The Messenger No. 178 - Quarter 4 | 2019 - European Southern Observatory

The Messenger No. 178 - Quarter 4 | 2019 - European Southern Observatory
ELT M4 — The Largest Adaptive Mirror Ever Built
A Celebration of GRAVITY Science
The ESO Summer Research Programme 2019
                                                                               The Messenger
                                                  No. 178 – Quarter 4 | 2019
The Messenger No. 178 - Quarter 4 | 2019 - European Southern Observatory
ESO, the European Southern Observa-                  Contents
tory, is the foremost intergovernmental
astronomy organisation in Europe. It is              Telescopes and Instrumentation
supported by 16 Member States: Austria,              Vernet E. et al. – ELT M4 — The Largest Adaptive Mirror Ever Built               3
­Belgium, the Czech Republic, Denmark,               Kasper M. et al. – NEAR: First Results from the Search for Low-Mass ­
 France, Finland, Germany, Ireland, Italy,             Planets in a Cen                                                               5
 the Netherlands, Poland, Portugal, Spain,           Arnaboldi M. et al. – Report on Status of ESO Public Surveys and
 Sweden, Switzerland and the United                    Current Activities                                                            10
 Kingdom, along with the host country of             Ivanov, V. D. et al. – MUSE Spectral Library                                    17
 Chile and with Australia as a Strategic
 Partner. ESO’s programme is focussed                GRAVITY Science
 on the design, construction and opera-              GRAVITY Collaboration – Spatially Resolving the Quasar Broad Emission
 tion of powerful ground-based observing              Line Region                                                                    20
 ­facilities. ESO operates three observato-          GRAVITY Collaboration – An Image of the Dust Sublimation Region in the
  ries in Chile: at La Silla, at P
                                 ­ aranal, site of    Nucleus of NGC 1068                                                            24
  the Very Large Telescope, and at Llano             GRAVITY Collaboration – GRAVITY and the Galactic Centre                         26
  de Chajnantor. ESO is the European                 GRAVITY Collaboration – Spatially Resolved Accretion-Ejection in
  ­partner in the Atacama Large Millimeter/           Compact Binaries with GRAVITY                                                  29
   submillimeter Array (ALMA). Currently             GRAVITY Collaboration – Images at the Highest Angular Resolution
   ESO is engaged in the construction of the          with GRAVITY: The Case of h Carinae                                            31
   Extremely Large ­Telescope.                       Wittkowski M. et al. – Precision Monitoring of Cool Evolved Stars:
                                                      Constraining Effects of Convection and Pulsation                               34
The Messenger is published, in hardcopy              GRAVITY Collaboration – Multiple Star Systems in the Orion Nebula               36
and electronic form, four times a year.              GRAVITY Collaboration – Probing the Discs of Herbig Ae/Be Stars at
ESO produces and distributes a wide                   Terrestrial Orbits                                                             38
variety of media ­connected to its activi-           GRAVITY Collaboration – Spatially Resolving the Inner Gaseous Disc of the
ties. For further information, including              Herbig Star 51 Oph through its CO Ro-vibration Emission                        40
postal subscription to The Messenger,                Davies C. L. et al. – Spatially Resolving the Innermost Regions of the
contact the ESO Department of Commu-                  Accretion Discs of Young, Low-Mass Stars with GRAVITY                          43
nication at:                                         Dong S. et al. – When the Stars Align — the First Resolved Microlensed Images   45
                                                     GRAVITY Collaboration – Hunting Exoplanets with Single-Mode
ESO Headquarters                                      Optical Interferometry                                                         47
Karl-Schwarzschild-Straße 2
85748 Garching bei München, Germany                  Astronomical News
Phone +498932006-0                                   Christensen L. L., Horálek P. – Light Phenomena Over ESO’s Observatories IV:                                     Dusk and Dawn                                                                51
                                                     Manara C. F. et al. – The ESO Summer Research Programme 2019                    57
The Messenger                                        Boffin H. M. J. et al. – Report on the ESO Workshop
Editor: Gaitee A. J. Hussain                            “Artificial Intelligence in Astronomy”                                       61
Layout, Typesetting, Graphics:                       Vieser W. et al. – Report on the IAU Conference
Jutta B
      ­ oxheimer, Mafalda Martins                       “Astronomy Education — Bridging Research & Practice”                         63
Design, P­ roduction: Jutta ­Boxheimer               Kokotanekova R., Facchini S., Hartke J. – Fellows at ESO                        67
Proofreading: Peter Grimley                          In Memoriam Cristian Herrera González                                           70
­w                             Personnel Movements                                                             71
                                                     Patat F. – Erratum: The Distributed Peer Review Experiment                      71
Printed by FIBO Druck- und Verlags GmbH
Fichtenstraße 8, 82061 Neuried, Germany

Unless otherwise indicated, all images in
The Messenger are courtesy of ESO,
except authored contributions which are
courtesy of the respective authors.
                                                     Front cover: Simulation of the orbits of stars very
                                                     close to the supermassive black hole at the heart of
© ESO 2019                                           the Milky Way, Sgr A*. One of these stars, S2, is the
ISSN 0722-6691                                       perfect laboratory to test Einstein’s general theory
                                                     of relativity as it passes very close to the black hole,
                                                     with an orbital period of 16 years. S2’s orbit has
                                                     been monitored with ESO’s telescopes since the
                                                     1990’s and continues at even greater precision with
                                                     GRAVITY. Credit: ESO/L. Calçada/

2            The Messenger 178 – Quarter 4 | 2019
The Messenger No. 178 - Quarter 4 | 2019 - European Southern Observatory
Telescopes and Instrumentation                                                                                DOI: 10.18727/0722-6691/5162

ELT M4 — The Largest Adaptive Mirror Ever Built

Elise Vernet 1                               (approximately a ninth of the full moon).

Michele Cirasuolo 1                          Thanks to the combined use of M4 and
Marc Cayrel 1                                M5, the optical system is capable of
Roberto Tamai 1                              ­correcting for atmospheric turbulence
Aglae Kellerer 1                              and the vibration of the telescope struc-
Lorenzo Pettazzi 1                            ture itself induced by motion and wind.
Paul Lilley 1
Pablo Zuluaga 1                              This adaptive capability is crucial to
Carlos Diaz Cano 1                           allowing the ELT to reach its diffraction
Bertrand Koehler 1                           limit, which is ~ 8 milliarcseconds (mas) in
Fabio Biancat Marchet 1                      the J-band (at λ ~ 1.2 μm) and ~ 14 mas
Juan Carlos Gonzalez 1                       in the K-band. In so doing the ELT will
Mauro Tuti 1                                 be able to yield images 15 times sharper
+ the ELT Team                               than the Hubble Space Telescope and
                                             with much greater sensitivity. Translated
                                             into astrophysical terms this means
    ESO                                      opening up new discovery spaces, from          Figure 1. Rendering of the M4 adaptive mirror unit
                                                                                            for the ELT.
                                             exoplanets closer to their stars, to black
                                             holes, to the building blocks of galaxies
The Extremely Large Telescope (ELT) is       both in the local Universe and billions of     consortium name of AdOptica. Many
at the core of ESO’s vision to deliver the   light years away. For example, the ELT         8-metre telescopes now have a metre-
largest optical and infrared telescope       will be able to detect and characterise        scale adaptive mirror. The same tech­
in the world. Continuing our series of       extrasolar planets in the habitable zone       nology is now being adapted to serve the
Messenger articles describing the opti-      around our closest star Proxima Centauri,      ELT, in order to produce a mirror with an
cal elements of the ELT, we focus here       or to resolve giant molecular clouds (the      area five times larger. The M4 mirror uses
on the quaternary mirror (M4), a true        building blocks of star formation) down to     the same principle as a loudspeaker; the
technological wonder; it is the largest      ~ 50 parsecs in distant galaxies at z ~ 2      mirror is made of a very thin shell levitating
deformable mirror ever made. In combi-       (and even smaller structures for sources       100 microns away from its reference sur-
nation with M5, M4 is vital to delivering    that are gravitationally lensed by fore-       face (this corresponds to the thickness
the sharp (diffraction-limited) images       ground clusters) with an unprecedented         of a standard A4 sheet of paper) and it
needed for science by correcting for         sensitivity.                                   acts like a membrane which deforms
atmospheric turbulence and the vibra-                                                       under the effect of about 5000 voice coil
tions of the telescope itself. Here we                                                      actuators. A voice coil actuator is a type
describe the main characteristics of M4,     The quaternary mirror (M4)                     of direct drive linear motor and the name
the challenges and complexity involved                                                      “voice coil” comes from one of its first
in the production of this unique adaptive    M4 is the main adaptive mirror of the tele-    historical applications, vibrating the paper
mirror, and its manufacturing status.        scope. The term “adaptive mirror” means        cone of a loudspeaker. It consists of a
                                             that its surface can be deformed to cor-       permanent magnetic field assembly and
                                             rect for atmospheric turbulence, as well       a coil assembly. The current flowing
Background: how the ELT works                as for the fast vibration of the telescope     through the coil assembly interacts with
                                             structure induced by its motion and the        the permanent magnetic field and gener-
Let’s briefly recall how the ELT works.      wind. In the case of M4, more than 5000        ates a force that can be reversed by
The optical design of the ELT is based on    actuators are used to change the shape         changing the polarity of the current.
a novel five-mirror scheme capable of        of the mirror up to 1000 times per second.
collecting and focusing the light from                                                      Depending on the current injected into
astronomical sources and feeding state-      In combination with the M5 mirror, M4          the coil the mirror can be pushed or
of-the-art instruments for the purposes of   forms the core of the adaptive optics of       pulled up to a distance of 90 microns
imaging and spectroscopy. The light is       the ELT. With a diameter of 2.4 metres,        from its mean position. With the help of
collected by the giant primary mirror        M4 will be the largest adaptive mirror ever    a very fast and precise set of capacitive
39 metres in diameter, relayed via the M2    built. By comparison, current adaptive         sensors and amplifiers that are co-located
and M3 mirrors (each of which has a          mirrors are just over 1 metre in diameter,     with the voice coil actuators, the mirror’s
diameter of ~ 4 metres) to the M4 and M5     for example the 1.1-m M2 adaptive sec-         position is measured 70 000 times per
mirrors that form the core of the adaptive   ondary on the VLT UT4 telescope (Yepun).       second to an accuracy of a few tens of
optics of the telescope; the light then                                                     nanometres (the size of the smallest virus)
reaches the instruments on one or other      Adaptive mirror technology was trans-          with the actuators being driven up to
of the two Nasmyth platforms. This           lated into an industrial product for astron-   1000 times per second.
design provides an unvignetted field of      omy more than two decades ago by
view (FoV) of 10 arcminutes in diameter      the Italian companies Microgate s.r.l and      M4 is made of several state-of-the-art
on the sky, ~ 80 square arcminutes           ADS, internationally known under the           components, the mirror and its reference

                                                                                            The Messenger 178 – Quarter 4 | 2019                 3
The Messenger No. 178 - Quarter 4 | 2019 - European Southern Observatory
Telescopes and Instrumentation                     Vernet E. et al., ELT M4 — The Largest Adaptive Mirror Ever Built

                                                                                                                             Figure 2. (Left) One of
                                                                                                                             the shell mirrors of the
                                                                                                                             M4 in Z
                                                                                                                                   ­ erodur©.

                                                                                                                             Figure 3. (Right) The
                                                                                                                             ­reference body in
                                                                                                                              silicon-­c arbide being
                                                                                                                              inspected after brazing
                                                                                                                              the six parts.

structure being two of the most critical           The back of the reference structure is

                                                                                                                                                        Mersen Boostec
ones. The mirror is an assembly of six             supported by a 12-point whiffletree and
optically polished thin shells, or petals,         laterally at six points on the mirror edge.
made of the low-expansion glass-ceramic            The overall M4 sub-system is mounted
Zerodur© (manufactured by Schott                   on six position actuators (a hexapod sys-
GmbH). The six petals are obtained from            tem), which provide the fine alignment
a 35 mm-thick blank, which is polished             of the mirror. It is further mounted on a
and thinned down to a thickness of less            rotating mechanism (called a switcher)
than 2 mm — necessary to achieve the               which is used to select the Nasmyth
desired flexibility for shaping the mirror —       focus to which the light will be directed.
and then finally cut into a precise shape
by Safran Reosc (France; see Figure 2).
                                                   Manufacturing the M4
In order to adjust the shapes of the thin
shells, a rigid and sufficiently accurate          Safran Reosc (France) started to manu-
flat reference structure is also needed to         facture the thin segment mirrors in 2017
hold the petals. This structure must be            and four thin shells are now ready for        Figure 4. Detail of the M4 reference body.
stiff enough to provide a good reference           integration in Italy. The remaining eight
surface, whatever the orientation of the           shells still need to be delivered in order
telescope. It also needs to hold all the           to have two sets of six shells each (during   The final integration will start at AdOptica
actuators, which will deform and change            ELT operation one set is integrated on        once the reference structure has been
the shape of the six petals.                       M4, while the other is being recoated).       delivered. Given the number of compo-
                                                   The reference body manufacturing also         nents that need to be assembled to a
The 2.7-metre diameter lightweight                 began in 2017 and six segments have           high degree of precision, the integration
­structure is made of Boostec® silicon             been brazed in the last few months. The       will be a lengthy task requiring proce-
 carbide, one of the stiffest materials            reference surface will need to be lapped      dures to ensure that the assembly and
 ­available (stiffer than steel, carbon fibre      to 5 microns flatness before being deliv-     calibration meet requirements. It should
  or beryllium). Its surface has more than         ered to Italy.                                take 1.5 years to fully integrate the M4
  5000 holes which will hold the actuators                                                       mirror and start the final calibration of
  (see Figure 4), while the back surface is        To have a mirror fully tested in Chile by     each mirror segment and their associated
  composed of several ribs to reinforce the        early 2024, AdOptica has to ensure            capacitive sensors. A test tower is being
  structure. Owing to its large dimensions,        the procurement and manufacture of all        specially developed to verify and test the
  the silicon carbide structure is made of         the other components, including all the       M4. It will be used in Europe to calibrate
  six parts brazed together, similar to the        voice coil actuators and more than 60%        the M4 unit before being transferred
  Herschel primary mirror which was man-           of the permanent magnets, which are           to Chile where it will be used before the
  ufactured more than a decade ago. The            already in house and are waiting to be        mirror is installed on the t­elescope and
  manufacture of the structure is signifi-         integrated. In addition, more than half of    kept on-site for any future major mainte-
  cantly challenging, not only because of          the electronics boards are either ready       nance activities that may be required.
  the depth, length, and thickness of the          or under calibration, and most of the
  ribs, but also given the requirements on         mechanical parts are ready, including the
  its straightness, as well as the number          reference structure cell support and its
  and accuracy of the actuator holes.              whiffletree.

4           The Messenger 178 – Quarter 4 | 2019
The Messenger No. 178 - Quarter 4 | 2019 - European Southern Observatory
Telescopes and Instrumentation                                                                               DOI: 10.18727/0722-6691/5163

 NEAR: First Results from the Search for Low-Mass
­Planets in a Cen

Markus Kasper 1                                ESO, in collaboration with the Break-         ESO 3.6-metre telescope at La Silla,
Robin Arsenault 1                              through Initiatives, has modified the VLT     was modified and used to carry out the
Ulli Käufl 1                                   mid-infrared imager VISIR to greatly          acceptance tests of the internal chopper.
Gerd Jakob 1                                   enhance its ability as a planet finder. It    This was followed by a performance eval-
Serban Leveratto 1                             has conducted a 100-hour observing            uation of the Annular Groove Phase Mask
Gerard Zins 1                                  campaign to search for low-mass plan-         (AGPM) coronagraph with a dedicated
Eric Pantin 2                                  ets around both components of the             optical setup incorporating a line-tunable
Philippe Duhoux 1                              binary a Centauri, part of the closest        CO2 laser, elliptical mirrors and germa-
Miguel Riquelme 1                              stellar system to the Earth. Using adap-      nium lenses. Four AGPM coronagraphs
Jean-Paul Kirchbauer 1                         tive optics and high-performance coro-        were tested, three specifically optimised
Johann Kolb 1                                  nagraphy, the instrument reached              for the NEAR filter (10–12.5 μm) and an
Prashant Pathak 1                              unprecedented contrast and sensitivity        older sample manufactured in 2012 and
Ralf Siebenmorgen 1                            allowing it to see Neptune-sized planets      optimised for wavelengths between 11
Christian Soenke 1                             in the habitable zone, if present. The        and 13.1 μm. Surprisingly, the older coro-
Eloy Fuenteseca 1                              experiment allowed us to characterise         nagraph performed best, with a rejection
Michael Sterzik 1                              the current limitations of the instrument.    ratio of up to 400 at 10.5 μm, and a con-
Nancy Ageorges 3                               We conclude that the detection of             trast level of < 10 – 4 at 3 λ/D.
Sven Gutruf 3                                  rocky planets similar to Earth in the
Dirk Kampf 3                                   habitable zone of the a Centauri System       After passing Provisional Acceptance
Arnd Reutlinger 3                              is already possible with 8-metre-class        Europe (PAE) in November 2018, the
Olivier Absil 4                                tele­scopes in the thermal infrared.          NEAR hardware was shipped to Paranal.
Christian Delacroix 4                                                                        At the same time, VISIR was dismounted
Anne-Lise Maire 4                                                                            from UT3 (Melipal) and brought to Para-
Elsa Huby 5                                    From an idea to the telescope                 nal’s New Integration Hall (NIH) in prepa-
Olivier Guyon 6, 7                                                                           ration for the on-site installation starting
Pete Klupar 7                                  The a Centauri system is uniquely suited      in early January 2019. As expected, three
Dimitri Mawet 8                                to the search for signatures of low-          cool-downs of VISIR were required to
Garreth Ruane 8                                mass planets in the thermal infrared. The     successfully implement all the new modi-
Mikael Karlsson 9                              N-band at around 10 μm is best suited         fications. First, the aperture wheel was
Kjetil Dohlen 10                               for such observations, because this           rearranged with the help of the Paranal
Arthur Vigan 10                                is where a planet with a temperature like     mechanical workshop to include two new
Mamadou N’Diaye 11                             Earth’s is brightest. The a Centauri          AGPMs and a special optical mask
Sascha Quanz 12                                binary consists of the solar-type stars       (ZELDA, N’Diaye et al., 2014) to measure
Alexis Carlotti 13                             a Centauri A and B, and the planet-­          and pre-compensate optical aberrations
                                               hosting (Anglada-Escudé et al., 2016)         in the instrument. New Lyot filters were
                                               M-dwarf star Proxima Centauri. In a           mounted and mechanically centred with
    ESO                                        ­previous Messenger article (Kasper et al.,   the cold stop of VISIR to an accuracy of
  	AIM, CEA, CNRS, Université Paris-           2017), we provided details of how we         better than 175 μm (i.e., 1% of the pupil
    Saclay, Université Paris Diderot,           planned to modify the existing VISIR         diameter). The internal chopper, the
    Sorbonne Paris Cité, Gif-sur-Yvette,        instrument to conduct the necessary          wavefront sensor arm and the calibration
    France                                      observations with the Very Large Tele-       unit were installed with the help of the
  	Kampf Telescope Optics (KT Optics),         scope (VLT). This article describes how      contractor KT Optics, and all units were
    Munich, Germany                             VISIR was moved to UT4, the innova­-         successfully tested. In particular, the
  	University of Liège, Liège, Belgium         tions and new technologies that were         alignment of the calibration unit, which
  	Observatoire de Paris-Meudon, France        implemented and how they work, con-          uses an elliptical mirror with an aberration-­
  	Subaru Telescope, Tokyo, Japan              cluding with the execution of the NEAR       free field of view of around 0.1 mm in
  	Breakthrough Initiatives, Mountainview,     (New Earths in the a Centauri Region)        diameter was laborious and required
    USA                                         experiment — a unique 100-hour obser-        some modifications of the mechanical
    Caltech, Pasadena, USA                      vation of the a Centauri system, which       mounts on-site.
    Uppsala University, Sweden                 took place in early June 2019.
   	L aboratoire d’Astrophysique Marseille,                                                 Following the completion of the assembly
    France                                     Three years were needed to develop the        integration and verification (AIV) activities,
   	Observatoire de la Côte d’Azur, Nice,     NEAR experiment from the initial idea,        VISIR was transported and mounted to
    France                                     from the Phase A review held in July 2016     UT4 (Yepun) in mid-March 2019 (see Fig-
   	Eidgenössische Technische Hochschule      to the observing campaign in June 2019.       ure 1). After measuring the expected
    Zürich, Switzerland                        Between January and July 2018, ESO’s          residual misalignment between the instru-
   	Institude de Planétologie et d’Astro­     mid-infrared detector test facility Thermal   ment and telescope pupil on-sky on
    physique de Grenoble, France               Infrared MultiMode Instrument (TIMMI2),       24 March, VISIR was taken off the tele-
                                               a decommissioned instrument from the          scope again for adjustment by tilting

                                                                                             The Messenger 178 – Quarter 4 | 2019        5
The Messenger No. 178 - Quarter 4 | 2019 - European Southern Observatory
Telescopes and Instrumentation                                             Kasper M. et al., NEAR: First Results from the Search for Low-Mass P
                                                                                                                                              ­ lanets in a Cen

the instrument, and some fine adjustment                                                                                                       Figure 1. (Left) VISIR

                                                  ESO/NEAR Collaboration
                                                                                                                                               mounted on UT4 and
of the wavefront sensor arm. On-sky
                                                                                                                                               ready for NEAR. The
commissioning started on 3 April 2019                                                                                                          alternative altitude cable
and lasted for 10 half-nights, during                                                                                                          wrap connecting the
which the various new functions were                                                                                                           instrument to the elec-
                                                                                                                                               tronics racks and helium
tested, and operational procedures were
                                                                                                                                               compressors on the
tuned.                                                                                                                                         ­a zimuth platform can be
                                                                                                                                                seen on the left hanging
                                                                                                                                                down from the mirror
Technical innovations, observing modes
and performance

NEAR implements several technologies
which are either completely new for
N-band astronomy or have not previously
been tested on-sky at this wavelength.
For example, the experiment confirmed
that atmospheric water vapour content
does not significantly impact the adaptive
optics (AO) corrected N-band image
quality, and that mid-infrared spectral fil-
ters can be overcoated with chromium
masks implementing Lyot stops or apo-
disers for the coronagraph. We also, for
the first time, implemented an alternative
altitude cable wrap (see Figure 1), which
could also greatly facilitate the operation
of other Cassegrain instruments.
                                                                                                                                               Figure 2. (Below) Illus-
                                                                                                                                               tration of the VISIR data
                                                                                                                                               acquisition of a Centauri
Chopping, internal and external                                                                                                                with chopping.

Among the new technologies is an internal
chopping device, the so-called Dicke
                                                                               a Cen B                                                    a Cen B
Switch, which is described in more detail
in Kasper et al. (2017). We tested the
Dicke Switch at chopping frequencies
                                                                                                                                                    a Cen A – a Cen B
up to 10 Hz during commissioning, and                                                    AGPM and WFS                                               (on AGM coro)
it substantially reduces the detector’s
Excess Low Frequency Noise (ELFN) as
                                                                                                                         a Cen A
foreseen. There is an expected mismatch                                                                                                                   a Cen A
in the spatial distribution of the sky and                                  Chop A                        Chop B                       Chop A – Chop B
internal background, but this mismatch
turns out to be stable in time and can be
well modelled or subtracted by nodding                                     (SPARTA)2 made sure that DSM chopping         the left and middle panels, and the
techniques. This device can be used when                                   observations are highly efficient and         ­chopping subtracted image of the two
external chopping is not possible — when,                                  almost transparent to the instrument. In       on the right.
for example, the source size exceeds the                                   addition, the a Centauri binary offers
throw range of an external chopper.                                        the possibility of chopping with an ampli-
                                                                           tude corresponding to the separation          Coronagraph modes and centring
The second option, external chopping                                       between the two stars of about 5 arcsec-
using the Deformable Secondary Mirror                                      onds in 2019, placing all the time a scien-   The light from the star at the location
(DSM), worked flawlessly. This option                                      tifically interesting target on the corona-   where we search for planets can be sup-
was initially deemed a risky approach,                                     graphic mask and doubling the efficiency.     pressed using two different concepts in
because the chopping action is seen by                                     Because of these advantages, we used          NEAR. The first is the AGPM, a technical
the AO and could have disturbed its                                        external chopping with the DSM for the        realisation of a Vortex coronagraph using
operation. However, the clever design of                                   a Centauri observations, and Figure 2         a sub-wavelength grating etched into a
the DSM and the Standard Platform for                                      illustrates the data as seen by the detec-    diamond substrate (Mawet et al., 2005).
Adaptive optics Real Time Applications                                     tor during the two chopping cycles on         The second is a shaped pupil mask

6          The Messenger 178 – Quarter 4 | 2019
The Messenger No. 178 - Quarter 4 | 2019 - European Southern Observatory
0.10                                                               9                                                                         120
Estimated pointing error (λ/D)

                                                                           y                                        8
                                                 0.05                                                               7

                                                                                                                                                                                                    Cumulative time (h)
                                                                                                                    6                                                                         80

                                                                                                   Time/night (h)
                                                 0.00                                                               5
                                                – 0.05                                                              3                                                                         40
                                                – 0.10                                                                                                                                        20
                                                                                                                    0                                                                         0
                                                 0.25                                                                   1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
Relative field selector position (arcseconds)

                                                                                                                                  Nights of 23 May–11 June, 26 June
                                                                                                                                    Time/night (h)           Cumulative time (h)
                                                                                                   Figure 3. (Left) Top: The shifting move-          Figure 4. (Above) Hours of open shutter time over
                                                                                                   ment of the star behind the AGPM is               the duration of the NEAR campaign. Only small
                                                 0.15                                              measured by the QACITS algorithm in               amounts of data could be collected between 24 and
                                                                                                   a closed loop during a NEAR observ-               28 May and between 5 and 11 June, owing to medio-
                                                                                                   ing night. The horizontal axis shows              cre (mostly cloudy) observing conditions.
                                                                                                   the hour angle and the colours refer to
                                                 0.10                                              the x and y directions. Over 4 hours,
                                                                                                   centered on meridian passage, the
                                                                                                   rms estimation is 0.015 λ/D for both              Figure 4 shows the campaign progress in
                                                                                                   the x and y directions. Bottom: The               data hours collected per night. The maxi-
                                                 0.05                                              relative positions of the field selector          mum possible time for which a Centauri
                                                                                                   recorded in the same night. The varia-
                                                                                                                                                     could be observed at an airmass smaller
                                                                                                   tion in the position of the field selector
                                                                                                   is due to differential atmospheric                than two is about seven hours in a good
                                                 0.00                                              refraction between the AO wave-                   observing night. Figure 4 shows, how-
                                                         –4   –2          0         2
                                                                                                   front-sensing channel and the science             ever, that there were several consecutive
                                                                                                   channel of VISIR. There was an AO
                                                                   Hour angle (h)                                                                    nights during the first and last weeks of
                                                                                                   interruption at hour angle ~ 2.5–3.
                                                                                                                                                     the campaign when either no or only small
                                                                                                                                                     amounts of data were recorded. These
  (­Carlotti et al., 2012), which does not sup-                                         This method estimates the offsets directly                   nights suffered from extended periods of
  press the overall light intensity, but modi-                                          from the images recorded on the detec-                       cloud coverage. Even thin high clouds,
  fies the light distribution in the focal plane                                        tor. The tests during the commissioning                      which can be acceptable for observa-
  so as to carve out a dark high-contrast                                               phase allowed us to optimise the QACITS                      tions in the near-infrared, are very detri-
  region at the relevant angular separation.                                            algorithm parameters and the observing                       mental for thermal infrared observations,
  Both concepts work well and improve                                                   strategy. It was shown that background                       because they lead to very high fluctua-
  the contrast by a factor of between 50 and                                            residuals after chopping have to be sub-                     tions in throughput and sky background.
  100. What tipped the balance towards                                                  tracted from the images analysed by
  the AGPM as the choice for the NEAR                                                   QACITS. After tuning, ­Q ACITS was able
  campaign was the higher throughput,                                                   to automatically centre the star on the                      Solid N2 on the coronagraph
  resulting in a moderately improved sensi-                                             AGPM and keep it there with an accuracy
  tivity overall and, more importantly, the                                             of 0.015 λ/D rms, almost one-hundredth                       There were, of course, a number of
  suppression of the high-intensity stellar                                             of a resolution ­element (Figure 3).                         smaller and larger problems during the
  image, thus avoiding detector electronics                                                                                                          long campaign and lots of stories to
  “ghosts”.                                                                                                                                          tell. Here is a particularly interesting one,
                                                                                        One hundred hours of observations                            which concerns one of the unknown
  As with all small inner working angle                                                                                                              unknowns that we encountered.
  coronagraphs, the AGPM performance                                                    ESO allocated 20 observing nights
  is sensitive to small offsets of the star                                             for the NEAR campaign between 23 May                         During the first few nights of the cam-
  behind the coronagraph (for example,                                                  and 11 June 2019 to observe the                              paign, we noticed that the contrast
  slow drifts). In order to actively control the                                        a Centauri system. Even though the                           ­provided by the coronagraph was less
  centring of a Centauri behind the AGPM                                                observing efficiency of NEAR is very                          effective than during commissioning, with
  during the observation, we implemented                                                high, with very small overheads for tele-                     a continuing slow degradation every other
  an algorithm called “Quadrant Analysis                                                scope sky offsets and data transfer                           night. While we were expecting a suppres-
  of Coronagraphic Images for Tip-tilt                                                  (well below 10%), the campaign struggled                      sion of the central point spread function
  Sensing” (QACITS; Huby et al., 2015).                                                 to collect the 100 hours of data desired.                     (PSF) by a factor of about 120, we started

                                                                                                                                                     The Messenger 178 – Quarter 4 | 2019                       7
The Messenger No. 178 - Quarter 4 | 2019 - European Southern Observatory
Telescopes and Instrumentation                     Kasper M. et al., NEAR: First Results from the Search for Low-Mass P
                                                                                                                      ­ lanets in a Cen

                                                                                                   Separation (λ/D)                                     Figure 5. Left: The deepest ever
                                                                                 0   5        10      15     20     25   30                             view of the habitable zone (indi-
                                                                         10   –2
                                                                                                                                                        cated by the dashed circle) around
                                                                                                                                                        a Centauri A; the 76-hour image
                                                                                                                                                        obtained during the NEAR cam-
                                                                                                                                                        paign — ~ 6 × 6 arcseconds.

                                                  Sensitivity (Jy, 5σ)
                                                                                                                                                        Right: Sensitivity and contrast
                                                                                                                                                        ­e stimated from the deep image as

                                                                         10 –3                                                       10 –5               a function of radial distance to the

                                                                                                    Affected strongly
                                                                                                    by ADI residuals

                                                                         10 –4                                                       10 –6

                                                                                 0       2         4         6       8          10
                                                                                             Separation (arcseconds)

with a factor of 80, which degraded to             out the mini-warmup and cool-down in                                       to the position between the two off-axis
only a factor of 40 about a week into the          the morning and be back in business in                                     PSFs and binned the surviving 76 hours
campaign. Somewhat frustratingly, none             less than 10 hours, sufficiently quick to                                  of data to 1-minute time resolution, which
of the typical external effects that degrade       be ready for the following observing                                       is short enough to avoid any noticeable
coronagraph efficiency (for example, Lyot          night. And it was a success! The corona-                                   smearing of the images because of field
stop misalignments or optical aberra-              graphic rejection recovered with each                                      rotation. This procedure compressed the
tions) could explain the shape of the              mini-warmup, and starting from 1 June                                      full campaign into ~ 4600 frames or 3 Gb.
residual image that we observed. It really         we repeated this procedure approximately
looked like an intrinsic degradation of the        every three days.                                                          A relatively simple high-contrast imaging
coronagraphic mask itself.                                                                                                    analysis can help evaluate the detection
                                                                                                                              limits reached during the campaign. By
Could air, entering the cryostat through           The data and preliminary results                                           sorting all the frames according to their
a known tiny leak, freeze out on the                                                                                          parallactic angle, we run a PSF calibration
20-Kelvin cold coronagraphic mask and              The campaign data were taken at a                                          procedure based on principal component
produce the loss of contrast? A back-of-           VISIR/NEAR detector frame rate of 166 Hz,                                  analysis using all the frames, i.e., pro-
the-envelope estimate showed that such             i.e., the detector integration time (DIT)                                  cessing the campaign as a whole rather
a leak could indeed build up an ice layer          was 6 ms. Chopping ran at 8.33 Hz for                                      than night-by-night. The calibrated images
of a few microns thickness every day.              most of the campaign, and each chop-                                       are then combined using noise-weighted
With the refractive index of solid nitrogen,       ping half-cycle thus lasted 60 ms. During                                  averages in order to properly take into
the main constituent of air, ice partly            this time span, 48 ms or 8 DITs were                                       account the rather large variations in the
entering the grooves of the AGPM coro-             averaged into a single frame, and 12 ms                                    sky background.
nagraphic mask could change the opti-              or 2 DITs were skipped for the transition
cal depth of the grooves sufficiently to           of the DSM between the two chopping                                        Figure 5 shows the result of this simple
degrade the performance.                           positions. Each 30-second data file                                        data reduction and the contrast sensitivity
                                                   ­consists of 500 half-cycle frames, and                                    achieved. The 5σ background-limited
So, how were we to test this theory, and            the 100 hours of data add up to 6 million                                 sensitivity far away from the star is of the
even more importantly, fix it during the            frames or 6 Tb.                                                           order of 100 μJy, which is consistent
campaign as it was running? Solid nitro-                                                                                      with our initial goals. At ~ 1.1-arcsecond
gen starts to sublimate at a sufficiently          Before entering advanced high-contrast                                     separation, i.e., at the angular size of
high rate to de-ice the coronagraph mask           imaging data reduction, some pre-pro-                                      the habitable zone around α Centauri A,
at temperatures that are only moderately           cessing was necessary to remove bad                                        the sensitivity is reduced to about 250 μJy
higher than the nominal 20 Kelvin. It              frames and reduce the data volume to a                                     mostly by the central glow of the AGPM.
turned out that the temperature after the          more manageable size. We removed                                           This does not yet mean that a point
first stage of the instrument warmup,              frames with extremely high or variable                                     source can readily be detected at this
­lasting just a few hours, is 35–40 Kelvin.        background produced, for example, by                                       level, but first estimates using a fake
 Tricking the PLC-controlled system into           thin clouds or low encircled energy for                                    injected source show that a planet of
 stopping the warmup sequence after the            the off-axis stars during ineffective AO                                   ~ 350 μJy brightness corresponding to a
 first stage and going into cooling again          correction, and frames with low corona-                                    temperate Neptune could indeed be seen.
 was risky (a glitch could have resulted in        graphic suppression through bad cen-
 a full warmup which would have taken              tring of the PSF on the coronagraph                                        No planet candidate of the size of Neptune
 out VISIR for several days), but it paid          mask. Finally, we cropped the images to                                    or larger was found in the data so far.
 off. A procedure was developed to carry           400 × 400 pixels, carefully centred them                                   While we were obviously hoping for a

8          The Messenger 178 – Quarter 4 | 2019
The Messenger No. 178 - Quarter 4 | 2019 - European Southern Observatory
detection, the result can also be seen as      While no planet candidates have been           telescopes, Gemini South and Magellan,
                         good news for the existence of rocky           found so far, NEAR is already a very           true Earth analogues could soon be
                         planets, which may therefore still exist in    ­successful collaboration between ESO,         discovered.
                         the habitable zone of a Centauri in a sta-      the Breakthrough Initiatives1 and many
                         ble orbit. There is also a roughly 35%         partners in the exoplanet and mid-infrared
                         chance that an existing planet would           astronomy communities. Several key             Acknowledgements
                         have been hidden by the star as the            technologies for mid-infrared high-contrast    The NEAR experiment greatly benefited, and still
                         result of an unfavourable projected orbital    imaging were successfully tested on-sky,       benefits, from the exchange with the exoplanet
                         position during our single-epoch obser-        and many important assumptions were            and mid-infrared scientific community on both sides
                         vation. In addition, the image in Figure 5     validated — for example, the scaling of        of the Atlantic. We would like to thank Derek Ives
                                                                                                                       for access to the Infrared Lab at ESO, Paranal’s
                         shows some straight lines connecting the       the achieved signal-to-noise ratio with the    mechanical workshop for the excellent support dur-
                         coronagraphic centre field with the off-       square-root of the observing time.             ing the integration on-site, and Rus Belikov, Eduardo
                         axis stellar image to the lower left. These                                                   Bendek, Anna Boehle, Bernhard Brandl, Christian
                         streaks appear because of a small per-         All raw data obtained during the 100-hour      Marois, Mike Meyer and Kevin Wagner for very help-
                                                                                                                       ful discussions and their interest in the data analysis.
                         sistence in the detector, i.e., the pixels     α Centauri campaign are publicly available,    Many thanks go also to our industrial partners KT
                         remember the stars being dragged over          and a condensed easy-to-use 3 Gb               Optics, Optoline and the Infrared Multilayer Labora-
                         the detector during the chopping transi-       package of all the good frames is availa-      tory of the University of Reading (now Oxford), for
                         tion. This feature is difficult to model and   ble on request 3. The on-sky contrast at       their R&D spirit and their willingness to stay with us
                                                                                                                       during the rapid development of the experiment.
                         may hide another 5–10% of the possible         3 λ/D and the N-band sensitivity are
                         planet orbits.                                 unprecedented in ground-based astron-
                                                                        omy by a large margin — more than one          References
                                                                        order of magnitude. The sensitivity limits
                                                                                                                       Anglada-Escudé, G. et al. 2016, Nature, 536, 437
                         Beyond NEAR                                    are well understood and could be               Carlotti, A. et al. 2012, Proc. SPIE, 8442, 844254
                                                                        improved further by a factor 2–2.5, mainly     Huby, E. et al. 2015, A&A, 584, A74
                         Preliminary results of the NEAR commis-        by removing the AGPM glow by introduc-         Ives, D. et al. 2014, Proc. SPIE, 9154, 91541J
                         sioning and experiment have triggered          ing a small optical relay incorporating a      Kasper, M. et al. 2017, The Messenger, 169, 16
                                                                                                                       Lagage, P. O. et al. 2004, The Messenger, 117, 12
                         substantial interest within the community      cold pupil stop in front of the AGPM. But      Mawet, D. et al. 2005, ApJ, 633, 1191
                         in this facility, and also for other astro-    this is still not the limit for mid-infrared   N’Diaye, M. et al. 2014, Proc. SPIE, 9148, 91485H
                         nomical observations. ESO therefore            observations from the ground. A novel
                         issued a call for Science Demonstration        lower-noise detector technology is
                         proposals, which received a lot of atten-      emerging, which promises to double the
                         tion and resulted in 26 proposals being        sensitivity once more. These next-gener-       1
                                                                                                                          reakthrough Initiatives webpage: http://break-
                         submitted for NEAR observing time. Two         ation detectors would allow the VLT to 
                         periods of Science Demonstration were          probe the rocky planet regime in the hab-        SPARTA:
                         allocated in September and December            itable zone around a Centauri. When            3
                                                                                                                          Data can be requested via e-mail from Prashant
                         2019 to conduct roughly half of the pro-       combined with similar instruments at the           Pathak ( or Markus Kasper
                         posed programmes.                              other southern hemisphere 8-metre-class            (
ESO/NEAR Collaboration

                                                                                                                       The NEAR experiment being mounted on the
                                                                                                                       Cassegrain focus of the VLT’s UT4 (Yepun).

                                                                                                                       The Messenger 178 – Quarter 4 | 2019                  9
The Messenger No. 178 - Quarter 4 | 2019 - European Southern Observatory
Telescopes and Instrumentation                                                                                    DOI: 10.18727/0722-6691/5164

Report on Status of ESO Public Surveys and
Current Activities

Magda Arnaboldi 1                                 ESO Public Surveys: overview of                  sis, including the timeline for the delivery
Nausicaa Delmotte 1                               engagement rules and status                      of science data products over the entire
Dimitri Gadotti 1                                                                                  duration of the survey project. The
Michael Hilker 1                                  By design, the ESO Public Surveys cover          approval of the survey management plan
Gaitee Hussain 1                                  a variety of research areas, from the            is further confirmed by the agreement
Laura Mascetti 2                                  detection of planets via micro-lensing,          signed between ESO’s Director General
Alberto Micol 1                                   through stellar variability and evolution, the   (DG) and the Principal Investigator (PI) of
Monika Petr-Gotzens 1                             Milky Way and Local Group galaxies, to           each survey.
Marina Rejkuba 1                                  extragalactic astronomy, galaxy evolution,
Jörg Retzlaff 1                                   the high-redshift Universe and cosmol-           The agreement between the DG and the
Chiara Spiniello 1, 3                             ogy. Differently from Large Programmes,          PI specifies the milestones for the data
Bruno Leibundgut 1                                these projects are planned to span more          releases and their content and responsi-
Martino Romaniello 1                              than four semesters and last for many            bility for the scientific quality and accu-
                                                  years. For example, the latest call for the      racy of the data products, which is to be
                                                  Cycle 2 survey projects for the Visible          warranted by the Public Survey team
   ESO                                            and Infrared Survey Telescope for Astron-        under the leadership of the PI. The agree-
   Terma GmbH, Darmstadt, Germany                 omy (VISTA) required them to span a time         ment states that a final release which
   Astronomical Observatory of                   interval of more than three years. These         includes the reprocessing of the entire
  ­Capodimonte, Naples, Italy a                   survey projects all have a legacy value for      data set is expected upon completion of
                                                  the community at large in addition to the        data acquisition for each survey. This final
                                                  science goals identified by the proposing        data release should take place within
This report on the status of the ESO              teams.                                           one year of completion of the data acqui-
Public Surveys includes a brief overview                                                           sition for any survey. The PSP was set up
of their legacy value and scientific                                                               to periodically review the progress of the
impact. Their legacy is ensured by their          ESO science policies for Public Surveys          surveys and to assess compliance with
homogeneity, sensitivity, large sky                                                                the specification of the survey products.
­coverage in multiple filters, large num-         The selection of ESO Public Surveys is           In May 2019, a PSP review took place to
 ber of targets, wavelength coverage              a two-step process which starts with the         evaluate the scientific impact of the active
 and spectral resolution, which make              submission of letters of intent. On the          Public Surveys.
 them useful for the community at large,          basis of these letters, the Public Survey
 extending beyond the scientific goals            Panel (PSP) formulates a coherent, well-
 identified by the survey teams. In May           balanced scientific programme that takes         Operations for ESO Public Surveys
 2019, as almost all first-generation             into account any synergies among teams
 imaging and spectroscopic surveys                in the community and the international           The ESO Public Survey observations —
 completed their observations and second-         survey projects. The PSP then provides           whether in service mode or visitor mode
 generation imaging surveys got well              recommendations to ESO including a list          — are carried out according to the pro-
 underway, the Public Survey Panel                of the teams that should be invited to           cess defined by the ESO Data Flow Sys-
 reviewed the scientific impact of these          submit full proposals on the basis of the        tem. The raw data acquired for ESO
 projects. The review was based on a              ranking of the descriptions of their sci-        ­Public Surveys are immediately public.
 quantitative assessment of the number            ence projects as provided in the letters of       Once the Public Survey teams have car-
 of refereed publications from the survey         intent. In so doing, the PSP fosters active       ried out data reduction to remove instru-
 teams and archive users. It included             collaborations within the community by            mental signatures, calibrate the data and
 the number of citations, the number of           asking independent teams to join, encour-         complete the measurements defined by
 data releases and statistics on access           aging them to optimise science goals              their scientific goals, ESO assists the
 to archive data by the user community.           and observing strategies, and sharing             ­survey teams to define and package their
 The ESO Users Committee also dis-                resources.                                         data products in a manner consistent
 cussed the availability and usage of                                                                with the ESO Science Archive and Virtual
 ESO Public Survey data by the commu-             Once the proposals have been recom-                Observatory standards and in agreement
 nity during their yearly meeting in April        mended for approval by the PSP and                 with the specifications in the survey
 2019. We describe the status of these            the Observing Programmes Committee,                ­management plans. The goal is to inte-
 projects with respect to their observing         data acquisition for each ESO Public                grate science data products from the
 plans, highlight the most recent data            ­Survey starts. This involves the review            Public Surveys into the ESO Archive,
 releases and provide links to the result-         and assessment of each survey manage-              together with the entire archive content
 ing science data products.                        ment plan by the ESO Survey Team.                  from the La Silla Paranal Observatory.
                                                   The survey management plan is an                   This is done via the Phase 3 process,
                                                   essential tool for the survey team, as well        which is an audit process that certifies
                                                   as for operations at ESO; it details the           the integrity, consistency and data quality
                                                   data acquisition plan, and the allocated           of the products available from the ESO
                                                   resources for data processing and analy-           Archive and ensures a homogeneous

10         The Messenger 178 – Quarter 4 | 2019
user experience once the data are pub-                                                 120%
lished through the Archive.

                                                 Percentage of completion (OB hours)

ESO Public Survey status
A total of twenty Public Surveys1 have
been carried out by consortia in the com-                                               60%
munity and are actively supported by
ESO. The majority have completed data
acquisition using ESO facilities and are                                                40%
in the process of publishing science data
products via the ESO archive.                                                           20%
ESO Public Surveys were launched in                                                                                                                      VPHAS+
2005 with an initial call for the optical
                                                                                           1 8
                                                                                           12 2
                                                                                           1 4
                                                                                           1 8
                                                                                          13 2
                                                                                          1 4
                                                                                           1 8
                                                                                          1 2
                                                                                          1 4
                                                                                           1 8
                                                                                          15 2
                                                                                          1 4
                                                                                           1 8
                                                                                          16 2
                                                                                          1 4
                                                                                           1 8
                                                                                           17 2
                                                                                           1 4
                                                                                          18 2
                                                                                          1 4
                                                                                           1 8
                                                                                          19 2
imaging surveys at the VLT Survey Tele-
                                                                                        20 1-1

                                                                                        20 2-1

                                                                                        20 3-1

                                                                                        20 4-1

                                                                                        20 5-1

                                                                                        20 17-1
                                                                                        20 6-1

                                                                                        20 8-1
                                                                                        20 - 0
                                                                                         20 1-0

                                                                                        20 2-0

                                                                                        20 - 0
                                                                                        20 3 - 0

                                                                                        20 4 - 0
                                                                                        20 4 - 0

                                                                                        20 - 0
                                                                                        20 5 - 0

                                                                                        20 - 0

                                                                                        20 - 0

                                                                                        20 - 0
                                                                                        20 6 - 0

                                                                                         20 7-0

                                                                                        20 8 - 0


scope (VST; Capaccioli & Schipani, 2011),

followed by a call for the near-infrared                                                                 Date
surveys (Cycle 1) in 2007 (Arnaboldi et al.,
2007) at VISTA (Sutherland et al., 2015).                                              120%
                                               Percentage of completion (OB hours)

Once the imaging surveys were under                                                                                                               VHS
way, ESO opened a first call for Public                                                100%                                                       UltraVISTA
Spectroscopic Surveys in 2011, followed                                                                                                           VIDEO
by a second call for the VIMOS Public                                                  80%                                                        VVV
Spectroscopic Surveys in 2015. The call                                                                                                           VMC
for Cycle 2 VISTA imaging Public Surveys                                               60%
was opened in 2015 and the selected                                                                                                               VVVx
surveys began in April 2017 (Arnaboldi et                                                                                                         G-CAV
                                                                                       40%                                                        VEILS
al., 2017). Four of the seven Cycle 2 VISTA
surveys exploit the time domain: for                                                                                                              SHARKS
example, following up exotic transients                                                20%                                                        UltraVISTA-New
like the optical-near-infrared echo of grav-                                                                                                      VISIONS
itational wave (GW) events (VinRouge);                                                  0%                                                        VinRouge
studying the 3D shape of the Milky Way
                                                                                          10 0
                                                                                           1 2
                                                                                           11 0
                                                                                           1 2
                                                                                           12 0
                                                                                           1 2
                                                                                          13 0
                                                                                           1 2
                                                                                           14 0
                                                                                           1 2
                                                                                          15 0
                                                                                           1 2
                                                                                          16 0
                                                                                           1 2
                                                                                           17 0
                                                                                            1 2
                                                                                          18 0
                                                                                           1 2
                                                                                          19 0
                                                                                          19 2
                                                                                        20 7-1
                                                                                        20 0-1

                                                                                        20 2-1
                                                                                        20 9-1

                                                                                        20 1-1

                                                                                        20 3-1

                                                                                        20 4-1

                                                                                        20 5-1

                                                                                        20 6-1

                                                                                        20 8-1
                                                                                        20 - 0

                                                                                         20 - 0

                                                                                        20 - 0

                                                                                        20 - 0

                                                                                        20 - 0

                                                                                        20 - 0

                                                                                        20 - 0

                                                                                         20 - 0

                                                                                        20 - 0

                                                                                        20 - 0

bulge (VVVX) via astrometry — to test

stellar evolution models and microlensing,                                                           Date
and to obtain proper motion membership
(VVVX, VISIONS); or detecting high-z                 Figure 1. (Upper) Cumulative curves for the comple-        Figure 2. (Lower) Cumulative curves of completion
                                                     tion of the VST surveys. The VST ATLAS survey was          for Cycle 1 and 2 VISTA Surveys. The cumulative
supernovae (SN) in cosmological deep
                                                     extended after the completion of the observing plan        curves of completion can reach values over 100% for
fields (VEILS). Two of the seven Cycle 2             to allow further coverage in the u’g’r’-bands (as out-     Public Surveys that were compensated for time cor-
surveys, VVVX and the Continuing Ultra-              lined in its Survey Management Plan). The VPHAS+           responding to low-quality observations (OBs with a
VISTA, follow up the successful Cycle 1              curve is below 100% because the PI requested that          D grade). Public Survey teams can ask for compen-
                                                     it be terminated early.                                    sation via reports submitted to the OPC.
surveys, very much in the spirit of other
surveys such as the Sloan Digital Sky
Survey (SDSS).                                       and 2018, except for the VHS South Pole                    observing time — in hours for the imag-
                                                     fields. The data acquisition for the Cycle 2               ing surveys and in nights for the spectro-
In the optical, the VST Public Surveys               VISTA imaging surveys is two-thirds                        scopic surveys.
completed their data acquisition in                  ­complete. In Figure 2 we show the cumu-
Period 104. The V-ATLAS survey was                    lative curves of data acquisition for all
granted an extension by the PSP to                    VISTA surveys. The data acquisition for                   Scientific impact of ESO Public Surveys
acquire the u’g’r’-band imaging of chosen             the four spectroscopic surveys, including
sub-areas. The data acquisition for this              the two that were carried out with the                    A standard reference metric for the
extension is ongoing and completion is                VIMOS spectrograph, has also been                         assessment of scientific impact is given
expected in Period 105. In Figure 1 we                completed.                                                by the number of refereed publications
show the cumulative curves of data acqui-                                                                       from the Public Survey teams. Given the
sition for the VST surveys. The data                 In Tables 1 and 2 we provide a summary                     legacy value of these projects and the
acquisition for the Cycle 1 VISTA imaging            of the observational parameters for the                    science data products readily available
surveys was completed between 2015                   twenty ESO Public Surveys and the total                    for download via the ESO Archive, other

                                                                                                                The Messenger 178 – Quarter 4 | 2019            11
Telescopes and Instrumentation                           Arnaboldi M. et al., Report on Status of ESO Public Surveys and Current Activities

VST Survey ID                                                  Science                 Area (square           Filters            Magnitude limits       Total time
                                                                                       degrees)                                                         (hours)
KiDS — Kilo-Degree Survey                                      Extragalactic           1350 b                 u’ g’ r’ i’        24.1 24.6 24.4         3421                                                                                                  23.4
(de Jong et al., 2013)
ATLAS                                                          Wide area/baryon        4700 c                 u’ g’ r’ i’ z      22.0 22.2 22.2         1585                     acoustic oscillations                                             21.3 20.5
(Shanks et al., 2013)
VPHAS+ — VST Photometric Hα Survey of the Southern             Stellar astrophysics    1800 d                 u’ g’ Hα r’ i’     21.8 22.5 21.6         1200
Galactic Plane                                                                                                                   22.5 21.8 (Drew et al., 2013)

VISTA Cycle 1                                                  Science                 Area (square           Filters            Magnitude limits       Total time
                                                                                       degrees)                                                         (hours)
UltraVISTA                                                     Deep high-z             1.7 Deep               Y J H Ks           25.7 25.5 25.1         1832                                            0.73 Ultra deep        NB118              24.5 26.7 26.6
(McCracken et al., 2013)                                                                                                         26.1 25.6 26.0
VHS — VISTA Hemisphere Survey                                  Southern sky            17 800                 Y J H Ks           21.2 21.1 20.6         4623                                                                                               20.0
(McMahon et al., 2013)
VIDEO — VISTA Deep Extragalactic Observations Survey           Deep high-z             12                     Z Y J H Ks         25.7 24.6 24.5         2073                                                                                 24.0 23.5
(Jarvis et al., 2013)
VVV — VISTA Variables in the Via Lactea                        Milky Way               560                    Z Y J H Ks         21.9 21.1 20.2         2205                                                                                                            18.2 18.1
(Hempel et al., 2014)
VIKING — VISTA Kilo-Degree Infrared Galaxy Survey              Extragalactic           1500                   Z Y J H Ks         23.1 22.3 22.1         2424                                                                                                       21.5 21.2
(Edge et al., 2013)
VMC — VISTA Magellanic Clouds Survey                           Resolved star           180                    Y J Ks             21.9 21.4 20.3         2047                           formation history
(Cioni et al., 2013)

VISTA Cycle 2                                                  Science                 Area (square           Filters            Magnitude limits       Total time
                                                                                       degrees)                                                         (hours)
VINROUGE* — Kilonova counterparts to gravitational wave        Kilonova                300                    Y J Ks             21.0 21.0 20.1         77
sources                                                        counterparts to                        GW sources
(Tanvir et al., 2017)
Cont. UltraVISTA — Completing the legacy of UltraVISTA         High-z                  0.75                   J H Ks             26.0 25.7 25.3         567
VVVX* — Extending VVV to higher Galactic latitudes             Milky Way               1700                   J H Ks             Ks = 17.5              1631
VEILS* — VISTA Extragalactic Infrared Legacy Survey            Galaxy evolution,       9                      J Ks               J < 23.5               847      AGN, SN                                                           Ks < 22.5
G-CAV — Galaxy Clusters At VIRCAM                           Galaxy clusters            30                     Y J Ks             24.5 24 23             440
VISIONS* — VISTA star formation atlas                          Star formation          550                    J H Ks             21.5 20.5 19.5         449                                   atlas
SHARKS — Southern Herschel-Atlas Regions                       Near-infrared           300                    Ks                 22.7                   929
Ks-band survey                                                 counterparts for                                     radio sources

Table 1. (Upper) VLT Survey Telescope Public Sur-        Table 2. (Centre) Cycle 1 VISTA Public Surveys;           Table 3. (Lower) Cycle 2 VISTA Public Surveys
veys. These projects began operations in October         these projects began operations in April 2010 and         began operations in April 2017. The four Cycle 2
2011 and data acquisition is now completed accord-       are now all completed but for the VHS subareas            VISTA surveys that explore the time domain are
ing to their survey management plans. The total          close to the South Galactic Pole. The total number        ­indicated by an asterisk in the table. The total
number of completed hours is reported to the 30          of completed hours is reported to the 30 September         ­number of completed hours by 30 September 2019
September 2019 date.                                     2019 date.                                                  is shown in the last column.

12             The Messenger 178 – Quarter 4 | 2019
Public Spectroscopic Survey ID                                                                  Science topic                     Number of                Spectral                Total time
and homepage                                                                                                                      targets/spectra          resolution              (nights)
Gaia–ESO                                                                                        Milky Way, stellar                200 000                  20 000                  282.5                                                                         populations
(Randich et al., 2013)
PESSTO — Public ESO Spectroscopic Survey of Transient Objects                                   Transient,                        150                      ~ 2500                  384.0                                                                          SN progenitors
(Smartt et al., 2013)
VANDELS                                                                                         Physics of galaxies in the        2700                     ~ 1500                  142.7                                                                          early universe CANDELS,
(McLure et al., 2017)                                                                           UDS & CDFS fields
LEGA-C — Large Early Galaxy Astrophysics Census                                                 Dynamics of galaxies              3100                     ~ 1500                  99.8                                                        at z = 0.6–1.0
(van der Wel et al., 2016)

Table 4. Public Spectroscopic Surveys. PESSTO                                         700
and Gaia–ESO began operations in 2012 and were
                                                   Number of publications/citations

completed in 2017. The surveys using the VIsible
                                                                                      600                                                                             Refereed publications
Multi Object Spectrograph (VIMOS), called                                                                                                                             Citations (since 2010)
­VANDELS and LEGA-C, began operations in 2015
                                                                                      500                                                                             Total ref. publications: 848
 and were completed in March 2018, before the
                                                                                                                                                                      172 (20.3%) from archive
 decommissioning of the VIMOS spectrograph.
                                                                                      400                                                                             84 (10%) are archive + PI

independent archives (for example, the                                                300
VISTA science archive, Vizier) or the
­Public Survey webpages have also been                                                200
 made available to those in the community
 interested in accessing data products for                                            100
 their independent scientific explorations.











The ESO library routinely monitors













­refereed publications, based on data

                                                                                       tr a







 acquired from ESO approved observing                                                                                          ESO Public Survey
 programmes. This includes papers
 ­published by PIs/co-investigators (CoIs)            Published data releases                                                               Figure 3. Histogram of the cumulative number of ref-
                                                                                                                                            ereed publications and citations (divided by 10) for
  as well as archive papers. Archive papers
                                                                                                                                            each ESO Public Survey.
  come in two flavours: archive only and              Because of the extensive amount of time
  archive plus PI publications. In archive            allocated using ESO facilities, the science
  only papers, none of the authors of these           policies for ESO Public Surveys entail
  refereed publications are listed as PIs             the submission and publication of the                                                  specific infrastructures. Five out of six
  or CoIs of the approved Public Survey               ­science data products from these pro-                                                 Cycle 1 VISTA surveys and two out of
  proposals. In the case of archive plus PI            jects into the ESO Archive. The publica-                                              three VST surveys received extensive
  publications, science data products from             tion process for science data products                                                support from the dedicated data centres,
  the ESO Public Surveys are used                      extends well after the completion of the                                              CASU 3 and WFAU 4. The deep UltraVISTA
  together with data owned by a PI or CoI              data acquisition. This additional time is                                             and Continuing UltraVISTA surveys relied
  of an ESO programme to achieve their                 used by the Public Survey teams to exe-                                               on dedicated support from CASU,
  scientific published results. In the case of         cute global calibrations of the entire data                                          ­TERAPIX 5 and CALET 6 centre at the IAP
  ESO Public Surveys, the total number of              volume and to carry out the relevant                                                  in Paris, while the KIDS survey is sup-
  refereed publications by teams and                   measurements required to achieve their                                                ported by Astro-WISE 7. The Cycle 2 VISTA
  archive users was 848 by 30 September                scientific goals. The ultimate publication                                            surveys have adopted different strategies
  2019. Of these refereed publications, 172            of the results of these steps is contained                                            compared to the first generation, with
  (20.3%) are archive only and 84 (10%) are            in the final catalogue release. All twenty                                            a larger number receiving tailored sup-
  archive plus PI since 2010 (from ESO tel-            ESO Public Surveys are currently involved                                             port to their data processing from their
  bib 2). The total number of citations from           in the publication of their science data                                              respective science institutes.
  ESO Public Survey refereed publications              products via the ESO Archive.
  is 26 266. In Figure 3 we provide the his-                                                                                                For the Public Spectroscopic Surveys,
  togram of the cumulative number of refer-           The Public Survey teams adopted a                                                     Gaia-ESO, PESSTO, LEGA-C and
  eed publications and citations per survey           range of strategies to deal with the data                                             ­VANDELS, the teams built their data
  project.                                            volumes from their respective surveys.                                                 reduction infrastructure based on previous
                                                      Some rely on the support of data centres                                               experience they had acquired through
                                                      while others have developed their own                                                  managing large programmes at ESO and

                                                                                                                                              The Messenger 178 – Quarter 4 | 2019                   13
Telescopes and Instrumentation                      Arnaboldi M. et al., Report on Status of ESO Public Surveys and Current Activities

scientific networks (for example,                   of the Small Magellanic Cloud. All data     of files downloaded by the community for
PESSTO–WISeREP).                                    releases were promptly advertised via the   each ESO Public Survey. The lower chart
                                                    Archive/Phase 3 web pages, followed up      shows the numbers of catalogues, the
All survey teams have successfully                  with specific announcements on the ESO      numbers of distinct users and the num-
­published several data releases for some           science page, the Science News­letter 9     bers of queries carried out using the ESO
 of their science data through the Phase 3          and the ESO archive community forum10.      catalogue query interface 12 to access
 process (Arnaboldi et al., 2014); an over-                                                     ESO Public Survey catalogues. On aver-
 view of these releases is available via            The most recent data releases join a        age, users of the ESO catalogue query
 this webpage8. Since January 2019, the             large number of data collections from the   interface carry out at least 21 independ-
 total volume of science data products              ESO Public Surveys that can be browsed      ent queries to access catalogue records.
 released from the ESO Public Surveys               using the Archive Science Portal11. The
 amounts to 27.4 Tb, including ancillary            science data products from the ESO          An enhanced archive capability allowing
 files. The data releases published this            Public Surveys amount to a total volume     programmatic access13 results in anony-
 year include: the fourth data release of           of 68.6 Tb (nearly 8.5 × 105 files) which   mous exploration and retrieval of cata-
 KIDS (> 1000 square degrees) and                   are currently accessible via the ESO        logue records (and other products) via
 ­UltraVISTA (deep stacked images of the            Archive. The science data products that     Virtual Observatory tools, for example,
  COSMOS field from observations                    can be actively queried and downloaded      Aladin and Topcat. This new service
  acquired between December 2009 and                amount to nearly 320 000 catalogue files,   allows users to repeat queries in an auto-
  June 2016); the proper motion of selected         half a million astrometrically and photo-   mated fashion, in order to perform more
  stars in the Milky Way disc and bulge             metrically calibrated images, and 56 000    complex queries by combining data from
  from the VVV near-infrared Astrometric            1D extracted spectra. In Figure 4 we        different surveys or other content of the
  Catalogue (VIRAC); accurate PSF-fitting           show a collection of on-sky footprints of   ESO Science Archive, thereby enhancing
  photometry of the 300 square degrees              the data releases published during the      the scientific use of the catalogue content
  around the Galactic centre; and the fifth         last year by the ESO Public Survey teams.   of the ESO Archive. One interesting
  data release of VMC with full coverage                                                        ­statistic is the number of distinct users —
                                                                                                 1583 users from 77 different countries —
                                                    Data download statistics                     who have downloaded ESO Public
Figure 4. Montage of the footprints of the data
releases from the ESO Public Surveys published by
                                                                                                 Survey science products published via
2019, as shown on the ESO Archive Science Portal    In Figure 5 we show the cumulative curves    the ESO Archive. To place this in context,
interactive interface.                              of the data volume (Gb) and the number       the fraction of distinct users who access

  VPHAS+ DR4                                        VEILS DR1                                   V-ATLAS DR4

  PSF Phot. MW DR1                                  VMC DR5                                     KIDS DR4

  VINROUGE DR1                           VISIONS DR1                      G-CAV DR1                        UltraVISTA DR4

14           The Messenger 178 – Quarter 4 | 2019
You can also read
Next part ... Cancel