The Messenger 185
The Messenger 185
The Messenger 185
ESO Headquarters
Karl-Schwarzschild-Straße 2
85748 Garching bei München, Germany
Phone +498932006-0
[email protected]
The Messenger
Editor: Mariya Lyubenova
Editorial assistant: Isolde Kreutle
Copy-editing, Proofreading:
Peter Grimley
Graphics: Lorenzo Benassi
Layout, Typesetting, Design, Production:
Jutta Boxheimer
www.eso.org/messenger/
Front cover: This bright cluster of stars is 47 Tucanae (NGC 104), shown here in
Printed by omb2 Print GmbH, an image taken by ESO’s VISTA (Visible and Infrared Survey Telescope for
Lindberghstraße 17, 80939 Munich, Astronomy) from the Paranal Observatory in Chile. This image was taken as part
of the VISTA Magellanic Cloud survey, a project that is scanning the region of
Germany the Magellanic Clouds, two small galaxies that are very close to our Milky Way.
Credit: ESO/M.-R. Cioni/VISTA Magellanic Cloud survey. Acknowledgement:
Unless otherwise indicated, all images in Cambridge Astronomical Survey Unit
The Messenger are courtesy of ESO,
Back cover: This poster shows 42 of the largest objects in the Solar System’s
except authored contributions which are asteroid belt, located between Mars and Jupiter (orbits not to scale). The
courtesy of the respective authors. images in the outermost circle of this infographic have been captured with the
Spectro-Polarimetric High-contrast Exoplanet REsearch (SPHERE) instrument
© ESO 2021 on ESO’s Very Large Telescope. The asteroid sample features 39 objects larger
than 100 kilometres in diameter, including 20 larger than 200 kilometres. The
ISSN 0722-6691 poster highlights a few of the objects, including Ceres (the largest asteroid in
the belt), Urania (the smallest one imaged), Kalliope (the densest imaged) and
Lutetia, which was visited by the European Space Agency’s Rosetta mission.
Elise Vernet 1 Background: how the ELT works and characterise, with unprecedented
Michele Cirasuolo 1 sensitivity, extrasolar planets in the habit-
Jean-François Pirard 1 The optical design of ESO’s Extremely able zone around our closest star,
Marc Cayrel 1 Large Telescope (ELT) is based on a Proxima b, or to resolve giant molecular
Roberto Tamai 1 novel five-mirror scheme capable of col- clouds (the building blocks of star forma-
Pablo Zuluaga Ramírez 1 lecting and focusing the light from astro- tion) down to ~ 50 parsecs in distant
Constanza Araujo Hauck 1 nomical sources and feeding state-of- galaxies at z ~ 2 (and even smaller struc-
Bertrand Koehler 1 the art instruments for the purposes of tures for sources that are gravitationally
Fabio Bianca Marchet 1 imaging and spectroscopy. The light is lensed by foreground clusters).
Juan-Carlos Gonzalez 1 collected by the giant primary mirror,
Mauro Tuti 1 39 metres in diameter, relayed via the M2
and the ELT team and M3 mirrors (each of which is around The quintenary mirror (M5)
4 metres in diameter) to the M4 and
M5 mirrors that are the core of the tele- M5 is the field stabilisation unit of the tel-
1
ESO scope’s adaptive optics; the light then escope. The term “field stabilisation”
reaches the instruments on one of the means that the mirror is moving in a rigid
two Nasmyth platforms. This design pro- way (tip-tilting) to steer the image and
The Extremely Large Telescope (ELT) is vides an unvignetted field of view (FoV) of correct for vibrations of the telescope
at the core of ESO’s vision to deliver the 10 arcminutes in diameter on the sky — structure induced by its motion and by
largest optical and infrared telescope an area of 80 square arcminutes (1/9 of the wind, as well as some of the atmos-
in the world. Following on from our pre- the full Moon’s area) — and thanks to the pheric turbulence. This is achieved by the
vious Messenger articles we continue combination of M4 and M5 it is capable M5 cell which is composed of three piezo
with the description of the optical ele- of correcting for both atmospheric turbu- actuators. It supports and moves the
ments of the ELT. In this article we lence and the vibration of the telescope M5 mirror up to 10 times per second.
focus on the quintenary mirror (M5), the structure itself induced by motion and M5 is a flat, elliptical mirror with a diame-
field stabilisation unit. In combination wind. This adaptive capability is crucial in ter of 2.2 metres on the minor axis and
with the M4 mirror, M5 is vital to deliver- allowing the ELT to reach its diffraction 2.7 metres on the major axis (see Figure 1).
ing the sharp diffraction limited images limit, which is ~ 8 milliarcseconds in the The role of the M5 unit (mirror + cell) is to
needed for science by correcting for the J band (at l ~ 1.2 μm) and ~ 14 milli- reduce the image movement down to a
vibrations of the telescope, wind shak- arcseconds in the K band, thereby pro- level where the M4 mirror can take over.
ing and the atmosphere. We describe viding images 15 times sharper than The M5 mirror assembly needs to be very
the main characteristics of the M5, as those from the NASA/ESA Hubble Space stiff and at the same time very light (less
well as the challenges and complexity Telescope and with much greater sensi- than 500 kilogrammes in total) to allow its
of this unique field stabilisation unit and tivity. Translated into astrophysical terms cell to move it fast enough whilst remain-
its design and manufacturing status. this means opening up new discovery ing flat when moving.
spaces, from exoplanets closer to their
stars, to black holes, to the building Initial studies have demonstrated that to
blocks of galaxies both in the local Uni- achieve this highly demanding level of
Figure 1. The ELT M5 mirror is a 2.7 × 2.2-metre
ellipsoidal mirror made of silicon carbide, supported
verse and billions of light-years away. performance only specialist materials
by the M5 cell (ESO renderings of SENER Aeroespa- For example, it will be possible to detect can be used for the mirror, either silicon
cial and Safran Reosc design). carbide (SiC) or ultra-low-expansion
CEDRAT TECHNOLOGIES
piezo actuator.
it extremely lightweight.
Raw material
Assembling
Brazing
CVD coating
Grinding
Sintering
> 2000 °C
Polishing
Figure 3. (Above) Full manufacturing process of the Figure 4. (Below) Early July 2021: three mirror
M5 mirror. segments with CVD cladding at Boostec.
Mersen Boostec
Before brazing them together, the seg- perturbations at the scale of a few tens net and the front-end electronics to the
ments must be aligned in such way that of milliarcseconds on the sky. M5 cell has also improved the design of
the positions of the axial support inter- the M5 cell.
faces are within the required tolerance SENER Aeroespacial is the company
and that the brazing joints meet the developing the M5 cell re-using experi- Another demanding requirement has
specified thickness. Once the six petals ence acquired with the development been the new and higher minimum eigen-
are well aligned, the unit is brazed. The of a functional M5 field stabilisation unit frequencies of the M5 cell, and this has
CVD layer requirements reduce the toler- demonstrator (Barriga et al., 2014), in had direct implications for the stiffness of
ance one can accept in the segment collaboration with CEDRAT TECHNOLO- the final model.
positioning as any misalignment of the GIES for the piezo actuators.
optical surface will reduce the final The technological challenge of both
CVD layer available for polishing. After The design of the M5 cell has been driven improving the stiffness and the stroke of
brazing, Mersen Boostec will perform by stricter requirements than those that the M5 tip-tilt actuators and developing
a final grinding of the flat optical surface applied to the demonstrator and this has a new compact actuator for the M5 align-
to make the blank ready for delivery to forced the SENER Aeroespacial team to ment stage has been met with good
Safran Reosc. apply state-of-the art methods and pro- results.
cedures whilst analysing different alterna-
The mirror blank will be delivered to tives and iterating them to obtain a robust SENER Aeroespacial, together with
Safran Reosc in the autumn of 2022 and result. Moreover, the addition of a new CEDRAT TECHNOLOGIES (for the APA),
once the axial and lateral supports are functionality — providing active alignment is presently finishing the qualification
mounted, polishing will last for two and a at low rate in piston and tip-tilt — has campaign of the different mechanisms.
half years. The mirror will be ready to be necessitated the division of the M5 cell The Final Design Review is expected to
shipped to the observatory by the end of into two stages, which has added a new take place in the autumn of 2021, well in
2025. twist to the design (see Figure 5). line with the schedule.
SENER Aeroespacial
M5 tip-tilt frame
M5 tip-tilt stage
M5 tip-tilt actuators
M5 cell
M5 base frame
M5 lateral constraint
M5 alignment stage
M5 alignment actuators
0.3
MODEL HST ELT JWST MAVIS
0.2
0.1
arcseconds
0.0
– 0.1
– 0.2
– 0.3
– 0.2 0.0 0.2 – 0.2 0.0 0.2 – 0.2 0.0 0.2 – 0.2 0.0 0.2 – 0.2 0.0 0.2
arcseconds arcseconds arcseconds arcseconds arcseconds
Resolution (λ/λΔ)
and post-focal instrumentation helped in 12 500 RED 10.2
NaD LGS
converging on a healthy design, in partic HR-
ular the optical design, which maximises 10 000 12.7
throughput and minimises optical aberra- 7500 17.0
D
tions and field distortion, whilst allowing UE L R- R E
easy alignment and modular integration 5000 L R- B L 25.5
(Viotto et al., 2020; Ellis et al., 2020).
2500 51.0
MUSE
The current MAVIS design can be sum- 0
marised as follows: 4000 5000 6000 7000 8000 9000 10 000
– A largely transmissive design across Wavelength (Å)
the board (AO module, imager, spec-
trograph), using only on-axis optics. Figure 3. (Above) The four MAVIS IFS spectral reso- Figure 4. (Below) The MAVIS focal plane, with eight
lution modes and corresponding dynamical scales, laser guide stars (orange) and three natural guide
The intention is to minimise field distor-
chosen to address key science areas. These include: stars (red). The imager covers/defines the “science
tions, as well as to make the optical the study of chemical abundances and radial veloci- field” (green). The integral field spectrograph (IFS)
alignment more tolerant. ties in crowded fields (HR-BLUE); exploring the evo- field is in purple (coarse mode) and blue (fine mode).
– A gravity-invariant design, in which the lution of ISM turbulence in galaxy discs and hunting The exact shape and location of the IFS field of view
intermediate-mass black holes (HR-RED); probing are still being optimised. Having the IFS off-centred
science and NGS WFSs are commonly
the extremes of mass and metallicity of young stars with respect to the technical field increases the
de-rotated optically and the LGS WFSs and untangling the complexity of ionised gas in gal- c apture range for natural guide stars, and thus the
use mechanical de-rotation to track the axies (LR-BLUE); and resolving Lyman-alpha emit- sky coverage.
LGS constellation. ters at z > 6.5, and studying stellar dynamics in z < 1
g alaxies (LR-RED). MUSE is shown in grey for com-
– Three deformable mirrors: the AOF sec-
parison.
ondary (1170 actuators) and two post-
focal deformable mirrors, optically con-
jugated to 6 and 13.5 kilometres, for a
grand total of 5420 actuators. )
ale
– Eight laser guide stars created from sc NGS1
o
tt
splitting the four existing AOF lasers,
o
(n
Arjen van der Wel 1 completed with the now-retired VIsible erties of specific galaxies, to utilise the
Rachel Bezanson 2 Multi-Object Spectrograph (VIMOS) full dataset to shed new light on old prob-
Francesco D’Eugenio 1 instrument at ESO’s Very Large Tele- lems, to support follow-up surveys at
Caroline Straatman 1 scope (VLT). Its aim is to characterise different wavelengths, and to test galaxy
Marijn Franx 3 with high precision and for a very large formation models.
Josha van Houdt 4 sample the stellar population and
Michael V. Maseda 3 kinematic properties of galaxies at red-
Anna Gallazzi 5 shifts 0.6 < z < 1, providing a first The data
Po-Feng Wu 6 assessment of the star formation histo-
Camilla Pacifici 7 ries, the absolute mass scale, and the For full technical details we refer to the
Ivana Barisic 4 stellar kinematic structure of galaxies at recently published Data Release 3 paper
Gabriel B. Brammer 8 large lookback times (7 gigayears ago). (van der Wel et al., 2021). Very briefly,
Juan Carlos Munoz-Mateos 9 This article coincides with the third and LEGA-C consists of several thousand
Sarah Vervalcke 1 final data release but mostly focuses galaxies at redshifts 0.6 < z < 1 selected
Stefano Zibetti 5 on the large variety of scientific results only by their Ks-band magnitude (van der
David Sobral 10 achieved so far. Wel et al., 2016), a proxy for stellar mass.
Anna de Graaff 3 As a result, the survey contains galaxies
Joao Calhau 11, 12 of all types and colours above an approx-
Yasha Kaushal 2 LEGA-C imate stellar mass limit of 3 × 1010 M⊙.
Adam Muzzin 13 Of course, the majority of galaxies in the
Eric F. Bell 14 Before LEGA-C, most of our understand- Universe have lower masses than this,
Pieter G. van Dokkum 15 ing of the stellar bodies of galaxies at but this mass limit does account for most
large lookback times derived from photo- of the mass (and star formation) budget
metric measurements. Tremendous pro- and therefore can be considered as rep-
1
Astronomical Observatory, University of gress had been achieved in that way, by resentative. The simplicity of the survey
Ghent, Belgium quantifying the evolution of the number of design allows for a precise determination
2
Department of Physics and Astronomy, stars as a function of cosmic time, their of sample selection effects and com-
University of Pittsburgh, USA distribution over galaxies with different pleteness, enabling accurate measure-
3
Leiden Observatory, Leiden University, masses, their structural properties and ments of ensemble properties.
the Netherlands their crude spectral properties in terms of
4
Max Planck Institute for Astronomy, colours. But these measurements are The observations were carried out from
Heidelberg, Germany merely indirect proxies for the quantities December 2014 to March 2018, with
5
INAF – Astronomical Observatory of that are the most relevant for constraining ~ 20-hour integrations for each of the
Arcetri, Florence, Italy galaxy formation models: galaxy mass, 32 slit masks. At a spectral resolution of
6
National Astronomical Observatory of angular momentum and star formation/ R ~ 3500 the typical spectrum has a
Japan, Tokyo, Japan assembly history. Moreover, sample sizes signal-to-noise ratio of ~ 15–20 per Ång-
7
Space Telescope Science Institute, are now very large, so we have run up strom. The data processing is described
Baltimore, Maryland, USA against the limit of systematic uncertain- in detail by Straatman et al. (2018) and
8
Cosmic Dawn Center (DAWN) ties instead of sample variance. Spectro- van der Wel et al. (2021), and the collated
Niels Bohr Institute, University of scopic observations of the stellar contin- spectra are shown in Figure 1. Broadly
Copenhagen, Denmark uum provide a more direct way of speaking these spectra serve two pur-
9
ESO estimating those quantities with less bias, poses: measuring stellar kinematic signa-
10
Department of Physics, Lancaster but it took more than 1000 hours of VLT tures (Bezanson et al., 2018b) and stellar
University, UK observing time to collect such expensive population characteristics (for example,
11
Astrophysics Institute of the Canaries, data for galaxies at large lookback times. Chauke et al., 2018; Wu et al., 2018a).
La Laguna, Tenerife, Spain The detailed characteristics of the spec-
12
Department of Astrophysics, U niversity This article briefly summarises the prop- trum and the basic modelling approach
of La Laguna, Spain erties and quality of the data that have are illustrated for just one example in Fig-
13
Department of Physics and Astronomy, now been published in full, along with ure 2. The LEGA-C survey represents a
York University, Toronto, Canada value-added catalogues, and the broad 30-fold increase in sample size of galax-
14
Department of Astronomy, University of range of scientific results published so ies with measured stellar kinematics and
Michigan, USA far. In a narrow sense, the LEGA-C sur- stellar population properties compared to
15
Astronomy Department, Yale University, vey has now been completed: the data all previous work combined and, equally
New Haven, USA have been collected, processed and importantly, samples the full galaxy pop-
published. But in a broad sense, the final ulation rather than a specifically selected
data release 1 marks only the beginning. sub-sample of very massive galaxies with
The Large Early Galaxy Astrophysics Our hope is that, for many years to come, mostly early-type morphologies.
Census (LEGA-C) survey is the final researchers will use these data for a wide
Public Spectroscopic Survey to be variety of purposes: to find detailed prop-
1.0
reflects the stellar population information content.
0.8
The figure is adapted from van der Wel et al. (2021).
0.6
0.4
0.2
K H Hδ CN Ca G Hγ Fe Ca Fe C Hβ Fe Mg
0.0
2500
2000
Number of spectra
1500
1000
300
Galaxy age, as a concept, is somewhat
Fλ (10 –18 erg s –1 cm –2 Å –1 )
–3.0 1.50 8
1.25
–3.5
6
Effective radius log10 (Re [kpc])
dal system. But the integrated velocity 300 Figure 4. Spatially resolved stellar
k inematic measurements with an
dispersion does not tell us about the
axisymmetric model. The radial veloc-
dynamical structure of a galaxy: a high s
–2
–2
50 100 150 200 250 300
Synthesis
50 100 150 200 250 300
Stellar velocity dispersion σ (km s –1)
*
Galaxy structure and star formation his-
tory are closely linked; the true power between morphology, velocity dispersion Figure 5. Hd absorption strength — indicative of
stellar population age — versus stellar velocity dis-
of LEGA-C is therefore the availability of and stellar population age (Figure 5).
persion. Hubble Space Telescope images reflect
both stellar kinematics and stellar pop We see that young galaxies generally the variation in morphology throughout this parame-
ulation information. We have only just have late-type morphologies with disc- ter space. The inset panel shows the Sérsic index
begun to explore this direction, and for like structures, whereas old galaxies (Sérsic, 1963), a structural property that distin-
guishes between disc- and bulge-dominated galax-
the general galaxy population we have so have early-type, bulge-dominated mor-
ies (n = 1 and n = 4, respectively). The figure is
far only shown the global correlation phologies. At a fixed (integrated) velocity adapted from van der Wel et al. (2021).
2.5
1 2
INAF – Astronomical Observatory of
A(Li)
3.0
Arcetri, Florence, Italy
2
Nicolaus Copernicus Astronomical
Center, Polish Academy of Sciences, 3.5 1
Warsaw, Poland
3
UTINAM Institute, CNRS UMR 6213, 4.0
University of Bourgogne Franche-
Comté and Besançon Observatory, 0
France 4.5
4
LAB CNRS, University of Bordeaux,
France 5.0
5
ESO –1
6 7000 6500 6000 5500 5000 4500 4000 3500
INAF – Astrophysics and Space Teff (K)
Science Observatory of Bologna, Italy
7
Institute of Astronomy, University of
Cambridge, UK Introduction the most complex origin and evolution
(see Romano et al., 2021). Li is one of the
Since the latest internal data release of few elements, along with H and He, pro-
After more than ten years and six data the Gaia–ESO spectroscopic survey1 in duced in the first instants of the Universe,
releases, the Gaia–ESO spectroscopic December 2020, many topics related by primordial nucleosynthesis after the
survey has come to an end. Gaia–ESO to lithium (Li) abundances have been Big Bang. However, the Li abundance we
provides an extremely rich database addressed by the Gaia–ESO collabora- measure today is only in part the primor-
of stellar parameters, radial velocities, tion. Recent publications include the dial one, because many destructive and
and chemical abundances of more than serendipitous discovery of a rare Li-rich constructive processes have occurred
100 000 stars, amongst which the abun- giant in the globular cluster NGC 1206 since (see, for example, Matteucci,
dance of lithium can be considered one (Sanna et al., 2020), the study of the D’Antona & Timmes, 1995; Romano et al.,
of the main products. Lithium is per- Galactic evolution of Li (Randich et al., 2021; Randich & Magrini, 2021).
haps the most enigmatic of the ele- 2020; Romano et al., 2021), the use of
ments, with several open issues regard- Li as an age tracer (Gutierrez Albarran et The production channels of Li include:
ing its nucleosynthesis and its evolution al., 2020; Binks et al., 2021), the exploi stellar nucleosynthesis through the inter-
in stars and in the Galaxy. Gaia–ESO tation of Li to put constraints on pre-main mediate production of 7Be, which is
observations are allowing such issues sequence evolutionary models and the carried by convection to cooler stellar lay-
to be addressed, by providing lithium effect of magnetic activity (Franciosini et ers where it decays to 7Li — the so-called
abundances in stars from the pre-main al., submitted to Astronomy and Astro- Cameron-Fowler (CF) mechanism
sequence, through the main sequence, physics), and the study of mixing pro- (Cameron & Fowler, 1971); thermonuclear
up to the red giant branch and the helium- cesses in giant stars using Li as a tracer runaways during classical nova explo-
burning red clump phase, over a wide (Magrini et al., 2021a,b). We refer to sions (Arai et al., 2021); Li production
range of masses. In the present work, Randich & Magrini (2021) for a description triggered by the flux of neutrinos emerg-
we discuss the journey of lithium on the of the potential of the Gaia–ESO Survey ing from the collapsing cores of explod-
surface of evolved stars, using Gaia– data for Li studies. Figure 1 plots surface ing massive stars (Sieverding et al., 2018);
ESO data for both field and cluster gravity against effective temperature and spallation of interstellar medium
stars. We focus on the impact of extra (sometimes called a Kiel diagram) for all atoms by high-energy Galactic cosmic
mixing and possible lithium enrichment stars in the final data release of the Gaia– rays (Meneguzzi, Audouze & Reeves,
during the helium-burning phase. We ESO Survey for which Li was measured. 1971). However, as stated by Romano et
briefly comment on the implications that The figure highlights the data coverage in al. (2021), “it is a little disconcerting
these results may have for models of terms of evolutionary stages. that none of the proposed 7Li production
the chemical evolution of lithium in the channels have been firmly assessed yet”.
Galaxy. Of all the elements in the periodic table, To shed light on the production of Li,
Li, in the form of its main isotope 7Li, has all these mechanisms are considered by
Age ≤ 120 Myr 120 < Age ≤ 400 Myr 400 < Age ≤ 1200 Myr 1200 < Age ≤ 1500 Myr 1500 < Age ≤ 1800 Myr
3
Lithium abundance A(Li)
–1
1800 < Age ≤ 2400 Myr 2400 < Age ≤ 3000 Myr 3000 < Age ≤ 4000 Myr 4000 < Age ≤ 5000 Myr Age > 5000 Myr
3
Lithium abundance A(Li)
–1
7000 6000 5000 4000 3000 7000 6000 5000 4000 3000 7000 6000 5000 4000 3000 7000 6000 5000 4000 3000 7000 6000 5000 4000 3000
Teff (K) Teff (K) Teff (K) Teff (K) Teff (K)
RC stars are higher than expected What happens after the RGB, in the which this enrichment happens. In Fig-
from the models, while the same models He-burning phase? ure 4 we show member stars in a repre-
fairly well reproduce A(Li) in the previous sentative cluster, Trumpler 5, separating
RGB phases. The data from open clusters in Fig- them into the three evolutionary phases
ure 3 suggest that a further Li enrichment (RGB before and after the bump, and
occurs just before or during the RC phase RC). The distribution of Li abundances in
in a number of RC stars. The important the RC stars shows values that are either
thing is to evaluate the frequency at similar to or even higher than those of
the upper RGB stars. In practice, in Fig- effectively compete with Li synthesis in Lagarde, N. et al. 2012, A&A, 543, A108
Magrini, L. et al. 2021a, A&A, 651, A84
ure 4 we would expect RC stars to be thermonuclear nova outbursts, and this
Magrini, L. et al. 2021b, arXiv:2108.11677
located in the region where A(Li) reaches would push the quest for increasingly Martell, S. et al. 2020, arXiv:2006.02106
a plateau, with values between 0 and –1. refined stellar and Galactic chemical evo- Matteucci, F., D’Antona, F. & Timmes, F. X. 1995,
lution models. A&A, 303, 460
Meneguzzi, M., Audouze, J. & Reeves, H. 1971, A&A,
Considering only stars for which we pro-
15, 337
vide measurements of A(Li), we find that Monaco, L. et al. 2011, A&A, 529, A90
from 35 to 50% of RC stars have A(Li) Acknowledgements Mori, K. et al. 2021, MNRAS, 503, 2746
higher than stars in the previous phase, This work is based on data products from observa-
Randich, S. & Magrini, L. 2021, Frontiers in Astron-
omy and Space Sciences, 8, 6
i.e., at the end of the RGB phase. Since tions made with ESO Telescopes at the La Silla Randich, S. et al. 2020, A&A, 640, L1
the presence of several upper limits Paranal Observatory under programme ID 188.B- Romano, D. et al. 2021, A&A, 653, A72
might hide much lower A(Li), we conclude 3002, 193.B-0936, 197.B-1074. These data products Sanna, N. et al. 2020, A&A, 639, L2
have been processed by the Cambridge Astronomy
that Li enrichment happens in a large Survey Unit (CASU) at the Institute of Astronomy,
Schwab, J. 2020, ApJL, 901, L18
Sieverding, A. 2018, ApJ, 865, 143
percentage of RC stars, but that it might University of Cambridge, and by the FLAMES/UVES Silva Aguirre, V. et al. 2014, ApJL, 784, L16
not be ubiquitous. The comparison reduction team at INAF – Astrophysical Observa- Singh, R., Reddy, B. E. & Kumar, Y. B. 2019,
with models that include additional mix- tory of Arcetri. The data have been obtained from the MNRAS, 482, 3822
Gaia–ESO Survey Data Archive, prepared and
ing processes after the RGB phase, for hosted by the Wide Field Astronomy Unit at the Insti-
Singh, R. et al. 2021, ApJL, 913, L4
Smiljanic, R. et al. 2018, A&A, 617, A4
instance He-flash mixing (Schwab, 2020) tute for Astronomy, University of Edinburgh, which is Yan, H.-L. et al. 2021, Nature Astronomy, 5, 86
and neutrino momentum mixing (Mori et funded by the UK Science and Technology Facilities
al., 2021), is very promising. Such models Council.
have the capacity to qualitatively explain Links
the behaviour of the RC stars. Whilst References 1
The Gaia–ESO spectroscopic survey:
they differ in terms of the description of https://www.gaia-eso.eu/
the processes and their timescales, they Anthony-Twarog, B. J. et al. 2018, AJ, 155, 138
Arai, A. et al. 2021, ApJ, 916, 44
do agree on the requirement for a mixing Binks, A. S. et al. 2021, MNRAS, 505, 1280
process during the He flash that is Notes
Bressan, A. et al. 2012, MNRAS, 427, 127
needed to activate the production of Li. Brown, J. A. et al. 1989, ApJS, 71, 293 a
The first dredge-up is the first episode of mixing
Cameron, A. G. W. & Fowler, W. A. 1971, ApJ, 164,
occurring when a star enters the RGB, and
111
With the new data from Gaia–ESO, the Casey, A. R. et al. 2016, MNRAS, 461, 3336
corresponding in Figure 2 to the point in the
journey of Li on the surface of evolved classical models at which we have no further
Casey, A. R. et al. 2019, ApJ, 880, 125
decrease of A(Li).
stars is seen to be complex. Even in evo- Charbonnel, C. & Balachandran, S. C. 2000, A&A, b
Thermohaline instability arises when an unstable
lutionary phases in which we might have 359, 563
gradient in composition is stabilised by a gradient
Charbonnel, C. et al. 2020, A&A, 633, A34
expected only destructive processes, we Charbonnel, C. et al. 2021, A&A, 649, L10
in temperature — it also happens on Earth, in
find that Li can still be recreated. Since oceanic regions where evaporation produces a
Deepak & Reddy, B. E. 2019, MNRAS, 484, 2000
warm layer of saltier water above a layer of less
low-mass stars are the dominant popula- Deepak & Lambert, D. L. 2021, MNRAS, 507, 205
salty, cooler water. The two layers can exchange
tion in the Galaxy that rules Li production Gutierrez Albarran, M. L. et al. 2020, A&A, 643, A71
heat though “salt fingers”, which allow the mixing.
Kumar, Y. B., Reddy, B. E. & Lambert, D. L. 2011,
at late times, this kind of production ApJL, 730, L12
c
The RGB bump is the evolutionary phase in which
could have remarkable effects on our Kumar, Y. B. & Reddy, B. E. 2020, Journal of Astro-
the shell reaches the discontinuity left by the
envelope causing a momentary drop in the stellar
understanding of the whole evolutionary physics and Astronomy, 41, 49
luminosity.
history of Li. In particular, we need to Kumar, Y. B. et al. 2020, Nature Astronomy, 4, 1059
Iben, I. 1967, ApJ, 147, 624
ascertain if Li production on the RC could
10
Male 0
2018 2019 2020 2021 2022
Female
the programme 1. We show in Figure 3 Whilst the numbers are already quite
a screenshot of the front page of the impressive, it is when watching the semi-
Hypatia Colloquium YouTube channel. nars that the real quality and impact of
the series can be fully appreciated. With
The channel analytics clearly attest to the Hypatia Colloquium, ESO allowed
the success of the series. Whilst each the young generation of astronomers to
event was followed live by a total of 40 to paint a remarkable and unique picture
Figure 3. Screenshot of the YouTube
60 participants spread equally between of the excellence of the science being channel of the Hypatia Colloquium.
Zoom and YouTube, the videos record (at done by early career astronomers. The All the talks in the 2021 series are
the time of writing) a total of 4400 views. videos available online represent an available on the dedicated playlist.
This photograph shows some of the antennas com- to 16 kilometres and is formed of 66 individual anten-
prising the Atacama Large Millimeter/submillimeter nas. Some features visible in the sky include the
Array (ALMA) against the backdrop of a panoramic constellation Crux (The Southern Cross) just above
view of the Milky Way. There is far more to ALMA and to the right of the nearest antenna, and the
than shown here; the array can span distances of up Carina Nebula slightly further to the right.
Elizabeth George 1 ate an open-source simulation tool that (a Python package for X-ray detector
Benoît Serra 1 allows astronomical instrument builders modelling), CERN’s Allpix Squared (silicon
Thibaut Prod’homme 2 to simulate their detectors at the design detector Monte Carlo simulations for
Matej Arko 2 and engineering stage as well as helping particle physics), and electron avalanche
Frédéric Lemmel 2 astronomers develop calibration or analy- photodiode (e-APD) saphira modelling. In
Bradley Kelman 3 sis strategies to address detector effects these sessions we had several presenta-
that may impact their science. The DeMo tions from fields outside astronomy (par
workshop was envisioned both as a way ticle physics, for example) which provided
1
ESO to share Pyxel with the community a different perspective on detector char-
2
European Space Agency, ESTEC, through hands-on tutorial sessions, and acteristics and modelling.
Noordwijk, the Netherlands as a forum for the exchange of ideas
3
Open University, Milton Keynes, UK around detector modelling for astronomy The largest number of talks covered
via a programme of contributed talks. measurements, models, and/or simula-
Figure 1 is the workshop poster. tions of individual detector effects.
The Detector Modelling (DeMo) work- These tended to be very detailed models
shops aim to bring together a commu- of single detector effects that astrono-
nity of scientists and engineers who Scientific programme mers and engineers had worked hard to
are interested in modelling detector understand in order to enable a specific
effects and simulating detectors for The scientific programme was based on science case. These included C3TM
astronomy. The first such workshop was the contributions of the participants and (radiation damage in CCDs), CosmiX
held online over three afternoons from consisted of 22 talks in seven thematic (charged particles in detectors), and
14 to 16 June 2021. The three after- blocks. These were: various models covering interference,
noons were organised around blocks –o ptical and infrared instrument simu non-linearity, inter-pixel capacitance
of contributed talks from the community lators (two blocks); effects, persistence, and luminescence
covering a wide range of detector –d etector measurements and models effects in mercury cadmium telluride
topics such as detector effects like per- (three blocks); (MCT) hybridised arrays, as used in many
sistence and radiation damage, instru- –X -ray instrument simulators; ESO instruments. Some of the detector
ments covering a wide range of astro- –p articles and radiation modelling. models presented are already integrated
nomical wavelengths from X-ray to the into Pyxel, and others will be added by
optical and infrared, and detectors used One of the most popular themes of the the speakers in the coming months as
in other fields like particle physics. conference was instrument simulators. the flexible nature of Pyxel allows users to
In addition to the scientific programme, Speakers presented ScopeSIM (the sim- add their own favourite models to the
the workshop featured a tutorial series ulator for the Multi-AO Imaging Camera simulation framework. A highlight of these
on how to use the Pyxel detector simu- for Deep Observations [MICADO] at talks was the many cases where preci-
lation framework and how to contribute ESO’s Extremely Large Telescope), sion laboratory data were combined with
to Pyxel. PLATOsim, the Euclid suite of instrument physical knowledge of the detectors
simulators, SIXTE (a generic X-ray instru- to create a model of detector behaviour.
ment simulation) toolkit, xifusim (the
Introduction Athena X-IFU instrument simulator), and The final theme covered by the scientific
PhoSim for Vera C. Rubin Observatory. programme was the impact of detector
As astronomical observations move ever These complete instrument simulators effects on science instruments and the
further into the realm of precision astron- often include modules for generating use of detector simulations in instrument
omy, systematic effects in instruments, astronomical scenes and transmission design and systems engineering. Speak-
especially detectors, are beginning to through the instrument optics, a simu ers presented Pandeia (the James Webb
dominate instrument error budgets. Under- lation of detector effects, and in some Space Telescope exposure time cal
standing, modelling, and correcting for cases also an analysis pipeline for the culator), persistence correction in ESO’s
detector effects are now necessary to resulting data. The goal of these simula- instruments, mitigation of tearing in the
achieve science goals ranging from char- tion tools is complete end-to-end model- Vera C. Rubin Observatory CCDs, and
acterising exoplanet atmospheres, to ling of an instrument, or even a complete NASA’s Spectro-Photometer for the
measuring chemical abundances in high- simulation of a specific instrument sci- History of the Universe, Epoch of Reioni-
redshift galaxies, to performing precision ence case including the analysis of syn- zation and Ices Explorer (SPHEREx)
astrometry of stellar fields. thetic data. mission. These talks provided valuable
context for why detector modelling is so
Engineers working on detectors at ESO Following the full instrument simulators, important for precision astronomical
and the European Space Agency (ESA) detector-specific simulations were pre- instruments, as the impacts of the effects
have been collaborating for several years, sented in several sessions covering simu- are clearly seen in the (synthetic) data.
under the umbrella of the ESA–ESO col- lation work for the China Space Station
laboration, on a detector simulation tool Telescope (CSST), the European Syn- Whilst the models and simulations pre-
named Pyxel 1. The goal of Pyxel is to cre- chrotron Radiation Facility’s X-DECIMO sented at DeMo were all different, some
The Science Organising Committee was Overall the DeMo2021 workshop was a
a small group made up of the Pyxel success in terms of the number of partic-
Florentin Millour 1 Very Large Telescope Inerferometer To adapt to the new situation we had
Anthony Meilland 1 (VLTI) schools have nearly a 20-year to swiftly reorganise the school schedule,
Alexis Matter 1 history and have trained a significant with the expertise and support of
Guillaume Mella 2 fraction of today’s optical inter the technical group of the Jean-Marie
Laurent Bourgès 2 ferometrists who use high-angular- Mariotti Center (JMMC), and we accepted
Claudia Paladini 3 resolution techniques on a regular the challenge of hosting 100% online
Isabelle Tallon-Bosc 4 basis. Very early in the development of practice sessions with students scattered
Michel Tallon 4 the VLTI, training was identified by all over the world. All the school’s teach-
Ferreol Soulez 4 the community as a necessary tool, as ers accepted the added complication
David Buscher 5 the expertise in optical long-baseline of giving their lectures either prerecorded
Antoine Mérand 3 interferometry was limited to a few or live (with recorded videos 3). Finally,
Myriam Benisty 12 groups in France and Germany (in those to allow students from most countries to
Roy van Boekel 6 early years the UK was not an ESO attend the school at decent hours, we
Pierre Cruzalèbes 1 member state). The first VLTI school extended its duration to 2 weeks and lim-
Denis Defrère 7 took place in Les Houches, France, in ited the mandatory online presence to
Armando Domiciano de Souza 1 2002 and since then VLTI schools 4 hours a day.
Mercedes Filho 8 have been organised in several loca-
Paulo Garcia 8 tions (France, Germany, Hungary, The school took place online in a spe-
Sebastian F. Hönig 9 Poland, Portugal) roughly every two cially tailored Gather (gather.town) space
Roxanne Ligi 1 years, the previous one being held for direct interactions (chat, voice and
Camille Maccotta 1 in 2018 in L
isbon. The VLTI schools are video) and practice sessions, combined
John McKean 10, 11 funded and coordinated through the with Zoom sessions for the lectures, and
Fabien Patru 1 European Interferometry Initiative (Eii). the Nuclino live wiki-like environment for
Karine Perraut 12 the exchange of information and files
Jörg-Uwe Pott 6 (see Figure 1). On the technical side, the
Alain Spang 1 Very Large Telescope Interferometer most challenging part was the setup
Sophie Rousset 1 (VLTI) schools1 (Garcia, 2009) have been of 21 virtual machines for the students to
held every two years since 2002, the run all the practice sessions, the students
9th being in 2018 in Lisbon (Garcia et al., being typically in groups of three. It
1
University of Côte d’Azur, Côte d’Azur 2018). Here we report on the 10th VLTI was possible thanks to the strong involve-
Observatory, CNRS, Lagrange Labora- school 2, which was organised by the ment of the JMMC technical group
tory, France J.-L. Lagrange laboratory in Nice, and and the resources of the Grenoble Alpes
2
University of Grenoble Alpes, CNRS, had its focus on the latest VLTI instru- Research Scientific Computing and Data
IRD, INRAE, Météo-France, OSUG, ment: the Multi-AperTure mid-Infrared Infrastructure (GRICAD) service.
France SpectroScopic Experiment (MATISSE;
3
ESO Lopez et al., 2021). MATISSE is a four- There were 63 registered students (par-
4
University of Lyon, ENS de Lyon, beam combiner operating in the mid- ticipating in both the lectures and the
CNRS, Lyon Centre for Astrophysical infrared, from 2.8 to 13 microns, with an practice sessions) plus 59 free listeners
Research, France emphasis on interferometry for planet (following the lectures only), 45 of whom
5
Cavendish Laboratory, University of ology. The GRAVITY instrument and other were MSc or PhD students, including
Cambridge, UK instruments from the Center for High 35 female participants. The registered
6
Max Planck Institute for Astronomy, Angular Resolution Astronomy (CHARA) students originated from 23 different
Heidelberg, Germany observatory were also presented. It was countries (see Figure 2): nine from France,
7
Institute of Astronomy, KU Leuven, aimed at graduate and PhD students, six each from Chile, Poland and the
Belgium as well as postdocs, who wish to learn USA, five from Germany, four each from
8
CENTRA/SIM, Department of Physics the theory and practicalities of optical Italy and Switzerland, three from Belgium,
Engineering, University of Porto, and infrared interferometry. It was origi- two each from Egypt, Great Britain,
Portugal nally intended to hold the school in India, the Netherlands and Turkey, and
9
School of Physics & Astronomy, Univer- June 2020, in the technology park named one each from Brazil, China, Greece,
sity of Southampton, UK Sophia Antipolis, near Nice in France, a Ireland, Iran, Iraq, Malta, Peru, Spain
10
Kapteyn Astronomical Institute, Univer- beautiful location in a pine forest, just and Sweden. That balance changes a bit
sity of Groningen, the Netherlands 10 minutes from sandy beaches. How- when account is taken of the free
11
ASTRON, Institute for Radio Astron- ever, the COVID-19 pandemic situation listeners, as there was a massive partici-
omy, Dwingeloo, the Netherlands forced us to shift the date, first to Sep- pation from Chile and China at the lec-
12
University of Grenoble Alpes, CNRS, tember 2020 and finally to June 2021. tures. The interferometric expertise of the
IPAG, France Given the unfavourable evolution of the students was diverse, allowing students
global pandemic situation in early 2021, to learn from each other too. The number
we finally decided to switch to a 100% of connections to the lectures was around
online format with lectures and tutorials. 60 on average with peaks up to 86.
Poland | 6 Chile | 6
USA | 6 France | 9
Germany | 5
Sweden | 1
Spain | 1
Peru | 1
Malta | 1
Italy | 4 Iraq | 1
Iran | 1
Ireland | 1
Greece | 1
Switzerland | 4 China | 1
Brazil | 1
Belgium | 3 Turkey | 2
Egypt | 2 The Netherlands | 2
United Kingdom | 2 India | 2
All the lectures and scientific presenta- opening hours. The lecturers and teach- Acknowledgements
tions are now available to anyone who ers were very impressed by the profes-
The 10th VLTI school received funding from the
wants to follow them at their pace 2. sionalism of the students, especially dur- European Union’s Horizon 2020 research and inno-
ing the proposal preparation session. vation programme under grant agreement
The school was designed to be very No 730890 (OPTICON). The European Interferometry
Initiative (Eii) 4 is an open association of institutes
practical, introducing the students to In conclusion, we hope that the 10th VLTI
and laboratories willing to collaborate in the exploita-
many tools commonly used in VLTI school will have provided the partici- tion and development of long-baseline interferome-
observation preparation and data reduc- pants with all the skills and knowledge try in optical/infrared astronomy. We are grateful to
tion. The Gather space was central as they need to make successful propo- the JMMC for their scientific, technical and moral
support along this school’s rough and bumpy organ-
it was used to connect to the lectures of sals and to publish, in the near future,
isational journey. We received support from the
the school through Zoom sessions. astonishing scientific results using optical GRICAD service for the virtual machines used during
Moreover, it was also used to allow inter- interferometry. the practice sessions. GRICAD is a support unit
actions between the students and the (UAR 3758) linked to the CNRS’s Institut national des
sciences mathématiques et de leurs interactions
teachers during the practice sessions on
(INSMI), to Grenoble Alpes University, to the
the virtual machines, as well as for the School environmental impact Grenoble Institute of Engineering and to the French
“coffee breaks” and informal discussions. national research institute for the digital sciences
The environmental impact of astronomi- (INRIA). The school was also supported by the
Lagrange astrophysics laboratory in Nice, the Côte
The lectures were exhaustive and varied, cal meetings is an increasing concern
d’Azur Observatory (OCA), and the CNRS through
with a thorough introduction to long- (Burtscher et al., 2020). To compare the its Action Spécifique Haute Résolution Angulaire
baseline interferometry, presentations of impact of a VLTI school held online and (ASHRA). Nuclino was kindly provided for free for the
facilities (the VLTI and CHARA), of instru- one held in person, we compared the school organisation.
ments (MATISSE, GRAVITY, Stellar impact of this school with the impact of
Parameters and Images with a Cophased the 2018 school in Lisbon. A relevant References
Array [SPICA], the Michigan InfraRed metric is the equivalent tonnage of CO2
Combiner-eXeter [MIRCX] and the produced by an activity; we can use Burtscher, L. et al. 2020, Nature Astronomy, 4, 823
Garcia, P. 2009, The Messenger, 135, 50
Michigan Young STar Imager at CHARA the numbers provided by Burtscher et al.
Garcia, P. J. V. et al. 2018, The Messenger, 173, 49
[MYSTIC]), introductions to data reduc- (2020). The VLTI schools have a similar Lopez, B. et al. 2021, accepted for publication
tion, model fitting, image reconstruc- attendance composition to the European in A&A
tion, and radiative transfer for astrophysi- Astronomical Society Annual Meetings
cal modelling, and also several science (EAS, formerly known as EWASS). Such
Links
courses on young stellar objects, exo an in-person meeting produces 1.5 t
planets and asteroids. These courses CO2e per capita, essentially from travel 1
List of all the previous VLTI schools:
were backed up by practice sessions on (1855 t CO2e for 1240 participants in http://www.jmmc.fr/english/training
2
the basics of interferometry and observa- the 2019 meeting), whilst an online meet- 10th VLTI school website: https://vltischool2021.
sciencesconf.org
tion preparation, a MATISSE data reduc- ing produces 328 g CO2e per capita 3
Course videos: https://pod.univ-cotedazur.fr/vlti-
tion session, model fitting and image (582 kg for 1777 participants at EAS school-2021
reconstruction sessions, and for the first 2020). Translated into VLTI schools, the 4
European Interferometry Initiative: https://
time a practice session on radiative trans- 2018 Lisbon school (52 participants in european-interferometry.eu
fer. Finally there was a proposal prepara- person) produced an equivalent of 75 t
tion homework that lasted throughout the of CO2, mainly the result of transportation
school, starting with a presentation of the (by plane or train). On the other hand,
principles of telescope time application, the 2021 online school produced an
and ending with the students’ proposal equivalent of 48 kg of CO2, mainly from
presentations in front of a mock observ- the activity of the computer servers
ing programme committee. dedicated to the school. In addition, we
mailed goodies to the students that
When polled after the school, the stu- account for ~ 950 kg CO2e, including
dents were very satisfied with the quality manufacturing (300 kg CO2e, the manu-
of the lectures and the practice sessions. facturer being in France) and mailing
Of course, the 100% online format of (650 kg CO2e), raising the total impact of
the school complicated the organisation the 2021 school to roughly 1 t CO2e.
and the interactions. Fortunately, how- This therefore confirms that conducting
ever, the Gather platform, with its persis- such a school in a 100% online format,
tent space including video chatting, even with goodies sent to the students at
helped a lot to allow frequent interactions home, reduces by a tremendous amount
between students and teachers during the environmental impact compared to a
the practice sessions and outside the normal in-person school.
Evanthia Hatziminaoglou 1 The EAS 2021 symposium talks ascimento, on the other hand, pre-
N
Gergö Popping 1 addressed five intertwined themes sented a galaxy at z = 6.8 that is rotation-
Martin Zwaan 1 spread over six sessions: ALMA’s ally supported, as derived from [CII]
unmatched potential to study high- observations. At z ~ 4 at least some gal-
redshift galaxies, its contribution to our axies have regular rotating discs (as
1
ESO understanding of the properties of explained by Francesca Rizzo); however,
dust in the early Universe, the properties Fernanda Roman de Oliveira showed
of the ISM, the feeding and feedback that there is a diversity in V/s, leading to
The properties of the interstellar processes at high redshift and the syner- the well established dominance of rota-
medium (ISM) of the highest-redshift gies with current and future facilities. tion by z ~ 2.
galaxies and quasars provide important Figure 1 shows a schematic representa-
indications of the complex interplay tion of some examples of science with Other than targeted observations of
between the accretion of baryons onto ALMA. Highlights from each theme are high-z galaxies, several ALMA extraga-
galaxies, the physics that drives the summarised below, underlining the lactic Large Programs1 have demon-
build-up of stars out of this gas, the power of ALMA and its remarkable con- strated ALMA’s capability to be used as
subsequent chemical evolution and tribution to groundbreaking discoveries. a survey instrument, despite its small
feedback processes and the reionisa- field of view. Indeed, many of the results
tion of the Universe. The Atacama Large presented in the symposium come from
Millimeter/submillimeter Array (ALMA) High-redshift galaxies as observed by such projects. Hanae Inami focused
continues to play a pivotal role in the ALMA her invited talk on the Reionization Era
characterisation of the ISM of high- Bright Emission Lines Survey (REBELS;
redshift galaxies. Observations of the Observations of “old” galaxies at z > 8, Bouwens et al., submitted to ApJ), a
dust continuum emission, atomic an epoch at which the peak of dust [CII] and [OIII] survey of 40 Lyman-break
fine-structure and molecular lines aris- emission falls in ALMA’s bands 7 and 8, galaxies (LBGs) at 6.5 < z < 9.5 and
ing from high-redshift galaxies are can provide information about the time reported dust continuum detections for
now carried out routinely, providing ever at which these very first galaxies were about half of the sources. The [CII] and
more constraints on the theoretical born. Indeed, ALMA observations of gal- dust emissions are spatially extended
models of galaxy formation and evolu- axies at z ~ 9 confirmed a formation and diverse (Inami, in preparation;
tion in the early Universe. The European redshift beyond ~ 11 (Hashimoto et al., also reported by Rebecca Bowler in her
Astronomical Society’s EAS 2021 sym- 2018; Laporte et al., 2021a), providing an talk). Serendipitous detections of dust-
posium dedicated to the exploration of indirect probe into the cosmic dawn. obscured galaxies, discussed by Pascal
the high-redshift Universe with ALMA Oesch, imply that normal star-forming
provided a forum for the observational According to Mahsa Kohandel, simula- galaxies existed in the EoR but were
and theoretical high-redshift ALMA tions predict that early galaxies could essentially missed until now, setting con-
communities to exchange their views form their discs as early as during the straints on the obscured star formation
and recent results in this rapidly evolv- epoch of reionisation (EoR) but Sander out to the EoR.
ing field. Schouws showed that the majority of
galaxies (~ 70%) at the EoR are disper- In a talk introducing the ALMA Large
sion-dominated, with stellar mass only Programme to INvestigatE (ALPINE;
Figure 1. ALMA’s windows on the Epoch of Reioni accounting for up to 10% of their dy- Le Fèvre et al., 2020), which carried out
zation (EoR) — from Joris Witstok’s talk. namical mass. Ana Carolina Posses [CII] observations of 122 main-sequence
NAOJ
[CII] 158 µm
(ALMA)
[OII] 88 µm
(ALMA)
zๆ6 zๆ4
Figure 3. Origin of fine structure and molecular line to be below or around one, indicating Overdensities around z ~ 6 quasars turn
emission from the various ISM components.
that the gas in the outflow might be in a out to be low-z (~ 1–2) objects (Meyer, in
Reproduced from van der Tak et al. (2018) with the
author’s permission. warm ionised phase, or that the surface preparation), in agreement with [CII] inci-
brightness of the outflows is low (Carniani dence rates of companion overdensities
et al., in preparation). reported in the literature.
[CII] emission with respect to that of [OIII]/
UV (Ginolfi et al., 2020a; Herrera-Camus The recent literature on the stacking of
et al., 2021). With the help of zoom-in Feeding and feedback at high redshift [CII] observations in high-z active galactic
cosmological simulations and ISM mod- nuclei (AGN) in search of evidence for
elling, the above can be put in a wider The circumgalactic medium (CGM) is a outflows reports contradictory results.
context, indicating ISM conditions point- place where physical processes such as Even though the different spatial res
ing towards an efficient conversion of mixing, cooling and heating happen, olutions might explain the tension if out
gas into stars. The high metallicity and modifying the properties of the gas that is flows extend out to several kiloparsecs,
gas density are indicative of a dense ISM moving in or out of the galaxies. One Roberto Maiolino postulated that the esti-
where bursts of star formation rapidly of the main challenges currently is to mated high-z cold outflows are not pow-
enrich the gas (Vallini et al., 2021). Simu- explain the coexistence of the cold gas erful enough to quench star formation via
lations by Ramos Padilla et al. (2021), (T < 104 K) in a medium filled with hot gas. ejection, indicating either the importance
combining EAGLE cosmological simula- “The lack of cold gas in CGM simulations of ionised outflows or a weaker coupling
tions with CLOUDY cooling tables, is a bias rather than a prediction,” to with the ISM.
showed that at z ~ 6 [CII], [OIII] and other quote Claudia Cicone. Higher-resolution
lines in fact trace different phases of the simulations lead to larger fractions of At the same time, ALPINE found signa-
ISM depending on metallicity and SFR. cold gas, but simulations are still not tures of star formation-driven outflows
converging. (i.e., broad wings) in the stacked profiles
Extended [CII] emission observed in gal- of [CII] in galaxies with SFRs > 25 M⊙ yr –1
axies beyond redshift ~ 6 (although not Owing to the lack of sensitivity of the (the median SFR of the ALPINE sample),
seen in a deep stack by Jean-Baptiste current facilities, molecular gas is not but no evidence for outflows amongst
Jolly) can be produced by cooling out- detected in the CGM at z ~ 0. ALMA, galaxies with lower SFR (Ginolfi et al.,
flows with mass loading factors larger however, can observe the CGM at high 2020a). The outflowing gas in the highest-
than three (for example, Pizzati et al., redshift. Claudia Cicone reported on SFR galaxies moves at a maximum
2020). The presence of outflows is sup- observations of a halo extending out velocity (vout ~ 500 km s –1) below the
ported by larger values of L[OIII]/L[CII] to a radius of 200 kpc around a z = 2.2 escape velocities (400–800 km s –1) and it
compared to those in the local Universe X-ray-selected quasar (Cicone et al., will be trapped in the CGM of these gal-
(large values also reported by Joris 2021). There is no evidence of an axies (Ginolfi et al., 2020b). Hydrogen
Witstok) seen in galaxies at z > 6. How- overdensity around it, making the inter- fluoride was reported by Maximilien
ever, the loading factors are observed pretation of this CO halo challenging. Franco to be another excellent probe of
outflowing molecular gas, as well as a the degeneracies between nebular enriched gas in the haloes surrounding
good tracer of H2 in galaxies. Finally, emission and Balmer break. Further galaxies at z > 4, indicative of CGM pollu-
ALMA observations of a z ~ 5.5 main- ALMA-Spitzer synergies are illustrated by tion by outflows. ALMA has furthermore
sequence galaxy showed ISM properties the Spitzer Matching survey of the Ultra- opened entire new windows onto the
similar to those of local starburst gal VISTA deep Stripes (SMUVS) project chemistry and ISM physics of galaxies
axies, with extended [CII] emission, evi- (using the Infrared Array Camera chan- during the EoR, contributing to key
dence for an outflow and a regular rotat- nels 1 & 2 to observe the Ultra-VISTA questions pertaining to the build-up and
ing disc (Herrera-Camus et al., 2021). deep stripes) presented by Tomoko reionisation of the early Universe. This
Suzuki, that studies the submillimetre symposium was a brilliant demonstration
properties of high-redshift galaxies in of these groundbreaking results and
Synergies with current and future the A3COSMOS catalogue (Liu et al., served as an inspiration to continue to
facilities 2019). Galaxies with ALMA counterparts push the envelope to deliver more excit-
are found to be systematically massive ing ALMA high-redshift results in the dec-
ALMA has undoubtedly revolutionised (M∗ > 1010 M☉) and tend to have larger ade to come.
our knowledge of the physics of the early dust extinction and higher star formation
Universe. Synergies with existing and activity. Additionally, ALMA, Multi Unit
future facilities help push ALMA’s poten- Spectroscopic Explorer (MUSE) and Acknowledgements
tial to its limits. From the discovery of HST observations of luminous galaxies at In addition to the authors of this article, the Scientific
the very first galaxies to galaxy kinemat- z > 6 presented by Jorryt Matthee indi- Organising Committee of the symposium included
ics at redshifts close to the EoR to metal- cate that high-redshift luminous galaxies Paola Andreani, Andy Biggs, Gabriela Calistro
licities at redshifts of six and beyond, likely reside in early ionised bubbles Rivera, Jackie Hodge, Rob Ivison, Kirsten Kraiberg
Knudsen, Renske Smit, Remco van der Burg
the possibilities for further groundbreak- (Matthee et al., 2020). and Eelco van Kampen. Thanks to them, and to all
ing discoveries are almost endless. the speakers and participants, the symposium was
Meanwhile, it remains unclear how the a great success.
One of ALMA’s challenges is to resolve supermassive black holes (with masses
the gas properties of galaxies with well often above 109 M☉) residing in the cen- References
resolved dynamics at high redshift and to tres of more than 200 known quasars at
decompose galaxies into their constitu- z > 6 have reached these sizes. Current Calzetti, D., Kinney, A. L. & Storchi-Bergmann, T.
ents (for example, Lelli et al., 2021). An theories predict a massive gas reservoir 1994, ApJ, 429, 582
Caputi, K. I. et al. 2021, ApJ, 908, 146
obvious synergy, brought out by Mark and ALMA-supported studies of these Carniani, S. et al. 2020, MNRAS, 499, 5136
Swinbank in his invited talk, is to improve objects point towards rapidly star-forming Cicone, C. et al. 2021, A&A, 654, L8
dynamical constraints of galaxies at high molecular gas and dust content da Cunha, E. et al. 2021, ApJ, 919, 30
high redshifts by combining their ALMA hosts. These are often accompanied by Dessauges-Zavadsky, M. et al. 2020, A&A, 643, A5
Drake, A. B. et al. 2020, ApJ, 902, 37
CO and [CII] maps with stellar mass submillimetre companions but simula- Faisst, A. L. et al. 2020, MNRAS, 498, 4192
maps measured by the James Webb tions presented by Fabio Di Mascia indi- Fujimoto, S. et al. 2021, ApJ, 911, 99
Space Telescope (JWST) and kiloparsec- cate that such systems might be, instead, Ginolfi, M. et al. 2020a, A&A, 633, A90
scale dynamics measured by instruments quasar merging systems. Alyssa Drake Ginolfi, M. et al. 2020b, A&A, 643, A7
Gruppioni, C. et al. 2020, A&A, 643, A8
like the Enhanced Resolution Imager and reported on MUSE Ly-a halo observa- Hashimoto, T. et al. 2018, Nature, 557, 392
Spectrograph (ERIS) on ESO’s Very tions and complementing ALMA cold gas Herrera-Camus, R. et al. 2021, A&A, 649, A31
Large Telescope and, in the future, the and gas continuum observations to Hodge, J. A. & da Cunha, E. 2020, RSOS, 7, 200556
High Angular Resolution Monolithic find decoupled kinematics between halo Jones, T. et al. 2020, ApJ, 903, 150
Laporte, N. et al. 2017, ApJL, 837, L21
Optical and Near-infrared Integral field gas and the ISM in high-z quasars (Drake Laporte, N. et al. 2021a, MNRAS, 505, 3336
spectrograph (HARMONI) on ESO’s et al., 2020). Follow-up observations Laporte, N. et al. 2021b, MNRAS, 505, 4838
Extremely Large Telescope. of ALMA-detected high-redshift galaxies Le Fèvre, O. et al. 2020, A&A, 643, A1
with the JWST or the Origins Space Tele- Lelli, F. et al. 2021, Science, 371, 713
Liu, D. et al. 2019, ApJ, 887, 235
The JWST is going to observe optical scope might reveal dust-obscured or faint Matthee, J. et al. 2020, MNRAS, 492, 1778
rest-frame spectra of z > 6 galaxies AGN residing in them. Pizzati, E. et al. 2020, MNRAS, 495, 160
for the first time. Combined observations Ramos Padilla, A. F. et al. 2021, A&A, 645, A133
of strong optical lines ([OII], [OIII], [SII] Rizzo, F. et al. 2021, MNRAS, in press
Roberts-Borsani, G. W., Ellis, R. S. & Laporte, N.
doublet) with the JWST and [OIII] 88 μm Conclusions 2020, MNRAS, 497, 3440
with ALMA will give direct estimates of Tamura, Y. et al. 2019, ApJ, 874, 27
metallicities at z > 6 (Jones et al., 2020). In its first decade of operation, ALMA has Vallini, L. et al. 2021, MNRAS, 505, 5543
revolutionised our understanding of van der Tak, F. F. S. et al. 2018, PASA, 35, 2
Zavala, J. A. 2021, ApJ, 909, 165
ALMA [OIII] 88 μm observations impose the high-redshift Universe. It has demon-
strong constraints on the spectral energy strated the presence of massive dusty
distribution fit of Spitzer/IRAC-selected systems that contain large gas reservoirs Links
z ~ 7–9 galaxy candidates (Roberts- fuelling vigorous star formation and 1
ALMA Large Programmes: https://almascience.
Borsani, Ellis & Laporte, 2020), especially accretion processes. ALMA has also org/alma-data/lp
prior to the arrival of the JWST, breaking exposed the presence of cold and metal-
Boris Häußler 1 together the low- and high-redshift extra- rounded programme. By mid-October
Emanuela Pompei 1 galactic communities to review where 2019 everything was set for a successful
Yara Jaffé 2 we stand and prepare for the challenges workshop, bringing together experts
ahead. from all over the globe to discuss the lat-
est and most exciting results.
1
ESO
2
Instituto de Física y Astronomía, Run-up However, around this time a wave of
Universidad de Valparaíso, Chile social unrest in Chile began. The associ-
We settled on a date in late 2019, to give ated protests worried many workshop
the largest and latest VIMOS surveys participants, so to avoid cancellations it
In April 2021 more than 200 participants (the Large Early Galaxy Astrophysics was decided to postpone the meeting
gathered on Zoom for a meeting to dis- Census [LEGA-C] and VANDELS) enough until a later date, aiming for March/April
cuss what we have learned about gal- time to produce some early scientific 2020, hoping that the unrest would have
axy evolution using spectroscopic sur- results using their final datasets and to subsided by then. After checking with
veys, and to pay a special tribute to present to the broader extragalactic com- the registered participants, we decided to
the Visible Multi-Object Spectrograph munity the vast set of raw and enhanced keep the same programme and the same
(VIMOS), which carried out some of the data that both surveys had publicly participants, so Galspec19 became
largest spectroscopic galaxy surveys released. The date also coincided excel- Galspec20, and we decided to hold it in
to date and helped to pave the way for lently with the planned arrival at ESO October 2020 to avoid direct competition
ongoing and future facilities. Despite of new MOS facilities, such as the Multi- or overlap with other meetings.
being delayed twice and having to be Object Optical and Near-infrared Spec-
adapted to an online format, the meet- trograph (MOONS) in 2021 and the We updated websites and started adver-
ing was well received and many high- 4-metre Multi-Object Spectroscopic Tele- tising the new dates. We had of course
impact results were presented by a scope (4MOST) in 2022, and elsewhere, heard in the news that there was a new
diverse body of participants. We pres- such as the Multi-Espectrógrafo en virus spreading, but like everybody else
ent the organisation, lessons learned, GTC de Alta Resolución para Astronomía we did not make much of it at first, so we
and legacy of this workshop. (MEGARA) at the Gran Telescopio started replanning the workshop. Before
Canarias and the WHT Enhanced Area we had even sent out anything concrete,
Velocity Explorer (WEAVE) at the William however, the COVID-19 pandemic had
Motivation Herschel Telescope. Further MOS and reached Chile in full swing and by mid-
IFU instruments for future 30–40-metre- March it was getting less and less likely
Our understanding of galaxy evolution class telescopes are also appearing on that we could confidently plan a (in-person)
has changed dramatically in the last dec- the horizon, such as the Multi-Object meeting for October, with the observatory
ade. One of the main advances has Spectrograph for Astrophysics, Interga- and much of the country going into
undoubtedly been the emergence of both lactic medium studies and Cosmology lockdown, and no possibility of making a
large-scale and and/or deep surveys with (MOSAIC) and the High Angular Resolu- decision in May or June. Instead, we
multi-object spectrographs (MOS) and tion Monolithic Optical and Near-infrared decided to postpone the workshop once
integral field units (IFU). These vast data- Integral field spectrograph (HARMONI) again and within a few months decided
sets have brought into focus the complex at ESO’s Exremely Large Telescope, the to aim for April 2021 for one last try at
interplay between the ages, metallicities Wide-Field Optical Spectrometer (WFOS) hosting this workshop, ideally in person
and kinematics of galaxies, as well as and the InfraRed Imaging Spectrograph but with the possibility of holding it online
their masses, sizes, structures, nuclear (IRIS) at the Thirty Meter Telescope and or elsewhere (Garching, for example) if
activity and environments. Still, many the GMT Multi-object Astronomical and the pandemic would not allow a meeting
open questions remain. Cosmological Spectrograph (GMACS) in Santiago. And so Galspec20 became
and GMT Consortium Large Earth Galspec21.
At the end of March 2018 the Visible Finder (G-CLEF) at the Giant Magellan
Multi-Object Spectrograph (VIMOS), one Telescope. By then the original workshop had been
of ESO’s workhorses in Paranal obser delayed by more than 18 months. As
vatory, was decommissioned from In June 2018 ESO approved financial that would mean that the programme and
Unit Telescope 3 after nearly 15 years of support for our proposed workshop enti- talks would be quite outdated, we
service. During this time, VIMOS had tled Extragalactic Spectroscopic Surveys: decided to start from scratch with a new
amassed over 9700 hours of science Past, Present and Future of Galaxy abstract deadline in December 2020
data, mostly devoted to spectroscopic Evolution (GalSpec19), to be hosted at and a final decision on whether the work-
surveys of galaxies across cosmic time. the ESO offices in Vitacura, Santiago, shop would be held in person or virtually
To commemorate this milestone, we Chile in November 2019. at around the same time. As the pan-
thought it appropriate to celebrate the demic rolled on, we decided in November
achievements of this amazing instrument We received well over 100 excellent to hold the meeting online, with a
with a five-day workshop. The goal of abstracts, and the scientific organising reduced, half-day schedule. However, to
this ESO workshop was also to bring committee (SOC) put together a well allow additional talks we decided to try a
sessions, our Slack channels, and – How do gas accretion and outflows
Gather for our poster sessions in a meet- shape galaxy properties?
ing that covered a wide range of topics: – A re rotation curves flat or declining at
– Stellar Populations and Star Formation z > 1?
History – How does environment impact the for-
– Stellar and Gas Kinematics mation and evolution of galaxies?
– Environment and Intergalactic Medium – What is the redshift evolution of galaxy
– Star Formation and Interstellar Medium properties and scaling relations?
– Upcoming Instrumentation – How are metals distributed in galaxies
and how are they modulated by accre-
All contributions to the workshop were tion of gas, outflows, feedback and star
excellent; picking out one or two high- formation?
lights would not do the other contribu-
tions justice and would present a very The workshop was an excellent example
biased view. However, all talks (except for of why we need these workhorse
the discussion sessions) have been instruments on large telescopes to carry
recorded and are available on YouTube1, out large surveys of statistically signifi-
posters and talk slides have all been cant samples over a large range of red-
uploaded to Zenodo 2, and additional shifts. Luckily, the next generation of
information is available on our website 3. such instruments, for example 4MOST,
MOONS and the blue-optimised inte-
The talks given and the posters pre- gral-field spectrograph BlueMUSE in the
sented highlighted well that not only did case of ESO instrumentation, and the
Figure 1. The Galspec21 workshop poster. VIMOS work efficiently as a redshift next generation of dedicated survey tele-
machine (see VIPERS, z-COSMOS, scopes (for example the Simonyi Survey
VVDS, VUDS, VANDELS, and other sur- Telescope at Vera C. Rubin Observatory,
veys), but was also extremely successful the EUCLID space telescope and others)
mixed scheme in which each session as an instrument used to derive proper- are already well advanced or in develop-
additionally featured several pre-recorded ties of galaxies (see, for example, Ses- ment. They will be starting to take data in
talks that participants could watch at sions 1, Stellar Populations and Star For- the coming years, opening yet another
their leisure, with discussion on these mation History, and 2, Stellar and Gas window onto the distant Universe, with
talks on Slack, along with long (~ 45 min- Kinematics) and to study the influence even larger statistical samples and more
utes) poster sessions on the video chat of environment on galaxy properties and interesting science to be done.
platform Gather, to allow as much per- active galactic nuclei (see, for example,
sonal interaction between participants as Sessions 3, Environment and Intergal
possible. The advantage of this setup actic Medium, and 4, Star Formation and Demographics
was that the financial support provided Interstellar Medium). VIMOS has left
by ESO more than covered the costs behind a vast dataset on the ESO archive, Although it is hard to tell who actually
of the entire workshop, so registration which will allow further studies to be car- participated in which sessions during an
was free for all participants. As travel ried out for years to come. online meeting, the 251 registered parti
costs also did not have to be covered, we cipants signed up from 26 different coun-
received more than 140 abstracts for However, as well as VIMOS data, results tries from all continents (48% Europe,
~ 50 talk and ~ 50 poster slots and more using other instruments were also dis- 21% South America, 13% North America,
than 250 registrations. We even had to cussed during all the sessions. The Multi 10% Asia, 7% Australia, and two parti
close registration to avoid the workshop’s Unit Spectroscopic Explorer (MUSE) cipants from Africa; see Figure 2). Of the
becoming too big to handle, as we featured prominently as an instrument 142 abstracts received, 64 (45%) were
wanted the in-person contact to be an well suited for galaxy surveys, and results submitted by women, 78 (55%) by men,
important element of the meeting, from the Atacama Large Millimeter/ 40 (28%) by students, 50 (35%) by post-
including discussion and Q&A sessions submillimeter Array (ALMA), the K-band docs/fellows, and 52 (37%) by senior
and person-to-person interaction during Multi Object Spectrograph (KMOS) and astronomers. Thanks to our amazing
our poster sessions. Figure 1 shows the other instruments were also discussed. SOC, these fractions are nearly exactly
workshop poster. These, and other upcoming instruments represented in the 96 poster and talk
(see Session 5, Upcoming Instrumenta- contributions to the workshop (see Fig-
tion) will help us to answer outstanding ures 3 & 4): 44 (46%) by women, 52 (54%)
Workshop and Legacy questions in galaxy evolution. Important by men, 28 (29%) by students, 39 (41%)
advances on future key issues were by postdocs/fellows, 29 (30%) by senior
Finally, in the second week of April 2021 already presented during the workshop: astronomers. Four out of the eight invited
and after 3 years of planning, up to – How do gas and stellar kinematics talks were given by women.
250 participants logged into our Zoom evolve with redshift?
United Kingdom | 25
Poland | 4
Mexico | 4
Chile | 49
Italy | 34
Figure 2. (Above) These pie charts show the coun- Figure 3. (Below left) Gender balance of presenters Figure 4. (Below right) Distribution of seniority of
tries hosting the workshop participants. (talks and posters), which very closely resembles the presenters (talks and posters), which very closely
gender distribution in the abstracts submitted. resembles the distribution in the submitted
abstracts.
100% 100%
90% 90% 2
40 13
82 15
80% 80% 1
124 64 26 19
4 6
70% 70%
60% 10 60%
50 23 5
74
50% 50% 8 12
16
40% 40%
30% 30%
127 78 30 22
4 6
20% 20% 95 52 19 4
10% 10% 10
2
0% 0%
s
rs
s
ts
rs
rs
rs
s
ts
C
C
C
C
ct
er
ct
er
an
ke
SO
ke
LO
an
SO
ke
ke
LO
ra
st
ra
st
ea
ea
ea
ea
ip
ip
Po
st
Po
st
ic
ic
sp
sp
sp
sp
Ab
Ab
rt
rt
Pa
Pa
d
d
te
te
te
te
vi
bu
vi
bu
In
In
tri
tri
on
on
C
C
Main conclusions & ways forward seemed to work very well, although it Acknowledgements
certainly required some getting used
Throughout these troublesome times and multiple
Overall, we found that the online setup of to. We had also outsourced the video re-starts of the organisation, the support from
the workshop worked surprisingly well, editing for YouTube, which could in princi- everybody was outstanding. Our SOC stayed
and we can recommend a similar combi- ple have been done by one of us but together and everybody was happy to contribute
despite being thrown around by us and the pan-
nation of tools for other workshops that would have been a full-time job. The
demic. Nearly all of our invited speakers were also
are held online in the future. We found, company we used was superb and got all happy to give their talks, despite our significantly
however, that it was critical to have very talks online within a few hours, despite changing their scope, given the reduced, half-day,
long poster sessions on Gather (see Fig- its being the end of the day at their loca- schedule. The support from ESO, both in Chile
and Garching, was also great throughout this time.
ure 5) despite the half-day schedule of tion, so people around the world could
Our funding was flexible enough to be moved by
the meeting, in order to foster in-person watch them before we even met for the two (!) financial years, while Claudio Melo (the ESO
contact and trigger discussion. This following day. representative in Chile at the time) and Rob Ivison
Figure 5. This picture shows a small screenshot No scientific workshop can be successful with- Olivier and to all the supporting staff for making a
from our space on Gather which was used for poster out interesting science and the people behind it. We dream instrument come true!
sessions, including two separate discussion groups. wish to remember the principal investigator of
VIMOS, Olivier Le Fèvre, who died in 2020 after a
serious illness. The extragalactic community and Links
his family and friends were deeply saddened by this
1
loss, but the news also hit those of us who met Galspec21 YouTube channel: https://www.youtube.
(ESO’s Director for Science) were very supportive O livier personally during the many VIMOS commis- com/channel/UC7tsM_Uwa48BJEYICjYumvA/
of all our decisions and gave us the background and sioning or first science runs and had the opportunity playlists
2
information that we needed to make them. Our to know and appreciate his passion for his work Zenodo for posters and slides: https://zenodo.org/
thanks go to everybody who supported this work- and his dedication. We were truly sorry that he was communities/galspec2021
3
shop throughout, not least the amazing speakers not with us to discuss both the excellent results Workshop website: https://www.eso.org/sci/
and participants, and who made this workshop a obtained with his instrument and also the interesting meetings/2021/galspec2021.html
success. challenges for the future. A big thank you goes to
ESO/PHANGS
Tony Mroczkowski 1
Andra Stroe 2
Stella-Maria Chasiotis-Klingner 1
1
ESO
2
Center for Astrophysics | Harvard &
Smithsonian, Cambridge,
Massachusetts, USA
important but understudied pathways for tions coming from within ESO and the during the workshop were primarily
the transformation of the member galax- wider community as represented by the through the Slack chat platform, which
ies. Jellyfish galaxies were discussed in workshop scientific organising committee we organised in the following way.
detail, addressing open questions as to (SOC). The meeting itself was hosted Each day was assigned its own Slack
how these disrupted galaxies are shaped on the popular online meeting platform channel, while general announcements,
by their environments. Not only is galaxy Zoom, and streamed live on YouTube personal introductions, job postings,
evolution heavily impacted in low-z clus- for those who could not participate SOC discussions, helpdesk issues, and
ters, but results showed that all types directly. Despite the distribution of partici- random conversations each had their
of active galactic nucleus activity can be pants across the globe, the number own channels. Each session of the work-
promoted in such overdense environ- of active attendees during the live Zoom shop had a main chair to moderate the
ments. The workshop revealed opportu- sessions was very high, fluctuating talks, and a Slack chair to monitor for rel-
nities for knowledge exchange between between 120 and 160 live Zoom partici- evant questions during the talk. During
the lower- and higher-redshift cluster pants throughout the five days of the the discussion session, the Slack chairs
communities, with the possibility of workshop. In order to accommodate constructed polls related to the main
unveiling and studying extended proto- more time zones around the world, we themes they identified for the session.
cluster-like structures and mergers at recorded all but two of the presentations These polls were disseminated using
lower redshift still undetected by current (subject to permissions) and posted Mentimeter. The result of using Slack and
instruments. The participants agreed them online on YouTube 3 after editing by Mentimeter was generally successful,
that new instruments such as Athena, an external professional video editing and may have encouraged more active
the 4-metre Multi-Object Spectroscopic service. The videos garnered significant participation by early-career scientists.
Telescope (4MOST), and Euclid will be views from participants during the work-
transformational in unveiling these hidden shop and reached a broad audience The second component to the online
populations. during and after the conference. As of interactions was through Gather, an
25 August 2021, the workshop YouTube online conference tool that participants
It was originally intended that GCF2021 channel has 115 subscribers, over
would be held in person in 2020, but 3300 views and a total watchtime of
it was delayed because of the COVID-19 almost 280 hours.
pandemic. A considerable effort was Figure 2. Conference photo montage produced
made to adapt the workshop to a suc- Interactions mainly took place in two by David Sobral using screenshots from Zoom and
cessful online format, with many sugges- ways. First, the questions and discussion Gather.
75% 75%
50% 50%
25% 25%
0% 0%
rs
irs
s
rs
irs
s
s
s
r
r
C
C
ke
ke
er
nt
er
nt
ke
ke
LO
SO
LO
SO
ha
ha
st
st
pa
pa
ea
ea
ea
ea
C
C
Po
Po
ci
ci
sp
sp
sp
sp
r ti
r ti
d
d
Pa
Pa
ed
ed
te
te
bu
bu
v it
v it
In
In
tr i
tr i
on
on
C
C
Male Female Non-binary Rather not say Blank Student Postdoc Staff Other Blank
Figure 3. Charts depicting the distributions of gen- Demographics names and identifying information from
der (left) and career stage (right) of the participants
the abstract submissions and abstained
in the workshop, where we note the categories
shown may not accurately reflect the complex reality. As was the case for GCF2017 and many from voting in the talk selection process.
other ESO science workshops, we The result was that the allocated talks
sought to ensure that the presenters and posters reflect the gender distri
were a fair representation of the commu- bution of those submitted. This also rep-
nity. Whilst we expected fewer than resented fairly the overall distribution
use to interact with a virtual space via 150 participants, in the end we had of the participants, as the majority of
an avatar cartoon, allowing one to visit 192 registered participants, of whom participants had either a talk or poster
poster sessions and meeting rooms, 67 were female, 118 were male, 3 were presentation.
meet colleagues, and generally wander non-binary, and 4 did not identify.
around as if in a real conference venue. This was a dramatic increase over the
Mixed feedback indicates it was generally 100 participants at GCF2017, and based Acknowledgements
successful, though a number of users on feedback from the participants was We are grateful for support from ESO’s Directorate
reported delays and that disproportion- likely facilitated by hosting the workshop for Science and for contributions from the Center for
ately heavy CPU resources were required online and not having a registration fee. Astrophysics | Harvard & Smithsonian that enabled
for both the software and web-based We note that only 72 of the participants the use of the widespread online meeting platform
Zoom. We are also indebted to the members of the
interfaces to the platform. A montage were senior scientists, whilst the major- science and local organising committees who made
snapshot of the participants in Zoom ity were early career scientists (graduate this workshop a success.
and their avatars in Gather is shown in students and postdocs). In order to
Figure 2. ensure active participation, we contacted
References
those who did not submit an abstract for
Similarly to GCF2017, the majority of a talk or poster to verify that they were Bassini, L. et al. 2020, A&A, 642, A37
the presentations were collected on a legitimately interested astronomers. We Dolag, K. et al. 2008, New Journal of Physics, 10,
voluntary basis, and are now available on note that only 18 registered participants 125006
Mroczkowski, T. et al. 2017, The Messenger, 170, 63
the Zenodo platform 4, which provides did not connect to the workshop.
Digital Object Identifier (DOI) links and is
indexed by the SAO/NASA Astrophysics Of the seven invited review talks selected Links
Data System (ADS). by the SOC, three were by female sci 1
GCF2017: https://www.eso.org/sci/meetings/2017/
entists, three were by male scientists, GCF2017.html
Using Mentimeter, we conducted an exit and one speaker did not self-identify. Of 2
GCF2021 workshop programme: https://www.eso.
poll after the workshop to gauge partici- the 50 accepted contributed talks, org/sci/meetings/2021/GCF2021.html
pants’ perspectives on it. The consensus 16 were by female speakers and 34 by 3
Workshop YouTube channel: https://www.youtube.
from the 45 respondents was that the male speakers, whilst for the 59 poster com/channel/UCntCTDx2OHfUbnpm4gUHtWg
4
Online proceedings: https://zenodo.org/
workshop was extremely well received, presenters, 20 identified as female, 36 as communities/gcf2021
with particular praise for the use of male and two as non-binary, and one
recordings that made the workshop more did not self-identify. The SOC itself com-
accessible, for the use of Slack to con- prised five female and seven male scien-
tinue discussions after the live sessions tists, with one female and one male
and for the interactions in Gather. Whilst co-chair. These simple demographics are
the consensus was that the conference depicted in Figure 3.
was a great success as an online event,
the vast majority of respondents hoped Following a similar methodology to that
that the next GCF workshop will take successfully applied for talk selection
place in person. in GCF2017, one SOC member removed
Fellows at ESO
Marianne Heida