The nature of the [C ii] emission in dusty star-forming galaxies from the SPT survey

Abstract

We present [CII] observations of 20 strongly lensed dusty star forming galaxies at 2.1 < z < 5.7 using APEX and Herschel. The sources were selected on their 1.4 mm flux (S 1.4 mm > 20 mJy) from the South Pole Telescope survey, with far-infrared (FIR) luminosities determined from extensive photometric data. The [CII] line is robustly detected in 17 sources, all but one being spectrally resolved. Eleven out of 20 sources observed in [CII] also have low-J CO detections from ATCA. A comparison with midand high-J CO lines from ALMA reveals consistent [CII] and CO velocity profiles, suggesting that there is little differential lensing between these species. The [CII], low-J CO and FIR data allow us to constrain the properties of the interstellar medium. We find [CII] to CO(1-0) luminosity ratios in the SPT sample of 5200 ± 1800, with significantly less scatter than in other samples. This line ratio can be best described by a medium of [CII] and CO emitting gas with a higher [CII] than CO excitation temperature, high CO optical depth τ CO (1-0) 1, and low to moderate [CII] optical depth τ [CII] 1. The geometric structure of photodissociation regions allows for such conditions.

ALMA observations of atomic carbon inz ∼ 4 dusty star-forming galaxies

“…The mean derived density produced by the Kaufman et al (1999) models is < log n >= 4.4 ± 0.4 cm 3 , a factor of ∼ 6 lower than that produced by 3d-pdr (< log n >= 5.2 ± 0.6 cm 3 ). This is also the reason for the higher derived densities in this work, compared to densities derived for SPT-DSFGs by Gullberg et al (2015) (who quote densities ranging from 2 < log(n)/cm −3 < 5). Gullberg et al (2015) measured densities using the older Kaufman et al (1999) models.…”

“…This is also the reason for the higher derived densities in this work, compared to densities derived for SPT-DSFGs by Gullberg et al (2015) (who quote densities ranging from 2 < log(n)/cm −3 < 5). Gullberg et al (2015) measured densities using the older Kaufman et al (1999) models. Because all previous works applying PDR modelling techniques to DSFGs have tended to use Kaufman et al (1999) models, it seems that the gas density in these extreme high-z objects may have been significantly underestimated.…”

“…We find a mean gas temperature for our sources of 25 K. In Fig. 11, we compare our derived gas temperatures with dust temperatures, taken from Weiß et al 2013, Gullberg et al 2015, and Spilker et al Figure 11. Gas temperature implied by the density (n) and UVfield strength (G 0 ) outputted by 3d-pdr (defined at A V = 3), plotted against dust temperature derived using SED fitting to far-IR photometry.…”

“…Observations were made at 345 GHz (for sources with redshifts 4.2 < z < 5.7) and 460GHz (3.1 < z < 3.8), with system temperatures of 230 K and 170 K respectively. Further details of the observations and data reduction can be found in Gullberg et al (2015).…”

“…[Cii] lines are taken from Gullberg et al (2015). CO(2 − 1) lines are taken from Aravena et al (2016) -other CO lines are taken from the program described in Weiss et al (2013).…”