1 Introduction

# An ALMA survey of submillimetre galaxies in the Extended Chandra Deep Field South: Spectroscopic redshifts

## Abstract

We present spectroscopic redshifts of S 2 mJy submillimetre galaxies (SMGs) which have been identified from the ALMA follow-up observations of 870-m detected sources in the Extended Chandra Deep Field South (the ALMA-LESS survey). We derive spectroscopic redshifts for 52 SMGs, with a median of  = 2.4  0.1. However, the distribution features a high redshift tail, with  23% of the SMGs at  3. Spectral diagnostics suggest that the SMGs are young starbursts, and the velocity offsets between the nebular emission and UV ISM absorption lines suggest that many are driving winds, with velocity offsets up to 2000 km s. Using the spectroscopic redshifts and the extensive UV-to-radio photometry in this field, we produce optimised spectral energy distributions (SEDs) using Magphys, and use the SEDs to infer a median stellar mass of  = (6  1)  10 M for our SMGs with spectroscopic redshift. By combining these stellar masses with the star-formation rates (measured from the far-infrared SEDs), we show that SMGs (on average) lie a factor  5 above the so-called “main-sequence” at  2. We provide this library of 52 template fits with robust and uniquely well-sampled SEDs available as a resource for future studies of SMGs, and also release the spectroscopic catalog of  2000 (mostly infrared-selected) galaxies targeted as part of the spectroscopic campaign.

galaxies: starburst, submillimetre: galaxies

## 1. Introduction

Submillimeter galaxies (SMGs) with 850 m fluxes of S 1 mJy represent a population of dusty starbursts whose space density peaked  10 Gyr ago. Although they are relatively rare, their far-infrared luminosities (L 2  10 L) imply high star-formation rates ( 300 M yr) and so SMGs appear to contribute at least 20% of the total cosmic star-formation rate density over  = 1–4 (e.g. Chapman et al., 2005; Barger et al., 2012; Casey et al., 2014; Swinbank et al., 2014). If they can maintain their star-formation rates, SMGs also have the potential to consume all their cold gas reservoir within just 100 Myr (e.g. Tacconi et al., 2008; Bothwell et al., 2013), and so double their stellar masses within their short but intense lifetime (e.g. Hainline et al., 2009; Magnelli et al., 2012). Their ability to form up to 10 M of stars within a short period of time makes them candidates of progenitors of  = 1–2 compact quiescent galaxies (Toft et al., 2014; Simpson et al., 2015a; Ikarashi et al., 2015) as well as local massive ellipticals (e.g. Lilly et al., 1999; Genzel et al., 2003; Simpson et al., 2014). These characteristics suggest that bright SMGs represent an essential population for models of galaxy formation and evolution (e.g. Efstathiou & Rowan-Robinson, 2003; Baugh et al., 2005; Swinbank et al., 2008; Narayanan et al., 2009; Davé et al., 2010; Hayward et al., 2011; Lacey et al., 2015).

However, to identify the physical processes that trigger the starbursts, measure the internal dynamics of the cold (molecular) and ionised gas, and infer stellar masses first requires accurate redshifts. To date, the largest such spectroscopic survey of 870 m-selected submillimetre sources was carried out by Chapman et al. (2005) who targeted a sample of 104 radio-identified, SCUBA-detected submillimetre sources spread across seven extragalactic survey fields. Using rest-frame UV spectroscopy with the Low-resolution Imaging Spectrograph (LRIS) on the Keck telescope, they derived spectroscopic redshifts for 73 submillimetre sources with a median redshift of  2.4 for the radio-selected sample (with a maximum redshift in their sample of  = 3.6).

Although the requirement for a radio detection in these previous surveys was a necessary step to identify the most probable galaxy counterpart responsible for the submillimetre emission, the radio wavelengths do not benefit from the same negative K-correction as longer submillimetre wavelengths and indeed, above  3.5, the 1.4 GHz flux of a galaxy with a star-formation rate of  100 M yr falls below  15 Jy and so below the typical sensitivity limit of deep radio surveys. This has the potential to bias the redshift distribution to  3.5, especially if a significant fraction of submillimetre sources do not have multi-wavelength counterparts. Indeed, in single-dish 850 m surveys, up to 50% of all submillimetre sources are undetected at radio wavelengths (e.g. Ivison et al., 2005, 2007; Biggs et al., 2011). Some progress can be made by targeting lensed sources whose multi-wavelength identifications are less ambiguous, and indeed spectroscopic redshifts have been derived for SMGs up to  5 (e.g. Weiß et al., 2013).

Due to the angular resolution and sensitivity of the ALMA interferometer, it has become possible to identify the counterparts of submillimetre sources to  0.3 accuracy without recourse to statistical asssociations at other wavelengths. To identify a sample of SMGs in a well studied field with a well defined selection function, Hodge et al. (2013) undertook an ALMA survey of 122 SMGs found in the Extended Chandra Deep Field South (ECDFS): the “ALESS” survey. This survey followed up 122 of the 126 submillimetre sources originally detected with the LABOCA instrument on the Atacama Pathfinder Experiment 12 metre telescope (APEX); the LABOCA ECDFS Sub-mm Survey (LESS) (Weiß et al., 2009). Each LESS submillimetre source was targeted with ALMA at 870 m (Band 7). The typical FWHM of the ALMA synthesised beam was  1.5 (significantly smaller than the LABOCA 19.2 beam), thus allowing us to directly pinpoint the position of the SMG precisely.

From these data, Karim et al. (2013) (see also Simpson et al., 2015b) showed that statistical identifications (e.g. using radio counterparts) were incorrect in  30% of cases, whilst the single-dish submillimetre sources also suffer from significant “multiplicity”, with  35% of the single-dish sources resolved into multiple SMGs brighter than  1 mJy. This flux limit corresponds approximately to a far-infrared luminosity of  10 L at  2, and so it appears that a large fraction of the single-dish submillimetre sources often contain two (or more) Ultra-Luminous Infrared Galaxies (ULIRGs). Consequently, a new ALESS SMG catalogue was defined comprising 131 SMGs (Hodge et al., 2013).

One of the primary goals of the ALESS survey is to provide an unbiased catalog of SMGs for which we can derive molecular gas masses, as well as measure spatially resolved dynamics of the gas and stars in order to identify the trigering mechanisms that cause the burst of star formation. The first necessary step in this process is to derive the precise spectroscopic redshifts. To this end, we have undertaken a spectroscopic survey of ALMA-identified SMGs using VLT, Keck and Gemini (supplemented by ALMA) and in this paper we describe the UV, optical and near-infrared spectroscopic follow-up. We use the resulting redshifts to investigate the redshift distribution, the environments and typical spectral features of these SMGs. In addition, we use these precise redshifts to better constrain the SED fitting from UV-to-radio wavelengths and provide template SEDs for the ALESS SMG population.

The structure of the paper is as follows. We discuss the observations and the data reduction in § 2, followed by redshift identification and sample properties in § 3. In § 4 we show the ALESS redshift distribution and discuss the spectroscopic completeness. In § 5 we discuss the velocity offsets of various different spectral lines, search for evidence of stellar winds and galaxy-scale outflows and investigate the environments of SMGs and the individual and composite spectral properties. We present our conclusions in § 6. In the Appendix, we give the table of ALESS SMG redshifts and provide information on individual SMGs from the sample.

Unless otherwise stated the quoted errors on the median values within this work are determined through bootstrap analysis and are quoted as the equivalent of 68.3% confidence limits. Throughout the paper we use a CDM cosmology with H = 72 km s Mpc,  = 0.27 and  = 1 -  (Spergel et al., 2003) and a Chabrier initial mass function (IMF; Chabrier 2003). Unless otherwise noted, all magnitudes are on the AB system.

## 2. Observations and reduction

### 2.1. Sample definition

The 870 m LESS survey (Weiß et al., 2009) was undertaken using the LABOCA camera on APEX, covering an area of 0.50.5 degrees centered on the ECDFS. The total exposure time for the survey was 310 hours, reaching a 1- sensitivity of  1.2 mJy beam with a beam of 19.2 FWHM. In total, we identified 126 submillimetre sources above a signal-to-noise of 3.7. Follow-up observations of the LESS sources were carried out with ALMA (the ALMA-LESS, ALESS programme). Details of the ALMA observations are described in Hodge et al. (2013) but in summary, the 120 s observations for each source were taken between October and November 2011 in the Cycle 0 Project #2011.1.00294.S. These submillimetre interferometric identifications confirmed some of the probabilistically determined counterparts (Biggs et al., 2011; Wardlow et al., 2011) but also revealed some mis-identified counterparts and a significant number of new counterparts. Therefore, the ALESS SMG catalogue was formed, comprising a main (hereafter main) catalogue of the 99 of the most reliable ALMA-identified SMGs (i.e. lying within the the primary beam FWHM of the best-quality maps). A supplementary (hereafter supp) catalogue was also defined comprising 32 ALMA-identified SMGs extracted from outside the ALMA primary beam, or in lower quality maps (Hodge et al., 2013). When searching for spectroscopic redshifts, we included both the main and supp sources, and in § 4 we demonstrate that the inclusion of supp sources makes very little quantitative difference to the statistics of the redshift distribution.

To search for spectroscopic redshifts, we initiated an observing campaign using the the FOcal Reducer and low dispersion Spectrograph (FORS2) and VIsible MultiObject Spectrograph (VIMOS) on VLT, but to supplement these observations, and in particular to increase the wavelength coverage and probability of determining redshifts, we also obtained observations with XSHOOTER on VLT, the Gemini Near-Infrared Spectrograph (GNIRS) and the Multi-Object Spectrometer for Infra-Red Exploration (MOSFIRE) on the Keck i telescope, all of which cover the near-infrared. As part of a spectroscopic campaign targeting Herschel-selected galaxies in the ECDFS, ALESS SMGs were included on DEep Imaging Multi-Object Spectrograph (DEIMOS) slit masks on Keck ii (e.g. Casey et al., 2012). These observations probe a similar wavelength range to FORS2 targeting some of the ALMA-identified SMGs that could not be targeted with VLT (due to slit collisions). In total, we observed 109 out of the 131 ALESS SMGs in the combined main and supp samples. In many cases we have ALESS SMGs with spectra from five different spectrographs covering a broad wavelength range and we can cross check the spectroscopic redshifts across all of the instruments. Next, we discuss the various instruments involved in our survey. We note that for all observations described below, flux calibration was carried out using standard stars to calibrate instrumental response.

### 2.2. Vlt Fors2 / Vimos

Our spectroscopic programme aimed to target as many of the ALESS SMGs as possible using a dual approach with FORS2 and VIMOS (for a typical SMG redshift of  1–3, we are sensitive to Ly and UV ISM lines with VIMOS or [Oii] 3727 with FORS2). In total, we observed for 100 hours each with VIMOS and FORS as part of programme 183.A-0666. We used deep exposures on ten (overlapping) VIMOS masks to cover the field, plus deep integrations for sixteen FORS masks (which cover a sub-set of the field but target the regions with the highest density of ALMA SMGs; Fig. 1). All of the FORS observations were carried out in grey time and all of the VIMOS observations carried out in dark time during service mode runs with seeing  0.8 and clear sky conditions (transparency variations below 10%). Our dual-instrument approach allowed us to probe a large wavelength range using VIMOS LR-Blue grism (4000–6700Å ) and FORS2 300I (6000–11000 AA ). When designing the slit masks, the first priority was always given to the SMGs, but we also infilled the masks with other mid- or far-infrared selected galaxies from the FIDEL Spitzer survey (Magnelli et al., 2009), the HerMES and PEP Herschel surveys of this field (Oliver et al., 2012; Lutz et al., 2011),  30 Jy radio sources and Chandra X-ray sources (Lehmer et al., 2005; Luo et al., 2008) or optical/near-infrared colour selected galaxies (see Table 3 and Fig. 15).

In Fig. 1 we show the spectroscopic coverage of the ECDFS from our FORS2 and VIMOS programmes, where the darkest areas demonstrate the areas with the longest total exposure time and the FORS2 pointings are overlaid. In total, we recorded 5221 galaxy spectra, targeting 2454 (unique) galaxies.

#### Fors2

FORS2 covers the the wavelength range  = 3300–11000Å and provides an image scale of 0.25 pix in the standard readout mode (22 binning). FORS2 was used in its multi-object spectroscopy mode with exchangeable masks (MXU). We varied the slit length and orientation for each target in order to observe the maximum number of sources on each mask (Fig. 1), but we consistently used a slit width of 1. We used  40–70 slits per mask and the OG590 order-sorting filter with the 300I grism which results in a wavelength range covering 6000–11000Å. The typical resolution in this configuration is  /  660. We used 16 pointings, although in a small number of cases, we moved slits between exposures if there were multiple sources within  5 which could not be simultaneously observed on a mask. Each mask was observed in blocks of 3  900 s with each exposure nodded up and down the slits by  1.0 to aid sky-subtraction and cosmic-ray removal when the images were combined. Each mask was typically observed six times (with a range of three to nine times depending on the number of SMGs on the mask and their median brightness), resulting in an on-source exposure time 4.5 hrs (with a range of 2.25–6.75 hr).

We reduced the data using the spectroscopic reduction package from Kelson (2003) adapted for use with FORS2 data FORS2 pipeline. The pipeline produces two-dimensional, bias-corrected, flat-fielded, wavelength-calibrated, sky-subtracted images. Individual exposures were combined in two-dimensions by taking a median of the frames and sigma clipping. We then extracted one-dimensional spectra over the full spatial-extent of the continuum/emission lines visible, or in the case where no emission was obvious in the two-dimensional image, we extracted data from the region around the expected source position.

#### Vimos

The VIMOS observations were undertaken in multi-object spectroscopy (MOS) mode. VIMOS consists of four quadrants each of a field-of-view of 7 8 with a detector pixel scale of 0.205 pix. Each observing block comprised 3  1200 s exposures dithering  1.0 along the slit. The exposure time per mask was 3–9 hr, again depending on the number of SMGs on the mask and their average brightness. Slit widths of 1.0 were used, for which the typical resolution is  180 and the dispersion is 5.3Å pix for the LR_blue grism with the OS_blue order sorting filter ( 4000–6700Å). We used 40–160 slits per quadrant, totalling 160–400 slits over the four quadrants. The data were reduced using the standard ESOREX pipeline package for VIMOS. The frames were stacked in two-dimensions before extracting the one-dimensional spectra. In a number of cases, the data suffer from overlapping spectra which results in a second order overlapping the adjacent spectrum (this can be seen in the VIMOS two-dimensional spectrum of ALESS 057.1 in Fig. 2).

### 2.3. Xshooter

To improve the wavelength coverage of our observations, we also obtained XSHOOTER observations of 20 ALESS SMGs. XSHOOTER simultaneously observes from UV to near-infrared wavelengths covering wavelength ranges of 3000–5600Å, 5500–10200Å and 10200–24800Å for the UV (UVB), visible (VIS) and near-infrared (NIR) arms respectively. Targets were prioritised for XSHOOTER follow-up based on their -band magnitudes. Our XSHOOTER observations were taken in visitor mode as part of programme 090.A-0927(A) from 2012 December 7–10 in dark time. We observed each source for  1 hr in generally clear conditions with a typical seeing of  1.0. Our observing strategy was 4  600 s exposures per source, nodding the source up and down the slit. The pixel scales were 0.16, 0.16 and 0.21 pix for the UVB, VIS and NIR arms respectively. The slits were all 11 long and 0.9 wide for the VIS and NIR arms and 1.0 wide for the UVB arm. The typical resolution was  4350, 7450, 5300 for the UVB, VIS and NIR arms respectively. The data reduction was carried out using the standard esorex pipeline package for XSHOOTER.

### 2.4. Mosfire

We also targeted 36 ALESS SMGs with the MOSFIRE spectrograph on Keck i (2012B_H251M, 2013B_U039M, and 2013B_N114M) in - (1.46–1.81 m) and -band (1.93–2.45 m). Observations were taken in clear or photometric conditions with the seeing varying from 0.4–0.9. In all cases we used slits of width 0.7. The pixel scale of MOSFIRE is 0.18 pix and the typical spectral resolution for this slit width is  3270. The total exposure time per mask was 2.2–3.6 ks which was split in to 120 s (-band) and 180 s (-band) exposures, with an ABBA sequence and a 1.5 nod along the slit between exposures. Data reduction was completed with mospy.

### 2.5. Deimos

We targeted 71 of the ALESS SMGs as “mask infill” during a Keck ii DEIMOS spectroscopy programme to measure redshifts for Herschel / SPIRE sources (programme 2012B_H251). The data were taken on 2012 December 9–10 in clear conditions with seeing between 1–1.3. We used a setup with the 600ZD (600 lines mm) grating with a 7200Å blaze angle and the GG455 blocking filter which resulted in a wavelength range of 4850–9550Å. Slit widths of 0.75 were used and the masks were filled with 40–70 slits per mask. The pixel scale of DEIMOS is 0.1185 pix and the typical resolution was  3000. Individual exposures were 1200 s, and the total integration times were 2–3 hrs. The data were reduced using the DEEP2 DEIMOS data reduction pipeline (Cooper et al., 2012; Newman et al., 2013).

### 2.6. Gnirs

The Gemini Near-Infrared Spectrograph (GNIRS) was used to target eight ALESS SMGs as (programme GN-2012B-Q-90) between 2012 November 10–15 and December 4–23. The targets were selected based on their -band magnitude and whether they had a photometric redshift that was predicted to place strong emission lines in the near-infrared. The instrument was used in cross-dispersing mode (via the SXD prism with 32 lines mm), using the short camera, with slit widths of 0.3, slit lengths of 7 and a pixel scale of 0.15 pix. The wavelength coverage with this setup is 9000–25600Å, typically with  1700. Our observing strategy comprised 200 s exposures and nodding up and down the slit by  1. Each observing block comprised eight coadds of three exposures, resulting in an exposure of  1.3 hr per source. The GNIRS data were reduced using the Gemini iraf package.

### 2.7. Alma

Spectroscopic redshifts for two of our SMGs, ALESS 61.1 and ALESS 65.1 were determined from serendipitous detections of the [Cii]158m line in the ALMA band (Swinbank et al., 2012). Although based on single line identifications, both redshifts have been confirmed by the identification of CO(1–0) emission using ATCA (Huynh et al. 2013; Huynh et al. 2017, submitted).

Once all of the data were collected from the different spectrographs, we collated the spectra for each ALESS SMG. The instruments used to observe each SMG are listed in Table 1.

## 3. Analysis

### 3.1. Redshift identification

To determine redshifts for the sample, the one- and two-dimensional spectra (for all  2000 galaxies) were independently examined by two investigators (AMS and ALRD). Any emission / absorption features that were identified were fit with a Gaussian profiles to determine their central wavelengths. In the FORS2, VIMOS and DEIMOS data the most commonly identified lines were Ly, Civ1548.89,1550.77 Å, Ciii1909 Å, Heii1640 AA and [Oii]3726.03,3728.82 Å. In the near-infrared, we typically detect H, Nii6583 and [Oiii] 4959, 5007 and in a small number of cases, H (see Tables 2 & 3). The optical / near-infrared counterparts of the SMGs are often faint and we detect continuum in only  50% of the 52 SMGs for which we determine a redshift, (compared to  75% for the radio-identified submillimetre sources in Chapman et al. 2005).

The spectra often only contain weak continuum, emission and / or absorption lines, making redshifts difficult to determine robustly. We therefore assign four quality flags to our spectroscopic data:

1. Q = 1 denotes a secure redshift where multiple features were identified from bright emission / absorption lines;

2. Q = 2 denotes a redshift but derived from one or two bright emission (or strong absorption) lines;

3. Q = 3 is a tentative redshift based on one (or sometimes two tentative) emission or absorption lines. In these cases, we often use the photometric redshift as a guide to identify the line. These redshifts are therefore not independent of the photometric redshifts and are thus highlighted in the analysis;

4. Q = 4 is assigned to galaxies with no emission lines or continuum detected and so no redshift could be determined.

Examples of spectra from which Q = 1, 2 & 3 redshifts are determined are shown in Fig. 2. Since the ECDFS has been the focus of extensive spectroscopic campaigns (although focusing mainly on bright optical/UV-selected galaxies) six of our ALMA SMGs have published archival spectroscopic redshifts, and we highlight these in Table 2. 31

The emission / absorption lines we are using to derive redshifts have a range of physical origins within the galaxies. For example, nebular emission lines arise from Hii regions and so are expected to trace the systemic redshift, whereas UV ISM lines can trace outflowing material and so can be offset from the systemic redshift by several 100 km s (e.g. Erb et al., 2006; Steidel et al., 2010). Ly emission, which is often used to derive spectroscopic redshifts, also suffers resonant scattering. As such, to derive redshifts for each galaxy we adopt the following approach:

1. Wherever possible, systemic redshifts are determined using nebular emission lines such as H, [Oii] 3726,3729, [Oiii] 4959,5007 and/or H. If none of these lines are available we use Heii or Ciii] 1909 in emission if they are narrow.

2. If no nebular emission lines are detected, we determine the mean of the redshifts from the UV ISM absorption lines of Cii1334.53, Siiv1393.76 and Siii1526.72, or other strong emission lines such as Nv1240, Mgii2800 and Heii.

3. If Ly is the only detected line then the redshift is determined from a fit to this line, although we caution that the velocity offset from the systemic can be up to  1000 km s. In most of the galaxies where a redshift is determined solely from Ly, the observations were taken with VIMOS using the low-resolution ( 180) grating, precluding any detailed analysis to determine the shape of the emission line and judge the influence of absorption on its observed profile. Similarly, where possible we avoid using Civ1549 for measuring the redshifts, since it can be strongly influenced by winds and frequently exhibits a profile which is a superposition of P-Cygni emission and absorption, nebular emission and interstellar absorption (or AGN activity).

For the ALESS SMGs,  30% of the redshifts are determined from a single line and generally these redshifts are allocated Q = 3 unless strong continuum features (such as breaks across Ly) are also identified, which leads to an unambiguous identification and a higher quality flag. Single line redshifts are typically backed up by either continuum breaks across Ly, the absence of other emission lines that would correspond to a different redshift, line profiles (i.e. asymmetric Ly profile or identifying the doublet of [Oii] 3726,3729Å emission). In seven cases, single line redshifts are based on detections of Ly; in three cases they are determined from H detections in near-infrared spectra and in five cases they are from detections of the [Oii] doublet.

We summarise the main spectroscopic features that we detect in Table 1 and provide detailed information on each of the 109 observed SMGs in Table 2.

In Fig. 3 we compare our precise spectroscopic measurements for the ALESS SMGs to the photometric redshift estimates for these SMGs from Simpson et al. (2014) who determine photometric redshifts for 77 of the ALESS SMGs which have 4–19 band photometry. We flag those sources with spectroscopic redshifts, but poor photometric coverage and we also highlight the spectroscopic Q = 3 redshifts since their spectroscopic identification is often guided by the photometric redshifts. Nevertheless, even if these Q = 3 SMGs are omitted, there is good agreement between the photometric and spectroscopic redshifts with a median  = 0.00  0.02 and a variance of  = 0.1. In four cases, there appear to be significant outliers, with  0.5. In these cases, the large offset between the photometric and spectroscopic redshifts appears to be associated with complex systems or incomplete photometric coverage, and we briefly discuss these here:

1. ALESS 006.1: the photometry of this ALESS SMG appears to contaminated by an adjacent low-redshift (and unassociated) AGN, and in this case it appears that the SMG is lensed. The photometry (and photometric redshift) is dominated by the foreground AGN.

2. ALESS 010.1: the Q = 1 spectroscopic redshift is significantly lower than predicted by the photometry. There is a blue source slightly offset ( 1) from the ALMA position and an IRAC source coincident with the ALMA position. HST imaging (Chen et al., 2015) reveals two galaxies and it is possible that the blue source is a lens, as confirmed by high-resolution,  0.1 ALMA band 7 follow-up observations; (Hodge et al., 2016).

3. ALESS 037.2: the Q = 3 spectroscopic redshift is significantly lower than the  4 predicted by the photometry. However, the spectroscopic redshift is based on two tentative line detections at the correct separation for H and [Sii] (see Fig. 2; [Nii], if present would lie under strong sky lines) and the photometric redshift is poorly constrained and based on detections in six bands and limits in a further six. Furthermore, the spectroscopic line identifications would not correspond to any common emission lines if the photometric redshift is correct.

4. ALESS 101.1: this has a Q = 2 redshift based on a single detection of Ly. It has poor constraints on the photometric redshift with photometric detections in only five bands and no detections below -band. Thus the spectroscopic redshift is significantly more reliable.

For a significant fraction of the ALMA sample targeted in our survey, we were unable to derive a spectroscopic redshift (these are assigned  = 4 in Table 2). To understand whether this is caused by magnitude limits or their redshifts, we first compare the photometric redshifts of the spectroscopic failures to those for the SMGs for which we were able to determine a spectroscopic redshift. The median photometric redshift of spectroscopic failures is  2.4  0.2, compared to  2.4  0.1 for the sources for which we were able to measure a spectroscopic redshift (these estimates use the best-fit photometric redshifts values, but they change by less than the quoted uncertainty if the full photometric redshift probability distributions are used instead). This suggests that the SMG with spectroscopic failures are not at much higher redshifts than those SMGs where we have succeeded in obtaining a redshift. Similarly there does not appear to be any correlation with submillimetre flux: for the 52 SMGs with spectroscopic redshifts, the median 870-m flux is  =  mJy, whereas those 57 SMGs where we could not determine a redshift have a median  =  mJy.

Next, we test the hypothesis that we were unable to measure spectroscopic redshifts for some ALMA SMGs simply due to their faint optical magnitudes. In Fig. 4 we show the distributions of the flux density, -band and 4.5m magnitudes and 1.4 GHz flux density for the 109 (out of 131) ALESS SMGs that were spectroscopically targeted. The median -band magnitude of the ALESS SMGs with spectroscopic redshifts is  24.0  0.2 whereas the median magnitude of those SMGs for which we could not measure a redshift is  1 magnitude fainter, at  25.0  0.4. Turning to longer wavelengths, in the mid-infrared, the median magnitude at 4.5 m is  20.9  0.2 for the ALESS SMGs with spectroscopic redshifts, as compared to a median of  21.7  0.2 for those targeted SMGs for which we could not derive a spectroscopic redshift. Hence, there is evidence that the ALESS SMGs for which we were unable to determine a spectroscopic redshift are marginally fainter in and than those for which we were able to measure a spectroscopic redshift (and also may have slightly redder colours).

In Fig. 5 we plot the redshifts of the ALESS SMGs versus their 4.5 m apparent magnitudes. At the typical redshift of SMGs ( 2.4), the 4.5 m emission provides the most reliable tracer of the underlying stellar mass, since it corresponds to rest-frame  1.6 m (-band). As a guide, to crudely test how the 4.5 m magnitude depend on redshift in our sample, we generate a non-evolving starburst track, based on the composite SED for the ALESS SMGs (shown in Simpson et al. 2014 but updated to contain the spectroscopic redshift information in Fig. 9). This model SED has been normalised to the median apparent 4.5 m magnitude for the spectroscopic and photometric redshift samples at the median redshift of  2.4. The dependence of 4.5 m flux with redshift for our spectroscopic sample is consistent with this track, although with a spread of  2 magnitudes at fixed redshift. However, the data do show a trend of decreasing 4.5 m flux with increasing redshift. Smail et al. (2004) (see also Serjeant et al., 2003) also identified a similarly large spread in -band magnitudes for SMGs.

Hence we see both a spread in the apparent rest-frame near-infrared luminosities within the SMG population, as well as the fainter optical apparent magnitudes (and redder colours) for those SMGs which we failed to obtain redshifts for and marginally higher photometric redshifts compared to those for which spectroscopic redshifts were measured. Each of these trends are weak, but they do suggest several factors may be driving the spectroscopic incompleteness: a range in stellar masses for SMGs at a fixed redshift (a demonstration of the diversity of the SMG population), varying levels of strong dust extinction and fainter apparent optical fluxes for SMGs at higher redshifts (due to the K correction and increasing distance).

In terms of the radio-detected sub-sample, from the entire main+supp ALESS catalogue, 53 / 131 ALESS SMGs are radio-detected, and we have targeted 52 with spectroscopy, measuring redshifts for 34. The median 1.4 GHz flux density of the SMGs with spectroscopic redshifts is  = 63 Jy compared to  = 39 Jy for those without spectroscopic redshifts (Fig. 4). Thus, SMGs for which we were unable to determine a spectroscopic redshift are fainter at radio wavelengths than those for which we measured a spectroscopic redshift.

## 4. Spectroscopic redshift distribution

The spectroscopic redshift distribution of the ALESS SMGs is shown in Fig. 6. In total 52 redshifts have been determined for the ALESS SMGs: 45 main catalogue SMGs and seven supp catalog SMGs. We also overlay the probability density function of the photometric redshift distribution of ALESS SMGs from Simpson et al. (2014), scaled to the same number of sources. The Q = 1 & 2 and Q = 1, 2 & 3 distributions are shown as individual histograms to test the effect of including the Q = 3 redshifts. The full redshift distribution ranges between  = 0.7–5.0, with a significant (but not dominant) tail at  3 for those distributions without a radio-selection.

In Fig. 7 we show the ALESS spectroscopic redshift distribution and compare this with the 1.1-mm selected (U)LIRGs from the recent ALMA surveys of the Hubble Ultra Deep Field (UDF) by ASPECS (Aravena et al., 2016; Walter et al., 2016) and Dunlop et al. (2016) Given the different selection wavelengths, flux limits and sample sizes between the ALESS SMGs and the ALMA / UDF galaxies, we caution against drawing strong conclusions about the differences between these redshift distributions (for a detailed discussion see Béthermin et al., 2015). Nevertheless, we note that all of these distributions peak at  2.0  0.5, with a suggestion that fainter sources may lie at lower redshifts on average.

Before continuing with the analysis, we briefly assess the effect on our sample of including the supp SMGs and those with only Q = 3 redshifts. Karim et al. (2013) demonstrate that up to  30% of the supp sources are likely to be spurious. However, supp sources which have an optical / near-infrared counterpart have a lower liklihood of being spurious sources. The median redshift of the main catalogue SMGs with Q = 1, 2 & 3 redshifts is  2.5  0.1 with an interquartile range of  2.1–3.4, whereas the median redshift of the main+supp catalogue with Q = 1, 2 & 3 redshifts is  2.4  0.1 with an interquartile range of  2.1–3.0. The median redshift of the Q = 1, 2 & 3 SMGs in the supp sample alone is  2.3  0.5. Thus, the median redshifts of these various samples are all consistent. Indeed, a two-sided Kolmogorov-Smirnov (K-S) test between the main and supp samples suggests only a 60% likelihood that they are drawn from different populations. Since the statistics of the samples do not vary strongly with the inclusion of the supp sources, we are therefore confident that including the supp sources in our analyses is unlikely to bias any of our results.

## 5. Discussion

Although the primary aim of this work is to determine the redshifts of unambiguously identified SMGs to support further detailed follow-up (e.g. CO or H dynamics, Huynh et al. e.g. 2013), there is also a wealth of information contained within the spectra themselves concerning the dynamics, chemical composition, and energetics of these SMGs. Furthermore, the redshifts can be used as constraints in SED models (e.g. constraining the star-formation history and so the stellar masses) and to investigate the environments in which these SMG reside.

### 5.1. Spectral diagnostics

#### Stacked spectral properties

Stacked spectra are a useful tool to detect weak features that are not visible in individual spectra and also for determining the average properties of the population. We therefore produce composite spectra over two different wavelength ranges, one covering Ly and UV ISM lines and one around the [Oii]3727 and Balmer break, and we use these to search for evidence of emission/absorption features and continuum breaks. To construct the composites, we first transform each spectrum to the rest-frame using the best redshift in Table 2. Where the sky subtraction leaves significant residuals, the region within  5Å of the sky lines are masked before stacking (and we use the OH line catalogue from Rousselot et al. (2000) to identify the bright sky lines in the near-infrared). We then sum the spectra, inverse weighted by the noise (measured as the standard deviation in the region of continuum over which they have been normalised). In the case of the 1000–2000Å composite (Fig. 8), we normalise the spectra by their median continuum value at  1250Å and in the case of the composite around 3400–4400Å (Fig. 9), we normalise by the median continuum value between 2900–3600Å. We note that when transforming the spectra to the rest-frame, in a number of cases, the UV ISM lines and Ly can be significantly offset in velocity from this systemic redshift (see Fig. 12). In the composite spectrum these spectral features may therefore appear broadened and offset.

We first discuss the composite spectra of the region around Ly, 1000–2000Å, see Fig. 8. We show a composite constructed from just the Q = 1 and 2 spectra which displays strong Ly and a continuum break at  1200Å. The spectrum also shows two Siii absorption lines and apparently offset Siiv absorption, as well as potentially weak Civ absorption and emission and Oi absorption. If the feature identified as Siiv is real, then it and the weaker Civ features, both of which show blueshifts, may be indicative of strong stellar winds. To illustrate the typical strength of the absorption features we also overlay the composite spectrum of  200 Lyman break galaxies (LBGs) from Shapley et al. (2003) (the LBG composite shown here corresponds to the quartile of 200 LBGs from the Shapley et al. (2003) sample that has the closest match in Ly equivalent width to our ALESS sample). We note that due to the different wavelength ranges of the different instruments used and the fact that we de-redshift and stack in the rest-frame, not all the spectra in our stack contribute to the full wavelength range.

We also construct a composite from the Q = 3 spectra and plot this in Fig. 8. The purpose of this is to test the reliability of the redshifts derived for the Q = 3 spectra by searching for weak spectral features which are undetected in the individual spectra, but become visible in the stacked spectrum due to the improved signal to noise. In addition to an emission line identified as Ly (which is frequently the feature used to derive the redshift for these sources), we see only a potential emission feature which would correspond to Ciii] 1909 and no evidence of a break in the continuum across the bluer emission line. If the Ciii] 1909 emission is real, then it may indicate that some of the Q = 3 redshifts are correct.

To search for continuum breaks and absorption lines in the rest-frame optical, and to determine if we can constrain the luminosity weighted age of the stellar populations in SMGs, we also produce a rest-frame composite of the Q = 1 and 2 spectra over the wavelength range of 3400–4400Å (removing the bright X-ray AGN from the sample; Wang et al. 2013) and show this in Fig. 9. We detect strong [Oii], and potentially also H absorption (Fig. 9). In addition, we see in this composite that continuum falls off bluewards of  3800Å. A break in this region could be due to the 4000Å  break, typically observed in older stellar systems, or more likely the Balmer break at  3656Å. The Balmer break arises in stellar populations which are either experiencing on-going star formation over the previous  100 Myr, or in post-starburst stellar populations, 0.3–1 Gyr after the strongest star formation has ended (Shapley, 2011). In the composite, the position discontinuity is more consistent with the Balmer break than a 4000Å break, as the continuum at 3500–3600Å is lower than it is at 3900–4000Å.

To try to place limits on the age of the visible stellar populations within the ALESS SMGs, we use the SED templates from Bruzual & Charlot (2003) to predict the spectra expected from a starburst of 100 Myr duration observed at ages of 10 Myr, 100 Myr and 1 Gyr (post-starburst). We redden the model spectra using the reddening law from Calzetti et al. (2000) adopting the median extinction of A = 2 for the ALESS SMGs, as derived from SED fitting (see § 5.1.2). As Fig. 9 shows, the stellar continuum emission seen in the composite spectrum is most similar to an on-going burst (i.e. undergoing star-formation on 10–100 Myr timescales), as expected for these strongly star-forming galaxies.

As well as stacking the spectra, we can also create a rest-frame broad-band SED for a “typical” SMG (or at least “typical” of the brighter/bluer examples for which redshifts can be measured). Simpson et al. (2014) and Swinbank et al. (2014) discuss the optical / near-infrared and far-infrared / radio photometry of the ALESS SMGs (see also da Cunha et al., 2015). By combining the multi-wavelength photometry with spectroscopic redshifts for the 52 ALESS SMGs, we create composite SEDs from the rest-frame UV to radio wavelengths. First, we transform the photometry to the rest-frame, and then stack the photometry (normalised by rest-frame -band luminosity; see §5.1.2). A running median is then calculated through the data to produce an average SED which we show in Fig. 9. We also overlay a hyper-z fit using a constant star-formation history, which indicates (as expected) a heavily dust reddened spectrum of these SMGs. Our best-fit constant star-formation model shows a slightly bluer continuum than that derived using the photometric redshift sample by Simpson et al. (2014), illustrating a modest bias to bluer restframe UV continuua in those SMGs for which we can measure spectroscopic redshifts for. Nevertheless, our spectroscopic composite SED still display a very red continuum shape and a clear break at  3800Å, as seen in the composite spectrum at this wavelength (Fig. 9).

From the sample, we derive a median extinction of A = 1.9  0.2 and far-infrared luminosity of  =(3.2  0.4)  10 L, both of which are consistent with previous estimates (for the same sample) derived using photometric redshifts (A = 1.7  0.2 and  = (3.5  0.4)  10 L respectively from Simpson et al. (2014)). In addition, magphys also returns estimates of the stellar masses (solving for the star formation histories and ages) and we derive a median stellar mass for our 52 SMGs with spectroscopic redshifts of  = (6  1)  10 M, in agreement with previous estimates for this sample using photometric redshifts and simple assumptions about the star formation histories by Simpson et al. (2014), see also da Cunha et al. (2015). This is also consistent with the stellar masses estimates for the radio-identified submillimetre sources in the Chapman et al. (2005) sample ( 7  10 M; Hainline et al. 2011). In Fig. 11 we plot the ALESS SMGs with spectroscopic redshifts on the stellar mass–star-formation rate plane. For comparison, we overlay the trends proposed for the so-called “main-sequence” of star-forming galaxies at  1, 2 & 3 and compare these to the SMGs in the same redshift slices. From this plot, it is clear that the SMGs in our sample lie (on average) a factor  5 above the so-called “main-sequence” at all three redshifts, with a median specific star-formation rates (sSFR) of sSFR = (6  1)  10 yr (see also e.g.  Magnelli et al., 2012; Simpson et al., 2014).

### 5.2. Velocity offsets between emission / absorption lines

Rest-frame UV optical spectroscopic analysis of high-redshift, star-forming galaxies have shown that redshifts derived from UV ISM absorption lines typically display systematic blue-shifted offsets from the systemic (nebular) redshifts (e.g. Erb et al., 2006; Steidel et al., 2010; Martin et al., 2012), whilst redshifts determined from Ly emission often show a systematic offset redward of the systemic. These velocity offsets are a consequence of large scale outflows (e.g. Pettini et al., 2002; Steidel et al., 2010), where the outflows material between the galaxy and the observer absorbs the UV and scatter Ly photons from the receeding outflow, redshifting them with respect to the neutral medium within the galaxies. For some of the ALESS SMGs we are able to determine nebular, UV ISM and Ly redshifts, allowing us to compare to the results for other star-forming populations.

In Table 2 we summarise the lines detected for each ALESS SMG and the redshift associated with fitting to each line. We show the velocity offsets between the Ly, UV ISM and nebular emission lines in Fig. 12. We also overlay the velocity offsets for the radio-identified counterparts to submillimetre sources studied by Chapman et al. (2005). Although the same trend is seen in the SMGs and LBGs (Ly is redshifted and the UV ISM lines are blueshifted with respect to the systemic redshift), the SMGs display significantly more scatter, with velocity offsets ranging between to  km s for the UV ISM-derived redshifts and between to  km s for the Ly-derived redshifts, as compared to to  km s and to  km s respectively for the LBGs in Steidel et al. (2010). The wide variation in the velocity offsets may be due to a spread in the viewing angle of the winds or the presence of multiple components (Chen et al. 2015 suggest that most SMGs are major mergers and so the spectra may have contributions from merging components), or the diversity of conditions within these SMGs, in particular with regard to the strength of large-scale winds. Since the wind must be accelerated by star formation or AGN activity, in Fig. 12 we plot the velocity offsets between lines as a function of bolometric luminosity (we note that only two SMGs in our sample are X-ray AGN; Wang et al. 2013 and neither of these show Ly and UV ISM lines with extreme offsets from the systemic redshift). Although there is significant scatter, within the ALESS sample the SMGs with lower bolometric luminosity tend to have wind velocities that are lower than those of the highest luminosity sources.

We note that the outliers in Fig. 12 are ALESS 088.5 and ALESS 049.1, with Ly offset from the systemic by  2000 km s. For both ALESS 088.5 and ALESS 049.1 the only line available to determine a nebular / systemic velocity was Heii1640, which, as we described previously can originate from the stellar winds from Wolf-Rayet stars, making it less reliable as a systemic velocity tracer than the typical nebular lines (e.g. H). It is important to note that the nebular lines such as H, [Oiii] and [Oii] may also be influenced by winds, however this is more typically observed as line broadening as opposed to a centroid shifting.

### 5.3. Environments

One of the key benefits from obtaining spectroscopic redshifts for SMGs is the capability they provide to study both the small- and larger-scale environments of these sources. Hence, we next use our spectroscopic redshift sample to search for physical associations between SMGs and between SMGs and other galaxy populations within the field. Various studies have investigated the environments of SMGs and suggested that at least some SMGs reside within overdense environments (e.g. Chapman et al., 2001; Blain et al., 2004; Chapman et al., 2009; Daddi et al., 2009; Capak et al., 2011; Walter et al., 2012; Ivison et al., 2013; Decarli et al., 2014; Smolcic et al., 2016). For example, Blain et al. (2004) (see also Chapman et al., 2009) identified an over-density of six SMGs and two radio galaxies at  1.99 within 1200 km s of each other in the GOODS-N field. Clustering analysis has also suggested that SMGs cluster on scales of 5–10  Mpc, while pair counting suggests SMGs have properties consistent with them evolving into the passive red galaxies at  1, and subsequently the members of rich galaxy groups or clusters at  0 (e.g. Blain et al., 2004; Hickox et al., 2012; Chen et al., 2016; Wilkinson et al., 2016).

A potentially related result was found by Karim et al. (2013), who demonstrated that single dish submillimetre sources suffer significant “multiplicity” (see also Simpson et al., 2015b), with  35% of the single dish sources resolved into multiple SMGs (where an SMG is a far-infrared bright galaxy with a 870 m flux brighter than 1 mJy). Simpson et al. (2015b) also showed that the number density of  2 mJy SMGs in ALMA maps of bright single-dish submillimetre sources is  80 times higher than that derived from blank-field counts. After taking into account the observational biases in their sample, they proposed that an over-abundance of faint SMGs of this magnitude is inconsistent with line-of-sight projections dominating multiplicity in the brightest SMGs, and strongly suggests that a significant proportion of these high-redshift ULIRGs are likely to be physically associated. These SMGs are typically separated by  6 which corresponds to  40–50 kpc if they lie at the same redshift.

With our survey, we can use a simple approach and exploit the spectroscopic redshifts to search for associations and overdensities in the ALESS SMG population. First, we search for physical associations between SMGs in the same ALMA map (i.e. within  18) where the SMGs lie within 2000 km s (although an offset of 2000 km s is larger than the typical velocity dispersion of rich clusters, even at  0, we broaden our search window to account for potential outflow-driven shifts in the spectral features used to derive the redshifts of many of the SMGs (see §5.2). Unfortunately, there are only three ALESS maps in which we were able to determine a reliable spectroscopic redshift for two or more of the SMGs (ALESS 017.1, 017.2; 075.1, 075.2; 088.1, 088.2, 088.5 and 088.11), and in none of these maps do we find any small-scale clustering of SMGs along the line of sight, the range of redshift offsets between these (previously blended) components is  0.06–1.25. Only in ALESS 067 do we have indirect evidence for an interacting pair of SMGs (ALESS 067.1 and ALESS 067.2) based on the morphology of the sources in HST imaging (Chen et al., 2015).

Next, we search for physical associated between SMGs across the whole ECDFS field (i.e. between the ALMA maps). We identify seven pairs of SMGs within 2000 km s of each other, with ALESS 075.2, ALESS 088.5 and ALESS 102.1 also appearing as a triple “association”. These pairs/triples of SMGs have an average offset of  4 Mpc in the plane of the sky (with a range of  2–15 Mpc). On these scales, the pairs (or triples) may lie within the same large-scale structure but are unlikely to lie within the same dark matter halos.

To determine whether these potential “associations” correlate with redshift peaks in other galaxy populations we compare the spectroscopic redshift distribution of the ALESS SMGs with that of the infill targets from our survey, as well as other archival surveys. Most of the spectroscopic redshifts for the other galaxy populations were taken from an updated version of the redshift compilation in Luo et al. (2011) listing  15,000 spectroscopic redshifts for galaxies in the ECDFS with a median redshift of  0.67 and an inter-quartile range of  0.3–1.0,33. From this catalog, we select only secure redshifts and remove duplicates (we also remove cases in which two secure but differing redshifts are given from two different references).

In Fig. 13 we plot the spectroscopic redshift distribution of the ALESS SMGs, together with the field population. In those cases where  2 SMGs lie within 2000 km s, these associations do not often statistically coincide with significant over-densities in the background galaxy population, although the two SMGs at  1.36 are coincident with a slight peak in the radio / MIPS sources at that redshift.

Finally, returning to Fig. 5 we have highlighted there the ten SMGs that are members of pairs (or triples) with spectroscopic redshift offsets between components of  2000 km s. The median apparent magnitude at 4.5 m for these ten SMGs is  20.4 as compared to a median of  21.1 for the 42 ALESS SMGs in the parent spectroscopic sample which are not in identified “associations”. We conclude that there is no evidence in the current sample that the SMGs in “associations” are any brighter (and thus potentially more massive) than those not in “associations”.

## 6. Conclusions

In this work we present the results from a redshift survey of ALMA-identified SMGs. Our main conclusions are:

• The redshift distribution for ALESS SMGs with spectroscopic redshifts is centered at  2.4  0.1, but with a full range of  0.7–5.0 and an interquartile range of  2.1–3.0. This is consistent with the photometric redshift distribution for these sources, and the median is consistent with previous estimates based on the radio-identified counterparts to submillimetre sources (Chapman et al., 2005). However, since we do not rely on a radio selection, our sample is not biased against higher redshift SMGs and indeed, 23% of the ALESS SMGs with spectroscopic redshifts lie at  3.

• We identify velocity offsets up to  3000 km s between the redshifts measured from nebular emission lines (i.e. H, [Oiii], H and [Oii]) and those measured from Ly or UV ISM absorption lines. We conclude that it is likely that the extreme SFRs within the SMGs (typically  300  30 M yr) are driving strong galaxy-scale outflows in many of these systems.

• Since many of our spectra of SMGs are too faint to exhibit any obvious emission or absorption features (continuum is only detected in  50% of the sources), we produce composite spectra over various wavelength ranges to search for weaker features in the “typical” ALESS SMG optical-to-near infrared spectrum. At rest-frame 1000–2000Å we see strong, asymmetric Ly emission and blueshifted Siii and potentially Siiv absorption suggestive of strong stellar winds. Our composite spectrum at rest-frame 3400–4400Å shows a Balmer break, indicative of on-going star formation. Comparing our composite to spectral models we suggest that it is most consistent with a young starburst with an age of  10 Myr.

• We use our precise spectroscopic redshifts to reduce the uncertainties when modelling the SEDs of our SMGs using magphys and find a large spread in the dust attenuation (A 0.5–7 magnitudes) with a median A = 1.9  0.2. We also derive a median stellar mass of  = (6  1)  10 M and by combining with our estimates of their star-formation rates, we show that SMGs lie (on average)  5 times above the so-called “main-sequence” at  1–3. We provide this library of template SEDs for 52 SMGS with precise redshifts and well-sampled photometry as a resource for future studies of SMGs.

This work has highlighted the challenges of measuring spectroscopic redshifts at optical-to-near infrared wavelengths for dusty star-forming galaxies identified by ALMA, and thus demonstrates the importance of alternative methods of measuring redshifts such as mid-infrared spectroscopy (e.g. Valiante et al., 2007) and the increasing importance of blind submillimetre / millimetre spectral searches with ALMA (e.g. Weiß et al., 2013).

Neverthless, we find that the SMG population is a diverse population of dusty galaxies most common at  2.4, with evidence of energetic outflows which are likely to be predominantly driven by star formation, although some may have a contribution from AGN. The main goal of this study was to provide redshifts for subsequent studies such as CO gas or further detailed integral field unit (IFU) follow-up observations. Such studies will allow us to separate out the relative contributions of star formation and AGN, to probe the conditions within the star-forming gas to better understand this extreme and diverse population of galaxies.

## Acknowledgments

We acknowledge the ESO programmes 183.A-0666 and 090.A-0927(A). The ALMA observations were carried out under programme 2011.0.00294.S. ALRD acknowledges an STFC studentship (ST/F007299/1) and an STFC STEP award. AMS gratefully acknowledges an STFC Advanced Fellowship through grant ST/H005234/1, STFC grant ST/L00075X/1 and the Leverhume foundation. IRS acknowledges support from STFC, a Leverhulme Fellowship, the ERC Advanced Investigator programme DUSTYGAL 321334 and a Royal Society/Wolfson Merit Award. WNB acknowledges STScI grant HST-GO-12866.01-A. CMC acknowledges support from a McCue Fellowship at the University of California, Irvineâs Center for Cosmology and the University of Texas at Austinâs College of Natural Science. JLW is supported by a European Union COFUND/Durham Junior Research Fellowship under EU grant agreement number 267209. AK acknowledges support by the Collaborative Research Council 956, sub-project A1, funded by the Deutsche Forschungsgemeinschaft (DFG). ALMA is a partnership of ESO (representing its member states), NSF (USA) and NINS (Japan), together with NRC (Canada) and NSC and ASIAA (Taiwan), in cooperation with the Republic of Chile. The Joint ALMA Observatory is operated by ESO, AUI/NRAO and NAOJ.

## Appendix A ALESS SMGs with literature redshifts

The following sources are ALESS SMGs with previously measured spectroscopic redshifts:

1. ALESS 018.1: is listed as ID 66 in Casey et al. (2011), with a redshift of  2.252 derived from an H detection with the Infrared Spectrometer And Array Camera (ISAAC) on the VLT;

2. ALESS 057.1: is listed as ID 112a in Szokoly et al. (2004) with a redshift of  2.940 derived from detections of Heii, Ovi and Nv with FORS1 / FORS2. It is classed as a QSO with strong high-ionisation emission lines;

3. ALESS 067.1: is listed as ECDFS-45 in Kriek et al. (2008) at  2.122, derived from emission lines in the near-infrared spectrum observed with GNIRS;

4. ALESS 073.1: is listed as GDS J033229.29275619.5 in the Vanzella, E. et al. (2008) compilation of 1019 spectroscopic redshifts for GOODS / CDFS. The redshift of  4.762 was determined via the detection of Ly and Nv using FORS2.

5. ALESS 098.1: is identified as ID J033129 in Casey et al. (2011). The redshift,  1.4982 is derived through a tentative detection of H, however, it is also spectroscopically-identified in the restframe UV in the same paper and therefore it is given a “secure” status. This redshift is, however, in disagreement with our Q = 1 redshift of  1.3735 derived from fitting to an [Oii] line in the FORS2 observations, with a tentative detection of H at the same redshift under a sky line in the XSHOOTER near-infrared spectrum. We use our redshift in the analysis in this work;

6. ALESS 122.1: is listed as radio ID 149 in Bonzini et al. (2012). The redshift of  2.03 is determined from UV ISM absorption features observed with VIMOS.

## Appendix B Notable Individual Sources

Since we have a wealth of spectroscopic data we can utilise the spectra not only for the purpose of determining redshifts but also to search for diagnostic features indicative of AGN activity, star formation, strong stellar winds etc. Here we highlight and discuss some of the most notable, high signal-to-noise spectra.

ALESS 057.1: This SMG hosts a luminous AGN which is detected in X-rays (Wang et al., 2013). The VIMOS spectrum (Fig. 2) exhibits strong, broad, symmetric Ly emission, broad Nv and Civ emission (FWHM  3700 km s) which is significantly blue-shifted ( 1600 km s) with respect to both Heii and Ly (which have velocities that are consistent within measurements errors). The Civ emission line also displays a P-Cygni profile.

ALESS 066.1: This SMG is listed as an X-ray AGN at  1.310 in Wang et al. (2013). However, our observations reveal the optical/near-infrared photometry and X-ray emission are dominated by a foreground QSO at  1.310 but our near-infrared spectroscopy with MOSFIRE identifies an emission line in -band slightly to the north of the QSO. At  = 2.333 m this line corresponds to H at  2.5542. Careful analysis of the ALMA and optical imaging reveals that the SMG is indeed  1 north of the QSO and hence is likely to be lensed by the foreground QSO.

ALESS 073.1: This SMG also hosts a luminous X-ray AGN (Vanzella, E. et al., 2008; Coppin et al., 2009; De Breuck et al., 2014; Wang et al., 2013) and the spectrum (Fig. 2) shows strong, broad Nv with a FWHM  3000 km s as compared to a relatively narrow and weak Ly (FWHM  700 km s).

ALESS 075.1: We have excellent spectroscopic coverage of this SMG and have strong detections of [Oii], [Oiii]4959, 5007, H and H with XSHOOTER. The H detection is narrow with FWHM  160 km s. The [Oiii] emission is not fit well with a single Gaussian as it is an asymmetric line with a red wing, possibly indicating an outflow (e.g. Alexander et al., 2010). Given the high [Oiii] luminosity and the lack of an X-ray detection, this outflow may be accelerated by an obscured AGN (i.e. outflows in high-redshift ULIRGs hosting AGN activity; Harrison et al. 2012).

ALESS 079.2: This SMG has strong detections of H and [Nii] with XSHOOTER. The one- and two-dimensional spectra show structured emission (see Fig. 14). In the one-dimensional spectrum the H and [Nii] lines are truncated at their red end and appear to be more extended towards lower velocities. The flux ratio of [Nii]6583/H is consistent with the ionising radiation arising from Hii regions as opposed to an AGN.

ALESS 087.1: Strong rest-frame UV continuum is detected in this SMG with ISM absorption lines, with reshifts consistent with the Ly emission line. However, the Ly is significantly offset northwards of the continuum in the two-dimensional spectrum. We therefore extract two spectra in Fig. 14 taken from the position of the Ly and the continuum. The Ly profile is marginally asymmetric with a truncated blue edge. The continuum spectrum shows an obvious break and relatively strong Siiv absorption. Unfortunately, there is very poor photometric coverage of this SMG (3.6–8 m only) so we are unable to say whether the offset Ly is due to a close companion or an interaction with another system, or a less-obscured part of a single galaxy.

ALESS 122.1: This SMG has very blue continuum with strong UV ISM absorption lines in both the FORS2 and VIMOS spectra (Fig. 14). There is very strong, broad Civ absorption (FWHM of  7000 km s). The Civ exhibits a strong, narrow component associated with interstellar absorption and a very broad red component associated with stellar winds. The strength of this redshifted component suggests the presence of a large number of very massive stars ( 30 M; Leitherer & Heckman 1995). Models show that Siiv is relatively weak for a continuous star formation history but yields a strong P-Cygni profile for bursty star formation. Detection of a P-Cygni profile for Siiv is therefore a good indicator that the burst duration is short relative to the age. The Siiv absorption feature is unusually broad ( 3000 km s). This is the blueshifted wind absorption. Swinbank et al. (2014) determine L L for this SMG which implies a star-formation rate of SFR  Myr (using Kennicutt 1998) which is higher than typical ALESS SMGs, SFR  Myr (Swinbank et al., 2014). We note that an AGN may also exhibit strong Civ absorption and given the very strong continuum and the large width of the Civ in this SMG, it is plausible that it may be a broad absorption line (BAL) AGN.

## Appendix C Ancillary Redshifts

When designing the slit masks, we in-filled the unused portions masks (not targeting the high-prioroty SMGs) with other candidate high-redshift galaxies, in particular with mid-, far-infrared or radio selected galaxies. Here, we provide the details of the galaxies targeted.

The ID for each galaxy relates to the input catalogue from which a target was selected. These are summarised as:

101–500: Statistically Robust or Tentative candidate LESS SMG multiwavelength counterparts from Biggs et al. (2011) (see also Wardlow et al. 2011) but which were later shown by ALMA observations to be incorrect IDs (Hodge et al., 2013).

500–700: Robust or tentative IDs for LESS sources with signal-to-noise of SNR = 2.7–3.7 in the original LESS map. These IDs for “faint SMGs” are derived using 1.4 GHz radio emission (Biggs et al., 2011) but have not yet been confirmed (or ruled out) by ALMA.

700–1000: Galaxies in the LESS submillimetre error circles which have photometric redshifts that are consistent with the ALESS photometric redshifts (Wardlow et al., 2011).

1000–3000: 24+70m-selected galaxies from the Spitzer FIDEL survey without pre-existing spectroscopic redshifts (Magnelli et al., 2009).

4000–4300: Chandra X-ray sources from the 2 Ms or 4 Ms surveys (e.g. Lehmer et al., 2005; Luo et al., 2008).

5000–6000: Galaxies from the Herschel / SPIRE images which peak at 350m (and which have been identified and deblended using the 24m positions as priors; Roseboom et al. 2010). Individual redshifts for these sources will be published in Oliver et al. (in prep), although we include the redshift distributio in Fig. 15.

6000-9800: Galaxies from the Herschel / SPIRE images which peak at 250m or 350m (and which have been identified and deblended using the 24m positions as priors; Roseboom et al. 2010. Individual redshifts for these sources will be published in Oliver et al. (in prep), although we include the redshift distributio in Fig. 15.

50000–51000: Optically faint radio galaxies (OFRGs) from the JVLA 1.4 GHz survey of this field. These radio sources are typically brighter than 20 Jy at 1.4 GHz but have optical magnitudes fainter than  = 22.

70000–72000: Optically (colour) selected galaxies. These comprise a mix of  2 Lyman emitting galaxies, BM/BX galaxies and Lyman break galaxies at .

80000–89999: Galaxies which were not in any of the other prior catalogs but which could still be placed on the masks.

90000–90200: - or -band drop-out galaxies (i.e. candidate  2.5 or  3.5 galaxies).

Any source that is labelled with a “b” suffix denotes a secondary galaxy that happened to lie on the slit, but is not the primary target.

We also note that the catalogs are not unique (a galaxy could be an ALMA source that is also in the FIDEL 24m catalog, a radio catalog, a BX/BM and also a Chandra X-ray source). In those instances, the object will only appear once in the table, but under the ID from which it was selected for slit placement (i.e. there are no RA / Dec repeats). As in Table 2, the instrument IDs are denoted by F = VLT / FORS2, V = VLT / VIMOS, X = VLT / XSHOOTER, M = Keck / MOSFIRE, D = Keck /, DEIMOS, and G = Gemini / GNIRS. The quality flag (Q) for the spectroscopic redshifts is Q = 1 for secure redshifts; Q = 2 for redshifts measured from only one or two strong lines; Q = 3 for tentative redshifts measured based on one or two very faint features; Q = 4 for those sources which were targeted but no redshift could be determined. The redshift distribution for each of these sub-samples is shown in Fig. 15.

### Footnotes

1. affiliation: Centre for Extragalactic Astronomy, Durham University, Department of Physics, South Road, Durham DH1 3LE, UK
2. affiliation: Centre for Extragalactic Astronomy, Durham University, Department of Physics, South Road, Durham DH1 3LE, UK
3. affiliation: Institute for Computational Cosmology, Durham University, South Road, Durham DH1 3LE, UK
4. affiliation: Email: a.m.swinbank@durham.ac.uk
5. affiliation: Centre for Extragalactic Astronomy, Durham University, Department of Physics, South Road, Durham DH1 3LE, UK
6. affiliation: Institute for Computational Cosmology, Durham University, South Road, Durham DH1 3LE, UK
7. affiliation: Centre for Extragalactic Astronomy, Durham University, Department of Physics, South Road, Durham DH1 3LE, UK
8. affiliation: Department of Astronomy, The University of Texas at Austin, 2515 Speedway Boulevard Stop C1400, Austin, TX 78712, USA
9. affiliation: Department of Physics and Astronomy, University of California, Irvine, Irvine, CA 92697, USA
10. affiliation: Department of Physics and Atmospheric Science, Dalhousie University, Halifax, NS B3H 4R2, Canada
11. affiliation: Research School of Astronomy and Astrophysics, Australian National University, Canberra, ACT 2611, Australia
12. affiliation: Leiden Observatory, Leiden University, P.O. Box 9513, 2300 RA Leiden, The Netherlands 0000-0001-5434-5942
13. affiliation: Max-Planck-Institut für Astronomie, Königstuhl 17, D-69117 Heidelberg, Germany
14. affiliation: Centre for Extragalactic Astronomy, Durham University, Department of Physics, South Road, Durham DH1 3LE, UK
15. affiliation: Centre for Extragalactic Astronomy, Durham University, Department of Physics, South Road, Durham DH1 3LE, UK
16. affiliation: Department of Astronomy and Astrophysics and the Institute for Gravitation and the Cosmos, The Pennsylvania State University
17. affiliation: European Southern Observatory, Karl Schwarzschild Straße 2, 85748, Garching, Germany
18. affiliation: Centre for Astrophysics Research, Science and Technology Research Institute, University of Hertfordshire, College Lane, Hatfield AL10 9AB, UK
19. affiliation: Universität Wien, Institut für Astrophysik, Türkenschanzstraße 17, 1180, Wien, Austria
20. affiliation: National Optical Astronomy Observatory, Tucson, AZ 85719, USA
21. affiliation: Centre for Extragalactic Astronomy, Durham University, Department of Physics, South Road, Durham DH1 3LE, UK
22. affiliation: Department of Physics and Astronomy, Rutgers University, Piscataway, NJ 08854-8019, USA
23. affiliation: European Southern Observatory, Karl Schwarzschild Straße 2, 85748, Garching, Germany
24. affiliation: Max-Planck-Institut für Radioastronomie, Auf dem Hügel 69, D-53121 Bonn, Germany
25. affiliation: Astronomy Department, University of Minnesota, MN 12345, USA
26. affiliation: Max-Planck-Institut für extraterrestrische Physik, Giessenbachstraße, 85748, Garching, Germany
27. affiliation: Max-Planck-Institut für Radioastronomie, Auf dem Hügel 69, D-53121 Bonn, Germany
28. affiliation: Leiden Observatory, Leiden University, P.O. Box 9513, 2300 RA Leiden, The Netherlands 0000-0001-5434-5942
29. affiliation: Max-Planck-Institut für Radioastronomie, Auf dem Hügel 69, D-53121 Bonn, Germany
30. affiliation: Leiden Observatory, Leiden University, P.O. Box 9513, 2300 RA Leiden, The Netherlands 0000-0001-5434-5942
31. Our goal is to provide a quality flag that allows users to gauge the likely success, (or interpret) follow up observations a source. For example, a non-detection of the CO emission in a Q = 1 source should be interpreted as CO faint, whereas a CO non-detection of a Q = 3 source may be due to the faintness of the CO emission, or due to a misidentified/spurious redshift.
32. The template SEDs are available from: http://astro.dur.ac.uk/$∼$ams/zLESS/
33. http://www.eso.org/sci/activities/garching/projects/goods/
MASTERCAT_v3.0.dat which includes redshifts from Cristiani et al. (2000); Croom et al. (2001); Bunker et al. (2003); Dickinson et al. (2004); Stanway et al. (2004a, b); Strolger et al. (2004); Szokoly et al. (2004); van der Wel et al. (2004); Le Fèvre et al. (2005); Doherty et al. (2005); Mignoli et al. (2005); Ravikumar et al. (2007); Vanzella, E. et al. (2008); Popesso et al. (2009); Balestra et al. (2010); Coppin et al. (2010); Silverman et al. (2010); Kurk et al. (2013); and redshifts also taken from Kriek et al. (2008); Boutsia et al. (2009); Taylor et al. (2009); Treister et al. (2009); Wuyts et al. (2009); Casey et al. (2011); Xia et al. (2011); Bonzini et al. (2012); Cooper et al. (2012); Coppin et al. (2012); Iwasawa et al. (2012); Mao et al. (2012); Le Fèvre et al. (2013); Georgantopoulos et al. (2013); De Breuck et al. (2014); Williams et al. (2014) and the 2dF Galaxy Redshift Survey (Colless et al., 2003)

### References

1. Alexander D. M., Brandt W. N., Smail I., Swinbank A. M., Bauer F. E., Blain A. W., Chapman S. C., Coppin et al., 2008, AJ, 135, 1968
2. Alexander D. M., Swinbank A. M., Smail I., McDermid R., Nesvadba N. P. H., 2010, MNRAS, 402, 2211
3. Aravena M., Decarli R., Walter F., Da Cunha E., Bauer F. E., Carilli C. L., Daddi E., Elbaz, D. et al., 2016, ArXiv:1607.06769
4. Balestra I. et al., 2010, A&A, 512, A12
5. Barger A. J., Wang W.-H., Cowie L. L., Owen F. N., Chen C.-C., Williams J. P., 2012, ApJ, 761, 89
6. Baugh C. M., Lacey C. G., Frenk C. S., Granato G. L., Silva L., Bressan A., Benson A. J., Cole S., 2005, MNRAS, 356, 1191
7. Béthermin M., De Breuck C., Sargent M., Daddi E., 2015, A&A, 576, L9
8. Biggs A. D., Ivison R. J., Ibar E., Wardlow J. L., Dannerbauer H., Smail I., Walter F., Weiß, A. et al., 2011, MNRAS, 413, 2314
9. Blain A. W., Chapman S. C., Smail I., Ivison R., 2004, ApJ, 611, 725
10. Bonzini M. et al., 2012, ApJS, 203, 15
11. Bothwell M. S., Smail I., Chapman S. C., Genzel R., Ivison R. J., Tacconi L. J., Alaghband-Zadeh S., Bertoldi, F. et al., 2013, MNRAS, 429, 3047
12. Boutsia K., Leibundgut B., Trevese D., Vagnetti F., 2009, A&A, 497, 81
13. Bruzual G., Charlot S., 2003, MNRAS, 344, 1000
14. Bunker A. J., Stanway E. R., Ellis R. S., McMahon R. G., McCarthy P. J., 2003, MNRAS, 342, L47
15. Calzetti D., Armus L., Bohlin R. C., Kinney A. L., Koornneef J., Storchi-Bergmann T., 2000, ApJ, 533, 682
16. Capak P. L. et al., 2011, Nature, 470, 233
17. Casey C. M., Berta S., Béthermin M., Bock J., Bridge C., Budynkiewicz J., Burgarella D., Chapin, E. et al., 2012, ApJ, 761, 140
18. Casey C. M., Chapman S. C., Smail I., Alaghband-Zadeh S., Bothwell M. S., Swinbank A. M., 2011, MNRAS, 411, 2739
19. Casey C. M., Narayanan D., Cooray A., 2014, ArXiv e-prints
20. Chabrier G., 2003, PASP, 115, 763
21. Chapman S. C., Blain A., Ibata R., Ivison R. J., Smail I., Morrison G., 2009, ApJ, 691, 560
22. Chapman S. C., Blain A. W., Smail I., Ivison R. J., 2005, ApJ, 622, 772
23. Chapman S. C., Richards E. A., Lewis G. F., Wilson G., Barger A. J., 2001, ApJL, 548, L147
24. Chen C.-C., Smail I., Ivison R. J., Arumugam V., Almaini O., Conselice C. J., Geach J. E., Hartley, W. G. e al., 2016, ApJ, 820, 82
25. Chen C.-C., Smail I., Swinbank A. M., Simpson J. M., Ma C.-J., Alexander D. M., Biggs A. D., Brandt W. N. e. a., 2015, ApJ, 799, 194
26. Colless M., Peterson B. A., Jackson C., Peacock J. A., Cole S., Norberg P., Baldry I. K., Baugh, C. M. et al., 2003, ArXiv.0306581
27. Cooper M. C., Newman J. A., Davis M., Finkbeiner D. P., Gerke B. F., 2012, spec2d: DEEP2 DEIMOS Spectral Pipeline. Astrophysics Source Code Library
28. Coppin K. E. K., Chapman S. C., Smail I., Swinbank A. M., Walter F., Wardlow J. L., Weiss A., Alexander, D. M. et al., 2010, MNRAS, 407, L103
29. Coppin K. E. K., Danielson A. L. R., Geach J. E., Hodge J. A., Swinbank A. M., Wardlow J. L., Bertoldi F., Biggs, A. et al., 2012, MNRAS, 427, 520
30. Coppin K. E. K. et al., 2009, MNRAS, 395, 1905
31. Cristiani S. et al., 2000, A&A, 359, 489
32. Croom S. M., Shanks T., Boyle B. J., Smith R. J., Miller L., Loaring N. S., Hoyle F., 2001, MNRAS, 325, 483
33. da Cunha E., Charlot S., Elbaz D., 2008, MNRAS, 388, 1595
34. da Cunha E. et al., 2015, ApJ, 806, 110
35. Daddi E., Dannerbauer H., Krips M., Walter F., Dickinson M., Elbaz D., Morrison G. E., 2009, ApJL, 695, L176
36. Davé R., Finlator K., Oppenheimer B. D., Fardal M., Katz N., Kereš D., Weinberg D. H., 2010, MNRAS, 404, 1355
37. De Breuck C., Williams R. J., Swinbank M., Caselli P., Coppin K., Davis T. A., Maiolino R., Nagao, T. et al., 2014, A&A, 565, A59
38. Decarli R., Walter F., Carilli C., Riechers D., Cox P., Neri R., Aravena M., Bell, E. et al., 2014, ApJ, 782, 78
39. Dickinson M., Stern D., Giavalisco M., Ferguson H. C., Tsvetanov Z., Chornock R., Cristiani S., Dawson, S. et al., 2004, ApJL, 600, L99
40. Doherty M., Bunker A. J., Ellis R. S., McCarthy P. J., 2005, MNRAS, 361, 525
41. Dunlop J. S., McLure R. J., Biggs A. D., Geach J. E., Michalowski M. J., Ivison R. J., Rujopakarn W., van Kampen, E. et al., 2016, ArXiv:1606.00227
42. Efstathiou A., Rowan-Robinson M., 2003, MNRAS, 343, 322
43. Erb D. K., Steidel C. C., Shapley A. E., Pettini M., Reddy N. A., Adelberger K. L., 2006, ApJ, 647, 128
44. Genzel R., Baker A. J., Tacconi L. J., Lutz D., Cox P., Guilloteau S., Omont A., 2003, ApJ, 584, 633
45. Georgantopoulos I., Comastri A., Vignali C., Ranalli P., Rovilos E., Iwasawa K., Gilli R., Cappelluti, N. et al., 2013, A&A, 555, A43
46. Hainline L. J., Blain A. W., Smail I., Alexander D. M., Armus L., Chapman S. C., Ivison R. J., 2011, ApJ, 740, 96
47. Hainline L. J., Blain A. W., Smail I., Frayer D. T., Chapman S. C., Ivison R. J., Alexander D. M., 2009, ApJ, 699, 1610
48. Harrison C. M. et al., 2012, MNRAS, 426, 1073
49. Hayward C. C., Kereš D., Jonsson P., Narayanan D., Cox T. J., Hernquist L., 2011, ApJ, 743, 159
50. Hickox R. C., Wardlow J. L., Smail I., Myers A. D., Alexander D. M., Swinbank A. M., Danielson A. L. R., Stott, J. P. et al., 2012, MNRAS, 421, 284
51. Hodge J. A., Karim A., Smail I., Swinbank A. M., Walter F., Biggs A. D., Ivison R. J., Weiss, A. et al., 2013, ApJ, 768, 91
52. Hodge J. A., Swinbank A. M., Simpson J. M., Smail I., Walter F., Alexander D. M., Bertoldi F., Biggs, A. D. et al., 2016, ArXiv:1609.09649
53. Huynh M. T., Norris R. P., Coppin K. E. K., Emonts B. H. C., Ivison R. J., Seymour N., Smail I., Swinbank, A. M. et al., 2013, MNRAS, 431, L88
54. Ikarashi S., Ivison R. J., Caputi K. I., Aretxaga I., Dunlop J. S., Hatsukade B., Hughes D. H., Iono, D. et al., 2015, ApJ, 810, 133
55. Ivison R. J., Greve T. R., Dunlop J. S., Peacock J. A., Egami E., Smail I., Ibar et al., 2007, MNRAS, 380, 199
56. Ivison R. J., Smail I., Dunlop J. S., Greve T. R., Swinbank A. M., Stevens J. A., Mortier A. M. J., Serjeant et al., 2005, MNRAS, 364, 1025
57. Ivison R. J., Swinbank A. M., Smail I., Harris A. I., Bussmann R. S., Cooray A., Cox P., Fu, H. et al, 2013, ApJ, 772, 137
58. Iwasawa K., Gilli R., Vignali C., Comastri A., Brandt W. N., Ranalli P., Vito F., Cappelluti, N. et al., 2012, A&A, 546, A84
59. Karim A., Swinbank A. M., Hodge J. A., Smail I. R., Walter F., Biggs A. D., Simpson J. M., Danielson, A. L. R. et al., 2013, MNRAS, 432, 2
60. Kelson D. D., 2003, PASP, 115, 688
61. Kennicutt R. C., 1998, ARAA, 36, 189
62. Kriek M., van Dokkum P. G., Franx M., Illingworth G. D., Marchesini D., Quadri R., Rudnick G., Taylor, E. N. et al., 2008, ApJ, 677, 219
63. Kurk J. et al., 2013, A&A, 549, A63
64. Lacey C. G. et al., 2015, ArXiv e-prints
65. Le Fèvre O., Cassata P., Cucciati O., Garilli B., Ilbert O., Le Brun V., Maccagni D., Moreau, C. et al., 2013, A&A, 559, A14
66. Le Fèvre O., Vettolani G., Garilli B., Tresse L., Bottini D., Le Brun V., Maccagni D., Picat, J. P. et al., 2005, A&A, 439, 845
67. Lehmer B. D., Brandt W. N., Alexander D. M., Bauer F. E., Schneider D. P., Tozzi P., Bergeron J., Garmire, G. P. et al., 2005, ApJS, 161, 21
68. Leitherer C., Heckman T. M., 1995, ApJS, 96, 9
69. Lilly S. J., Eales S. A., Gear W. K. P., Hammer F., Le Fèvre O., Crampton D., Bond J. R., Dunne L., 1999, ApJ, 518, 641
70. Luo B., Bauer F. E., Brandt W. N., Alexander D. M., Lehmer B. D., Schneider D. P., Brusa M., Comastri et al. A., 2008, ApJS, 179, 19
71. Luo B. et al., 2011, ApJ, 740, 37
72. Lutz D., Poglitsch A., Altieri B., Andreani P., Aussel H., Berta S., Bongiovanni A., Brisbin D. et al, 2011, A&A, 532, A90
73. Magnelli B., Elbaz D., Chary R. R., Dickinson M., Le Borgne D., Frayer D. T., Willmer C. N. A., 2009, A&A, 496, 57
74. Magnelli B., Lutz D., Santini P., Saintonge A., Berta S., Albrecht M., Altieri B., Andreani, P. et al., 2012, A&A, 539, A155
75. Mao M. Y. et al., 2012, MNRAS, 426, 3334
76. Martin C. L., Shapley A. E., Coil A. L., Kornei K. A., Bundy K., Weiner B. J., Noeske K. G., Schiminovich D., 2012, ApJ, 760, 127
77. Mignoli M. et al., 2005, A&A, 437, 883
78. Narayanan D., Cox T. J., Hayward C. C., Younger J. D., Hernquist L., 2009, MNRAS, 400, 1919
79. Newman J. A., Cooper M. C., Davis M., Faber S. M., Coil A. L., Guhathakurta P., Koo D. C., Phillips, A. C. et al., 2013, ApJS, 208, 5
80. Oliver S. J., Bock J., Altieri B., Amblard A., Arumugam V., Aussel H., Babbedge T., Beelen, A. et al., 2012, MNRAS, 424, 1614
81. Pettini M., Rix S. A., Steidel C. C., Adelberger K. L., Hunt M. P., Shapley A. E., 2002, ApJ, 569, 742
82. Popesso P. et al., 2009, A&A, 494, 443
83. Ravikumar C. D., Puech M., Flores H., Proust D., Hammer F., Lehnert M., Rawat A., Amram, P. et al., 2007, A&A, 465, 1099
84. Roseboom I. G., Oliver S. J., Kunz M., Altieri B., Amblard A., Arumugam V., Auld R., Aussel, H. et al., 2010, MNRAS, 409, 48
85. Rousselot P., Lidman C., Cuby J.-G., Moreels G., Monnet G., 2000, A&A, 354, 1134
86. Serjeant S., Farrah D., Geach J., Takagi T., Verma A., Kaviani A., Fox M., 2003, MNRAS, 346, L51
87. Shapley A. E., 2011, ARAA, 49, 525
88. Shapley A. E., Steidel C. C., Pettini M., Adelberger K. L., 2003, ApJ, 588, 65
89. Silverman J. D., Mainieri V., Salvato M., Hasinger G., Bergeron J., Capak P., Szokoly G., Finoguenov, A. et al., 2010, ApJS, 191, 124
90. Simpson J. M., Smail I., Swinbank A. M., Almaini O., Blain A. W., Bremer M. N., Chapman S. C., Chen, C.-C., 2015a, ApJ, 799, 81
91. Simpson J. M., Smail I., Swinbank A. M., Chapman S. C., Geach J. E., Ivison R. J., Thomson A. P., Aretxaga, I. et al., 2015b, ApJ, 807, 128
92. Simpson J. M., Swinbank A. M., Smail I., Alexander D. M., Brandt W. N., Bertoldi F., de Breuck C., Chapman, S. C., 2014, ApJ, 788, 125
93. Smail I., Chapman S. C., Blain A. W., Ivison R. J., 2004, ApJ, 616, 71
94. Smolcic V., Miettinen O., Tomicic N., Zamorani G., Finoguenov A., Lemaux B. C., Aravena M., Capak, P. et al., 2016, ArXiv e-prints
95. Spergel D. N., Verde L., Peiris H. V., Komatsu E., Nolta M. R., Bennett C. L., Halpern M., Hinshaw, G., et al., 2003, ApJS, 148, 175
96. Stanway E. R., Bunker A. J., McMahon R. G., Ellis R. S., Treu T., McCarthy P. J., 2004a, ApJ, 607, 704
97. Stanway E. R. et al., 2004b, ApJL, 604, L13
98. Steidel C. C., Erb D. K., Shapley A. E., Pettini M., Reddy N., Bogosavljević M., Rudie G. C., Rakic O., 2010, ApJ, 717, 289
99. Strolger L.-G., Riess A. G., Dahlen T., Livio M., Panagia N., Challis P., Tonry J. L., Filippenko, A. V. et al., 2004, ApJ, 613, 200
100. Swinbank A. M., Karim A., Smail I., Hodge J., Walter F., Bertoldi F., Biggs A. D., de Breuck, C. et al., 2012, MNRAS, 427, 1066
101. Swinbank A. M., Lacey C. G., Smail I., Baugh C. M., Frenk C. S., Blain A. W., Chapman S. C., Coppin et al., 2008, MNRAS, 391, 420
102. Swinbank A. M., Simpson J. M., Smail I., Harrison C. M., Hodge J. A., Karim A., Walter F., Alexander, D. M. et al., 2014, MNRAS, 438, 1267
103. Szokoly G. P. et al., 2004, ApJS, 155, 271
104. Tacconi L. J., Genzel R., Smail I., Neri R., Chapman S. C., Ivison R. J., Blain A., Cox, P., et al., 2008, ApJ, 680, 246
105. Taylor E. N., Franx M., van Dokkum P. G., Quadri R. F., Gawiser E., Bell E. F., Barrientos L. F., Blanc, G. A. et a., 2009, ApJS, 183, 295
106. Toft S., Smolčić V., Magnelli B., Karim A., Zirm A., Michalowski M., Capak P., Sheth, K. et al., 2014, ApJ, 782, 68
107. Treister E. et al., 2009, ApJ, 693, 1713
108. Valiante E., Lutz D., Sturm E., Genzel R., Tacconi L. J., Lehnert M. D., Baker A. J., 2007, ApJ, 660, 1060
109. van der Wel A., Franx M., van Dokkum P. G., Rix H.-W., 2004, ApJL, 601, L5
110. Vanzella, E. et al., 2008, A&A, 478, 83
111. Walter F., Decarli R., Aravena M., Carilli C., Bouwens R., da Cunha E., Daddi E., Ivison, R. J. et al., 2016, ArXiv:1607.06768
112. Walter F., Decarli R., Carilli C., Bertoldi F., Cox P., da Cunha E., Daddi E., Dickinson, M. et al., 2012, Nature, 486, 233
113. Wang S. X. et al., 2013, ApJ, 778, 179
114. Wardlow J. L., Smail I., Coppin K. E. K., Alexander D. M., Brandt W. N., Danielson A. L. R., Luo B., Swinbank, A. M. et al., 2011, MNRAS, 415, 1479
115. Weiß A., De Breuck C., Marrone D. P., Vieira J. D., Aguirre J. E., Aird K. A., Aravena M., Ashby, M. L. N. et al., 2013, ApJ, 767, 88
116. Weiß A., Kovács A., Coppin K., Greve T. R., Walter F., Smail I., Dunlop J. S., Knudsen, K. K. et al., 2009, ApJ, 707, 1201
117. Wilkinson A., Almaini O., Chen C.-C., Smail I., Arumugam V., Blain A., Chapin E. L., Chapman, S. C. et al., 2016, ArXiv e-prints
118. Williams R. J., Maiolino R., Santini P., Marconi A., Cresci G., Mannucci F., Lutz D., 2014, MNRAS, 443, 3780
119. Wuyts S., van Dokkum P. G., Franx M., Förster Schreiber N. M., Illingworth G. D., Labbé I., Rudnick G., 2009, ApJ, 706, 885
120. Xia L. et al., 2011, AJ, 141, 64
You are adding the first comment!
How to quickly get a good reply:
• Give credit where it’s due by listing out the positive aspects of a paper before getting into which changes should be made.
• Be specific in your critique, and provide supporting evidence with appropriate references to substantiate general statements.
• Your comment should inspire ideas to flow and help the author improves the paper.

The better we are at sharing our knowledge with each other, the faster we move forward.
The feedback must be of minumum 40 characters