We present near-infrared and optical spectroscopic observations of a sample of 450$\mu$m and 850$\mu$m-selected dusty star-forming galaxies (DSFGs) identified in a 400 arcmin$^2$ area in the COSMOS field. Thirty-one sources of the 102 targets were spectroscopically confirmed at $0.2<z<4$, identified primarily in the near-infrared with Keck MOSFIRE and some in the optical with Keck LRIS and DEIMOS. The low rate of confirmation is attributable both to high rest-frame optical obscuration in our targets and limited sensitivity to certain redshift ranges. The high-quality photometric redshifts available in the COSMOS field allow us to test the robustness of photometric redshifts for DSFGs. We find a subset (11/31$\approx35$%) of DSFGs with inaccurate ($\Delta z/(1+z)>0.2$) or non-existent photometric redshifts; these have very distinct spectral energy distributions from the remaining DSFGs, suggesting a decoupling of highly obscured and unobscured components. We present a composite rest-frame 4300--7300\AA\ spectrum for DSFGs, and find evidence of 200$\pm$30 km s$^{-1}$ gas outflows. Nebular line emission for a sub-sample of our detections indicate that hard ionizing radiation fields are ubiquitous in high-z DSFGs, even more so than typical mass or UV-selected high-z galaxies. We also confirm the extreme level of dust obscuration in DSFGs, measuring very high Balmer decrements, and very high ratios of IR to UV and IR to H$\alpha$ luminosities. This work demonstrates the need to broaden the use of wide bandwidth technology in the millimeter to the spectroscopic confirmations of large samples of high-z DSFGs, as the difficulty in confirming such sources at optical/near-infrared wavelengths is exceedingly challenging given their obscuration.

We present the first results from our on-going Australia Telescope Compact Array survey of CO(1-0) in ALMA-identified submillimetre galaxies in the Extended Chandra Deep Field South. Strong detections of CO(1-0) emission from two submillimetre galaxies, ALESS 122.1 (z = 2.0232) and ALESS 67.1 (z = 2.1230), were obtained. We estimate gas masses of M_gas ~ 1.3 \times 10^{11} M_odot and M_gas ~ 1.0 \times 10^{11} M_\odot for ALESS 122.1 and ALESS 67.1, respectively, adopting alpha_CO = 1.0. Dynamical mass estimates from the kinematics of the CO(1-0) line yields M_dyn (sin i)^2 = 2.1 +- 1.1 \times 10^{11} M_odot and (3.2 +- 0.9) \times 10^{11} M_\odot for ALESS 122.1 and ALESS 67.1, respectively. This is consistent with the total baryonic mass estimates of these two systems. We examine star formation efficiency using the L_FIR versus L'_CO(1-0) relation for samples of local ULIRGs and LIRGs, and more distant star-forming galaxies, with CO(1-0) detections. We find some evidence of a shallower slope for ULIRGs and SMGs compared to less luminous systems, but a larger sample is required for definite conclusions. We determine gas-to-dust ratios of 170 +- 30 and 140 +- 30 for ALESS 122.1 and ALESS 67.1, respectively, showing ALESS 122.1 has an unusually large gas reservoir. By combining the 38.1 GHz continuum detection of ALESS 122.1 with 1.4 and 5.5 GHz data, we estimate that the free-free contribution to radio emission at 38.1 GHz is 34 +- 17 microJy, yielding a star formation rate (1400 +- 700 M_\odot yr^{-1}) consistent with that from the infrared luminosity.

We describe the search for Lyman-break galaxies (LBGs) near the sub-millimeter bright starburst galaxy HFLS3 at $z$$=$6.34 and a study on the environment of this massive galaxy during the end of reionization.We performed two independent selections of LBGs on images obtained with the \textit{Gran Telescopio Canarias} (GTC) and the \textit{Hubble Space Telescope} (HST) by combining non-detections in bands blueward of the Lyman-break and color selection. A total of 10 objects fulfilling the LBG selection criteria at $z$$>$5.5 were selected over the 4.54 and 55.5 arcmin$^2$ covered by our HST and GTC images, respectively. The photometric redshift, UV luminosity, and the star-formation rate of these sources were estimated with models of their spectral energy distribution. These $z$$\sim$6 candidates have physical properties and number densities in agreement with previous results. The UV luminosity function at $z$$\sim$6 and a Voronoi tessellation analysis of this field shows no strong evidence for an overdensity of relatively bright objects (m$_{F105W}$$<$25.9) associated with \textit{HFLS3}. However, the over-density parameter deduced from this field and the surface density of objects can not excluded definitively the LBG over-density hypothesis. Moreover we identified three faint objects at less than three arcseconds from \textit{HFLS3} with color consistent with those expected for $z$$\sim$6 galaxies. Deeper data are needed to confirm their redshifts and to study their association with \textit{HFLS3} and the galaxy merger that may be responsible for the massive starburst.

Numerical simulations of cosmological structure formation show that the Universe's most massive clusters, and the galaxies living in those clusters, assemble rapidly at early times (2.5 < z < 4). While more than twenty proto-clusters have been observed at z > 2 based on associations of 5-40 galaxies around rare sources, the observational evidence for rapid cluster formation is weak. Here we report observations of an asymmetric, filamentary structure at z = 2.47 containing seven starbursting, submillimeter-luminous galaxies and five additional AGN within a comoving volume of 15000 Mpc$^{3}$. As the expected lifetime of both the luminous AGN and starburst phase of a galaxy is ~100 Myr, we conclude that these sources were likely triggered in rapid succession by environmental factors, or, alternatively, the duration of these cosmologically rare phenomena is much longer than prior direct measurements suggest. The stellar mass already built up in the structure is $\sim10^{12}M_{\odot}$ and we estimate that the cluster mass will exceed that of the Coma supercluster at $z \sim 0$. The filamentary structure is in line with hierarchical growth simulations which predict that the peak of cluster activity occurs rapidly at z > 2.

Galaxies' rest-frame ultraviolet (UV) properties are often used to directly infer the degree to which dust obscuration affects the measurement of star formation rates. While much recent work has focused on calibrating dust attenuation in galaxies selected at rest-frame ultraviolet wavelengths, locally and at high-$z$, here we investigate attenuation in dusty, star-forming galaxies (DSFGs) selected at far-infrared wavelengths. By combining multiwavelength coverage across 0.15--500\,$\mu$m in the COSMOS field, in particular making use of {\it Herschel} imaging, and a rich dataset on local galaxies, we find a empirical variation in the relationship between rest-frame UV slope ($\beta$) and ratio of infrared-to-ultraviolet emission ($L_{\rm IR}/L_{\rm UV}\equiv\,IRX$) as a function of infrared luminosity, or total star formation rate, SFR. Both locally and at high-$z$, galaxies above SFR$\gt$50\,M$_\odot$\,yr$^{-1}$ deviate from the nominal $IRX-\beta$ relation towards bluer colors by a factor proportional to their increasing IR luminosity. We also estimate contamination rates of DSFGs on high-$z$ dropout searches of $\ll1$\%\ at $z\lt4-10$, providing independent verification that contamination from very dusty foreground galaxies is low in LBG searches. Overall, our results are consistent with the physical interpretation that DSFGs, e.g. galaxies with $>50$\,M$_\odot$\,yr$^{-1}$, are dominated at all epochs by short-lived, extreme burst events, producing many young O and B stars that are primarily, yet not entirely, enshrouded in thick dust cocoons. The blue rest-frame UV slopes of DSFGs are inconsistent with the suggestion that most DSFGs at $z\sim2$ exhibit steady-state star formation in secular disks.

We discuss the restframe UV emission from the starbursting galaxy HFLS3 at z=6.34, discovered in Herschel/SPIRE data due to its red color in the submm wavelengths from 250-500 um. The apparent inst. SFR of HFLS3 inferred from the total FIR luminosity measured with over 15 photometric data points between 100 to 1000 um is 2900 Msun/yr. Keck/NIRC2 Ks band adaptive optics imaging data showed two potential NIR counterparts near HFLS3. Previously, the northern galaxy was taken to be in the foreground at z=2.1 while the southern galaxy was assumed to HFLS3's NIR counterpart. New HST/WFC3 and ACS imaging data show both optically bright galaxies are in the foreground at z<6. A new lensing model based on HST data and mm-wave continuum emission yields a magnification of 2.2+/-0.3. The lack of multiple imaging constrains the lensing magnification to be lower than either 2.7 or 3.5 at the 95% confidence level for the two scenarios, which attribute one or two components to HFLS3 in the source plane. Correcting for gravitational lensing, the inst. SFR is 1320 Msun/yr with the 95% confidence lower limit around 830 Msun/yr. Using models for the restframe UV to FIR SED, the ave. SFR over the last 100 Myr is around 660 Msun/yr. The dust and stellar masses of HFLS3 from the same SED models are 3x10^8 Msun and ~5x10^10 Msun, respectively, with large systematic uncertainties on assumptions related to the SED model. With HST/WFC3 images we also find diffuse NIR emission about 0.5" (~3 kpc) SW of HFLS3 that remains undetected in the ACS data. The emission has a photometric redshift consistent with either z~6 or a dusty galaxy template at z~2. If at the same redshift as HFLS3 the detected diffuse emission could be part of the complex merger system that could be triggering the starburst. Alternatively, it could be part of the foreground structure at z~2.1 that is responsible for lensing of HFLS3.

We use the James Clerk Maxwell Telescope's SCUBA-2 camera to image a 400 arcmin^2 area surrounding the GOODS-N field. The 850 micron rms noise ranges from a value of 0.49 mJy in the central region to 3.5 mJy at the outside edge. From these data, we construct an 850 micron source catalog to 2 mJy containing 49 sources detected above the 4-sigma level. We use an ultradeep (11.5 uJy at 5-sigma) 1.4 GHz image obtained with the Karl G. Jansky Very Large Array together with observations made with the Submillimeter Array to identify counterparts to the submillimeter galaxies. For most cases of multiple radio counterparts, we can identify the correct counterpart from new and existing Submillimeter Array data. We have spectroscopic redshifts for 62% of the radio sources in the 9 arcmin radius highest sensitivity region (556/894) and 67% of the radio sources in the GOODS-N region (367/543). We supplement these with a modest number of additional photometric redshifts in the GOODS-N region (30). We measure millimetric redshifts from the radio to submillimeter flux ratios for the unidentified submillimeter sample, assuming an Arp 220 spectral energy distribution. We find a radio flux dependent K-z relation for the radio sources, which we use to estimate redshifts for the remaining radio sources. We determine the star formation rates (SFRs) of the submillimeter sources based on their radio powers and their submillimeter and find that they agree well. The radio data are deep enough to detect star-forming galaxies with SFRs >2000 solar masses per year to z~6. We find galaxies with SFRs up to ~6,000 solar masses per year over the redshift range z=1.5-6, but we see evidence for a turn-down in the SFR distribution function above 2000 solar masses per year.

We present a method for selecting $z>4$ dusty, star forming galaxies (DSFGs) using Herschel/SPIRE 250/350/500 $\mu m$ flux densities to search for red sources. We apply this method to 21 deg$^2$ of data from the HerMES survey to produce a catalog of 38 high-$z$ candidates. Follow-up of the first 5 of these sources confirms that this method is efficient at selecting high-$z$ DSFGs, with 4/5 at $z=4.3$ to $6.3$ (and the remaining source at $z=3.4$), and that they are some of the most luminous dusty sources known. Comparison with previous DSFG samples, mostly selected at longer wavelengths (e.g., 850 $\mu m$) and in single-band surveys, shows that our method is much more efficient at selecting high-$z$ DSFGs, in the sense that a much larger fraction are at $z>3$. Correcting for the selection completeness and purity, we find that the number of bright ($S_{500\,\mu m} \ge 30$ mJy), red Herschel sources is $3.3 \pm 0.8$ deg$^{-2}$. This is much higher than the number predicted by current models, suggesting that the DSFG population extends to higher redshifts than previously believed. If the shape of the luminosity function for high-$z$ DSFGs is similar to that at $z\sim2$, rest-frame UV based studies may be missing a significant component of the star formation density at $z=4$ to $6$, even after correction for extinction.

Dust-obscured galaxies (DOGs) are a UV-faint, IR-bright galaxy population that reside at z~2 and are believed to be in a phase of dusty star-forming and AGN activity. We present far-IR observations of a complete sample of DOGs in the 2 deg2 of COSMOS. The 3077 DOGs have <z>=1.9+/-0.3 and are selected from 24um and r+ observations using a color cut of r+ - [24]>=7.5 (AB mag) and S24>=100uJy. Based on the near-IR SEDs, 47% are star-formation dominated and 10% are AGN-dominated. We use SPIRE far-IR photometry from HerMES to calculate the IR luminosity and characteristic dust temperature for the 1572 (51%) DOGs that are detected at 250um (>=3{\sigma}). For the remaining 1505 (49%) that are undetected, we perform a median stacking analysis to probe fainter luminosities. Detected and undetected DOGs have average IR luminosities of (2.8+/-0.4) x 1012 LSun and (0.77+-0.08)x10^12LSun, and dust temperatures of 34+/-7 K and 37+/-3 K, respectively. The IR luminosity function of DOGs with S24>=100uJy is calculated, using far-IR observations and stacking. DOGs contribute 10-30% to the total star formation rate density of the Universe at z=1.5-2.5, dominated by 250um detected and bump DOGs. For comparison, DOGs contribute 30% to the star-formation rate density of the Universe for all galaxies at z=1.5-2.5 with S24>=100uJy. DOGs have a large scatter about the star-formation main sequence and their specific star-formation rates show that the observed phase of star-formation could be responsible for their observed stellar mass at z~2.

Stellar archeology shows that massive elliptical galaxies today formed rapidly about ten billion years ago with star formation rates above several hundreds solar masses per year (M_sun/yr). Their progenitors are likely the sub-millimeter-bright galaxies (SMGs) at redshifts (z) greater than 2. While SMGs' mean molecular gas mass of 5x10^10 M_sun can explain the formation of typical elliptical galaxies, it is inadequate to form ellipticals that already have stellar masses above 2x10^11 M_sun at z ~ 2. Here we report multi-wavelength high-resolution observations of a rare merger of two massive SMGs at z = 2.3. The system is currently forming stars at a tremendous rate of 2,000 M_sun/yr. With a star formation efficiency an order-of-magnitude greater than that of normal galaxies, it will quench the star formation by exhausting the gas reservoir in only ~200 million years. At a projected separation of 19 kiloparsecs, the two massive starbursts are about to merge and form a passive elliptical galaxy with a stellar mass of ~4x10^11 M_sun. Our observations show that gas-rich major galaxy mergers, concurrent with intense star formation, can form the most massive elliptical galaxies by z ~ 1.5.

ABRIGED Herschel/SPIRE has provided confusion limited maps of deep fields at 250, 350, and 500um, as part of the HerMES survey. Due to confusion, only a small fraction of the Cosmic Infrared Background can be resolved into individually-detected sources. Our goal is to produce deep galaxy number counts and redshift distributions below the confusion limit, which we then use to place strong constraints on the origins of the cosmic infrared background and on models of galaxy evolution. We individually extracted the bright SPIRE with a method using the positions, the flux densities, and the redshifts of the 24um sources as a prior, and derived the number counts and redshift distributions of the bright SPIRE sources. For fainter SPIRE sources, we reconstructed the number counts and the redshift distribution below the confusion limit using the deep 24um catalogs associated with photometric redshift and information provided by the stacking of these sources into the deep SPIRE maps. Finally, by integrating all these counts, we studied the contribution of the galaxies to the CIB as a function of their flux density and redshift. Through stacking, we managed to reconstruct the source counts per redshift slice down to ~2 mJy in the three SPIRE bands, which lies about a factor 10 below the 5sigma confusion limit. None of the pre-existing population models are able to reproduce our results at better than 3sigma. Finally, we extrapolate our counts to zero flux density in order to derive an estimate of the total contribution of galaxies to the CIB, finding 10.1, 6.5, and 2.8 nW/m2/sr at 250, 350, and 500um, respectively. These values agree well with FIRAS absolute measurements, suggesting our number counts and their extrapolation are sufficient to explain the CIB. Finally, combining our results with other works, we estimate the energy budget contained in the CIB between 8 and 1000um: 26 nW/m2/sr.

We present Keck spectroscopic observations and redshifts for a sample of 767 Herschel-SPIRE selected galaxies (HSGs) at 250, 350, and 500um, taken with the Keck I Low Resolution Imaging Spectrometer (LRIS) and the Keck II DEep Imaging Multi-Object Spectrograph (DEIMOS). The redshift distribution of these SPIRE sources from the Herschel Multitiered Extragalactic Survey (HerMES) peaks at z=0.85, with 731 sources at z<2 and a tail of sources out to z~5. We measure more significant disagreement between photometric and spectroscopic redshifts (<delta_z>/(1+z)>=0.29) than is seen in non-infrared selected samples, likely due to enhanced star formation rates and dust obscuration in infrared-selected galaxies. We estimate that the vast majority (72-83%) of z<2 Herschel-selected galaxies would drop out of traditional submillimeter surveys at 0.85-1mm. We estimate the luminosity function and implied star-formation rate density contribution of HSGs at z<1.6 and find overall agreement with work based on 24um extrapolations of the LIRG, ULIRG and total infrared contributions. This work significantly increased the number of spectroscopically confirmed infrared-luminous galaxies at z>>0 and demonstrates the growing importance of dusty starbursts for galaxy evolution studies and the build-up of stellar mass throughout cosmic time. [abridged]

We present near-infrared spectroscopic observations from VLT ISAAC of thirteen 250\mu m-luminous galaxies in the CDF-S, seven of which have confirmed redshifts which average to <z > = 2.0 \pm 0.4. Another two sources of the 13 have tentative z > 1 identifications. Eight of the nine redshifts were identified with H{\alpha} detection in H- and K-bands, three of which are confirmed redshifts from previous spectroscopic surveys. We use their near-IR spectra to measure H{\alpha} line widths and luminosities, which average to 415 \pm 20 km/s and 3 \times 10^35 W (implying SFR(H{\alpha})~200 M_\odot /yr), both similar to the H{\alpha} properties of SMGs. Just like SMGs, 250 \mu m-luminous galaxies have large H{\alpha} to far-infrared (FIR) extinction factors such that the H{\alpha} SFRs underestimate the FIR SFRs by ~8-80 times. Far-infrared photometric points from observed 24\mu m through 870\mu m are used to constrain the spectral energy distributions (SEDs) even though uncertainty caused by FIR confusion in the BLAST bands is significant. The population has a mean dust temperature of Td = 52 \pm 6 K, emissivity {\beta} = 1.73 \pm 0.13, and FIR luminosity LFIR = 3 \times 10^13 L_\odot. Although selection at 250\mu m allows for the detection of much hotter dust dominated HyLIRGs than SMG selection (at 850\mu m), we do not find any >60 K 'hot-dust' HyLIRGs. We have shown that near-infrared spectroscopy combined with good photometric redshifts is an efficient way to spectroscopically identify and characterise these rare, extreme systems, hundreds of which are being discovered by the newest generation of IR observatories including the Herschel Space Observatory.

Using the IRAM Plateau-de-Bure Interferometer (PdBI) we have searched for the upper fine structure line of neutral carbon (CI(2->1), nu_rest = 809 GHz) and 12CO(J=7->6) (nu_rest=806GHz) towards the submillimetre galaxies (SMGs) GN20 (SMM J123711.9+622212, z = 4.055) and GN20.2 (SMM J123708.8+622202, z = 4.051). The far-infrared (FIR) continuum is detected at 8sigma significance in GN20, with a flux density of S_1.8mm = 1.9+-0.2mJy, while no continuum is detected in GN20.2. Both sources are statistically undetected in both CI and CO(7-6) lines; we derive line luminosity limits for both CI and CO of L'<~2x10^10K km s^-1 pc^2. Assuming carbon excitation temperatures of T_ex = 30K (the galaxies' measured dust temperatures), we infer CI mass limits of M_CI < 5.4x10^6 Msun (GN20) and M_CI < 6.8x10^6 Msun (GN20.2). The derived CI abundance limits are <1.8x10^-5 for GN20 and <3.8x10^-5 for GN20.2 implying that the systems have Milky Way level neutral carbon enrichment (X[CI]/X[H2]) or lower, similar to high-redshift carbon-detected systems (at 5x10^-5) but about 50 times less than the neutral carbon enrichment of local starburst galaxies. Observations of GN20 and GN20.2 in high-resolution MERLIN+VLA radio maps of GOODS-N are used to further constrain the sizes and locations of active regions. We conclude that the physical gas properties of young rapidly evolving systems like GN20 and GN20.2 are likely significantly different than starburst/ULIRG environments in the local Universe yet similar to z~2 SMGs. Unless gravitationally amplified examples can be found, observations of galaxies like GN20 will require the order of magnitude increase in sensitivity of the Atacama Large Millimetre Array (ALMA) to constrain their CI and high-J CO content, despite the fact that they are the brightest systems at z~4.

We identify eight z>1 radio sources undetected at 850um but robustly detected at 70um, confirming that they represent ultraluminous infrared galaxies (ULIRGs) with hotter dust temperatures (T_dust=52+-10 K) than Submillimetre Galaxies (SMGs) at similar luminosities and redshifts. These galaxies share many properties with SMGs: ultra-violet (UV) spectra consistent with starbursts, high stellar masses and radio luminosities. We can attribute their radio emission to star formation since high-resolution MERLIN radio maps show extended emission regions (with half light radii of 2-3kpc), which are unlikely to be generated by AGN activity. These observations provide the first direct confirmation of hot, dusty ULIRGs which are missed by current submillimetre surveys. They have significant implications for future observations from the Herschel Space Observatory and SCUBA2, which will select high redshift luminous galaxies with less selection biases.

We outline a strategy to select faint (i<24.5) type 1 AGN candidates down to the Seyfert/QSO boundary for spectroscopic targeting in the COSMOS field, picking candidates by their nonstellar colors in broadband ground-based photometry and morphological properties extracted from HST-ACS. AGN optical color selection has not been applied to such faint magnitudes in such a large continuous part of the sky. Hot stars are known to be the dominant contaminant for bright AGN candidate selection at z<2, but we anticipate the highest color contamination at all redshifts to be from faint starburst and compact galaxies. Morphological selection via the Gini Coefficient separates most potential AGN from these faint blue galaxies. Recent models of the quasar luminosity function are used to estimate quasar surface densities, and studies of stellar populations in the COSMOS field infer stellar contamination. We use 292 spectroscopically confirmed type 1 AGN and quasar templates to predict AGN colors with redshift, and contrast those predictions with the colors of known contaminating populations. The motivation of this study and subsequent spectroscopic follow-up is to populate and refine the faint end of the QLF where the population of type 1 AGN is presently not well known. The anticipated AGN observations will add to the ~300 already known AGN in the COSMOS field, making COSMOS a densely packed field of quasars to be used to understand supermassive black holes and probe the structure of the intergalactic medium in the intervening volume.