Space and Ground-Based Telescopes Cooperate to Gain Deep Cosmological Insights

Using the ESA XMM-Newton satellite, a team of European and Chilean
astronomers [2] has obtained the world’s deepest “wide-field” X-ray image of
the cosmos to date. This penetrating view, when complemented with
observations by some of the largest and most
efficient ground-based optical telescopes, including the ESO Very Large
Telescope (VLT), has resulted in the discovery of several large clusters of

These early results from an ambitious research programme are extremely
promising and pave the way for a very comprehensive and thorough census of
clusters of galaxies at various epochs. Relying on the foremost astronomical
technology and with an unequalled observational efficiency, this project is
set to provide new insights into the structure and evolution of the distant

The full text of this Press Release, with four photos (ESO PR Photos
19a-d/03) and all related links, is available at:

The universal web

Unlike grains of sand on a beach, matter is not uniformly spread throughout
the Universe. Instead, it is concentrated into galaxies which themselves
congregate into clusters (and even clusters of clusters). These clusters are
“strung” throughout the Universe in a web-like
structure, cf. ESO PR 11/01.

Our Galaxy, the Milky Way, for example, belongs to the so-called Local Group
which also comprises “Messier 31”, the Andromeda Galaxy. The Local Group
contains about 30 galaxies and measures a few million light-years across.
Other clusters are much larger.

The Coma cluster contains thousands of galaxies and measures more than 20
million light-years. Another well known example is the Virgo cluster,
covering no less than 10 degrees on the sky !

Clusters of galaxies are the most massive bound structures in the Universe.
They have masses of the order of one thousand million million times the mass
of our Sun. Their three-dimensional space distribution and number density
change with cosmic time and
provide information about the main cosmological parameters in a unique way.

About one fifth of the optically invisible mass of a cluster is in the form
of a diffuse hot gas in between the galaxies. This gas has a temperature of
the order of several tens of million degrees and a density of the order of
one atom per liter. At such
high temperatures, it produces powerful X-ray emission.

Observing this intergalactic gas and not just the individual galaxies is like
seeing the buildings of a city in daytime, not just the lighted windows at
night. This is why clusters of galaxies are best discovered using X-ray

Using previous X-ray satellites, astronomers have performed limited studies
of the large-scale structure of the nearby Universe. However, they so far
lacked the instruments to extend the search to large volumes of the distant

The XMM-Newton wide-field observations

Marguerite Pierre (CEA Saclay, France), with a European/Chilean team of
astronomers known as the XMM-LSS consortium [2], used the large field-of-view
and the high sensitivity of ESA’s X-ray observatory XMM-Newton to search for
remote clusters of galaxies
and map out their distribution in space. They could see back about 7,000
million years to a cosmological era when the Universe was about half its
present size and age, when clusters of galaxies were more tightly packed.

Tracking down the clusters is a painstaking, multi-step process, requiring
both space and ground-based telescopes. Indeed, from X-ray images with XMM,
it was possible to select several tens of cluster candidate objects,
identified as areas of enhanced X-radiation (cf PR Photo 19b/03).

But having candidates is not enough ! They must be confirmed and further
studied with ground-based telescopes. In tandem with XMM-Newton, Pierre uses
the very-wide-field imager attached to the 4-m Canada-France-Hawaii
Telescope, on Mauna Kea, Hawaii, to take an optical snapshot of
the same region of space. A tailor-made computer programme then combs the
XMM-Newton data looking for concentrations of X-rays that suggest large,
extended structures. These are the clusters and represent only about 10% of
the detected X-ray sources. The others are mostly distant active galaxies.

Back to the Ground

When the programme finds a cluster, it zooms in on that region and converts
the XMM-Newton data into a contour map of X-ray intensity, which is then
superimposed upon the CFHT optical image (PR Photo 19c/03). The astronomers
use this to check if anything is visible within the area of extented X-ray emission.

If something is seen, the work then shifts to one of the world’s prime
optical/infrared telescopes, the European Southern Observatory’s Very Large
Telescope (VLT) at Paranal (Chile). By means of the FORS multi-mode
instruments, the astronomers zoom-in on
the individual galaxies in the field, taking spectral measurements that
reveal their overall characteristics, in particular their redshift and hence,

Cluster galaxies have similar distances and these measurement ultimately
provide, by averaging, the cluster’s distance as well as the velocity
dispersion in the cluster. The FORS instruments are among the most efficient
and versatile for this type of work, taking on the average spectra of 30
galaxies at a time.

The first spectroscopic observations dedicated to the identification and
redshift measurement of the XMM-LSS galaxy clusters took place during three
nights in the fall of 2002.

As of March 2003, there were only 5 known clusters in the literature at such
a large redshift with enough spectroscopically measured redshifts to allow an
estimate of the velocity dispersion. But the VLT allowed obtaining the
dispersion in a distant cluster in 2 hours only, raising great expectations
for future work.

700 spectra…

Marguerite Pierre is extremely content : “Weather and working conditions at
the VLT were optimal. In three nights only, 12 cluster fields were observed,
yielding no less than 700 spectra of galaxies. The overall strategy proved
very successful. The high observing efficiency of the VLT and FORS
support our plan to perform follow-up studies of large numbers of distant
clusters with relatively little observing time. This represents a most
substantial increase in efficiency compared to former searches.”

The present research programme has begun well, clearly demonstrating the
feasibility of this new multi-telescope approach and its very high
efficiency. And Marguerite Pierre and her colleagues are already seeing the
first tantalising results: it seems to confirm that the number of clusters
7,000 million years ago is little different from that of today. This
particular behaviour is predicted by models of the Universe that expand
forever, driving the galaxy clusters further and further apart.

Equally important, this multi-wavelength, multi-telescope approach developed
by the XMM-LSS consortium to locate clusters of galaxies also constitutes a
decisive next step in the fertile synergy between space and ground-based
observatories and is therefore a basic building block of the forthcoming
Virtual Observatory.

More information

This work is based on two papers to be published in the professional
astronomy journal, Astronomy and Astrophysics (The XMM-LSS survey : I.
Scientific motivations, design and first results by Marguerite Pierre et al.,
astro-ph/0305191 and The XMM-LSS survey : II. First high redshift galaxy
clusters: relaxed and collapsing systems by Ivan Valtchanov et al.,


[1]: This a coordinated ESO/ESA release.

[2]: The XMM-LSS consortium is led by the Service d’Astrophysique du CEA
(France) and consists of institutes from the UK, Ireland, Denmark, The
Netherlands, Belgium, France, Italy, Germany, Spain and Chile. The
homepage of the XMM-LSS project can be found at

[3]: In astronomy, the “redshift” denotes the fraction by which the
lines in the spectrum of an object are shifted towards longer
wavelengths. Since the redshift of a cosmological object increases with
distance, the observed redshift of a remote galaxy also provides an
estimate of its distance.


Marguerite Pierre
CEA Saclay, France
Phone: (33) 169 08 34 92