ALMA Lensing Cluster Survey: a strongly lensed multiply imaged dusty system at z\geq6
N. Laporte, A. Zitrin, R. S. Ellis, S. Fujimoto, G. Brammer, J. Richard, M. Oguri, G. B. Caminha, K. Kohno, Y. Yoshimura, Y. Ao, F. E. Bauer, K. Caputi, E. Egami, D. Espada, J. González-López, B. Hatsukade, K. K. Knudsen, M. M. Lee, G. Magdis, M. Ouchi, F. Valentino, T. Wang
MMNRAS , 1–10 (2015) Preprint 7 January 2021 Compiled using MNRAS L A TEX style file v3.0
ALMA Lensing Cluster Survey: a strongly lensed multiply imageddusty system at 𝑧 ≥ N. Laporte, , ★ A. Zitrin, R. S. Ellis, S. Fujimoto, G. Brammer, J. Richard, M. Oguri, , , G. B. Caminha, K. Kohno, , Y. Yoshimura, Y. Ao, , F. E. Bauer, , K. Caputi, , E. Egami, D. Espada, , , J. González-López , B. Hatsukade, K. K. Knudsen, M. M. Lee, G. Magdis, M. Ouchi, , F. Valentino, T. Wang. , Kavli Institute for Cosmology, University of Cambridge, Madingley Road, Cambridge CB3 0HA, UK Cavendish Laboratory, University of Cambridge, 19 JJ Thomson Avenue, Cambridge CB3 0HE, UK Physics Department, Ben-Gurion University of the Negev, P.O. Box 653, Beer-sheva 8410501, Israel Department of Physics and Astronomy, University College London, Gower Street, London WC1E 6BT, UK Cosmic Dawn Center (DAWN), Copenhagen, Denmark Univ Lyon, Univ Lyon1, Ens de Lyon, CNRS, Centre de Recherche Astrophysique de Lyon UMR5574, F-69230, Saint-Genis-Laval,France Kavli Institute for the Physics and Mathematics of the Universe (WPI), The University of Tokyo, 5-1-5 Kashiwanoha, Kashiwa-shi, Chiba, 277-8583, Japan Department of Physics, University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033 Japan Research Center for the Early Universe, University of Tokyo, Tokyo 113-0033, Japan Kapteyn Astronomical Institute, University of Groningen,Postbus 800, 9700 AV Groningen, The Netherlands Institute of Astronomy, Graduate School of Science, The University of Tokyo, 2-21-1 Osawa, Mitaka, Tokyo 181-0015, Japan Institute for Cosmic Ray Research, The University of Tokyo, 5-1-5 Kashiwanoha, Kashiwa, Chiba 277ĂŞ8582, Japan Purple Mountain Observatory and Key Laboratory for Radio Astronomy, Chinese Academy of Sciences, Nanjing, China School of Astronomy and Space Science, University of Science and Technology of China, Hefei, China National Astronomical Observatory of Japan, 2-21-1 Osawa, Mitaka, Tokyo 181-8588, Japan Department of Space, Earth and Environment, Chalmers University of Technology, Onsala Space Observatory, SE-43992 Onsala, Sweden Instituto de Astrofísica, Facultad de Fśica, Pontificia Universidad Catolica de Chile Av. Vicuña Mackenna 4860, 782-0436 Macul,Santiago, Chile Millennium Institute of Astrophysics (MAS), Nuncio Monse nor Santero Sanz 100, Providencia, Santiago, Chile Department of Astronomical Science, SOKENDAI (The Graduate University of AdvancedStudies), 2-21-1 Osawa, Mitaka, Tokyo 181-8588, Japan SKA Organization, Lower Withington, Macclesfield, Cheshire SK11 9DL, UK Steward Observatory, University of Arizona, 933 North Cherry Avenue, Tucson, AZ 85721, USA Max-Planck-Institut für Extraterrestrische Physik (MPE), Giessenbachstr., D-85748 Garching, Germnay Núcleo de Astronomía de la Facultad de Ingeniería y Ciencias, Universidad Diego Portales, Av. Ejército Libertador 441, Santiago, Chile Las Campanas Observatory, Carnegie Institution of Washington, Casilla 601, La Serena, Chile
Accepted XXX. Received YYY; in original form ZZZ
ABSTRACT
We report the discovery of an intrinsically faint, quintuply-imaged, dusty galaxy MACS0600-z6 at a redshift 𝑧 = 𝑧 =0.46). A (cid:39) 𝜎 dustdetection is seen at 1.2mm as part of the ALMA Lensing Cluster Survey (ALCS), an on-goingALMA Large program, and the redshift is secured via [C II] 158 𝜇 m emission described ina companion paper. In addition, spectroscopic follow-up with GMOS/Gemini-North shows abreak in the galaxy’s spectrum, consistent with the Lyman break at that redshift. We use adetailed mass model of the cluster and infer a magnification 𝜇 (cid:38)
30 for the most magnifiedimage of this galaxy, which provides an unprecedented opportunity to probe the physicalproperties of a sub-luminous galaxy at the end of cosmic reionisation. Based on the spectralenergy distribution, we infer lensing-corrected stellar and dust masses of 2 . + . − . × and4 . + . − . × M (cid:12) respectively, a star formation rate of 9 . + . − . M (cid:12) yr − , an intrinsic size of0 . + . − . kpc, and a luminosity-weighted age of 200 ±
100 Myr. Strikingly, the dust productionrate in this relatively young galaxy appears to be larger than that observed for equivalent,lower redshift sources. We discuss if this implies that early supernovae are more efficient dustproducers and the consequences for using dust mass as a probe of earlier star formation.
Key words: galaxies: formation — galaxies: evolution — galaxies: high-redshift gravitationallensing: strong ★ E-mail: [email protected] © a r X i v : . [ a s t r o - ph . GA ] J a n N. Laporte et al.
Understanding the process of cosmic reionisation, during which in-tergalactic hydrogen transforms from a neutral to an ionised state inless than a billion years, represents a major challenge in extragalac-tic astronomy and cosmology. Although star-forming galaxies areconsidered to be the primary ionising sources, those which can bedirectly observed during the reionisation era appear to be insuffi-cient in number given their likely ionisation output (e.g. Robertsonet al. 2015). As a consequence, it is popular to appeal to contribu-tions from a larger number of sub-luminous sources, which can onlybe partially probed via current deep surveys (e.g., Bouwens et al.2015, Livermore et al. 2017, Bhatawdekar et al. 2019, Kikuchiharaet al. 2020 ). With current facilities, the only way to characterisethe physical properties of this sub-luminous population is to usethe magnification afforded by gravitational lensing (for a reviewsee Kneib & Natarajan 2011). Although several surveys have har-nessed the lensing power of foreground clusters, such as the ClusterLensing And Supernova survey with Hubble (CLASH ;Postmanet al. 2012), the
Hubble
Frontier Fields (Lotz et al. 2017) andmost recently REionization LensIng Cluster Survey (RELICS ; Coeet al. 2019), the effective surface area explored for the most highly-magnified background sources remains small. As a consequence,very few high-redshift sources ( 𝑧 ≥
6) with magnifications 𝜇 ≥ < (cid:12) perevent, e.g. Indebetouw et al. 2014). However, little is known aboutdust production rates at early times and thus correlating dust prop-erties of early galaxies over the full range of stellar masses withstellar ages inferred independently from spectral energy distribu-tions (SEDs) would be highly valuable. To date there are very fewconvincing dust detections beyond 𝑧 (cid:39) uv > − 𝜆 ∼ 𝜎 lensing-corrected rms of ≤ (or 0.01 mJy over the 0.5 arcmin for the high-est magnification regions). In this paper, we report the discovery ofdust-emission from a multiply-imaged lensed system MACS0600-z6 at 𝑧 = 𝜇 (cid:38)
30 for themost magnified arc. Section 4 presents spectroscopic follow-up withGemini-North of several images, and we discuss the physical prop-erties of this system in Section 5. Throughout this paper we assumeH = . s − . Mpc − , Ω 𝑚 =0.3 and Ω Λ =0.7. All magnitudes arein AB system (Oke & Gunn 1983). Image 𝜆 𝑐 t 𝑒𝑥𝑝 ZP 2 𝜎 -depth[Å] [ks] [AB] [AB]F435W 4317.5 1.95 25.66 27.4F606W 5924.7 2.18 26.50 28.3F814W 8210.3 3.56 25.94 28.0F105W 10530.9 1.41 26.27 27.1F125W 12495.7 0.7 26.23 26.5F140W 13976.1 0.74 26.45 26.5F160W 15433.1 1.96 25.95 27.4Ks 21440.4 2.16 30.05 24.4IRAC1 35465.6 15.2 23.9 26.0IRAC2 45024.3 15.2 23.9 25.6 Table 1.
Properties of data used in this study. The 2 𝜎 -depth is measuredin a 0.4” diameter aperture for HST , 0.8” radius aperture for Ks, and 1.2”radius aperture for
Spitzer . The cluster MACSJ0600.1-2008 ( 𝑧 =0.46) was observed as partof the ALMA Lensing Cluster Survey (ID: 2018.1.00035.L, P.I.Kohno, K.) in band 6 frequency in January 2019 with a total ex-posure time of 3.3 hrs. The data was reduced using the ALMApipeline (v. 5.6.1-8) with a natural weighting. A 2 (cid:48)(cid:48) uv -taper wasused to maximise the detection of faint sources. The final beamsize is 2 . (cid:48)(cid:48) × . (cid:48)(cid:48)
99 and the rms on the reduced mosaic is 80 𝜇 Jy/beam. 13 sources with a signal-to-noise (SNR) ≥ − . (cid:48)(cid:48)
3. By extracting the fluxon a line-free image at 1.2mm using the CASA imfit task, thepeak flux is S peak1 . =366 ± 𝜇 Jy . beam − and the integrated flux isS int1 . =484 ± 𝜇 Jy on an emission line subtracted image.A search for an ultraviolet (UV) rest-frame counterpart for thissource was undertaken on deep
HST images from the RELICSsurvey with a search radius of 2 (cid:48)(cid:48) . The data reduction of thisdataset is described in Coe et al. (2019). Photometric catalogueswere built using version 2.19.5 of
SExtractor
Bertin & Arnouts(1996) in dual image mode on psf matched images. The extractionparameters were defined to maximise the detection of faint objects(DETECT_THRESH = 2 𝜎 ; ANALYSIS_THRESH = 2 𝜎 ; DE-TECT_MINAREA= 5 px) on a detection picture composed by thesum of all WFC3 images. The final catalogue contains 25200 de-tections. To estimate the depth of each image, we masked the brightsources and measured the rms in hundreds non-overlapping 0.4”radius aperture. Table 1 summarises the data properties. Two over-lapping sources were detected 0 . (cid:48)(cid:48) phot =0.82 + . − . . The multiple image MACS0600-arc isabsent from that pipeline-generated catalogue, most likely becauseof its faintness and elongated shape.Noting the overlapping foreground object, in order to extractthe photometry of MACS0600-arc we model each source with aSersic profile using GALFIT (Peng et al. 2002) on the WFC3-NIR and IRAC images, which accounts for PSF difference betweenHST, VLT and Spitzer. Error bars are measured on the resid-ual image in a 0 . (cid:48)(cid:48) MNRAS , 1–10 (2015) usty lensed system at 𝑧 ≥ F435W F606W F814W F105W F125W F140W F160W K s Figure 1.
Thumbnail images of MACS0600-arc comprising images from ACS/
HST (F435W, F606W, F814W), WFC3-IR/HST (F105W, F125W, F140W andF160W), HAWK-I/VLT(K 𝑠 ), IRAC/ Spitzer (3.6 and 4.5 𝜇 m) and ALMA (Band 6). Each image is 7 (cid:48)(cid:48) × (cid:48)(cid:48) . Note the overlap between the foreground object(detected at F606W and F814W) and the lensed galaxy. The ALMA contours are drawn on the F160W images in 1 𝜎 from 3 𝜎 . The ALMA beam size is at thebottom right corner of the last column. ACS data where only the foreground source is detected and com-pared the GALFIT photometry with that derived using the standardMAG_AUTO photometry in single image mode with SExtractor(Bertin & Arnouts 1996). In F606W, our extracted photometry(m
GALFITF606W = . ± .
35) is in excellent agreement with the
SEx-tractor photometry (m
SExtractorF606W = . ± . 𝜎 depth in 1.2” radius aperture of24.4. We also searched in the Spitzer archive for deep IRAC 3.6 and4.5 𝜇 m exposures. This cluster was part of two observing programs(ID: 90218, P.I. : E. Egami and ID: 12005, P.I. : M. Bradac), wecombined all exposures and measured the depth in a 1.2” radiusaperture distributed over the field of view. We present the finalphotometry in Table 3.We determine the photometric redshift of the foregroundsource from a template based SED-fitting method with Hyperz(Bolzonella et al. 2000). We used the standard list of templatesincluding evolutionary SEDs with Chabrier IMF (Chabrier 2003)and solar metallicity from Bruzual & Charlot (2003), empiricalSEDs built by Coleman et al. (1980) and two starburst galaxiesfrom Kinney et al. (1996). We then searched for a solution between0 < 𝑧 <
8, with an extinction range 0 . < A v < . 𝑧 ∈ [0.0:3.0]. The best SED-fitfor the foreground source is obtained at z phot = . + . − . and mod-est reddening (A v =1.5mag). Allowing a larger redshift range, theMACS0600-arc data provides a solution at 𝑧 𝑝ℎ𝑜𝑡𝑜 =6 . + . − . withan extinction A v =0.5mag, broadly consistent with the spectroscopicvalue (Figure 2). The elongated shape of the prominent 𝑧 ∼ in preparation ),and here we provide only a brief summary. For convenience in thissection we drop the prefix MACS0600 in describing the multipleimages.The three models used here are produced using GLAFIC Figure 2.
Spectral Energy Distribution (SED) of MACS0600-z6 showingthe best-fit from MAGPHYS and the best physical properties corrected formagnification. (Oguri 2010), Lenstool (Jullo et al. 2007), and Light-Traces-Mass(LTM; Zitrin et al. 2015). The first two models can be regarded asparametric, in the sense that both the galaxies and their dark matterhalos are described using independent analytic forms (e.g., com-binations of pseudo isothermal elliptical mass distributions) andadopt empirical scaling relations for cluster members. The LTMmethod is similar, but here the dark matter is assumed to follow thelight distribution and thus is modeled as a smoothed version of theluminosity-weighted galaxy distribution. The parametric modelingof MACS0600 typically included three to four dark matter halosand, whereas for the LTM models, the masses of several brightcluster members were modeled independently. All models are mini-mized using available multiple image constraints to find the best-fitsolution and its associated errors. Four modellers (MO, JR, GC,AZ) voted on the choice of multiply-imaged systems which, besidesthe 𝑧 ∼ 𝑧 ∼ 𝑧 ∼ MNRAS , 1–10 (2015)
N. Laporte et al. fifth image on the other side of the cluster to the west (z6.5). Thestamps of the 5 images are shown in Fig. 3, and the configurationderived from the LTM model for example, including all five mul-tiple images, is shown on a
HST image of the cluster in Fig. 4. Toconfirm the reliability of this system, we first check that the colorsare similar for the 4 images within the error bars. We only includeone of the fifth image candidates in this analysis (z6.5c) becauseHST coverage is missing for the other two. Colors measured onWFC3 images are similar with averaged values of F125W-F160W= 0.32 ± ± < . (cid:48)(cid:48) HST data, image z6.3consists of a bright "bulge" referred to here as z6.3a and an ad-ditional faint component z6.3b which, according to the three lensmodels, is likely the multiply-imaged counterpart of the arc, and onwhich the ALMA signal is centered. It is unclear whether imagez6.3a is truly part of the lensed source despite not being quintuplyimaged as well, and – while we acknowledge the possibility thatz6.3a may be related – we consider only z6.3b a genuine multi-ple image of this system. In the ALMA data, z6.3b also shows afaint dust detection with a S/N ∼ peak = ± 𝜇 Jy . beam − as measured with imfit). The 1.2mm flux ratio between 1.3b andthe arc is 0.62 + . − . in excellent agreement with ratios observed inthe near infrared and with the predicted lens-model magnifications(see below). [CII]158 𝜇 m emission lines at 𝑧 = .
07 are detected inall images of this system that are covered by ALMA’s FOV: the arc,z6.3b and z6.4, strengthening further the adopted lensing configura-tion, again with line ratios that commensurate with the expectationsfrom the lens models. More extensive details on these detectionsare given in a companion paper (Fujimoto et al., submitted).While all models reassuringly agree on the positions of the fivemultiple images, there is larger scatter ( ∼ (cid:48)(cid:48) ) in the predicted loca-tion of the fifth image, z6.5. The GLAFIC and LTM models predictimage z6.5 lies outside the region covered by WFC3 data, makingits identification more challenging, and two objects undetected onthe ACS images but seen on the HAWK-I data are considered aspotential candidates (z6.5a and z6.5b). The Lenstool model pre-dicts the fifth image within the WFC3 field, and one clear dropoutis identified as a potential candidate using the ACS, WFC3 andHAWK-I data (MACS0600-z6.5c).Magnification factors 𝜇 for the five images were estimatedas follows. The two parametric models (Lenstool and GLAFIC)extracted values by first planting a compact source in the expectedposition, and adjusting its exact position with respect to the criticalcurves to match the observed [CII] 158 𝜇 m line flux ratio. The LTMmagnification values were derived in a slightly different manner:here the source was not planted in the source plane but formed thereby directly delensing the arc, and the magnification of the arc wasthen extracted by comparing the model’s prediction for the threeother multiple image sizes with their absolute magnification values.The magnification for the five images from the GLAFIC model is 𝜇 ∼
33 for the arc, 𝜇 ∼
35 for z6.3, 𝜇 ∼ . 𝜇 ∼
14 for z6.5. The Lenstool model yields 𝜇 ∼ . 𝜇 ∼ 𝜇 ∼ . 𝜇 ∼ . 𝜇 ∼ . 𝜇 ∼ . 𝜇 ∼ . 𝜇 ∼ . 𝜇 ∼ −
200 are obtained,although such estimates are typically more uncertain.We also take benefit from the detection of the dust continuumin two of the multiple images (z6.3 and the arc) to estimate thephysical offset between the FIR continuum and the UV continuum.The position of the UV continuum was estimated on the F160Wimage using the centroid of the counterpart. In the image plan wemeasured an observed offset of 0 . (cid:48)(cid:48)
24 and 0 . (cid:48)(cid:48)
35 respectively forz6.3b and the arc. Accounting for the magnification, we estimate aphysical offset of < The detection of an emission line within the ALMA band 6 datacubesuggests a redshift 𝑧 𝑠 𝑝𝑒𝑐 =6.07, assuming the line is [CII]158 𝜇 m(Fujimoto et al. submitted). However, with only one line detected,a low/intermediate redshift interloper cannot be excluded. The rel-atively shallow MUSE data used to constrain the mass models didnot show any features for the multiple-images of this system. There-fore we conducted a spectroscopic follow-up campaign with theGemini Multi-Object Spectrograph (GMOS - Hook et al. 2004)installed on Gemini-North on three images of the system, namelyMACS0600-arc, MACS0600-z6.3a and MACS0600-z6.4. Obser-vations were done in service mode on the 18 𝑡ℎ and 19 𝑡ℎ October2020 (ID : GN-2020B-Q-903 ; P.I. : A. Zitrin). We secured 4.5hrsreaching a 1-sigma sensitivity of 8.9 × − erg/s/cm over thewavelength range 505nm to 980nm. We reduced the data using theGemini IRAF package, as recommended by the instrument team.We follow the standard reduction procedure including bias subtrac-tion, flat fielding, wavelength calibration from the illumination ofour mask by the CuAr lamp and flux calibration from the whitedwarf Wolf 1346.Interestingly, the continuum of the more compact and brightestimage (MACS0600-z6.3a) is visible on the Gemini spectra. Weextracted the spectrum in a 1.5 × seeing diameter aperture and a breakis clearly identified between 850nm and 900nm. To improve thesignal, we binned the spectra in the spectral direction with a binningfactor of 30 (see Figure 5). We determined the spectroscopic redshiftby fitting our observed spectra with a stacked spectra coming from81 LBGs (Jones et al. 2012). We searched for the best-fit using a 𝜒 minimization technique over a redshift range between 𝑧 = 𝑧 =
8. The best fit is found at 𝑧 𝑠 𝑝𝑒𝑐 =6.19 + . − . , consistent with boththe photometric redshift and the [CII]158 𝜇 m emission line detectedin the ALMA data. One can also argue that if the break seen in theGemini data is the Balmer Break at 𝑧 ∼ 𝑧 ∼
6, we would have detected the [OII]3727,3729 doublet,which makes the high-redshift identification for this system robust.Moreover, after a careful visual inspection of the 2D spectra, noemission line is detected in any of the 3 images allowing us to placea firm upper limit on the Ly- 𝛼 luminosity in this 𝑧 ∼ 𝛼 , 2 𝜎 ) < × erg/s, corresponding to a rest-frame EW of < 𝜇 and EW<3.6Å assuming a magnification factor of 𝜇 =31. MNRAS , 1–10 (2015) usty lensed system at 𝑧 ≥ F435W F606W F814W F105W F125W F140W F160W 4.5μm3.6μm
Arc z6.3a-z6.3b z6.4z6.5
Figure 3.
Thumbnail images of all images of the quintuply lensed galaxy presented in this paper. Each stamp is 7 . (cid:48)(cid:48) × . (cid:48)(cid:48) arcz6.3a – z6.3bz6.4z6.5 : . . . . - : : . . . . z6.3a-z6.3b z6.4arc z6.5 Figure 4. ( Right ) Position of the five images of MACS0600-z6 superimposed on a colour HST image (F435W - blue ; F814W - green ; F160W - red). Thewhite contours from the LTM model are overplotted. (
Left ) Labelled F160W stamps of four multiple images. MACS0600-z6.1 and MACS0600-z6.2 (top left)are a merging pair crossing the critical line, and seen on HST images as an elongated arc and referred in the text as MACS0600-arc. MACS0600-z6.3 comprisesa bright object (z6.3a) and a fainter, fuzzy part (z6.3b), which is the quintuply imaged part as indicated by the lens models, and on which the line emissionseems to be concentrated. ALMA contours are overplotted on the arc stamp from 3 𝜎 and from 2 𝜎 for z6.3b.MNRAS , 1–10 (2015) N. Laporte et al.
Figure 5. ( Top: ) 2D GMOS/Gemini spectrum of MACS0600-z6.3asmoothed with a boxcar of 5pixels. The red rectangle shows the positionof our object in the slit. The continuum is clearly visible at the largerwavelength. (
Bottom: ) The blue line displays the extracted spectrum ofMACS0600-z6.3a within a 1.5 × seeing diameter aperture. The grey spec-trum is the best-fit of the Jones et al. (2012) spectra to our data. The best fitis found at 𝑧 𝑠𝑝𝑒𝑐 =6.19 + . − . . We determine the intrinsic (lensing-corrected) physical propertiesfor those two images with a clear dust detection (MACS0600-arcand MACS0600-z6.3b) using MAGPHYS (da Cunha et al. 2008a),which includes FIR models to fit the dust properties, while we useBAGPIPES (Carnall et al. 2018) to provide constraints for the re-maining images. For the BAGPIPES runs, we assume a constantStar Formation History (SFH) with a stellar mass ranging from 10 to 10 M (cid:12) , and an age ranging from 0.0 Gyr to the age of the Uni-verse at 𝑧 = .
07 . For each image we adopt the magnification fromthe LTM model, based upon the spectroscopic redshift, although asseen in the previous section the magnifications from the three mod-els seem to agree fairly well. Reassuringly, we determine that allimages have similar properties. Excluding image MACS0600-z6.3a,which does not seem to contribute significantly to the flux comparedto MACS0600-z6.3b, the mean stellar mass is 2.9 + . − . × M (cid:12) ,the mean SFR is 9.7 + . − . M (cid:12) /yr and the mean dust mass (com-puted only from images MACS0600-arc and MACS0600-z6.3b) is4.8 + . − . × M (cid:12) , where error bars take into account uncertaintieson the magnification factor. The latter is obtained assuming a dusttemperature of 30K, a dust emissivity of 𝛽 =2.0 and the mean mag-nification. Including Herschel /SPIRE 3 𝜎 upper limit(13.9mJy/ 𝜇 at250 𝜇 m), we can rule out T 𝑑𝑢𝑠𝑡 ≥
85K (Sun et al. in prep). We mea-sure the UV slopes of all images from the best SED-fit and find anaverage 𝛽 = − ± SExtractor half light radius and the
Hubble
PSF. We apply thismethod on z6.4 since this image is not blended with other sources.After correcting for magnification, we find an intrinsic size of 0 . (cid:48)(cid:48) + . − . kpc at 𝑧 ∼ Source M ★ M dust SFR 𝛽 𝜇 [ × M (cid:12) ] [ × M (cid:12) ] [M (cid:12) / yr]arc 2.2 + . − . + . − . + . − . − ± + − z6.3a 2.7 + . − . - 15.3 + . − . − ± (13 + − ) z6.3b 1.1 + . − . + . − . + . − . − ± + − z6.4 3.7 + . − . - 10.3 + . − . − ± + . − . z6.5 4.7 + . − . - 10.7 + . − . − ± + . − . Table 2.
Physical properties of all multiple images of MACS0600-z6 com-puted with MAGPHYS (da Cunha et al. 2008a) for sources with dust detec-tion (arc and z6.3b) and BAGPIPES (Carnall et al. 2018) for the remaining.Uncertainties for the LTM model magnification represent the range of val-ues predicted by four other trial LTM models. The central value is obtainedby averaging the parameter value of the best fit-model obtained assumingthe mean, min and max magnification value and the error bars includeduncertainties on the magnification factor
Figure 6.
Dust to stellar mass ratio as a function of the stellar mass. The reddot shows the properties of MACS0600-z6, blue dots represent previouslyobserved galaxies with dust constraints (detections or upper limits) with aspectroscopic redshift between 𝑧 ∼ 𝑧 ∼ 𝑧 = . We can compare the above results for MACS0600-z6 to similarspectroscopically confirmed galaxies at 𝑧 ∼ MNRAS , 1–10 (2015) usty lensed system at 𝑧 ≥ mass ratio is somewhat larger than that observed in other 𝑧 ∼ 𝜇 m bands,since the 3.6 𝜇 m band could be contaminated by strong [O III] emis-sion (Labbé et al. 2013). Following the methodology discussed inRoberts-Borsani et al. (2020), we estimate an age of 200 ±
100 Myr,indicative of a formation redshift of 𝑧 form =7.2 + . − . , comparable tovalues estimated for the other 𝑧 ∼ + . − . M (cid:12) of dust. Although thisestimate is highly uncertain given the uncertain production rate ofSNe in early metal poor galaxies and possible dust destruction andejection processes in low mass galaxies at high redshift, it is perhapssomewhat larger than that derived locally ( < (cid:12) , see e.g. Cher-chneff & Dwek 2010, Indebetouw et al. 2014, Gomez et al. 2012,Gall et al. 2011). Recent chemical evolution models predict dust-to-mass ratios similar to that observed for MACS0600-z6 could beachieved in ≤
200 Myr (Calura et al. 2017). Even higher values havealso been observed recently in MAMBO-9 (Casey et al. 2019), anintensely star-forming sub-mm galaxy whose properties are other-wise quite distinct from those of MACS0600-z6. Applying the samemethod, MAMBO-9 may be even more efficient to produce dust,but the different nature of these two objects makes the comparisonof their dust production efficiency difficult.The ratio between the IR and UV luminosity, often referredto as IRX=log ( 𝐿 𝐼 𝑅 / L UV ), can be compared to the UV continuumslope 𝛽 to offer insight into the dust extinction law at early cos-mic epochs. Bouwens et al. (2016) found that 𝑧 > [ 𝑀 ★ < 𝑧 ≥ 𝑧 ∼ 𝑧 ∼ 𝑧 ∼ Recent studies have suggesteded that the dust temperature may behigher at high-redshift than what is measured at lower redshift. Ob-taining an accurate measurement of the dust temperature in high-redshift galaxies requires to constrain the red slope and the peakof the FIR emission (see Figure 1 of da Cunha et al. 2008b). Theevolution of the dust temperature with redshift can be studied atintermediate redshift combining data from
Herschel and ALMA.Schreiber et al. (2018) demonstrate that the dust temperature goesfrom 25K at 𝑧 ≤ 𝑧 ∼
5. At 𝑧 ≥ 𝑧 ≥ Figure 7.
The IR/UV luminosity ratio IRX as a function of the UV slope 𝛽 . Black points represent galaxies at 𝑧 ≥
7, blue points are 𝑧 ∼ low-redshift, but deeper data at 𝜆 ≥ 𝑧 ≥ 𝑇 𝑑𝑢𝑠𝑡 =45K for 0.1 𝜇 m grain, and could reach as high temperatureas 𝑇 𝑑𝑢𝑠𝑡 ≥
60K for smaller grains (Ferrara et al. 2017). In view ofthis, we explore whether a higher dust temperature would changeour conclusions. For a temperature of 85K, the estimated dust masswould decrease to 𝑀 𝑑𝑢𝑠𝑡 =6.3 + . . × M (cid:12) , and its dust-to-stellarmass ratio will be comparable to previous findings at high-redshift.Further ALMA observations at longer wavelength (e.g. in band 5)are needed to constrain the dust temperature and to refine our dustmass estimates. Moreover, adopting a higher dust temperature tostudy the IRX- 𝛽 𝑢𝑣 relation confirms our conclusion on the dust lawpreferred by our object, since an increase in dust temperature willincrease the dust luminosity without changing the UV luminosity,and will therefore tend to higher IRX, consistent with the Calzettilaw. Such intrinsically faint sources ( 𝑀 𝑢𝑣 =-19.9 ± 𝛼 emission line (either from LBG or LAE) using MUSE/VLT(e.g. de La Vieuville et al. 2019, de La Vieuville et al. 2020) or theHyper Suprime-Cam/Subaru (e.g. Konno et al. 2018) show that theluminosity of Ly- 𝛼 at 𝑧 ≥ to 10 . erg/s.The Ly- 𝛼 properties in MACS0600-z6 (L(2 𝜎 ) < × erg/s andEW 𝑅𝐹 <
113 Å ) are therefore not particularly exceptional andcomparable in luminosity and Ly- 𝛼 EW to the bulk of the galaxypopulation at 𝑧 ∼ 𝑧 = MNRAS , 1–10 (2015)
N. Laporte et al. physical properties we deduced from a detailed SED analysis showthat this galaxy has a stellar mass of ∼ M (cid:12) , a dust mass of ∼ M (cid:12) respectively, a small star formation rate ( < (cid:12) . yr − ) and asize ( ∼ ≤ ACKNOWLEDGEMENTS
We thank the referee for providing useful comments which improvedthe quality of this paper, the Gemini Helpdesk team for their helpwith the reduction of Gemini data and Ian Smail for useful commentson this manuscript. NL acknowledges support from the Kavli Foun-dation. RSE acknowledges funding from the European ResearchCouncil (ERC) under the European Unions Horizon 2020 researchand innovation program(grant agreement No. 669253). MO is sup-ported by World Premier International Research Center Initiative(WPI Initiative), MEXT, Japan, as well as KAKENHI Grant-in-Aidfor Scientific Research (A) (17H01114, 19H00697, and 20H00180)through Japan Society for the Promotion of Science (JSPS). KK andTW are supported by JSPS KAKENHI Grant Number JP17H06130and by the NAOJ ALMA Scientific Research Grant Number 2017-06B. FEB acknowledges support from ANID-Chile Basal AFB-170002, FONDECYT Regular 1200495 and 1190818, and Millen-nium Science Initiative ICN12_009. KKK acknowledges supportfrom the Knut and Alice Wallenberg Foundation. Y.A. acknowl-edges support by NSFC grant 11933011. GB and KC acknowledgefunding from the European Research Council through the award ofthe Consolidator Grant ID 681627-BUILDUP.This paper makes use of the following ALMA data:ADS/JAO.ALMA
DATA AVAILABILITY
The data underlying this article will be shared on reasonable requestto the corresponding author.
REFERENCES
Atek H., et al., 2015, ApJ, 814, 69 Bakx T. J. L. C., et al., 2020, MNRAS, 493, 4294Behrens C., Pallottini A., Ferrara A., Gallerani S., Vallini L., 2018, MNRAS,477, 552Bertin E., Arnouts S., 1996, A&AS, 117, 393Bhatawdekar R., Conselice C. J., Margalef-Bentabol B., Duncan K., 2019,MNRAS, 486, 3805Bolzonella M., Miralles J. M., Pelló R., 2000, A&A, 363, 476Bouwens R. J., et al., 2015, ApJ, 803, 34Bouwens R. J., et al., 2016, ApJ, 833, 72Bouwens R. J., Illingworth G. D., Oesch P. A., Atek H., Lam D., StefanonM., 2017, ApJ, 843, 41Bradač M., et al., 2017, ApJ, 836, L2Bruzual G., Charlot S., 2003, MNRAS, 344, 1000Calura F., et al., 2017, MNRAS, 465, 54Calzetti D., Armus L., Bohlin R. C., Kinney A. L., Koornneef J., Storchi-Bergmann T., 2000, ApJ, 533, 682Capak P. L., et al., 2015, Nature, 522, 455Carnall A. C., McLure R. J., Dunlop J. S., Davé R., 2018, MNRAS, 480,4379Casey C. M., et al., 2019, ApJ, 887, 55Chabrier G., 2003, PASP, 115, 763Cherchneff I., Dwek E., 2010, ApJ, 713, 1Coe D., et al., 2013, ApJ, 762, 32Coe D., et al., 2019, ApJ, 884, 85Coleman G. D., Wu C. C., Weedman D. W., 1980, ApJS, 43, 393Dudzeviči¯ut˙e U., et al., 2020, MNRAS, 494, 3828Ebeling H., Edge A. C., Henry J. P., 2001, ApJ, 553, 668Faisst A. L., Fudamoto Y., Oesch P. A., Scoville N., Riechers D. A., PavesiR., Capak P., 2020, arXiv e-prints, p. arXiv:2005.07716Ferrara A., Hirashita H., Ouchi M., Fujimoto S., 2017, MNRAS, 471, 5018Gall C., Andersen A. C., Hjorth J., 2011, A&A, 528, A13Gomez H. L., et al., 2012, ApJ, 760, 96Hashimoto T., et al., 2019, PASJ, 71, 71Hook I. M., Jørgensen I., Allington-Smith J. R., Davies R. L., Metcalfe N.,Murowinski R. G., Crampton D., 2004, PASP, 116, 425Indebetouw R., et al., 2014, ApJ, 782, L2Jones T., Stark D. P., Ellis R. S., 2012, ApJ, 751, 51Jullo E., Kneib J. P., Limousin M., Elíasdóttir Á., Marshall P. J., VerdugoT., 2007, New Journal of Physics, 9, 447Kanekar N., Wagg J., Chary R. R., Carilli C. L., 2013, ApJ, 771, L20Katz H., Laporte N., Ellis R. S., Devriendt J., Slyz A., 2019, MNRAS, 484,4054Kawamata R., Ishigaki M., Shimasaku K., Oguri M., Ouchi M., TanigawaS., 2018, ApJ, 855, 4Kikuchihara S., et al., 2020, ApJ, 893, 60Kinney A. L., Calzetti D., Bohlin R. C., McQuade K., Storchi-Bergmann T.,Schmitt H. R., 1996, ApJ, 467, 38Kneib J.-P., Natarajan P., 2011, A&ARv, 19, 47Knudsen K. K., Richard J., Kneib J.-P., Jauzac M., Clément B., Drouart G.,Egami E., Lindroos L., 2016, MNRAS, 462, L6Knudsen K. K., Watson D., Frayer D., Christensen L., Gallazzi A.,Michałowski M. J., Richard J., Zavala J., 2017, MNRAS, 466, 138Konno A., et al., 2018, PASJ, 70, S16Labbé I., et al., 2013, ApJ, 777, L19Laporte N., et al., 2017a, A&A, 604, A132Laporte N., et al., 2017b, ApJ, 837, L21Laporte N., et al., 2019, MNRAS, 487, L81Livermore R. C., Finkelstein S. L., Lotz J. M., 2017, ApJ, 835, 113Lotz J. M., et al., 2017, ApJ, 837, 97Marrone D. P., et al., 2018, Nature, 553, 51Oesch P. A., et al., 2010, ApJ, 709, L21Oguri M., 2010, PASJ, 62, 1017Oke J. B., Gunn J. E., 1983, ApJ, 266, 713Ouchi M., et al., 2013, ApJ, 778, 102Peng C. Y., Ho L. C., Impey C. D., Rix H.-W., 2002, AJ, 124, 266Postman M., et al., 2012, ApJS, 199, 25Roberts-Borsani G., Ellis R. S., Laporte N., 2020, arXiv e-prints, p.arXiv:2002.02968 MNRAS , 1–10 (2015) usty lensed system at 𝑧 ≥ Robertson B. E., Ellis R. S., Furlanetto S. R., Dunlop J. S., 2015, ApJ, 802,L19Salmon B., et al., 2018, ApJ, 864, L22Salmon B., et al., 2020, ApJ, 889, 189Schaerer D., Boone F., Zamojski M., Staguhn J., Dessauges-Zavadsky M.,Finkelstein S., Combes F., 2015, A&A, 574, A19Schreiber C., Elbaz D., Pannella M., Ciesla L., Wang T., Franco M., 2018,A&A, 609, A30Scoville N., et al., 2016, ApJ, 820, 83Sklias P., et al., 2014, A&A, 561, A149Smit R., et al., 2018, Nature, 553, 178Tamura Y., et al., 2019, ApJ, 874, 27Watson D., Christensen L., Knudsen K. K., Richard J., Gallazzi A.,Michałowski M. J., 2015, Nature, 519, 327Zitrin A., et al., 2015, ApJ, 801, 44da Cunha E., Charlot S., Elbaz D., 2008a, MNRAS, 388, 1595da Cunha E., Charlot S., Elbaz D., 2008b, MNRAS, 388, 1595de La Vieuville G., et al., 2019, A&A, 628, A3de La Vieuville G., et al., 2020, arXiv e-prints, p. arXiv:2009.12824MNRAS , 1–10 (2015) N . L apo r t ee t a l . ID RA DEC F435W F606W F814W F105W F125W F140W F160W K 𝑠 𝜇 m 4.5 𝜇 m z phot Foreground 06:00:09.13 − > ± ± ± ± ± ± ± ± ± + . − . MACS0600-arc ∗ − > > > ± ± ± ± > ± ± + . − . MACS0600-z6.3a ∗ − > > ± ± ± ± ± > ± ± + . − . MACS0600-z6.3b 06:00:09.53 − > > > ± ± ± ± > ± ± + . − . MACS0600-z6.4 ∗ − > > ± ± ± ± ± > > > + . − . MACS0600-z6.5a † − > > > ± > > † − > > > ± > > − > > ± ± ± ± ± ± > > + . − . Table 3.
Photometry of the multiple images of this 𝑧 ≥ (cid:48)(cid:48) radius aperture, on HAWK-I data in a 0.8 (cid:48)(cid:48) radius aperture (takinginto account the seeing of the data), and in a 1.2 (cid:48)(cid:48) radius aperture for IRAC images. For the photometric extraction of z6.3a and z6.3b, we used GALFIT on ACS and WFC3 images assuming a Sersic profile for both.On the IRAC data, where the separation is not clearly resolved, we used the flux ratio observed at 1.6 𝜇 m to separate the contribution of each source. ★ spectroscopically confirmed at 𝑧 = .
07 (see Fujimoto et al. submitted) † Candidates for the fifth image of this system. The absence of WFC3 data at the expected position makes difficult the clear identification of the image.This paper has been typeset from a TEX/L A TEX file prepared by the author. M N R A S , (2015