In the paper Improving the identification of high-z Herschel sources with position priors and optical/NIR and FIR/mm photometric redshifts (A&A 2010, 518, L15), we analyzed the Herschel data obtained with both the PACS and SPIRE instruments for the Bullet cluster. Our main goal was to study the star formation in distant galaxies, benefiting from the gravitational lensing created by the cluster to detect and analyze faint sources at high redshift.

In this Letter, included in an Astronomy & Astrophysics Special Issue dedicated to the Herschel Space Telescope, we presented a preliminary analysis of the Science Demonstration Phase (SDP) data taken for the Herschel Lensing Survey Project, and Open Time Key Programme led by Eiichi Egami from the University of Arizona and awarded with nearly 300 hours of Herschel observing time.

The Letter presented the detection by PACS and SPIRE of 2 very interesting sources in the Bullet cluster field: the lensed arc at redshift z=3.24 presented in Mehlert et al. (2001) and the sub-mm galaxy detected by LABOCA and cataloged as source #10 in Johansson et al. (2010). Postage stamps in different photometric bands are shown in the images above. On the left, we show that both the Spitzer Space Telescope and Herschel were able to detect the IR dust emission from the Southern tip of the arc with the MIPS (at 24 μm), PACS (at 160 μm), and SPIRE (at 250 and 350 μm) instruments. The sub-mm galaxy was also detected in several bands from 24 μm to 500 μm, even though the source is very faint in the optical (cf. HST ACS-F850LP image on the right image above).

The work presented in the Astronomy & Astrophysics Herschel Special Issue consisted on using the Spitzer MIPS 24 μm imaging to recover robust fluxes and deblend faint Herschel sources near the confusion noise, i.e., using the positions given in the MIPS 24 μm catalog as priors to dig into the noise of the PACS and SPIRE data and measure FIR fluxes for those sources. We discussed the advantages of this method, and the general properties of the Herschel data: detection levels, source densities, fraction of blended sources, etc... The Letter also illustrated the power of using independent estimates of the photometric redshift based on UV-to-NIR data and IR data alone to validate the individual values (see the photo-z probability distributions in the figures below). The procedure is also useful for achieving more reliable identifications of the IR emitters. For example, for the LABOCA source at z~2.6 (right image above), we calculated photometric redshifts based on UV-to-NIR data for the three close neighbors detected by MIPS within 10" of the central source. The galaxy to the North lies at z~0.3 (a cluster member), that to the south at z~1.9, and the closest companion has a photo-z compatible with the central source (z=2.3±0.3), thus implying that the two sources are a possible interacting pair.

Once we had reliable source catalogs for the Herschel images, we discussed the properties of two interesting galaxies in the field. We were able to estimate with high reliability the total IR luminosity of the arc MIPS06:58:31.1−55:56:39.2, which turned to be a LIRG at z=3.24 (once you take into account the magnification of this source). This estimation is very precise, since it is based on up to 7 photometric data points in the MIR, FIR, and (sub-)mm spectral ranges (using also LABOCA and AzTEC data) covering the peak of the Spectral Energy Distribution (SED), which dominates the integrated IR luminosity. This demonstrates that the use of the gravitational lensing effect allows to detect faint IR sources at high redshift which will not be recovered by the typical Herschel Cosmological Surveys (see, e.g., Elbaz et al. 2010). We also showed that for this source the TIR luminosity obtained with 24 μm alone was a good estimation compared to the value obtained with the FIR and (sub-) data using the Rieke et al. (2009) dust emission templates. In contrast, several papers have shown problems in the estimation of L(TIR) when only using MIPS24 for ULIRGs at high redshift, indicating an evolution on the typical SED (and thus, the dust and/or star formation properties) of this type of galaxies (see, e.g., Papovich et al. 2007, Rigby et al. 2008, Rex et al. 2010, Elbaz et al. 2010, Barro et al. 2010), but probably not for LIRGs.

This SED evolution is also clear in the SED of the LABOCA source MIPS06:58:45.3−55:58:46.5 presented in the Letter Pérez-González et al. (2010), a HyLIRG (assuming no magnification by the cluster) at z~2.6 (average value from independent estimations of the photometric redshift obtained from UV-to-NIR data and MIR-to-mm data) for which there is no template that can reproduce the MIPS24 emission jointly with the FIR and (sub-)mm fluxes. Clearly, there is a higher emission at rest-frame wavelengths around 6-7 μm than that predicted by any template fitting the FIR-to-mm range. This could be solved with a more prominent PAH emission, or with an underlying MIR continuum emitted by relatively large amounts of warm/hot dust (e.g., heated by an obscured AGN).

Previous papers in this webpage

Exploring the evolutionary paths of the most massive galaxies since z~2 (Jun 2009)

The Stellar Mass Assembly of Galaxies from z = 0 to z = 4: Analysis of a Sample Selected in the Rest-Frame Near-Infrared with Spitzer (Jan 2009)

Created on Fri Jul 16 10:00:00 CEST 2010
Back to top