The ELAIS/SWIRE Survey

People involved at OAB : Ciliegi, Comastri, Gruppioni, Mignoli, Pozzetti, Pozzi, Zamorani.

ELAIS is a large European project, involving 19 different institutes, initially aimed at studying the nature and evolution of the extragalactic sources detected by the Infrared Space Observatory (ISO) in selected areas of the sky. At present the collaboration has been enlarged and the survey coverage has been extended to other wavelengths, so that now the ELAIS southern area S1 (4 square degrees) is one of the best studied fields due to its extensive multiwavelength coverage.

ELAIS-S1 is one of the main fields of the larger survey carried out with ISO (ELAIS, covering 12 sq. deg. at $15 \mu$m): its size is 4 sq. deg. and the $15 \mu$m catalogue published by Lari et al. (2001) contains 329 extragalactic sources over the flux range $0.5-100$ mJy. The entire area is covered by radio observations obtained with the ATCA (Australian Telescope Compact Array) down to $S_{\rm 1.4-GHz} \simeq 0.2$ mJy (Gruppioni et al. 1999). These radio data have been used to investigate and derive the radio-IR correlation for the first time at faint flux densities and for a sample of that size ($\sim 100$ ISOCAM-radio associations with measured $z$; Gruppioni et al. 2003). Recently, a new very deep radio observation at 1.4 GHz has been obtained with ATCA (PI : Boyle, CoI: Ciliegi, Condon, Lonsdale) in the ELAIS region S1 down to an rms noise of $10 \mu$Jy (see section 2.3.7 on the ATLAS radio survey). S1 is also covered by R-band CCD exposures reaching $R \sim 23$ (obtained at the ESO/Danish 1.5m telescope in collaboration with F. La Franca, I. Matute, C. Lari et al.), where 82% of the IR extragalactic sample has a likely counterpart. From the spectroscopic identifications two main classes are found to dominate the MIR extragalactic population: star-forming galaxies mainly at $z < 0.5$, which account for $\approx 75$% of the sources, and Active Galactic Nuclei (AGN; both type 1 and 2), which account for $\approx 25$% of the sources (La Franca, Gruppioni et al. 2004). Thanks to these unique data-sets, F. Pozzi, C. Gruppioni and G. Zamorani, in collaboration with S. Oliver (Sussex Univ.), I. Matute, F. La Franca, C. Lari, A. Franceschini (Univ. of Padova) and M. Rowan-Robinson (ICSTM), have obtained the first direct determination of the $15 \mu$m luminosity function and its cosmic evolution for galaxies and AGN from the ELAIS survey (Pozzi, et al. 2004; Matute et al. 2005).

The ELAIS-S1 field is one of the targets selected by the Spitzer Legacy Programme SWIRE (PI C. Lonsdale, Caltech) and has been observed in January 2005. The SWIRE project is the largest Survey project performed with Spitzer, covering a total of about 50 sq.deg. (5 of which on the S1 region) at all available wavelengths (from 3 to 200 $\mu$m). Our $15 \mu$m ISOCAM data are important for the calibrations and interpretation of the Spitzer data, since they fill the gap between the short- (IRAC: $3.6-8\mu$m) and long-wavelength (MIPS: $24-160\mu$m) instruments on board of Spitzer. Moreover, as it has been recently shown, the $24/15 \mu$m colour is a powerful tool to constrain the MIR SEDs of distant galaxies (Gruppioni et al. 2005).

In the framework of the SWIRE collaboration, we have obtained deep $B$, $V$ and $R$ images with the WFI at the ESO 2.2m telescope (ESO Large Programme ESIS: PI A. Franceschini) down to about $B=24.7$, $V=24.0$ and $R=24.5$ and in the $I$ and $Z$ band with VIMOS-VLT (Berta et al. 2006). Moreover, about 1 sq. deg. of S1 has been covered by deep $K^{'}$ band exposures with SOFI at the ESO NTT telescope (ESO Large Programme: PI A. Cimatti, INAF-Arcetri Observatory) and in the X-ray band with XMM-Newton (4 pointings of about 100 ksec each have been obtained in the central area of S1: PI F. Fiore, INAF-Rome Observatory) with about 500 sources detected in the $0.5-10$ keV band down to a flux of $2-3 \times 10^{-15}$ cgs (Puccetti et al. 2006). ELAIS-S1 is also one of the targets selected by the GALEX Deep Survey, which has deeply covered in the far- and near-UV the central part of S1.

Finally, 60 hours at the ESO VLT telescope with the VIMOS spectrograph have been allocated in the ELAIS S1 field (PI: F. La Franca) with the aim of obtaining spectroscopic identification for the fainter $15 \mu$m ISO sources, for the new XMM, $K^{'}$ and 24-$\mu$m sources and for the radio sources. About 2500 spectra have been taken and data reduction is still ongoing.

Thanks to this unique multi-wavelength data-set, we have recently undertaken the analysis of the broad band Spectral Energy Distribution (SED) of both mid-IR and X-ray selected samples. In particular, C. Gruppioni and F. Pozzi, in collaboration with M. Polletta of the San Diego University (California) and the SWIRE consortium, are performing a detailed study of the SEDs of the $15 \mu$m sources with spectroscopic identification considered for deriving the MIR LF of galaxies and AGN. These SEDs extend over the whole wavelength range from the far-UV to the far-IR ($0.1-200 \mu$m) and are of fundamental importance to understand the physical processes dominating galaxy evolution in the different bands. In particular, there are indications that at MIR wavelengths an AGN component dominates the emission of a significant fraction of the $15 \mu$m galaxies, but is negligible in the other bands. A similar study is being performed for the XMM-selected sample, by C. Gruppioni and A. Comastri in collaboration with C.  Vignali (Astronomy Dept., Univ. of Bologna). The SED of the X-ray sources, mostly AGN, are being analised and their broad-band properties are compared with the spectral classification (if available) and the X-ray informations. Particular attention is given to those sources showing high X-ray to optical ratios, which show spectra rising steeply from NIR to FIR, often fitted by template SEDs of known prototypical dust-enshrouded AGN at $z > 1-1.5$, in agreement with the indications of them being high-redshift obscured nuclei.