M. Magliocchetti

The evolution of the dust and gas content in galaxies

We use deep Herschel observations taken with both PACS and SPIRE imaging cameras to estimate the dust mass of a sample of galaxies extracted from the GOODS-S, GOODS-N and the COSMOS fields. We divide the redshift–stellar mass (Mstar)–star formation rate (SFR) parameter space into small bins and investigate average properties over this grid. In the first part of the work we investigate the scaling relations between dust mass, stellar mass and SFR out to zxa0=xa02.5. No clear evolution of the dust mass with redshift is observed at a given SFR and stellar mass. We find a tight correlation between the SFR and the dust mass, which, under reasonableassumptions, is likely a consequence of the Schmidt-Kennicutt (S-K) relation. The previously observed correlation between the stellar content and the dust content flattens or sometimes disappears when considering galaxies with the same SFR. Our finding suggests that most of the correlation between dust mass and stellar mass obtained by previous studies is likely a consequence of the correlation between the dust mass and the SFR combined with the main sequence, i.e., the tight relation observed between the stellar mass and the SFR and followed by the majority of star-forming galaxies. We then investigate the gas content as inferred from dust mass measurements. We convert the dust mass into gas mass by assuming that the dust-to-gas ratio scales linearly with the gas metallicity (as supported by many observations). For normal star-forming galaxies (on the main sequence) the inferred relation between the SFR and the gas mass (integrated S-K relation) broadly agrees with the results of previous studies based on CO measurements, despite the completely different approaches. We observe that all galaxies in the sample follow, within uncertainties, the same S-K relation. However, when investigated in redshift intervals, the S-K relation shows a moderate, but significant redshift evolution. The bulk of the galaxy population at zxa0~xa02 converts gas into stars with an efficiency (star formation efficiency, SFE = SFR/Mgas, equal to the inverse of the depletion time) about 5 times higher than at zxa0~xa00. However, it is not clear what fraction of such variation of the SFE is due to an intrinsic redshift evolution and what fraction is simply a consequence of high-z galaxies having, on average, higher SFR, combined with the super-linear slope of the S-K relation (while other studies find a linear slope). We confirm that the gas fraction (fgasxa0=xa0Mgas/(Mgasxa0+xa0Mstar)) decreases with stellar mass and increases with the SFR. We observe no evolution with redshift once Mstarand SFR are fixed. We explain these trends by introducing a universal relation between gas fraction, stellar mass and SFR that does not evolve with redshift, at least out to zxa0~xa02.5. Galaxies move across this relation as their gas content evolves across the cosmic epochs. We use the 3D fundamental fgas–Mstar–SFR relation, along with the evolution of the main sequence with redshift, to estimate the evolution of the gas fraction in the average population of galaxies as a function of redshift and as a function of stellar mass: we find that Mstarxa0xa0≳xa01011xa0xa0M⊙ galaxies show the strongest evolution at zxa0≳xa01.3 and a flatter trend at lower redshift, while fgas decreases more regularly over the entire redshift range probed in Mstar ≲xa01011xa0xa0M⊙ galaxies, in agreement with a downsizing scenario.

The PEP survey: clustering of infrared-selected galaxies and structure formation at z ∼ 2 in GOODS-South

This paper presents the first direct estimate of the 3D clustering properties of far-infrared sources up to z∼ 3. This has been possible thanks to the PACS Evolutionary Probe (PEP) survey of the GOODS-South field performed with the PACS instrument on board the Herschel satellite. 550 and 502 sources were detected respectively in the 100- and 160-μm channels down to fluxes xa0mJy and xa0mJy, cuts that ensure >80 per cent completeness of the two catalogues. More than 65 per cent of these sources have an (either photometric or spectroscopic) redshift determination from the MUSIC catalogue; this percentage rises to ∼95 per cent in the inner portion of GOODS-South which is covered by data at other wavelengths. An analysis of the deprojected two-point correlation function w(θ) over the whole redshift range spanned by the data reports for the (comoving) correlation length, r0∼ 6.3 and ∼6.7xa0Mpc, respectively at 100 and 160 μm, corresponding to dark matter halo masses M≳ 1012.4xa0M⊙, in an excellent agreement with previous estimates obtained for mid-IR selected sources in the same field. Objects at z∼ 2 instead seem to be more strongly clustered, with r0∼ 19 and ∼17xa0Mpc in the two considered PACS channels. This dramatic increase of the correlation length between z∼ 1 and ∼2 is connected with the presence, more visible at 100 μm than in the other band, of a wide (at least 4xa0Mpc across in projection), M≳ 1014xa0M⊙, filamentary structure which includes more than 50 per cent of the sources detected at z∼ 2. An investigation of the properties of such sources indicates the possibility of a boosted star-forming activity in those which reside within the overdense environment with respect to more isolated galaxies found in the same redshift range. If confirmed by larger data sets, this result can be explained as due to the combined effect of large reservoirs of gas available at high redshifts in deep potential wells such as those associated with large overdensities and the enhanced rate of encounters between sources favoured by their relative proximity. Lastly, we also present our results on the evolution of the relationship between luminous and dark matter in star-forming galaxies between z∼ 1 and ∼2. We find that the increase in (average) stellar mass in galaxies 〈M*〉 between z∼ 1 and ∼2 is about a factor of 10 lower than that of the dark matter haloes hosting such objects (〈M*〉z∼ 1/〈M*〉z∼ 2∼ 4 × 10−1 versus Mz∼ 1halo/Mhaloz∼ 2∼ 4 × 10−2). When compared with recent results taken from the literature, our findings agree with the evolutionary picture of downsizing whereby massive galaxies at z∼ 2 were more actively forming stars than their z∼ 1 counterparts, while at the same time they contained a lower fraction of their mass in the form of luminous matter.

The PEP survey: infrared properties of radio-selected AGN

By exploiting the VLA-COSMOS and the Herschel-PEP surveys, we investigate the Far Infrared (FIR) properties of radio-selected AGN. To this purpose, from VLA-COSMOS we considered the 1537, F[1.4 GHz]>0.06 mJy sources with a reliable redshift estimate, and sub-divided them into star-forming galaxies and AGN solely on the basis of their radio luminosity. The AGN sample is complete with respect to radio selection at all z<~3.5. 832 radio sources have a counterpart in the PEP catalogue. 175 are AGN. Their redshift distribution closely resembles that of the total radio-selected AGN population, and exhibits two marked peaks at z~0.9 and z~2.5. We find that the probability for a radio-selected AGN to be detected at FIR wavelengths is both a function of radio power and redshift, whereby powerful sources are more likely to be FIR emitters at earlier epochs. This is due to two distinct effects: 1) at all radio luminosities, FIR activity monotonically increases with look-back time and 2) radio activity of AGN origin is increasingly less effective at inhibiting FIR emission. Radio-selected AGN with FIR emission are preferentially located in galaxies which are smaller than those hosting FIR-inactive sources. Furthermore, at all z<~2, there seems to be a preferential (stellar) mass scale M ~[10^{10}-10^{11}] Msun which maximizes the chances for FIR emission. We find such FIR (and MIR) emission to be due to processes indistinguishable from those which power star-forming galaxies. It follows that radio emission in at least 35% of the entire AGN population is the sum of two contributions: AGN accretion and star-forming processes within the host galaxy.

The PEP survey: evidence for intense star-forming activity in the majority of radio-selected AGN at z ≳ 1

In order to investigate the FIR properties of radio-active AGN, we have considered three different fields where both radio and FIR observations are the deepest to-date: GOODS-South, GOODS-North and the Lockman Hole. Out of a total of 92 radio-selected AGN, ~64% are found to have a counterpart in Herschel maps. The percentage is maximum in the GOODS-North (72%) and minimum (~50%) in the Lockman Hole, where FIR observations are shallower. Our study shows that in all cases FIR emission is associated to star-forming activity within the host galaxy. Such an activity can even be extremely intense, with star-forming rates as high as ~10^3-10^4 Msun/yr. AGN activity does not inhibit star formation in the host galaxy, just as on-site star-formation does not seem to affect AGN properties, at least those detected at radio wavelengths and for z>~1. Furthermore, physical properties such as the mass and age distributions of the galaxies hosting a radio-active AGN do not seem to be affected by the presence of an ongoing star-forming event. Given the very high rate of FIR detections, we stress that this refers to the majority of the sample: most radio-active AGN are associated with intense episodes of star-formation. However, the two processes proceed independently within the same galaxy, at all redshifts but in the local universe, where powerful enough radio activity reaches the necessary strength to switch off the on-site star formation. Our data also show that for z>~1 the hosts of radio-selected star-forming galaxies and AGN are indistinguishable from each other both in terms of mass and IR luminosity distributions. The two populations only differentiate in the very local universe, whereby the few AGN which are still FIR-active are found in galaxies with much higher masses and luminosities.

On the radio properties of the highest redshift quasars

We present deep radio observations of the most distant complete quasar sample drawn from the Sloan Digital Sky Survey. Combining our new data with those from literature, we obtain a sample which is ∼100 per cent complete down to S1.4xa0GHz= 60xa0μJy over the redshift range 3.8 ≤z≤ 5. The fraction of radio detections is relatively high (∼43 per cent), similar to what observed locally in bright optical surveys. Even though the combined radio and optical properties of quasars remain overall unchanged from z∼ 5 to the local Universe, there is some evidence for a slight overabundance of radio-loud (RL) sources at the highest redshifts when compared with the lower-z regime. n n n nExploiting the deep radio Very Large Array observations, we present the first attempt to directly derive the radio luminosity function of bright quasars at z≳ 4. The unique depth – both in radio and optical – allows us to thoroughly explore the population of optically bright Fanaroff–Riley type II (FR II) quasars up to z∼ 5 and opens a window on the behaviour of the brightest FR I sources. A close investigation of the space density of RL quasars also suggests a differential evolution, with the more luminous sources showing a less-pronounced cut-off at high z when compared with the less-luminous ones.

The clustering properties of radio-selected AGN and star-forming galaxies up to redshifts z ∼ 3

We present the clustering properties of a complete sample of 968 radio sources detected at 1.4xa0GHz by the Very Large Array (VLA)-COSMOS survey with radio fluxes brighter than 0.15xa0mJy. 92xa0perxa0cent have redshift determinations from the Laigle etxa0al. catalogue. Based on their radio luminosity, these objects have been divided into 644 AGN and 247 star-forming galaxies. By fixing the slope of the autocorrelation function to γxa0=xa02, we find

The Herschel-PEP survey: evidence for downsizing in the hosts of dusty star-forming systems

r_0=11.7^{+1.0}_{-1.1}

Spectroscopic Cosmological Surveys in the Far-IR

xa0Mpc for the clustering length of the whole sample, while