The Cosmic Assembly Near-infrared Deep Extragalactic Legacy Survey (CANDELS) is designed to document the first third of galactic evolution, over the approximate redshift (z) range 8-1.5. It will image >250,000 distant galaxies using three separate cameras on the Hubble Space Telescope, from the mid-ultraviolet to the near-infrared, and will find and measure Type Ia supernovae at z > 1.5 to test their accuracy as standardizable candles for cosmology. Five premier multi-wavelength sky regions are selected, each with extensive ancillary data. The use of five widely separated fields mitigates cosmic variance and yields statistically robust and complete samples of galaxies down to a stellar mass of 10 9 M to z ≈ 2, reaching the knee of the ultraviolet luminosity function of galaxies to z ≈ 8. The survey covers approximately 800 arcmin 2 and is divided into two parts. The CANDELS/Deep survey (5σ point-source limit H = 27.7 mag) covers ∼125 arcmin 2 within Great Observatories Origins Deep Survey (GOODS)-N and GOODS-S. The CANDELS/Wide survey includes GOODS and three additional fields (Extended Groth Strip, COSMOS, and Ultra-deep Survey) and covers the full area to a 5σ pointsource limit of H 27.0 mag. Together with the Hubble Ultra Deep Fields, the strategy creates a three-tiered "wedding-cake" approach that has proven efficient for extragalactic surveys. Data from the survey are nonproprietary and are useful for a wide variety of science investigations. In this paper, we describe the basic motivations for the survey, the CANDELS team science goals and the resulting observational requirements, the field selection and geometry, and the observing design. The Hubble data processing and products are described in a companion paper.

We present evidence for very high gas fractions and extended molecular gas reservoirs in normal, near-infrared selected (BzK) galaxies at z∼1.5. Our results are based on multi-configuration CO[2-1] observations obtained at the IRAM Plateau de Bure Interferometer. All six star forming galaxies observed were detected at high significance. High spatial resolution observations resolve the CO emission in four of them, implying sizes of the gas reservoirs of order of 6-11 kpc and suggesting the presence of ordered rotation. The galaxies have UV morphologies consistent with clumpy, unstable disks, and UV sizes that are consistent with those measured in CO. The star formation efficiencies are homogeneously low within the sample and similar to those of local spirals -the resulting gas depletion times are ∼ 0.5 Gyr, much higher than what is seen in high-z submm galaxies and quasars. The CO luminosities can be predicted to within 0.15 dex from the observed star formation rates and stellar masses, implying a tight correlation of the gas mass with these quantities. We use new dynamical models of clumpy disk galaxies to derive dynamical masses for our sample. These models are able to reproduce the peculiar spectral line shapes of the CO emission. After accounting for the stellar and dark matter masses we derive molecular gas reservoirs with masses of 0.4-1.2×10 11 M ⊙ . The implied conversion (CO luminosity-to-gas mass) factor is very high: α CO = 3.6 ± 0.8, consistent with a Galactic conversion factor but four times higher than that of local ultra-luminous IR galaxies that is typically used for high-redshift objects. The gas mass in these galaxies is comparable to or larger than the stellar mass, and the gas accounts for an impressive 50-65% of the baryons within the galaxies' half light radii. We are thus witnessing truly gasdominated galaxies at z ∼ 1.5, a finding that explains the high specific SFRs observed for z > 1 galaxies. The BzK galaxies can be viewed as scaled-up versions of local disk galaxies, with low efficiency star formation taking place inside extended, low excitation gas disks. These galaxies are markedly different than local ULIRGs and high-z submm galaxies and quasars, where higher excitation and more compact gas is found.

We use the first systematic data sets of CO molecular line emission in z ∼ 1-3 normal star-forming galaxies (SFGs) for a comparison of the dependence of galaxy-averaged star formation rates on molecular gas masses at low and high redshifts, and in different galactic environments. Although the current high-z samples are still small and biased towards the luminous and massive tail of the actively star-forming 'main-sequence', a fairly clear picture is emerging. Independent of whether galaxy-integrated quantities or surface densities are considered, low-and high-z SFG populations appear to follow similar molecular gas-star formation relations with slopes 1.1 to 1.2, over three orders of magnitude in gas mass or surface density. The gas-depletion time-scale in these SFGs grows from 0.5 Gyr at z ∼ 2 to 1.5 Gyr at z ∼ 0. The average corresponds to a fairly low star formation efficiency of 2 per cent per dynamical time. Because star formation depletion times are significantly smaller than the Hubble time at all redshifts sampled, star formation rates and gas fractions are set by the balance between gas accretion from the halo and stellar feedback.In contrast, very luminous and ultraluminous, gas-rich major mergers at both low and high z produce on average four to 10 times more far-infrared luminosity per unit gas mass. We show that only some fraction of this difference can be explained by uncertainties in gas mass or luminosity estimators; much of it must be intrinsic. A possible explanation is a top-heavy stellar mass function in the merging systems but the most likely interpretation is that the star formation relation is driven by global dynamical effects. For a given mass, the more compact merger systems produce stars more rapidly because their gas clouds are more compressed with shorter dynamical times, so that they churn more quickly through the available gas reservoir than the typical normal disc galaxies. When the dependence on galactic dynamical Based on observations with the Plateau de Bure millimetre interferometre, operated by the Institute for Radio Astronomy in the Millimetre Range (IRAM), which is funded by a partnership of INSU/CNRS (France), MPG (Germany) and IGN (Spain).

This paper provides an update of our previous scaling relations (Genzel et al. 2015) between galaxy integrated molecular gas masses, stellar masses and star formation rates, in the framework of the star formation main-sequence (MS), with the main goal to test for possible systematic effects. For this purpose our new study combines three independent methods of determining molecular gas masses from CO line fluxes, far-infrared dust spectral energy distributions, and ~1mm dust photometry, in a large sample of 1444 star forming galaxies (SFGs) between z=0 and 4. The sample covers the stellar mass range log(M*/M)=9.0-11.8, and star formation rates relative to that on the MS, δMS=SFR/SFR(MS), from 10 -1.3 to 10 2.2 . Our most important finding is that all data sets, despite the different techniques and analysis methods used, follow the same scaling trends, once method-to-method zero point offsets are minimized and uncertainties are properly taken into account. The molecular gas depletion time tdepl, defined as the ratio of molecular gas mass to star formation rate, scales as (1+z) -0.6 × (δMS) -0.44 , and is only weakly dependent on stellar mass. The ratio of molecular-to-stellar mass μgas depends on (1+z) 2.5 × (δMS) 0.52 × (M*) -0.36 , which tracks the evolution of the specific star formation rate. The redshift dependence of μgas requires a curvature term, as may the mass-dependences of tdepl and μgas. We find no or only weak correlations of tdepl and μgas with optical size R or surface density once one removes the above scalings, but we caution that optical sizes may not be appropriate for the high gas and dust columns at high-z.

We present PHIBSS, the IRAM Plateau de Bure high-z blue sequence CO 3-2 survey of the molecular gas properties in normal star forming galaxies (SFGs) near the cosmic star formation peak. PHIBSS provides 52 CO detections in two redshift slices at z~1.2 and 2.2, with log(M * (M  )) ≥10.4 and log(SFR(M  /yr)) ≥ 1.5.Including a correction for the incomplete coverage of the M * -SFR plane, we infer average gas fractions of ~0.33 at z~1.2 and ~0.47 at z~2.2. Gas fractions drop with stellar mass, in agreement with cosmological simulations including strong star formation feedback. Most of the z~1-3 SFGs are rotationally supported turbulent disks. The sizes of CO and UV/optical emission are comparable. The molecular gasstar formation relation for the z=1-3 SFGs is near-linear, with a ~0.7 Gyrs gas depletion timescale; changes in depletion time are only a secondary effect. Since this timescale is much less than the Hubble time in all SFGs between z~0 and 2, fresh gas must be supplied with a fairly high duty cycle over several billion years. At given z and M * , gas fractions correlate strongly with the specific star formation rate. The variation of specific star formation rate between z~0 and 3 is mainly controlled by the fraction of baryonic mass that resides in cold gas.

2Stars form from cold molecular interstellar gas. Since this is relatively rare in the local Universe, galaxies like the Milky Way form only a few new stars per year. Typical massive galaxies in the distant Universe formed stars an order of magnitude more rapidly 1,2 . Unless star formation was significantly more efficient, this difference suggests that young galaxies were much more gas rich. Molecular gas observations in the distant Universe have so far been largely restricted to very luminous, rare objects, including mergers and quasars 3,4,5 . Here we report the results of a systematic survey of molecular gas in samples of typical massive star forming galaxies at ~1.2 and 2.3, when the Universe was 40% and 24% of its current age. Our measurements provide empirical evidence that distant star forming galaxies indeed were gas rich, and that the star formation efficiency is not strongly dependent on cosmic epoch. The average fraction of cold gas relative to total galaxy baryonic mass at z= 2.3 and z=1.2 is ~44% and 34%, three to ten times higher than in today's massive spiral galaxies 6 . The slow decrease between z~2 and 1 probably requires a mechanism of semi-continuous replenishment of fresh gas to the young galaxies.Direct observations of molecular gas in galaxies as a function of cosmic epoch are required to understand how galaxies have turned their gas into stars. To explore the evolution of cold gas fractions, we selected two samples of star forming galaxies (SFGs) spanning similar ranges in stellar mass and star formation rates: one at redshift z~1.2 (t o~5 .5 Gyr after the Big Bang) and the other at z~2.3 (t 0~3 Gyr). With recent improvements in instrumental sensitivity we can now sample the massive tail of the 3 typical, or 'normal' star forming galaxies in this epoch (Supplementary Information, section 1). These 'main-sequence' galaxies are believed to have high (40-70%) duty cycles of star formation, and most are probably not 'starbursts' in a brief period of activity, such as major dissipative mergers 1,2 . Figure 1 displays the source-integrated spectra in the CO J=3-2 transition for 19 of the galaxies observed, 10 at z~2 and 9 at z~1. For 14 SFGs we have solid (>4σ) detections in both redshift ranges, and for the first time for z>2 SFGs. In 5 galaxies the emission is marginally or not detected, or may be continuum rather than line emission. Table 1 summarizes the observed and derived galaxy properties.For EGS1305123 (z=1.12) and EGS1207881 (z=1.17) we also obtained high quality spatially resolved maps with FWHM resolutions of 0.65" and 1". Figure 2c shows the integrated CO 3-2 emission in EGS1305123, superposed on an optical image from the Hubble Space Telescope (HST). This map resolves for the first time the cold molecular gas distribution in a clearly non-merging high-z SFG. This system looks like a scaled-up version (in terms of star formation rate and gas mass) of normal z~0 gas rich disk galaxies. In the optical images EGS1305123 is a nearly face-on, large spiral disk. The clumpy CO emissio...

This paper describes the Hubble Space Telescope imaging data products and data reduction procedures for the Cosmic Assembly Near-IR Deep Extragalactic Legacy Survey (CANDELS). This survey is designed to document the evolution of galaxies and black holes at z ∼ 1.5 − 8, and to study Type Ia SNe beyond z > 1.5. Five premier multi-wavelength sky regions are selected, each with extensive multiwavelength observations. The primary CANDELS data consist of imaging obtained in the Wide Field Camera 3 / infrared channel (WFC3/IR) and UVIS channel, along with the Advanced Camera for Surveys (ACS). The CANDELS/Deep survey covers ∼ 125 square arcminutes within GOODS-N and GOODS-S, while the remainder consists of the CANDELS/Wide survey, achieving a total of ∼ 800 square arcminutes across GOODS and three additional fields (EGS, COSMOS, and UDS). We summarize the observational aspects of the survey as motivated by the scientific goals and present a detailed description of the data reduction procedures and products from the survey. Our data reduction methods utilize the most up to date calibration files and image combination procedures. We have paid special attention to correcting a range of instrumental effects, including CTE degradation for ACS, removal of electronic bias-striping present in ACS data after SM4, and persistence effects and other artifacts in WFC3/IR. For each field, we release mosaics for individual epochs and eventual mosaics containing data from all epochs combined, to facilitate photometric variability studies and the deepest possible photometry. A more detailed overview of the science goals and observational design of the survey are presented in a companion paper.

We present evidence that bona fide disks and starburst systems occupy distinct regions in the gas mass versus star formation (SF) rate plane, both for the integrated quantities and for the respective surface densities. This result is based on CO observations of galaxy populations at low and high redshifts, and on the current consensus for the CO luminosity to gas mass conversion factors. The data suggest the existence of two different star formation regimes: a long-lasting mode for disks and a more rapid mode for starbursts, the latter probably occurring during major mergers or in dense nuclear SF regions. Both modes are observable over a large range of SF rates. The detection of CO emission from distant near-IR selected galaxies reveals such bimodal behavior for the first time, as they allow us to probe gas in disk galaxies with much higher SF rates than are seen locally. The different regimes can potentially be interpreted as the effect of a top-heavy IMF in starbursts. However, we favor a different physical origin related to the fraction of molecular gas in dense clouds. The IR luminosity to gas mass ratio (i.e., the SF efficiency) appears to be inversely proportional to the dynamical (rotation) timescale. Only when accounting for the dynamical timescale, a universal SF law is obtained, suggesting a direct link between global galaxy properties and the local SF rate.