[en] Of the almost 40 star-forming galaxies at $z\gtrsim <!--\; ???\; -->5$ (not counting quasi-stellar objects) observed in $[\mathrm{C}\mathrm{I}\mathrm{I}]$ to date, nearly half are either very faint in $[\mathrm{C}\mathrm{I}\mathrm{I}]$ or not detected at all, and fall well below expectations based on locally derived relations between star formation rate and $[\mathrm{C}\mathrm{I}\mathrm{I}]$ luminosity. This has raised questions as to how reliable $[\mathrm{C}\mathrm{I}\mathrm{I}]$ is as a tracer of star formation activity at these epochs and how factors such as metallicity might affect the $[\mathrm{C}\mathrm{I}\mathrm{I}]$ emission. Combining cosmological zoom simulations of galaxies with SÍGAME (SImulator of GAlaxy Millimeter/submillimeter Emission), we modeled the multiphased interstellar medium (ISM) and its emission in $[\mathrm{C}\mathrm{I}\mathrm{I}]$, as well as in [O i] and [O iii], from 30 main-sequence galaxies at $z\simeq <!--\; ???\; -->6$ with star formation rates ∼3–23 $M_{\odot <!--\; ???\; -->}{\mathrm{y}\mathrm{r}}^{-<!--\; ???\; -->1}$, stellar masses $\sim <!--\; ???\; -->(0.7\text{--}8)\times <!--\; ??\; -->{10}^9{M}_{\odot <!--\; ???\; -->}$, and metallicities $\sim <!--\; ???\; -->(0.1\text{--}0.4)\times <!--\; ??\; -->Z_{\odot <!--\; ???\; -->}$. The simulations are able to reproduce the aforementioned $[\mathrm{C}\mathrm{I}\mathrm{I}]$ faintness of some normal star-forming galaxy sources at $z⩾<!--\; ???\; -->5$. In terms of [O i] and [O iii], very few observations are available at $z\gtrsim <!--\; ???\; -->5$, but our simulations match two of the three existing $z\gtrsim <!--\; ???\; -->5$ detections of [O iii] and are furthermore roughly consistent with the [O i] and [O iii] luminosity relations with star formation rate observed for local starburst galaxies. We find that the $[\mathrm{C}\mathrm{I}\mathrm{I}]$ emission is dominated by the diffuse ionized gas phase and molecular clouds, which on average contribute ∼66% and ∼27%, respectively. The molecular gas, which constitutes only $\sim <!--\; ???\; -->10\mathrm{\%<!--\; \%\; -->}$ of the total gas mass, is thus a more efficient emitter of $[\mathrm{C}\mathrm{I}\mathrm{I}]$ than the ionized gas, which makes up ∼85% of the total gas mass. A principal component analysis shows that the $[\mathrm{C}\mathrm{I}\mathrm{I}]$ luminosity correlates with the star formation activity of a galaxy as well as its average metallicity. The low metallicities of our simulations together with their low molecular gas mass fractions can account for their $[\mathrm{C}\mathrm{I}\mathrm{I}]$ faintness, and we suggest that these factors may also be responsible for the $[\mathrm{C}\mathrm{I}\mathrm{I}]$-faint normal galaxies observed at these early epochs.

[en] We present SÍGAME simulations of the $[\mathrm{C}\mathrm{I}\mathrm{I}]$ 157.7 μm fine structure line emission from cosmological smoothed particle hydrodynamics simulations of seven main sequence galaxies at z = 2. Using sub-grid physics prescriptions the gas in our simulations is modeled as a multi-phased interstellar medium comprised of molecular gas residing in giant molecular clouds, an atomic gas phase associated with photo-dissociation regions (PDRs) at the cloud surfaces, and a diffuse, ionized gas phase. Adopting logotropic cloud density profiles and accounting for heating by the local FUV radiation field and cosmic rays by scaling both with local star formation rate (SFR) volume density, we calculate the $[\mathrm{C}\mathrm{I}\mathrm{I}]$ emission using a photon escape probability formalism. The $[\mathrm{C}\mathrm{I}\mathrm{I}]$ emission peaks in the central $\lesssim <!--\; ???\; -->1\mathrm{k}\mathrm{p}\mathrm{c}$ of our galaxies as do the SFR radial profiles, with most $[\mathrm{C}\mathrm{I}\mathrm{I}]$ ($\gtrsim <!--\; ???\; -->70\mathrm{\%<!--\; \%\; -->}$) originating in the molecular gas phase, whereas further out ($\gtrsim <!--\; ???\; -->2\mathrm{k}\mathrm{p}\mathrm{c}$), the atomic/PDR gas dominates ($\gtrsim <!--\; ???\; -->90\mathrm{\%<!--\; \%\; -->}$) the $[\mathrm{C}\mathrm{I}\mathrm{I}]$ emission, no longer tracing ongoing star formation. Throughout, the ionized gas contribution is negligible ($\lesssim <!--\; ???\; -->3\mathrm{\%<!--\; \%\; -->}$). The $[\mathrm{C}\mathrm{I}\mathrm{I}]$ luminosity versus SFR ($[\mathrm{C}\mathrm{I}\mathrm{I}]$–SFR) relationship, integrated as well as spatially resolved (on scales of 1 kpc), delineated by our simulated galaxies is in good agreement with the corresponding relations observed locally and at high redshifts. In our simulations, the molecular gas dominates the $[\mathrm{C}\mathrm{I}\mathrm{I}]$ budget at $\mathrm{S}\mathrm{F}\mathrm{R}\gtrsim <!--\; ???\; -->20M_{\odot <!--\; ???\; -->}{\mathrm{y}\mathrm{r}}^{-<!--\; ???\; -->1}$ (${\mathrm{\Sigma <!--\; ??\; -->}}_{\mathrm{S}\mathrm{F}\mathrm{R}}$ $\gtrsim <!--\; ???\; -->0.5M_{\odot <!--\; ???\; -->}{\mathrm{y}\mathrm{r}}^{-<!--\; ???\; -->1}{\mathrm{k}\mathrm{p}\mathrm{c}}^{-<!--\; ???\; -->2}$), while atomic/PDR gas takes over at lower SFRs, suggesting a picture in which $[\mathrm{C}\mathrm{I}\mathrm{I}]$ predominantly traces the molecular gas in high-density/pressure regions where star formation is ongoing, and otherwise reveals the atomic/PDR gas phase.

[en] We present the first predictions for the L _{[C ii]}–SFR relation and [Cii] luminosity function (LF) in the epoch of reionization (EOR) based on cosmological hydrodynamics simulations using the simba suite plus radiative transfer calculations via sígame. The sample consists of 11,137 galaxies covering halo mass log M _halo ∈ [9, 12.4] M _⊙, star formation rate SFR ∈ [0.01, 330] M _⊙ yr⁻¹, and metallicity 〈Z _gas〉_SFR ∈ [0.1, 1.9] Z _⊙. The simulated L _{[C ii]}–SFR relation is consistent with the range observed, but with a spread of ≃0.3 dex at the high end of SFR (>100 M _⊙ yr⁻¹) and ≃0.6 dex at the lower end, and there is tension between our predictions and the values of L _{[C ii]} above 10^8.5 L _⊙ observed in some galaxies reported in the literature. The scatter in the L _{[C ii]}–SFR relation is mostly driven by galaxy properties, such that at a given SFR galaxies with higher molecular mass and metallicity have higher L _{[C ii]}. The [Cii] LF predicted by simba is consistent with the upper limits placed by the only existing untargeted flux-limited [Cii] survey at the EOR and those predicted by semianalytic models. We compare our results with existing models and discuss the differences responsible for the discrepant slopes in the L _{[C ii]}–SFR relation.

[en] We determine star formation rates (SFRs) in a sample of color-selected, star-forming (sBzK) galaxies (K_AB < 21.8) in the Extended Chandra Deep Field-South. To identify and avoid active galactic nuclei, we use X-ray, IRAC color, and IR/radio flux ratio selection methods. Photometric redshift-binned, average flux densities are measured with stacking analyses in Spitzer-MIPS IR, BLAST and APEX/LABOCA submillimeter, VLA and GMRT radio, and Chandra X-ray data. We include averages of aperture fluxes in MUSYC UBVRIz'JHK images to determine UV-through-radio spectral energy distributions. We determine the total IR luminosities and compare SFR calibrations from FIR, 24 μm, UV, radio, and X-ray wavebands. We find consistency with our best estimator, SFR_IR+UV, to within errors for the preferred radio SFR calibration. Our results imply that 24 μm only and X-ray SFR estimates should be applied to high-redshift galaxies with caution. Average IR luminosities are consistent with luminous infrared galaxies. We find SFR_IR+UV for our stacked sBzKs at median redshifts 1.4, 1.8, and 2.2 to be 55 ± 6 (random error), 74 ± 8, and 154 ± 17 M_☉ yr^–1, respectively, with additional systematic uncertainty of a factor of ∼2.

[en] We study the $r_{31}=L_{\mathrm{C}\mathrm{O}(3\text{--}2)}^{\mathrm{\prime <!--\; ???\; -->}}/L_{\mathrm{C}\mathrm{O}(1\text{--}0)}^{\mathrm{\prime <!--\; ???\; -->}}$ luminosity line ratio in a sample of nearby (z < 0.05) galaxies: 25 star-forming galaxies (SFGs) from the xCOLD GASS survey, 36 hard X-ray-selected active galactic nucleus (AGN) host galaxies from the BAT AGN Spectroscopic Survey, and 37 infrared-luminous galaxies from the SCUBA Local Universe Galaxy Survey. We find a trend for r ₃₁ to increase with star formation efficiency (SFE). We model r ₃₁ using the UCL-PDR code and find that the gas density is the main parameter responsible for the variation of r ₃₁, while the interstellar radiation field and cosmic-ray ionization rate play only a minor role. We interpret these results to indicate a relation between SFE and gas density. We do not find a difference in the r ₃₁ value of SFGs and AGN host galaxies, when the galaxies are matched in SSFR (〈r ₃₁〉 = 0.52 ± 0.04 for SFGs and 〈r ₃₁〉 = 0.53 ± 0.06 for AGN hosts). According to the results of the UCL-PDR models, the X-rays can contribute to the enhancement of the CO line ratio, but only for strong X-ray fluxes and for high gas density (n _H > 10⁴ cm⁻³). We find a mild tightening of the Kennicutt–Schmidt relation when we use the molecular gas mass surface density traced by CO(3–2) (Pearson correlation coefficient R = 0.83), instead of the molecular gas mass surface density traced by CO(1–0) (R = 0.78), but the increase in correlation is not statistically significant (p-value = 0.06). This suggests that the CO(3–2) line can be reliably used to study the relation between SFR and molecular gas for normal SFGs at high redshift and to compare it with studies of low-redshift galaxies, as is common practice.

[en] We performed a comprehensive stacking analysis on ∼14,200 quiescent galaxy (QG) candidates at z = 0–3 across mid-, far-infrared (MIR and FIR), and radio wavelengths. Identified via their rest-frame NUV − r and r − J colors, the QG candidates ($M_{\star <!--\; ???\; -->}={10}^{9.8\text{--}12.2}{M}_{\odot <!--\; ???\; -->}$) have drastically different IR and radio properties depending on their 24 μm emission strength. The fraction of QG candidates with strong 24 μm emission (equivalent to inferred star formation rates SFR${}_{24}⩾<!--\; ???\; -->100M_{\odot <!--\; ???\; -->}{yr}^{-<!--\; ???\; -->1}$, hereafter “IR-bright”) increases with redshift and peaks at 15%, and their stacked MIPS 24 μm, Herschel (PACS and SPIRE) and VLA emissions are consistent with being star-forming galaxies (SFGs). In contrast, the majority of QG candidates are faint or undetected at 24 μm individually (i.e., SFR₂₄ < 100 M_⊙ yr⁻¹, hereafter “IR-faint”). Their low dust-obscured SFRs derived from Herschel stacking (SFR_H ≲  3, 15, 50 M_⊙ yr⁻¹ out to z ∼ 1, 2, 3) are >2.5–12.5× lower than compared to SFGs. This is consistent with the quiescence, as expected from their low unobscured SFRs, as inferred from modeling their ultraviolet-to-NIR photometry. The discrepancy between the L_IR derived from stacking Herschel and 24 μm indicates that IR-faint QGs have dust SEDs that are different from those of SFGs. For the most massive ($M_{\star <!--\; ???\; -->}⩾<!--\; ???\; -->{10}^{11}{M}_{\odot <!--\; ???\; -->}$) IR-faint QGs at z < 1.5, the stacked 1.4 GHz emission is in excess of that expected from other SFR indicators, suggesting a widespread presence of low-luminosity active galactic nuclei. Our results reaffirm the existence of a significant population of QGs out to z = 3, thus corroborating the need to quench star formation in galaxies at early epochs.

[en] We analyze an extremely deep 450 μm image (1σ = 0.56 mJy beam⁻¹) of a ≃300 arcmin² area in the CANDELS/COSMOS field as part of the Sub-millimeter Common User Bolometric Array-2 Ultra Deep Imaging EAO Survey. We select a robust (signal-to-noise ratio ≥4) and flux-limited (≥4 mJy) sample of 164 submillimeter galaxies (SMGs) at 450 μm that have K-band counterparts in the COSMOS2015 catalog identified from radio or mid-infrared imaging. Utilizing this SMG sample and the 4705 K-band-selected non-SMGs that reside within the noise level ≤1 mJy beam⁻¹ region of the 450 μm image as a training set, we develop a machine-learning classifier using K-band magnitude and color–color pairs based on the 13-band photometry available in this field. We apply the trained machine-learning classifier to the wider COSMOS field (1.6 deg²) using the same COSMOS2015 catalog and identify a sample of 6182 SMG candidates with similar colors. The number density, radio and/or mid-infrared detection rates, redshift and stellar-mass distributions, and the stacked 450 μm fluxes of these SMG candidates, from the S2COSMOS observations of the wide field, agree with the measurements made in the much smaller CANDELS field, supporting the effectiveness of the classifier. Using this SMG candidate sample, we measure the two-point autocorrelation functions from z = 3 down to z = 0.5. We find that the SMG candidates reside in halos with masses of ≃(2.0 ± 0.5) × 10¹³ h ⁻¹ M _☉ across this redshift range. We do not find evidence of downsizing that has been suggested by other recent observational studies.