P. D. Hurley

A dust-obscured massive maximum-starburst galaxy at a redshift of 6.34

Massive present-day early-type (elliptical and lenticular) galaxies probably gained the bulk of their stellar mass and heavy elements through intense, dust-enshrouded starbursts—that is, increased rates of star formation—in the most massive dark-matter haloes at early epochs. However, it remains unknown how soon after the Big Bang massive starburst progenitors exist. The measured redshift (z) distribution of dusty, massive starbursts has long been suspected to be biased low in z owing to selection effects, as confirmed by recent findings of systems with redshifts as high as ∼5 (refs 2–4). Here we report the identification of a massive starburst galaxy at z = 6.34 through a submillimetre colour-selection technique. We unambiguously determined the redshift from a suite of molecular and atomic fine-structure cooling lines. These measurements reveal a hundred billion solar masses of highly excited, chemically evolved interstellar medium in this galaxy, which constitutes at least 40 per cent of the baryonic mass. A ‘maximum starburst’ converts the gas into stars at a rate more than 2,000 times that of the Milky Way, a rate among the highest observed at any epoch. Despite the overall downturn in cosmic star formation towards the highest redshifts, it seems that environments mature enough to form the most massive, intense starbursts existed at least as early as 880 million years after the Big Bang.

Far-infrared Fine-Structure Line Diagnostics of Ultraluminous Infrared Galaxies

We present Herschel observations of 6 fine-structure lines in 25 ultraluminous infrared galaxies at z < 0.27. The lines, [O III]52 μm, [N III]57 μm, [O I]63 μm, [N II]122 μm, [O I]145 μm, and [C II]158 μm, are mostly single Gaussians with widths <600 km s^(–1) and luminosities of 10^7-10^9 L_☉. There are deficits in the [O I]63/L_(IR), [N II]/L_(IR), [O I]145/L_(IR), and [C II]/L_(IR) ratios compared to lower luminosity systems. The majority of the line deficits are consistent with dustier H II regions, but part of the [C II] deficit may arise from an additional mechanism, plausibly charged dust grains. This is consistent with some of the [C II] originating from photodissociation regions or the interstellar medium (ISM). We derive relations between far-IR line luminosities and both the IR luminosity and star formation rate. We find that [N II] and both [O I] lines are good tracers of the IR luminosity and star formation rate. In contrast, [C II] is a poor tracer of the IR luminosity and star formation rate, and does not improve as a tracer of either quantity if the [C II] deficit is accounted for. The continuum luminosity densities also correlate with the IR luminosity and star formation rate. We derive ranges for the gas density and ultraviolet radiation intensity of 10^1 < n < 10^(2.5) and 10^(2.2) < G_0 < 10^(3.6), respectively. These ranges depend on optical type, the importance of star formation, and merger stage. We do not find relationships between far-IR line properties and several other parameters: active galactic nucleus (AGN) activity, merger stage, mid-IR excitation, and SMBH mass. We conclude that these far-IR lines arise from gas heated by starlight, and that they are not strongly influenced by AGN activity.

Gaussian process classification of Alzheimer's disease and mild cognitive impairment from resting-state fMRI

Multivariate pattern analysis and statistical machine learning techniques are attracting increasing interest from the neuroimaging community. Researchers and clinicians are also increasingly interested in the study of functional-connectivity patterns of brains at rest and how these relations might change in conditions like Alzheimers disease or clinical depression. In this study we investigate the efficacy of a specific multivariate statistical machine learning technique to perform patient stratification from functional-connectivity patterns of brains at rest. Whilst the majority of previous approaches to this problem have employed support vector machines (SVMs) we investigate the performance of Bayesian Gaussian process logistic regression (GP-LR) models with linear and non-linear covariance functions. GP-LR models can be interpreted as a Bayesian probabilistic analogue to kernel SVM classifiers. However, GP-LR methods confer a number of benefits over kernel SVMs. Whilst SVMs only return a binary class label prediction, GP-LR, being a probabilistic model, provides a principled estimate of the probability of class membership. Class probability estimates are a measure of the confidence the model has in its predictions, such a confidence score may be extremely useful in the clinical setting. Additionally, if miss-classification costs are not symmetric, thresholds can be set to achieve either strong specificity or sensitivity scores. Since GP-LR models are Bayesian, computationally expensive cross-validation hyper-parameter grid-search methods can be avoided. We apply these methods to a sample of 77 subjects; 27 with a diagnosis of probable AD, 50 with a diagnosis of a-MCI and a control sample of 39. All subjects underwent a MRI examination at 3T to obtain a 7minute and 20second resting state scan. Our results support the hypothesis that GP-LR models can be effective at performing patient stratification: the implemented model achieves 75% accuracy disambiguating healthy subjects from subjects with amnesic mild cognitive impairment and 97% accuracy disambiguating amnesic mild cognitive impairment subjects from those with Alzheimers disease, accuracies are estimated using a held-out test set. Both results are significant at the 1% level.

Mid- to far-infrared properties of star-forming galaxies and active galactic nuclei

We study the mid- to far-IR properties of a 24 μm-selected flux-limited sample (S_(24)> 5 mJy) of 154 intermediate redshift (⟨ z ⟩ ~ 0.15), infrared luminous galaxies, drawn from the 5 Milli-Jansky Unbiased Spitzer Extragalactic Survey. By combining existing mid-IR spectroscopy and new Herschel SPIRE submm photometry from the Herschel Multi-tiered Extragalactic Survey, we derived robust total infrared luminosity (L_(IR)) and dust mass (M_(dust)) estimates and infered the relative contribution of the AGN to the infrared energy budget of the sources. We found that the total (8−1000 μm) infrared emission of galaxies with weak 6.2 μm PAH emission (EW_(6.2) ≤ 0.2 μm) is dominated by AGN activity, while for galaxies with EW_(6.2)> 0.2 μm more than 50% of the L_(IR) arises from star formation. We also found that for galaxies detected in the 250–500 μm Herschel bands an AGN has a statistically insignificant effect on the temperature of the cold dust and the far-IR colours of the host galaxy, which are primarily shaped by star formation activity. For star-forming galaxies we reveal an anti-correlation between the L_(IR)-to-rest-frame 8 μm luminosity ratio, IR8 ≡ L_(IR)/L_8  and the strength of PAH features. We found that this anti-correlation is primarily driven by variations in the PAHs emission, and not by variations in the 5−15 μm mid-IR continuum emission. Using the [Ne iii]/[Ne ii] line ratio as a tracer of the hardness of the radiation field, we confirm that galaxies with harder radiation fields tend to exhibit weaker PAH features, and found that they have higher IR8 values and higher dust-mass-weighted luminosities (L_(IR)/M_(dust)), the latter being a proxy for the dust temperature (T_d). We argue that these trends originate either from variations in the environment of the star-forming regions or are caused by variations in the age of the starburst. Finally, we provide scaling relations that will allow estimating L_(IR), based on single-band observations with the mid-infrared instrument, on board the upcoming James Webb Space Telescope.

Herschel-SPIRE Fourier transform spectroscopy of the nearby spiral galaxy IC 342

We present observations of the nearby spiral galaxy IC342 with the Herschel Spectral and Photometric Imaging Receiver (SPIRE) Fourier Transform Spectrometer. The spectral range afforded by SPIRE, 196-671 m, allows us to access a number of 12 CO lines from J=4‐3 to J=13‐12 with the highest J transitions observed for the first time. In addition we present measurements of 13 CO, [CI] and [NII]. We use a radiative transfer code coupled with Bayesian likelihood analysis to model and constrain the temperature, density and column density of the gas. We find two 12 CO components, one at 35 K and one at 400 K with CO column densities of 6.3 10 17 cm 2 and 0.4 10 17 cm 2 and CO gas masses of 1.26 10 7 M and 0.15 10 7 M , for the cold and warm components, respectively. The inclusion of the high-J 12 CO line observations, indicate the existence of a much warmer gas component ( 400 K) confirming earlier findings from H2 rotational line analysis from ISO and Spitzer. The mass of the warm gas is 10% of the cold gas, but it likely dominates the CO luminosity. In addition, we detect strong emission from [NII] 205 m and the 3 P1! 3 P0 and 3 P2! 3 P1 [CI] lines at 370 and 608 m, respectively. The measured 12 CO line ratios can be explained by Photon-dominated region (PDR) models although additional heating by e.g. cosmic rays cannot be excluded. The measured [CI] line ratio together with the derived [C] column density of 2.1 10 17 cm 2 and the fact that [CI] is weaker than CO emission in IC342 suggests that [CI] likely arises in a thin layer on the outside of the CO emitting molecular clouds consistent with PDRs playing an important role.

Star formation rates in luminous quasars at 2 <z< 3

We investigate the relation between star formation rates (M_s) and AGN properties in optically selected type 1 quasars at 2 < z < 3 using data from Herschel and the SDSS. We find that M_s remains approximately constant with redshift, at 300 ± 100 M⊙ yr^(−1). Conversely, M_s increases with AGN luminosity, up to a maximum of ∼ 600 M⊙ yr^(−1), and with C IV FWHM. In context with previous results, this is consistent with a relation between M_s and black hole accretion rate (M_(bh)) existing in only parts of the z−M_s–M_(bh) plane, dependent on the free gas fraction, the trigger for activity, and the processes that may quench star formation. The relations between M_s and both AGN luminosity and C IV FWHM are consistent with star formation rates in quasars scaling with black hole mass, though we cannot rule out a separate relation with black hole accretion rate. Star formation rates are observed to decline with increasing C IV equivalent width. This decline can be partially explained via the Baldwin effect, but may have an additional contribution from one or more of three factors; M_i is not a linear tracer of L_(2500), the Baldwin effect changes form at high AGN luminosities, and high C IV EW values signpost a change in the relation between M_s and M_(bh). Finally, there is no strong relation between M_s and Eddington ratio, or the asymmetry of the C IV line. The former suggests that star formation rates do not scale with how efficiently the black hole is accreting, while the latter is consistent with C IV asymmetries arising from orientation effects.

Dust properties of Lyman-break galaxies at z ~ 3

Context. Since the mid-1990s, the sample of Lyman-break galaxies (LBGs) has been growing thanks to the increasing sensitivities in the optical and in near-infrared telescopes for objects at z \textgreater 2.5. However, the dust properties of the LBGs are poorly known because the samples are small and/or biased against far-infrared (far-IR) or submillimeter (submm) observations. Aims. This work explores from a statistical point of view the far-IR and submm properties of a large sample of LBGs at z similar to 3 that cannot be individually detected from current far-IR observations. Methods. We select a sample of 22, 000 LBGs at 2.5 \textless z \textless 3.5 in the COSMOS field using the dropout technique. The large number of galaxies included in the sample allows us to split it into several bins as a function of UV luminosity (L-FUV), UV continuum slope (beta(UV)), and stellar mass (M-*) to better sample their variety. We stack in PACS (100 and 160 mu m) images from PACS Evolution Probe survey (PEP), SPIRE (250, 350 and 500 mu m) images from the Herschel Multi-tied Extragalactic Survey (HerMES) programs, and AzTEC (1.1 mm) images from the Atacama Submillimeter Telescope Experiment (ASTE). Our stacking procedure corrects the biases induced by galaxy clustering and incompleteness of our input catalogue in dense regions. Results. We obtain the full infrared spectral energy distributions (SED) of subsamples of LBGs and derive the mean IR luminosity as a function of L-FUV, beta(UV), and M-*. The average IRX (or dust attenuation) is roughly constant over the L-FUV range, with a mean of 7.9 (1.8 mag). However, it is correlated with beta(UV), A(FUV) = (3.15 +/- 0.12) + (1.47 +/- 0.14) beta(UV), and stellar mass, log (IRX) = (0.84 +/- 0.11) log (M-*/10(10.35)) + 1.17 +/- 0.05. We investigate using a statistically controlled stacking analysis as a function of (M-*, beta(UV)), the dispersion of the IRX-beta(UV) and IRX-M-* plane. On the one hand, the dust attenuation shows a departure of up to 2.8 mag above the mean IRX-beta(UV) relation when log(M-* [M-circle dot]) increases from 9.75 to 11.5 in the same beta(UV) bin. This strongly suggests that M-* plays an important role in shaping the IRX-beta(UV) plane. On the other hand, the IRX-M-* plane is less dispersed for variation in the beta(UV). However, the dust attenuation shows a departure of up to 1.3 mag above the mean IRX-M-* relation, when beta(UV) increases from 1.7 to 0.5 in the same M-* bin. The low stellar mass LBGs (log(M-* [M-circle dot]) \textless 10.5) and red beta(UV) (beta(UV) \textgreater -0.7), 15% of the total sample, present a high dust attenuation than the mean IRX-M-*, but they are still in agreement with the mean IRX-beta(UV) relation. We suggest that we have to combine both the IRX-beta(UV) and IRX-M-* relations to obtain the best estimation of the dust attenuation from the UV and NIR properties of the galaxies (L-FUV, beta(UV), M-*). Our results enable us to study the average relation between star formation rate (SFR) and stellar mass, and we show that our LBG sample lies on the main sequence of star formation at z similar to 3. we demonstrate that the SFR is underestimate for LBGs with high stellar mass, but it give a good estimation for LBGs with lower stellar mass when we calculate the SFR by correcting the L-FUV using the IRX-beta(UV) relation.

Galaxy and mass assembly (GAMA): Colour- and luminosity-dependent clustering from calibrated photometric redshifts

We measure the two-point angular correlation function of a sample of 4289 223 galaxies with r < 19.4 mag from the Sloan Digital Sky Survey (SDSS) as a function of photometric redshift, absolute magnitude and colour down to Mr - 5 log h = -14 mag. Photometric redshifts are estimated from ugriz model magnitudes and two Petrosian radii using the artificial neural network package ANNz, taking advantage of the Galaxy And Mass Assembly (GAMA) spectroscopic sample as our training set. These photometric redshifts are then used to determine absolute magnitudes and colours. For all our samples, we estimate the underlying redshift and absolute magnitude distributions using Monte Carlo resampling. These redshift distributions are used in Limbers equation to obtain spatial correlation function parameters from power-law fits to the angular correlation function. We confirm an increase in clustering strength for sub-L* red galaxies compared with ˜L* red galaxies at small scales in all redshift bins, whereas for the blue population the correlation length is almost independent of luminosity for ˜L* galaxies and fainter. A linear relation between relative bias and log luminosity is found to hold down to luminosities L ˜ 0.03L*. We find that the redshift dependence of the bias of the L* population can be described by the passive evolution model of Tegmark & Peebles. A visual inspection of a random sample from our r < 19.4 sample of SDSS galaxies reveals that about 10 per cent are spurious, with a higher contamination rate towards very faint absolute magnitudes due to over-deblended nearby galaxies. We correct for this contamination in our clustering analysis.

HELP: XID+, The Probabilistic De-blender for Herschel SPIRE maps

We have developed a new prior-based source extraction tool, xid+, to carry out photometry in the Herschel SPIRE (Spectral and Photometric Imaging Receiver) maps at the positions of known sources. xid+ is developed using a probabilistic Bayesian framework that provides a natural framework in which to include prior information, and uses the Bayesian inference tool Stan to obtain the full posterior probability distribution on flux estimates. In this paper, we discuss the details of xid+ and demonstrate the basic capabilities and performance by running it on simulated SPIRE maps resembling the COSMOS field, and comparing to the current prior-based source extraction tool desphot. Not only we show that xid+ performs better on metrics such as flux accuracy and flux uncertainty accuracy, but we also illustrate how obtaining the posterior probability distribution can help overcome some of the issues inherent with maximum-likelihood-based source extraction routines. We run xid+ on the COSMOS SPIRE maps from Herschel Multi-Tiered Extragalactic Survey using a 24-μm catalogue as a positional prior, and a uniform flux prior ranging from 0.01 to 1000 mJy. We show the marginalized SPIRE colour–colour plot and marginalized contribution to the cosmic infrared background at the SPIRE wavelengths. xid+ is a core tool arising from the Herschel Extragalactic Legacy Project (HELP) and we discuss how additional work within HELP providing prior information on fluxes can and will be utilized. The software is available at https://github.com/H-E-L-P/XID_plus. We also provide the data product for COSMOS. We believe this is the first time that the full posterior probability of galaxy photometry has been provided as a data product.

Herschel observations and a model for IRAS 08572+3915: a candidate for the most luminous infrared galaxy in the local (z < 0.2) Universe

We present Herschel photometry and spectroscopy, carried out as part of the Herschel ultraluminous infrared galaxy (ULIRG) survey, and a model for the infrared to submillimetre emission of the ULIRG IRAS 08572+3915. This source shows one of the deepest known silicate absorption features and no polycyclic aromatic hydrocarbon 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 per cent of the total luminosity of 1.1 × 10 13 L� , which is about a factor of 5 higher than previous estimates. The large correction of the luminosity is due to the anisotropy of the emission of the best-fitting torus. Similar corrections may be necessary for other local and high-z analogues. This correction implies that IRAS 08572+3915 at a redshift of 0.058 35 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 [C II] to IR luminosity (log L[C II]/LIR < −3.8) and a [O I]63µ mt o [CII]158µm line ratio of about 1 that supports the model presented in this Letter.