Science Topic 2: Stars and Galaxies
Chair: Michael Merrifield, University of Nottingham, UK
This session looked at some of the scientific questions about the
relationship between stars and galaxies that might be addressed using
an Extremely Large Telescope. Specifically, we have considered:
Following on from these science-driven issues, we also asked:
As with the other summaries to the Workshop, this document is not
intended to provide a definitive description of the astounding
opportunities for innovative science that will become possible with an
Extremely Large Telescope; rather, it seeks to provide a summary of
the issues discussed at the Workshop, and a starting point for
A link to a quite comprehensive paper on this area of ELT work by Rosie
Wyse is available HERE (or will be when Rosie is happy with her document).
When and How Were Stars Formed?
In nearby galaxies, we will be studying resolved stars, allowing us to
From such data, we will be able to quantify the ages and metallicity
distributions of stars in different environments, and their initial
mass functions. Amongst the issues that could be addressed with such
data is quantifying the heavy element abundances of the most
metal-poor stars in the local Universe.
- Colour-magnitude diagrams down to the main sequence turn-off
- Spectra of red giant branch stars
- High dispersion spectra of the upper red giant branch
To make any definitive statement about the stellar populations of
galaxies, it is vital to sample galaxies of all Hubble types. In
order to investigate elliptical galaxies satisfactorily, it is necessary to
carry out the above observations to at least the distance of the Virgo or
Fornax clusters. This constraint imposes magnitude limits of:
spectroscopy (R ~ 5000)||(infrared OK?)|
spectroscopy (R ~ 30000)||(infrared OK?)|
In order to resolve stars at small radii in Local Group galaxies (to
test for spatial variations in their properties), we require
diffraction-limited imaging from a 70-metre ELT.
With diffraction-limited imaging to mV ~ 36, we will
be able to study intermediate-mass stars (clump and AGB) out to the
distance of the Coma cluster. Calibrating such data using
observations of nearby galaxies (from 8-metre telescopes), it will be
possible to obtain definitive ages and metallicities for galaxies in
this rich environment.
Similar image quality will also resolve high-mass stars out to a
redshift of z ~ 1, providing a direct probe of the variation in
star formation rate and initial mass function with radius. Issues of
confusion, with many O/B stars forming in each star formation region,
may compromise such studies, even with a 70-metre telescope. However,
we can certainly resolve typical integrated properties of star
formation regions (e.g. their luminosities) as a function of redshift.
When and How Were Galaxies Assembled?
With diffraction-limited spectra, it will be possible to measure the
kinematics of individual stars in nearby galaxies. These data will
also provide the >1000 objects per system that one requires in order
to model both the global distribution function of the galaxy and the
form of the gravitational potential that contains it. Each of these
factors provides important evidence for any study of galaxy evolution:
the distribution of stellar orbits reflects the process by which the
baryonic component formed, while the potential provides information on
the arrangement of dark matter in the system.
Any substructure in the phase-space distribution of stars, such as
moving groups, star streams, etc, will also be apparent in these data.
This fine structure provides "archeological" access to more recent
events in these systems' evolution such as minor mergers.
Where confusion becomes an issue (in the centres of nearby systems and
the outer parts of more distant galaxies), it will still be possible
to "kinematically resolve" individual stars: high dispersion spectra
will resolve the absorption lines of separate stars with different
Doppler shifts, allowing their kinematics to be disentangled.
Spatially-resolved galaxy kinematics will be observable to z ~
1, in both emission lines (gas) or absorption lines (stars). Studies
of the changes in the Tully-Fisher relation and fundamental plane with
redshift will provide direct observations of galaxy evolution, and the
evolving relationship between star formation (luminosity) and dark
matter (mass inferred from kinematics).
Such spectral studies will also provide spatially-resolved chemical
abundances out to this redshift, measuring the star formation history
of galaxies, their recycling and feedback efficiencies, etc.
Out to a redshift z ~ 1, investigation of the galaxy luminosity
function down to dwarf spheroidal luminosities (mV ~
32) will show directly whether these systems form the basic "building
blocks" from which galaxies are constructed. The variation in luminosity
function with redshift offers a direct picture of the hierarchical
merging of systems. Spectra of these building blocks will reveal
their masses and chemical properties.
How Do Stars Interact with Other Galaxy Components?
Central Black Holes
Diffraction-limited imaging and spectroscopy will resolve the
environments of the central black holes in nearby galaxies, allowing
us to study their impact on star formation and the mass flows in these
Studies of the central regions of dwarf galaxies will allow us to
explore the bottom end of the central black hole mass function: is
there a cut-off, or do the central black holes just continue to scale
down in mass as one might extrapolate from larger galaxies?
Black hole demographics can also be studies as a function of
environment, to see whether their masses and environments differ
between clusters and the field. To resolve the
dynamically-interesting scale of ~1 pc around the central black holes
in the nearest cluster environment (Fornax/Virgo) requires a 70-metre
Dynamical studies of dwarf ellipticals in different environments (field,
cluster) will address such basic issues as whether the smallest
galaxies are embedded in dark halos.
Microlensing studies with an Extremely Large Telescope will produce
spatially-resolved events, and even allow direct observation of the
lensing object (in the Local Group). At larger distances (out to
Virgo/Fornax), the "pixel lensing" technique will offer a direct probe
of any dark matter component consisting of massive objects.
Big Bang Nucleosynthesis
At present, the low primordial deuterium abundance derived from quasar
absorption spectra is inconsistent with Li7 abundances seen
in Galactic halo stars. To see if this inconsistency arises from a
localized enhancement in Li7, we need to determine its
abundance in external galaxies using observations of low-mass stars
(where the element will not have been heavily processed).
What Are the Implications for an ELT?
In order to carry out most of the above science programme, we would
place the following design constraints on an Extremely Large Telescope:
- 70-100 metre telescope strongly favoured over 50-metre instrument
- Optical capability important for much science
- Adaptive optics vital for much science (high dispersion
spectroscopy with an image slicer is the best bet in natural seeing)
- Stable point spread function vital for photometry of the requisite
- Large field of view not critical for most projects in this area
- Telescope can be sited in either hemisphere