Contractor is a consortium composed of academic and research institutes of four countries, Italy, Portugal, Spain and Switzerland, as listed in Table 4. ESO instrumentation division is included in the consortium as a construction partner. The difference between ESO partner and the consortium institutes is due to the non-return of GTO, ESO rules prohibiting such an internal return.

Table 4 The consortium institutes

Astronomical Observatory, University of Geneva OGE Switzerland Physics Institute, University of Bern SRPS Switzerland Instituto de Astrofísica de Canarias, Tenerife IAC Spain Osservatorio di Trieste INAF OATS Italy Osservatorio di Brera INAF Brera Italy Centro de Astrofísica da Universidade do Porto CAUP Portugal Faculdade de Ciências da Universidade de Lisboa SIM Portugal Faculdade de Ciências da Universidade de Lisboa LOLS-CAAUL Portugal

The Astronomical Observatory of the University of Geneva is the lead institute of the consortium, where the principal investigator (PI) and the project manager (PM) are located. The system engineer (SE) is located in the Brera Observatory and the quality engineer (QE) in the Canarian Institute of Astronomy. The project scientist (PS) and the software system engineer (SSE) are both from the Trieste Observatory. Each consortium country is represented in the managerial structure by a PI or a co-PI, which is the single point of reference for the institutes located in his country and responsible for the financial contribution of his country and the related institutes. ESO partner is represented by a Co-PI with the same rights and tasks as the other, with the exception of the sharing of the scientific return.

Table 5 Project key persons

PI Switzerland PEPE Francesco OGE Co-PI Italy CRISTIANI Stefano INAF OATS Co-PI Spain REBOLO Rafael IAC Co-PI Portugal SANTOS Nuno CAUP Co-PI ESO DEKKER Hans ESO PM MEGEVAND Denis OGE SE ZERBI Filippo INAF Brera QE AMATE Manuel IAC SSE DI MARCANTONIO Paolo INAF OATS PS MOLARO Paolo INAF OATS

Each workpackage (WP) is attributed to a WP manager (WPM), and the tasks linked to this WP are executed in the WPM institute as much as possible, or subcontracted by this institute. This structure has the advantage to minimize the interfaces between institutes, as they are only external to the WP. The various WPs are listed in Table 6.

Table 6 Workpackages with the responsible individuals and institutes

Paranal Interface AVILA Gerardo ESO Coudé Train Optics CABRAL Alexandre LOLS-CAAUL Coudé Train Mechanics MOITINHO André SIM Coudé Train Electronics and SW ABREU Manuel LOLS-CAAUL Front-End Unit, RIVA Marco INAF Brera Calibration Unit BROEG Christopher SRPS Laser Frequency Comb KERBER Florian ESO Fiber Link RASILLA José Luis IAC Thermal and Vacuum System HUGHES Ian OGE Optical Design and Exposure Meter CONCONI Paolo INAF Brera Spectrometer Main Optics Procurement RASILLA Jose Luis IAC APSU Procurement TOSO Giorgio INAF Brera Cameras Procurement DEKKER Hans ESO Echelle Grating Assembly LIZON Jean-Louis ESO Spectrometer Opto-mechanics TENEGI Fabio IAC Scientific Detectors System IWERT Olaf ESO Dewar System LIZON Jean-Louis ESO Control Electronics CORETTI Igor INAF OATS Control Software DI MARCANTONIO Paolo INAF OATS Observation Software CIRAMI Roberto INAF OATS Dataflow System SOSNOWSKA Danuta OGE Observation Preparation Software SOSNOWSKA Danuta OGE Data Reduction Software LOVIS Christophe OGE Data Analysis Software D'ODORICO Valentina INAF OATS

The project started by a conceptual study called Phase A during which the Consortium translated the scientific requirements into general technical requirements. This phase ended in March 2010 by a review during which the baseline design of the instrument and general SW functions were presented to the ESO management. The real project was attributed to the consortium and started by two kick-off meetings (KOM), a formal one and a technical one (TKO) beginning of 2011. The milestones of the project are presented in Table 7. The detailed design will be finished in April 2013 by the Final Design Review (FDR), preceded for the Long-Lead Items (LLI) by two advanced final design reviews.

Table 7 Project milestones

Milestone From TP Planned Date Site TKO TP = 2011/02/01 January 17th – 18th, 2011 Lisbon PDR TP + 10 months November 29th – 30th, 2011 ESO Garching

FDR: multiple TP + 17-26 months June 18th, 2012 – April 2013 Porto & Garching IRR: multiple TP + 38-50 months April 2014 - April 2015 Consortium premises PAE: multiple TP + 52-56 months June - October 2015 Consortium premises

PAC TP + 70 months December 2016 ESO Paranal FA TP + 94 months December 2018 ESO

After one-year long in the manufacturing phase, an Integration Readiness Meeting (IRM) will be organized to prove the readiness of the integration plans. Before the integration phase, Integration Readiness Reviews (IRR) will start the

integration phases of the various subsystems. In 2015, the Preliminary Acceptance Europe reviews (PAE) will conclude these integrations. Transportation and further final integration in Chile will take place in 2016, before the Preliminary Acceptance Chile (PAC), which will conclude the project and leave the instrument ready for scientific usage. Its final acceptance (FA) will follow after two years of usage demonstrating the fulfillment of the scientific requirements.

6. CONCLUSIONS In its 4UT-MR mode, ESPRESSO will be the first instrument on a 16-meter telescope, the largest visible photon collector until the extremely large telescopes (ELT) will be available. In its single(U)HR modes, it will be a super-HARPS, combining the highest resolution for a spectrometer on a 10-meter class telescope with an unprecedented stability, leading to an instrument producing the cleanest and best-quality spectra, both at high and low signal-to-noise ratio. Combined with the extended spectral coverage from 380 to 780 nm in one shot, and its quasi real-time data processing, it will allow the astronomers to gather very high scientific results on the spot.

REFERENCES

1. Mayor, M., et al., “Setting new standards with HARPS”, “The ESO Messenger”, 114, 20 (2003) 2. Perryman, M., Dravins, D., Hainaut, O., Leger, A., Quirrenbach, A. and Rauer, H., in “ESA-ESO Working Group on Extra-Solar Planets” (2005) 3. Peacock, J. A., Schneider, P., Efstathiou, G., Ellis, J.R., Leibundgut, B., Lilly, S.J., and Mellier, Y., in “ESA-ESO Working Group on Fundamental Cosmology” (2006) 4. Mayor, M., Udry, S., Lovis, C., et al., “The HARPS search for southern extra-solar planets. XIII. A planetary system with 3 super-Earths”, A&A, 493, 639 (2009) 5. Mordasini, C., Alibert, Y., and Benz, W., “Extrasolar planet population synthesis. I. Method, formation tracks, and mass-distance distribution”, A&A 501, 1161 (2009) 6. Mayor, M., et al., “The HARPS search for southern extra-solar planets XXXIV. Occurrence, mass distribution and orbital properties of super-Earths and Neptune-mass planets”, ArXiv e-prints (2011) 7. Batalha, N., et al., “Planetary Candidates Observed by Kepler, III: Analysis of the First 16 Months of Data”, ArXiv e-prints (2012) 8. Webb, J. et al., “Search for Time Variation of the Fine Structure Constant”, Phy. Rev L. 82, 884 (1999) 9. Murphy, M. T., et al., in “Astrophysics, Clocks and Fundamental Constants”, eds. S. G. Karshenboim and E. Peik (Springer- Verlag: Berlin, Heidelberg), p. 131 (2004) 10. Ivanchik, A., Petitjean, P., Varshalovich, et al., “A new constraint on the time dependence of the proton-to-electron mass ratio. Analysis of the Q 0347-383 and Q 0405-443 spectra”, A&A 440, 45 (2005) 11. King et al, “Spatial variation in the fine-structure constant - new results from VLT/UVES”, MNRAS 422, 3370, Webb et al Phys. Rev L 107, 1101 (2012) 12. Thompson, R. I., “Constraints on quintessence and new physics from fundamental constants”, MNRAS 422, L67 (2012) 13. Cabral, A. et al., “ESPRESSO – Design and analysis of a Coudé-Train for a stable and efficient optical feeding from the four VLT unit telescopes”, in “Ground-based and Airborne Telescopes IV”, Proc. SPIE 8444-167 (2012) 14. Riva, M., et al., “ESPRESSO front end opto-mechanical configuration”, in “Ground-based and Airborne Instrumentation for Astronomy IV”, Proc. SPIE 8446-362 (2012) 15. Landoni, M., et al., “ESPRESSO front-end guiding algorithm”, in “Software and Cyberinfrastructure for Astronomy II”, Proc SPIE 8451-115 (2012) 16. Rasilla, J. L., et al, “ESPRESSO fiber link”, in “Ground-based and Airborne Instrumentation for Astronomy IV”, Proc SPIE 8446-335 (2012) 17. Spanò, P., et al., “Very high-resolution spectroscopy for the combined focus of the VLT: the ESPRESSO spectrograph optical design”, in “Ground-based and Airborne Instrumentation for Astronomy IV”, Proc SPIE 8446-303 (2012) 18. Cirami, R. et al., “Adoption of new software and hardware solutions at the VLT: the ESPRESSO control architecture case”, in “Software and Cyberinfrastructure for Astronomy II”, Proc. SPIE 8451-26 (2012) 19. Di Marcantonio, P., “Challenges and peculiarities of ESPRESSO Data Flow Cycle: from target choice to scientific results”, in “Observatory Operations: Strategies, Processes and Systems IV”, Proc. SPIE 8448-62 (2012)