The Messenger No. 178 - Quarter 4 | 2019 - European Southern Observatory
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
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
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:
information@eso.org 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 ww.eso.org/messenger/ 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/spaceengine.org
2 The Messenger 178 – Quarter 4 | 2019
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).
AdOptica/ESO
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
1
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
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
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
1
ESO previous Messenger article (Kasper et al., the cold stop of VISIR to an accuracy of
2
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
3
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
4
University of Liège, Liège, Belgium tions and new technologies that were alignment of the calibration unit, which
5
Observatoire de Paris-Meudon, France implemented and how they work, con- uses an elliptical mirror with an aberration-
6
Subaru Telescope, Tokyo, Japan cluding with the execution of the NEAR free field of view of around 0.1 mm in
7
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
8
Caltech, Pasadena, USA vation of the a Centauri system, which mounts on-site.
9
Uppsala University, Sweden took place in early June 2019.
10
L aboratoire d’Astrophysique Marseille, Following the completion of the assembly
France Three years were needed to develop the integration and verification (AIV) activities,
11
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-
12
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-
13
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
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
cell.
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
0.10 9 120
Estimated pointing error (λ/D)
x
y 8
100
0.05 7
Cumulative time (h)
6 80
Time/night (h)
0.00 5
60
4
– 0.05 3 40
2
– 0.10 20
1
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)
0.20
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
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
Contrast
10 –3 10 –5 a function of radial distance to the
centre.
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
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
Links
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-
B
submitted for NEAR observing time. Two ation detectors would allow the VLT to throughinitiatives.org
2
periods of Science Demonstration were probe the rocky planet regime in the hab- SPARTA: https://www.eso.org/sci/facilities/
develop/ao/tecno/sparta.html
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 (ppathak@eso.org) or Markus Kasper
posed programmes. other southern hemisphere 8-metre-class (mkasper@eso.org
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
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
1
ESO and Infrared Survey Telescope for Astron- under the leadership of the PI. The agree-
2
Terma GmbH, Darmstadt, Germany omy (VISTA) required them to span a time ment states that a final release which
3
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 | 2019user experience once the data are pub- 120%
lished through the Archive.
Percentage of completion (OB hours)
100%
ESO Public Survey status
80%
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%
ATLAS
KIDS
ESO Public Surveys were launched in VPHAS+
0%
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
8
18 2
1 4
1 8
19 2
4
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
-0
1
scope (VST; Capaccioli & Schipani, 2011),
20
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%
VIKING
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
6
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
-0
bulge (VVVX) via astrometry — to test
0
20
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 11Telescopes 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
http://kids.strw.leidenuniv.nl/ 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
http://astro.dur.ac.uk/Cosmology/vstatlas/ 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
http://www.vphas.eu (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
http://home.strw.leidenuniv.nl/~ultravista/ 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
http://www.ast.cam.ac.uk/~rgm/vhs/ 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
https://www.eso.org/sci/observing/PublicSurveys/ 24.0 23.5
sciencePublicSurveys.html
(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
http://vvvsurvey.org/ 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
http://www.astro-wise.org/ 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
http://star.herts.ac.uk/~mcioni/vmc/ 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
http://www.star.le.ac.uk/nrt3/VINROUGE/ 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
http://home.strw.leidenuniv.nl/~ultravista/
VVVX* — Extending VVV to higher Galactic latitudes Milky Way 1700 J H Ks Ks = 17.5 1631
http://vvvsurvey.org/
VEILS* — VISTA Extragalactic Infrared Legacy Survey Galaxy evolution, 9 J Ks J < 23.5 847
http://www.ast.cam.ac.uk/~mbanerji/VEILS/veils_index.html AGN, SN Ks < 22.5
G-CAV — Galaxy Clusters At VIRCAM Galaxy clusters 30 Y J Ks 24.5 24 23 440
http://www.oats.inaf.it/index.php/en/2014-09-12-12-59-22/
tematiche-di-ricerca/macroarea-1-en/670-galaxy_cluster.html
VISIONS* — VISTA star formation atlas Star formation 550 J H Ks 21.5 20.5 19.5 449
https://visions.univie.ac.at atlas
SHARKS — Southern Herschel-Atlas Regions Near-infrared 300 Ks 22.7 929
Ks-band survey counterparts for
https://www.iac.es/sharks/ 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 | 2019Public 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
http://www.gaia-eso.eu/ populations
(Randich et al., 2013)
PESSTO — Public ESO Spectroscopic Survey of Transient Objects Transient, 150 ~ 2500 384.0
http://www.pessto.org/ SN progenitors
(Smartt et al., 2013)
VANDELS Physics of galaxies in the 2700 ~ 1500 142.7
http://vandels.inaf.it early universe CANDELS,
(McLure et al., 2017) UDS & CDFS fields
LEGA-C — Large Early Galaxy Astrophysics Census Dynamics of galaxies 3100 ~ 1500 99.8
http://www.mpia.de/home/legac/index.html 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.
0
O
G
e
+
S
C
X
AS
A
S
EO
TO
LS
AC
ES
ug
The ESO library routinely monitors
ST
AS
V/
D
VH
VM
N
DE
SS
L
KI
G
D
KI
VV
Ro
a–
VI
AT
H
VI
LE
VI
refereed publications, based on data
N
PE
tr a
VP
ai
N
VA
T
G
VI
Ul
VS
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 13Telescopes 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 | 2019You can also read
