The mid-infrared (MIR) spectra observed with the textit{Spitzer} Infrared Spectrograph (IRS) provide a valuable dataset for untangling the physical processes and conditions within galaxies. This paper presents the first attempt to blindly learn fun
damental spectral components of MIR galaxy spectra, using non-negative matrix factorisation (NMF). NMF is a recently developed multivariate technique shown to be successful in blind source separation problems. Unlike the more popular multivariate analysis technique, principal component analysis, NMF imposes the condition that weights and spectral components are non-negative. This more closely resembles the physical process of emission in the mid-infrared, resulting in physically intuitive components. By applying NMF to galaxy spectra in the Cornell Atlas of Spitzer/IRS sources (CASSIS), we find similar components amongst different NMF sets. These similar components include two for AGN emission and one for star formation. [... ABBREVIATED...] We show an NMF set with seven components can reconstruct the general spectral shape of a wide variety of objects, though struggle to fit the varying strength of emission lines. We also show that the seven components can be used to separate out different types of objects. We model this separation with Gaussian Mixtures modelling and use the result to provide a classification tool. We also show the NMF components can be used to separate out the emission from AGN and star formation regions and define a new star formation/AGN diagnostic which is consistent with all mid-infrared diagnostics already in use but has the advantage that it can be applied to mid-infrared spectra with low signal to noise or with limited spectral range. The 7 NMF components and code for classification are made public on arxiv and are available at: url{https://github.com/pdh21/NMF_software/}
We present Herschel photometry and spectroscopy, carried out as part of the Herschel ULIRG survey (HERUS), and a model for the infrared to submillimetre emission of the ultraluminous infrared galaxy IRAS 08572+3915. This source shows one of the deepe
st known silicate absorption features and no polycyclic aromatic hydrocarbon (PAH) emission. The model suggests that this object is powered by an active galactic nucleus (AGN) with a fairly smooth torus viewed almost edge-on and a very young starburst. According to our model the AGN contributes about 90% of the total luminosity of 1.1 x 10^13 Lo, which is about a factor of five higher than previous estimates. The large correction of the luminosity is due to the anisotropy of the emission of the best fit torus. Similar corrections may be necessary for other local and high-z analogs. This correction implies that IRAS 08572+3915 at a redshift of 0.05835 may be the nearest hyperluminous infrared galaxy and probably the most luminous infrared galaxy in the local (z < 0.2) Universe. IRAS 08572+3915 shows a low ratio of [CII] to IR luminosity (log L_[CII]/L_{IR} < -3.8) and a [OI]63um to [CII]158um line ratio of about 1 that supports the model presented in this letter.
We report on our observations of the 79 and 119um doublet transitions of OH for 24 local (z<0.262) ULIRGs observed with Herschel-PACS as part of the Herschel ULIRG Survey (HERUS). Some OH119 profiles display a clear P-Cygni shape and therefore imply
outflowing OH gas, other profiles are predominantly in absorption or are completely in emission. We find that the relative strength of the OH emission component decreases as the silicate absorption increases. This locates the OH outflows inside the obscured nuclei. The maximum outflow velocities for our sources range from less than 100 to 2000 km/s, with 15/24 (10/24) sources showing OH absorption at velocities exceeding 700 km/s (1000 km/s). Three sources show maximum OH outflow velocities exceeding that of Mrk231. Since outflow velocities above 500-700 km/s are thought to require an active galactic nucleus (AGN) to drive them, about 2/3 of our ULIRG sample may host AGN-driven molecular outflows. This finding is supported by the correlation we find between the maximum OH outflow velocity and the IR-derived bolometric AGN luminosity. No such correlation is found with the IR-derived star formation rate. The highest outflow velocities are found among sources which are still deeply embedded. We speculate that the molecular outflows in these sources may be in an early phase of disrupting the nuclear dust veil before these sources evolve into less obscured AGN. Four of our sources show high-velocity wings in their [C II] fine-structure line profiles implying neutral gas outflow masses of at least 2-4.5 x 10^8 Msun.
We present the Spitzer Extragalactic Representative Volume Survey (SERVS), an 18 square degrees medium-deep survey at 3.6 and 4.5 microns with the post-cryogenic Spitzer Space Telescope to ~2 microJy (AB=23.1) depth of five highly observed astronomic
al fields (ELAIS-N1, ELAIS-S1, Lockman Hole, Chandra Deep Field South and XMM-LSS). SERVS is designed to enable the study of galaxy evolution as a function of environment from z~5 to the present day, and is the first extragalactic survey both large enough and deep enough to put rare objects such as luminous quasars and galaxy clusters at z>1 into their cosmological context. SERVS is designed to overlap with several key surveys at optical, near- through far-infrared, submillimeter and radio wavelengths to provide an unprecedented view of the formation and evolution of massive galaxies. In this paper, we discuss the SERVS survey design, the data processing flow from image reduction and mosaicing to catalogs, as well as coverage of ancillary data from other surveys in the SERVS fields. We also highlight a variety of early science results from the survey.
Cosmic magnification is due to the weak gravitational lensing of sources in the distant Universe by foreground large-scale structure leading to coherent changes in the observed number density of the background sources. Depending on the slope of the b
ackground source number counts, cosmic magnification causes a correlation between the background and foreground galaxies, which is unexpected in the absence of lensing if the two populations are spatially disjoint. Previous attempts using submillimetre (sub-mm) sources have been hampered by small number statistics. The large number of sources detected in the {it Herschel} Multi-tiered Extra-galactic Survey (HerMES) Lockman-SWIRE field enables us to carry out the first robust study of the cross-correlation between sub-mm sources and sources at lower redshifts. Using ancillary data we compile two low-redshift samples from SDSS and SWIRE with <z> ~ 0.2 and 0.4, respectively, and cross-correlate with two sub-mm samples based on flux density and colour criteria, selecting galaxies preferentially at z ~ 2. We detect cross-correlation on angular scales between ~1 and 50 arcmin and find clear evidence that this is primarily due to cosmic magnification. A small, but non-negligible signal from intrinsic clustering is likely to be present due to the tails of the redshift distribution of the sub-mm sources overlapping with those of the foreground samples.
We present mid-infrared spectra of six FeLoBAL QSOs at 1<z<1.8, taken with the Spitzer space telescope. The spectra span a range of shapes, from hot dust dominated AGN with silicate emission at 9.7 microns, to moderately obscured starbursts with stro
ng Polycyclic Aromatic Hydrocarbon (PAH) emission. The spectrum of one object, SDSS 1214-0001, shows the most prominent PAHs yet seen in any QSO at any redshift, implying that the starburst dominates the mid-IR emission with an associated star formation rate of order 2700 solar masses per year. With the caveats that our sample is small and not robustly selected, we combine our mid-IR spectral diagnostics with previous observations to propose that FeLoBAL QSOs are at least largely comprised of systems in which (a) a merger driven starburst is ending, (b) a luminous AGN is in the last stages of burning through its surrounding dust, and (c) which we may be viewing over a restricted line of sight range.
We apply methods from Bayesian inferencing and graph theory to a dataset of 102 mid-infrared spectra, and archival data from the optical to the millimeter, to construct an evolutionary paradigm for z<0.4 infrared-luminous galaxies (ULIRGs). We propos
e that the ULIRG lifecycle consists of three phases. The first phase lasts from the initial encounter until approximately coalescence. It is characterized by homogeneous mid-IR spectral shapes, and IR emission mainly from star formation, with a contribution from an AGN in some cases. At the end of this phase, a ULIRG enters one of two evolutionary paths depending on the dynamics of the merger, the available quantities of gas, and the masses of the black holes in the progenitors. On one branch, the contributions from the starburst and the AGN to the total IR luminosity decline and increase respectively. The IR spectral shapes are heterogeneous, likely due to feedback from AGN-driven winds. Some objects go through a brief QSO phase at the end. On the other branch, the decline of the starburst relative to the AGN is less pronounced, and few or no objects go through a QSO phase. We show that the 11.2 micron PAH feature is a remarkably good diagnostic of evolutionary phase, and identify six ULIRGs that may be archetypes of key stages in this lifecycle.
We present mid-infrared spectra of thirty two high redshift ultraluminous infrared galaxies, selected via the stellar photospheric feature at rest-frame 1.6um, and an observed-frame 24um flux of >500muJy. Nearly all the sample reside in a redshift ra
nge of <z>=1.71+/-0.15, and have rest-frame 1-1000um luminosities of 10^12.9 - 10^13.8 Lsun. Most of the spectra exhibit prominent polycyclic aromatic hydrocarbon emission features, and weak silicate absorption, consistent with a starburst origin for the IR emission. Our selection method appears to be a straightforward and efficient way of finding distant, IR-luminous, star-forming galaxies in narrow redshift ranges. There is however evidence that the mid-IR spectra of our sample differ systematically from those of local ULIRGs; our sample have comparable PAH equivalent widths but weaker apparent silicate absorption, and (possibly) enhanced PAH 6.2um/7.7um and 6.2um/11.2um flux ratios. Furthermore, the composite mid-IR spectrum of our sample is almost identical to that of local starbursts with IR luminosities of 10^10-10^11 Lsun rather than that of local ULIRGs. These differences are consistent with a reduced dust column, which can plausibly be obtained via some combination of (1) star formation that is extended over spatial scales of 1-4Kpc, and (2) star formation in unusually gas-rich regions.
(Abridged) We present R~600, 10-37um spectra of 53 ULIRGs at z<0.32, taken using the IRS on board Spitzer. All of the spectra show fine structure emission lines of Ne, O, S, Si and Ar, as well as molecular Hydrogen lines. Some ULIRGs also show emissi
on lines of Cl, Fe, P, and atomic Hydrogen, and/or absorption features from C_2H_2, HCN, and OH. We employ diagnostics based on the fine-structure lines, as well as the EWs and luminosities of PAH features and the strength of the 9.7um silicate absorption feature (S_sil), to explore the power source behind the infrared emission in ULIRGs. We show that the IR emission from the majority of ULIRGs is powered mostly by star formation, with only ~20% of ULIRGs hosting an AGN with a comparable or greater IR luminosity than the starburst. The detection of the 14.32um [NeV] line in just under half the sample however implies that an AGN contributes significantly to the mid-IR flux in ~42% of ULIRGs. The emission line ratios, luminosities and PAH EWs are consistent with the starbursts and AGN in ULIRGs being more extincted, and for the starbursts more compac
We present measurements of the spatial clustering of ultraluminous infrared galaxies in two redshift intervals, 1.5<z<2.0 and 2<z<3. Both samples cluster strongly, with r_0=14.40+/-1.99 h^-1 Mpc for the 2<z<3 sample, and r_0=9.40+/-2.24 h^-1 Mpc for
the 1.5<z<2.0 sample, making them among the most biased galaxies at these epochs. These clustering amplitudes are consistent with both populations residing in dark matter haloes with masses of ~6x10^13 solar masses. We infer that a minimum dark matter halo mass is an important factor for all forms of luminous, obscured activity in galaxies at z>1. Adopting plausible models for the growth of DM haloes with redshift, then the haloes hosting the 2<z<3 sample will likely host the richest clusters of galaxies at z=0, whereas the haloes hosting the 1.5<z<2.0 sample will likely host poor to rich clusters at z=0.
