No Arabic abstract
We describe deep radio imaging at 1.4-GHz of the 1.3 square degree Subaru/XMM-Newton Deep Field (SXDF), made with the Very Large Array in B and C configurations. We present a radio map of the entire field, and a catalogue of 505 sources covering 0.8 square degrees to a peak flux density limit of 100 microJy. Robust optical identifications are provided for 90% of the sources, and suggested IDs are presented for all but 14 (of which 7 are optically blank, and 7 are close to bright contaminating objects). We show that the optical properties of the radio sources do not change with flux density, suggesting that AGNs continue to contribute significantly at faint flux densities. We test this assertion by cross-correlating our radio catalogue with the X-ray source catalogue and conclude that radio-quiet AGNs become a significant population at flux densities below 300 microJy, and may dominate the population responsible for the flattening of the radio source counts if a significant fraction of them are Compton-thick.
We present spectroscopic and eleven-band photometric redshifts for galaxies in the 100-uJy Subaru/XMM-Newton Deep Field radio source sample. We find good agreement between our redshift distribution and that predicted by the SKA Simulated Skies project. We find no correlation between K-band magnitude and radio flux, but show that sources with 1.4-GHz flux densities below ~1mJy are fainter in the near-infrared than brighter radio sources at the same redshift, and we discuss the implications of this result for spectroscopically-incomplete samples where the K-z relation has been used to estimate redshifts. We use the infrared--radio correlation to separate our sample into radio-loud and radio-quiet objects and show that only radio-loud hosts have spectral energy distributions consistent with predominantly old stellar populations, although the fraction of objects displaying such properties is a decreasing function of radio luminosity. We calculate the 1.4-GHz radio luminosity function (RLF) in redshift bins to z=4 and find that the space density of radio sources increases with lookback time to z~2, with a more rapid increase for more powerful sources. We demonstrate that radio-loud and radio-quiet sources of the same radio luminosity evolve very differently. Radio-quiet sources display strong evolution to z~2 while radio-loud AGNs below the break in the radio luminosity function evolve more modestly and show hints of a decline in their space density at z>1, with this decline occurring later for lower-luminosity objects. If the radio luminosities of these sources are a function of their black hole spins then slowly-rotating black holes must have a plentiful fuel supply for longer, perhaps because they have yet to encounter the major merger that will spin them up and use the remaining gas in a major burst of star formation.
As a step toward investigating the parsec-scale properties of faint extragalactic radio sources, the Very Long Baseline Array (VLBA) was used at 5.0 GHz to obtain phase-referenced images of 76 sources in the NOAO Bootes field. These 76 sources were selected from the FIRST catalog to have peak flux densities above 10 mJy at 5 arcsec resolution and deconvolved major diameters of less than 3 arcsec at 1.4 GHz. Fifty-seven of these faint radio sources were identified with accretion-powered radio galaxies and quasars brighter than 25.5 mag in the optical I band. On VLA scales at 1.4 GHz, a measure of the compactness of the faint sources (the ratio of the peak flux density from FIRST to the integrated flux density from the NVSS catalog) spans the full range of possibilites arising from source-resolution effects. Thirty of the faint radio sources, or 39 +9/-7%, were detected with the VLBA at 5.0 GHz with peak flux densities above 6 sigma ~2 mJy at 2 mas resolution. The VLBA detections occur through the full range of compactness ratios. The stronger VLBA detections can themselves serve as phase-reference calibrators, boding well for opening up much of the radio sky to VLBA imaging. For the adopted cosmology, the VLBA resolution correponds to 17 pc or finer. Most VLBA detections are unresolved or slightly resolved but one is diffuse and five show either double or core-jet structures; the properties of these latter six are discussed in detail. Three VLBA detections are unidentified and fainter than 25.5 mag in the optical I band; their properties are highlighted because they likely mark optically-obscured active nuclei at high redshift.
The Australia Telescope Hubble Deep Field-South (ATHDFS) survey of the Hubble Deep Field South reaches sensitivities of ~10 miceoJyJy at 1.4, 2.5, 5.2 and 8.7 GHz, making the ATHDFS one of the deepest surveys ever performed with the Australia Telescope Compact Array. Here we present the optical identifications of the ATHDFS radio sources using data from the literature. We find that ~66% of the radio sources have optical counterparts to I = 23.5 mag. Deep HST imaging of the area identifies a further 12% of radio sources. We present new spectroscopic observations for 98 of the radio sources, and supplement these spectroscopic redshifts with photometric ones calculated from 5-band optical imaging. The host galaxy colors and radio-to-optical ratios indicate that low luminosity (or radio quiet) AGN make up a significant proportion of the sub-mJy radio population, a result which is in accordance with a number of other deep radio studies. The radio-to-optical ratios of the bright (S_1.4GHz > 1 mJy) sources is consistent with a bimodal distribution.
The up-turn in Euclidean normalised source counts below 1mJy at 1.4GHz is well established in many deep radio surveys. There are strong reasons, observationally and theoretically, to believe that this up-turn is due to strong evolution of the starforming population up to z=2. However this hypothesis needs further confirmation spectroscopically and the examples in the literature are sparse. Theoretically the up-turn is well modelled by the evolution of the local radio starforming population and is consistent with the up-turn seen in recent mid-infrared source counts at 15um (ISOCAM) and 24um (Spitzer) and the tight correlation of the radio and MIR Luminosities of starforming galaxies.
We carried out new observations of two fields in the northern ring of M31 with XMM-Newton with two exposures of 100 ks each and obtained a complete list of X-ray sources down to a sensitivity limit of ~7 x 10^34 erg s^-1 (0.5 - 2.0 keV). The major objective of the observing programme was the study of the hot phase of the ISM in M31. The analysis of the diffuse emission and the study of the ISM is presented in a separate paper. We analysed the spectral properties of all detected sources using hardness ratios and spectra if the statistics were high enough. We also checked for variability. We cross-correlated the source list with the source catalogue of a new survey of the northern disc of M31 carried out with Chandra and Hubble (Panchromatic Hubble Andromeda Treasury, PHAT) as well as with other existing catalogues. We detected a total of 389 sources, including 43 foreground stars and candidates and 50 background sources. Based on the comparison to the Chandra/PHAT survey, we classify 24 hard X-ray sources as new candidates for X-ray binaries (XRBs). In total, we identified 34 XRBs and candidates and 18 supernova remnants (SNRs) and candidates. Three of the four brightest SNRs show emission mainly below 2 keV, consistent with shocked ISM. The spectra of two of them also require an additional component with a higher temperature. The SNR [SPH11] 1535 has a harder spectrum and might suggest that there is a pulsar-wind nebula inside the SNR. We find five new sources showing clear time variability. We also studied the spectral properties of the transient source SWIFT J004420.1+413702, which shows significant variation in flux over a period of seven months (June 2015 to January 2016) and associated change in absorption. Based on the likely optical counterpart detected in the Chandra/PHAT survey, the source is classified as a low-mass X-ray binary.