---

Index - ... - The Radio Follow-up - The NIR Follow-up - The FIR Follow-up - ...

---

Full infrared coverage of an 8x8 square degree field is not currently technically feasible and will have to await the availability of wide-field IR detectors. In order to optimize the scientific return from our multi-wavelength approach, we will thus adopt the strategy for near-infrared (NIR) follow-up described below.

7.1 - Distant clusters of galaxies

One of the goals of the XMM-LSS survey (in the main area, with 10-ks exposures) will be to investigate the existence of X-ray bright (i.e. massive) clusters of galaxies at redshifts 1 < z < 2. These clusters will be readily identified as such if their X-ray emission is extended together with an absence of conspicuous galaxy overdensity in the optical bands. However, in the case of low X-ray luminosity objects, the discrimination relying on the extension criterion will be difficult. Proper identification will therefore require a combination of ground-based imaging techniques, prior to obtaining spectroscopic redshifts. High z clusters are expected to contain a population of early-type galaxies that are best detected at NIR wavelengths. The best method for identifying these clusters will thus be to obtain deep high-resolution multicolor NIR imaging combined with BVRI; this may also yield some preliminary redshift estimates. Moreover, NIR information will provide luminosity functions little affected by absorption and at wavelengths were k-corrections are weakly dependent on galaxy type. Finally, the NIR observations, combined with optical and UV information from the Optical Monitor, will provide a unique opportunity to investigate the evolutionary properties of the galaxy population in high-redshift X-ray clusters.

We expect some 2-3 10² clusters above z = 1 to be detected by XMM (Fig. 2; the exact number is a sensitive function of the unknown LSS evolution and cosmology - and will provide indeed, strong constraints on the cosmology ). The NIR exposure times required for successful identifications depend mainly on the luminosity limit of the faint galaxies that we wish to detect (relative to an L* galaxy) and on the redshift of the cluster. We propose to use ISAAC/VLT to carry out an initial photometric survey - on extended X-ray sources having no optical counterpart - with the following limiting magnitudes for galaxies:

Js = 22.4,

H = 21.5,

Ks = 20.6

(JsAB = 23.3,

HAB = 22.9,

KsAB = 22.4)1

These sensitivities are sufficient to detect a cluster sequence at z = 1 and enough galaxies at z = 2 to confirm the existence of a cluster. A single cluster can be imaged in ~ 8 min in the Js band, 11 min in the H band and 12 min in Ks for a 5 detection, at the above given magnitude limits. Allowing for overheads, we expect to expend a total of about 1 hour per field. A number of order 10 VLT nights are then needed to complete observations of the sample of about 10² targets.

Image reduction and source detection will be performed by dedicated TERAPIX/NIRMOS/EIS pipelines developed by the consortium.

7.2 - Homogeneous sampling of the X-ray population

The seventeen overlapping 20-ks fields which constitute the sub-area of our Guaranteed Time Survey (Fig.1) will be imaged using the VLT/NIRMOS in the future. The NIRMOS field of view consisting of four 6'x8' quadrants, the 20'x26' central area of each XMM pointing will be covered by a 2x2 raster. The sensitivities and corresponding exposure times per raster position will be:

Sensitivity Exposure time* J 23 7 min H 22.5 19 min

(*) Exposure times are estimated via the SOFI/NTT calculator (i.e. 1.5h and 3h in the J and H bands respectively) for a point source and a S/N of 5. We assume a gain of NIRMOS relative to SOFI of a factor of 5.5 due to mirror size and a factor of 2 due to better seeing.

implying total integration times four times longer at the center of the raster.
The observing time will amount to ~30 hours for a total covered area of ~2 deg².

This NIR survey will provide a unique view of the infrared properties of the X-ray source population, and especially of X-ray-hard, heavily-obscured, SWIRE sources and objects which are undetectable in the optical band. From ROSAT counts, we expect some 200 such objects within 2 deg² at our sensitivity. This population consists mainly of strongly absorbed Seyfert galaxies at various redshifts and of clusters of galaxies beyond z > 1 (Lehmann et al 2001). A follow-up of the reddest pointlike hard X-ray sources will then be performed using the VLT/ISAAC, consisting of deep K-band imaging and spectroscopy in order to characterize the properties of this extreme population.

This survey will probe the 1 < z < 2 redshift range to a much better completeness than any of the present or forthcoming cluster survey as it is X-ray flux limited over a large sky area. The combination of the X-ray, UV, optical, IR and spectroscopic data will provide unique insights into the physics of the deep potential wells of the universe and their environment properties at presently unexplored stages.

¹ For a completeness limit 2-2.5 mag above the 1-sigma detection limit, this corresponds to 5-10 sigma detections. For a cluster of passively-evolving galaxies at z = 1, we will reach 2.7 mag in J and 3.3 mag in H below L*. At z = 2 the corresponding values are 0.7 mag and 1.5 mag.