A tentative emission line at z=5.8 from a 3mm-selected galaxy
DDraft version February 17, 2021
Typeset using L A TEX default style in AASTeX63
A tentative emission line at z=5.8 from a 3 mm-selected galaxy
Jorge A. ZavalaThe University of Texas at Austin, 2515 Speedway Blvd, Austin, TX 78712, USA
ABSTRACTI report a tentative ( ∼ σ ) emission line at ν = 100 .
84 GHz from “COS-3mm-1”, a 3 mm-selectedgalaxy reported by Williams et al. (2019) that is undetected at optical and near infrared wavelengths.The line was found in the ALMA Science Archive after re-processing ALMA band 3 observationstargeting a different source. Assuming the line corresponds to the CO(6 →
5) transition, this tenta-tive detection implies a spectroscopic redshift of z = 5 . Keywords:
High-redshift galaxies — Redshift surveys — Galaxy evolution — Molecular gas —Millimeter Astronomy — Ultraluminous infrared galaxies INTRODUCTIONThe three millimeter window has recently been exploited to search for high-redshift dusty star-forming galaxies(DSFGs) through the detection of their dust continuum emission (e.g. Zavala et al. 2018a, 2021; Gonz´alez-L´opezet al. 2019). Thanks to the negative K -correction at this wavelength, 3 mm-selected galaxies are expected to lieat z > α = 10:02:36.8, δ = +02:08:40.6), was serendipitously discovered ∼ (cid:48)(cid:48) away from the phase center of an ALMA band3 observation aimed at detecting CO in a galaxy at z ∼ . σ with a 3 mm flux density of S = 150 µ Jy, and it is just marginally detectedat the ≈ − σ level in Spitzer /IRAC, SCUBA-2 850 µ m, and VLA 3 GHz (after deblending the original maps withthe ALMA positional prioir). The system drops out of deep optical imaging, including HST /CANDELS, Subaru,UltraVISTA, and it is also undetected in
Spitzer /MIPS and in all the
Herschel bands (Williams et al. 2019). All thesephotometry datapoints were used by Williams et al. (2019) to constrain the SED of the galaxy and its photometricredshift. The SED-fitting results suggest a high-redshift solution of z phot = 5 . +1 . − . , with a 90% probability of lying at z > . ARCHIVAL DATAWith the aim of finding additional data to further constrain the redshift of this galaxy, I search for public ALMAobservations around the position of “COS-3mm-1” using the ALMA Science Archive Query. Besides the original datain which the source was identified (ALMA project code: 2018.1.01739.S; PI: C. Williams), an additional program
Corresponding author: Jorge A. [email protected] a r X i v : . [ a s t r o - ph . GA ] F e b Figure 1.
Band 3 spectrum extracted at the position of “COS-3mm-1”. The orange line represents the best-fit Gaussianfunction to the tentative emission line. The center frequency of the line, the line-width, and the integrated line flux density arealso indicated in the figure. The inset (black and white) plot shows the 3 mm continuum detection from Williams et al. (2019)with blue contours indicating the moment zero map of the line at 1, 2, and 3 σ . (2015.1.00861.S; PI: J. Silverman) was publicly available in the archive (see Silverman et al. 2018 for details on thedata). These ALMA observations are centered ∼ (cid:48)(cid:48) away from the position of the source of interest. The primarybeam response at the position of “COS-3mm-1” is around 0 .
29. The data, which cover the frequencies 85 . − . . − . − channel width was around 0.05 mJy/beam. RESULTSA tentative emission line was found at ν = 110 .
84 GHz (Figure 1). The peak SNR is estimated to be around 3 . ≈ .
7. The line is relatively well fit with a Gaussian function with line-widthof 550 ±
110 km s − and integrated flux density of 0 . ± .
02 Jy km s − . As revealed by the line moment-0 map shownin Figure 1, The spatial location of the line is coincident with the 3 mm continuum detection from Williams et al.(2019), implying that, if real, this emission line arises from the 3 mm-selected galaxy. Assuming the line is real andthat it corresponds to a transition of carbon monoxide (the most common emission line in DSFGs), the most likelyredshift solutions based on its photometric redshift constraints are z = 4 . . . CO(5 → CO(6 → CO(7 →
6) transitions, respectively. Here, I adopt z = 5 . ± . as theredshift of the source since it is the closest solution to the maximum of the posterior redshift distribution found byWilliams et al. 2019. Interestingly, “MAMBO-9”, another DSFGs ∼ . z = 5 .
850 (Casey et al. 2019). The confirmation of more galaxies with similar redshifts within this region of the sky might imply theexistence of a large-scale galaxy proto-cluster structure.
The CO(6 →
5) line luminosity of L (cid:48) CO(6 → = (2 . ± . × K km s − pc can be used to calculate themolecular gas mass of “COS-3mm-1” by, first, estimating the CO(1 →
0) line luminosity and, then, assuming a CO-to-H conversion factor. I assume the line luminosity ratio reported by Bothwell et al. (2013) ( L (cid:48) CO(6 → /L (cid:48) CO(1 → =0 . ± .
04) and a CO-to-H conversion factor of α CO = 4 . (cid:12) K km s − pc . Based on these assumptions, I infer agas mass of M gas = (6 ± × M (cid:12) , which implies a depletion time scale ( t dep ≡ M gas / SFR) of ∼
200 Myr (adoptingthe star formation rate of ∼
300 M (cid:12) yr − derived by Williams et al. 2019). This value is higher than the typical gasdepletion timescales estimated for the bright population of high-redshift DSFGs, with values on the order of a few tensof Myrs (e.g. Aravena et al. 2016). Hence, “COS-3mm-1” is probably undergoing a long-lived star formation phaseinstead of the typical short-lived starburst episodes triggered by galaxy mergers (see discussion by Jim´enez-Andradeet al. 2020). This relatively long gas depletion timescale is in better agreement with those estimated for isolatedstar-forming disks at lower redshifts, and might imply that the population of faint DSFGs (with SFR (cid:46)
500 M (cid:12) yr − )forms stars through a smooth star formation mode over hundreds of Myrs.Nevertheless, the low significance of this detection prevents a robust confirmation of the redshift of this source and itsphysical properties. Follow-up observations are thus required not only to confirm this line, but also to identify a secondline to unambiguously constrain its redshift. At z = 5 .
85 the CO(5 →
4) transition is redshifted to ∼ . CO(5 →
4) isbetter suited for NOEMA (NOrthern Extended Millimeter Array) or LMT (Large Millimeter Telescope) observations,although its detection might require exposure times of several hours. The CO(7 →
6) transition is redshifted to ∼ . µ m transition will be observed at ∼ . µ m or the[OI]-63 µ m, which have previously been detected in high-redshift galaxies (Hashimoto et al. 2018; Rybak et al. 2020),also lie within the ALMA coverage, but their expected line luminosities are more uncertain and the lines are redshiftedinto the high-frequency ALMA bands which require very good observing conditions. Alternatively, the redshift of thissource can be confirmed with JWST observations, through the detection of rest-frame ultraviolet and optical emissionlines, such as Ly α , [OII]3727˚ A , H β , [OIII]4959 , A , and/or H α . CONCLUSIONSThe tentative emission line reported in this work suggests that, if real, “COS-3mm-1” is the highest redshift galaxyselected at 3 mm and one of the highest DSFGs known to-date. The only three DSFGs identified via (sub-)millimeterobservations with higher spectroscopic redshifts are all gravitationally lensed and, with the exception of G09-83808(Zavala et al. 2018b), they are rare extreme galaxies with SFRs exceeding 1000 M (cid:12) yr − (Riechers et al. 2013; Strandetet al. 2017). Therefore, this source could, potentially, serve as an important benchmark for z > z = 5 .
85 in this field. The confirmation ofits redshift and its physical properties is thus imperative.ACKNOWLEDGMENTSJorge A. Zavala thanks Christina Williams, Caitlin Casey, and Sinclaire Manning for their feedback on thismanuscript. DATA AVAILABILITYThis paper makes use of the following ALMA data: ADS/JAO.ALMA85 in this field. The confirmation ofits redshift and its physical properties is thus imperative.ACKNOWLEDGMENTSJorge A. Zavala thanks Christina Williams, Caitlin Casey, and Sinclaire Manning for their feedback on thismanuscript. DATA AVAILABILITYThis paper makes use of the following ALMA data: ADS/JAO.ALMA