Euclid preparation: VIII. The Complete Calibration of the Colour-Redshift Relation survey: VLT/KMOS observations and data release
Euclid Collaboration, V.Guglielmo, R.Saglia, F.J.Castander, A.Galametz, S.Paltani, R.Bender, M.Bolzonella, P.Capak, O.Ilbert, D.C.Masters, D.Stern, S.Andreon, N.Auricchio, A.Balaguera-Antolínez, M.Baldi, S.Bardelli, A.Biviano, C.Bodendorf, D.Bonino, E.Bozzo, E.Branchini, S.Brau-Nogue, M.Brescia, C.Burigana, R.A.Cabanac, S.Camera, V.Capobianco, A.Cappi, C.Carbone, J.Carretero, C.S.Carvalho, R.Casas, S.Casas, M.Castellano, G.Castignani, S.Cavuoti, A.Cimatti, R.Cledassou, C.Colodro-Conde, G.Congedo, C.J.Conselice, L.Conversi, Y.Copin, L.Corcione, A.Costille, J.Coupon, H.M.Courtois, M.Cropper, A.Da Silva, S.de la Torre, D.Di Ferdinando, F.Dubath, C.A.J.Duncan, X.Dupac, S.Dusini, M.Fabricius, S.Farrens, P.G.Ferreira, S.Fotopoulou, M.Frailis, E.Franceschi, M.Fumana, S.Galeotta, B.Garilli, B.Gillis, C.Giocoli, G.Gozaliasl, J.Graciá-Carpio, F.Grupp, L.Guzzo, H.Hildebrandt, H.Hoekstra, F.Hormuth, H.Israel, K.Jahnke, E.Keihanen, S.Kermiche, M.Kilbinger, C.C.Kirkpatrick, T.Kitching, B.Kubik, M.Kunz, H.Kurki-Suonio, R.Laureijs, S.Ligori, P.B.Lilje, I.Lloro, D.Maino, E.Maiorano, C.Maraston, O.Marggraf, N.Martinet, F.Marulli, R.Massey, S.Maurogordato, E.Medinaceli, S.Mei, M.Meneghetti, R.Benton Metcalf, et al. (48 additional authors not shown)
AAstronomy & Astrophysics manuscript no. main © ESO 2020July 7, 2020
Euclid preparation: VIII. The Complete Calibration of theColour–Redshift Relation survey: VLT/KMOS observations anddata release
Euclid Collaboration: V. Guglielmo (cid:63) , R. Saglia , , F.J. Castander , , A. Galametz , S. Paltani , R. Bender , ,M. Bolzonella , P. Capak , , O. Ilbert , D.C. Masters , D. Stern , S. Andreon , N. Auricchio ,A. Balaguera-Antolínez , , M. Baldi , , , S. Bardelli , A. Biviano , , C. Bodendorf , D. Bonino , E. Bozzo ,E. Branchini , S. Brau-Nogue , M. Brescia , C. Burigana , , , R.A. Cabanac , S. Camera , , ,V. Capobianco , A. Cappi , , C. Carbone , J. Carretero , C.S. Carvalho , R. Casas , , S. Casas , M. Castellano ,G. Castignani , S. Cavuoti , , , A. Cimatti , , R. Cledassou , C. Colodro-Conde , G. Congedo ,C.J. Conselice , L. Conversi , , Y. Copin , L. Corcione , A. Costille , J. Coupon , H.M. Courtois , M. Cropper ,A. Da Silva , , S. de la Torre , D. Di Ferdinando , F. Dubath , C.A.J. Duncan , X. Dupac , S. Dusini ,M. Fabricius , S. Farrens , P. G. Ferreira , S. Fotopoulou , M. Frailis , E. Franceschi , M. Fumana , S. Galeotta ,B. Garilli , B. Gillis , C. Giocoli , , , G. Gozaliasl , , J. Graciá-Carpio , F. Grupp , L. Guzzo , , ,H. Hildebrandt , H. Hoekstra , F. Hormuth , H. Israel , K. Jahnke , E. Keihanen , S. Kermiche ,M. Kilbinger , , C. C. Kirkpatrick , T. Kitching , B. Kubik , M. Kunz , H. Kurki-Suonio , R. Laureijs ,S. Ligori , P. B. Lilje , I. Lloro , D. Maino , , , E. Maiorano , C. Maraston , O. Marggraf , N. Martinet ,F. Marulli , , , R. Massey , S. Maurogordato , E. Medinaceli , S. Mei , M. Meneghetti , R. Benton Metcalf , ,G. Meylan , M. Moresco , , L. Moscardini , , , E. Munari , R. Nakajima , C. Neissner , S. Niemi ,A.A. Nucita , , C. Padilla , F. Pasian , L. Patrizii , A. Pocino , , M. Poncet , L. Pozzetti , F. Raison ,A. Renzi , , J. Rhodes , G. Riccio , E. Romelli , M. Roncarelli , , E. Rossetti , A.G. Sánchez , D. Sapone ,P. Schneider , V. Scottez , A. Secroun , S. Serrano , , C. Sirignano , , G. Sirri , F. Sureau , P. Tallada-Crespí ,D. Tavagnacco , A.N. Taylor , M. Tenti , I. Tereno , , R. Toledo-Moreo , F. Torradeflot , A. Tramacere ,L. Valenziano , , T. Vassallo , Y. Wang , N. Welikala , M. Wetzstein , L. Whittaker , , A. Zacchei ,G. Zamorani , J. Zoubian , E. Zucca (A ffi liations can be found after the references) Received xxx; accepted yyy
ABSTRACT
The Complete Calibration of the Colour–Redshift Relation survey (C3R2) is a spectroscopic e ff ort involving ESO and Keck facilities designedspecifically to empirically calibrate the galaxy colour–redshift relation — P ( z | C ) to the Euclid depth ( i AB = .
5) and is intimately linked to thesuccess of upcoming Stage IV dark energy missions based on weak lensing cosmology. The aim is to build a spectroscopic calibration sample thatis as representative as possible of the galaxies of the
Euclid weak lensing sample. In order to minimise the number of spectroscopic observationsnecessary to fill the gaps in current knowledge of the P ( z | C ), self-organising map (SOM) representations of the galaxy colour space have beenconstructed. Here we present the first results of an ESO (cid:64) VLT Large Programme approved in the context of C3R2, which makes use of the two VLToptical and near-infrared multi-object spectrographs, FORS2 and KMOS. This data release paper focuses on high-quality spectroscopic redshiftsof high-redshift galaxies observed with the KMOS spectrograph in the near-infrared H - and K -bands. A total of 424 highly-reliable redshifts aremeasured in the 1 . ≤ z ≤ . H -band and 32.8% in the K -band. The newly determined redshiftsfill 55% of high (mainly regions with no spectroscopic measurements) and 35% of lower (regions with low-resolution / low-quality spectroscopicmeasurements) priority empty SOM grid cells. We measured H α fluxes in a 1 (cid:48)(cid:48) . E ( B − V ), total magnitudes, and SFRs. We combine the results obtained from the spectra with those derived via SED fitting, andwe show that the spectroscopic failures come from either weakly star-forming galaxies (at z < .
7, i.e. in the H -band) or low S / N spectra (in the K -band) of z > Key words. astronomical databases: catalogs - astronomical databases: surveys - cosmology: observations - galaxies: distances and redshifts (cid:63) e-mail: [email protected]
1. Introduction
The existence of a direct connection between cosmic shear andthe presence of gravitational fields created by the distribution ofmatter along the line of sight motivated the development of a
Article number, page 1 of 21 a r X i v : . [ a s t r o - ph . GA ] J u l & A proofs: manuscript no. main number of weak lensing cosmological surveys. These are bothspace based, such as
Euclid (Laureijs et al. 2011) and WFIRST(Spergel et al. 2015), and ground based, such as the ongoingKilo-Degree Survey (KiDS, de Jong et al. 2013), Dark EnergySurvey (DES, Dark Energy Survey Collaboration et al. 2016),Hyper Suprime-Cam Subaru Strategic Programme (HSC SSP,Aihara et al. 2017), and the future Vera C. Rubin Observa-tory survey (LSST, LSST Science Collaboration et al. 2009).The main advantage of space missions with respect to ground-based ones is the absence of atmospheric turbulence, which leadsto images with smaller and more stable point-spread functions(PSFs), allowing cosmological analyses at higher redshifts. Be-sides turbulence, space is key for near-infrared observations,thanks to the lower background, which makes it possible to reachhigher redshift than the ground-based surveys.The aims of the aforementioned projects are to determinegalaxy shape distortions, make use of weak lensing principles tomeasure the geometry of the Universe, and trace the evolution oflarge-scale structure (LSS) to shed light on the complex relationbetween galaxies and the dark components of the Universe. Inthis respect, the outcome of these ambitious programmes heavilydepends on the precise determination of the true ensemble red-shift distribution, or N ( z ), and thus an accurate reconstruction ofthe 3D distribution of galaxies. To the lowest order, weak lens-ing is primarily sensitive to the mean redshift and the width ofthe redshift distribution in tomographic bins (Amara & Réfrégier2007).Moreover, the sensitivity of weak lensing tomography to thedark energy equation of state cannot disregard the ability tomeasure the growth of structure by dividing the source sam-ples by redshift. The di ffi culty of finding optimal tomographicredshift bins for cosmic shear analysis has been studied in re-cent works, and solutions based on dimensionality reductionapproach through self-organising maps (SOM, Kohonen 2001)have been explored (Kitching et al. 2019).In the case of Euclid , this translates into stringent require-ments on the knowledge of the redshift distribution of sourcesevaluated in terms of (1) the precision of individual redshifts,which must be σ z < . + z ), and (2) the mean redshift (cid:104) z (cid:105) ofeach tomographic bin, which must be constrained at the level of ∆ (cid:104) z (cid:105) ≤ . + (cid:104) z (cid:105) ).The Euclid satellite, scheduled for launch in 2022, will ob-serve galaxies out to at least z = by meansof two instruments: VIS, an optical imager that will reach anAB magnitude depth of 24.5 with a single broad r + i + z fil-ter, and NISP, a combined near-infrared imager (in Y , J and H )and slitless spectrograph. The estimated number of weak lensingsource galaxies that will be imaged from Euclid makes their sys-tematic spectroscopic follow-up unfeasible; this mission is thuscritically dependent upon the determination of accurate photo-metric redshifts ( z phot ). However, the accuracy of current pho-tometric redshifts based on multi-band optical surveys is to theorder of σ z / (1 + z ) = . − . , and the fraction of catastrophicoutliers — defined as objects whose z phot di ff ers from their spec-troscopic redshift ( z spec ) by more than 0 . + z ) is to the order ofa few tens of percent (Ma et al. 2006; Hildebrandt et al. 2010).While small changes in z phot precision per source have a rela-tively small impact on cosmological parameter estimates, smallsystematic errors in z phot can dominate all other uncertainties forthese experiments.In this work, we present the results of all the redshift mea-surements on z > α fluxes and stellar masses and investigate their lo-cation in the star formation rate stellar mass (SFR- M (cid:63) ) plane;finally, we present our conclusions in Sect. 9.Throughout the paper, we assume H =
70 km s − Mpc − , Ω m = . Ω Λ = .
7. We adopt a Chabrier (2003) initial massfunction (IMF) in the mass range 0.1 – 100 M (cid:12) .
2. Mapping the colour–redshift relation withspectroscopy
In order to overcome the limitations of current techniques usedto estimate n(z), a complete calibration set of spectroscopic datais required. This spectroscopic calibration sample should be rep-resentative of the entire range of galaxy types and redshifts thatare going to be exploited by a given weak lensing survey. P ( z | C ) calibration In order to shed light on our current knowledge of the galaxypopulation for weak lensing measurements, and in particularfor
Euclid , Masters et al. (2015, hereafter M15) made use ofa SOM to map the high-dimensional galaxy colour space ontoa 2D plane. We used the SOM to group galaxies according tothe similarity of their colours (i.e. of their spectral energy dis-tributions; SEDs) in order to unveil which regions of the galaxycolour space (represented by cells in the plane) are not repre-sented in currently available spectroscopic surveys. This group-ing strategy allows us, in turn, to minimise the number of ad-ditional spectroscopic redshifts necessary to build a completeand representative calibration sample. The underlying assump-tion of this methodology is that, for a dense enough SOM anda su ffi ciently high-dimensional colour space, there is a uniqueand non-degenerate relation between the position occupied by agalaxy in a multi-colour space and its redshift — P ( z | C ). Simi-lar dimensionality reduction approaches in the context of weaklensing cosmological surveys have been used in recent works,using, for example, absolute magnitudes instead of colours inorder to calibrate photometric redshifts (Wright et al. 2019, inpress). The authors stress the importance of using magnitudesas a reference to an absolute flux scale in order to calibrate then(z) for Euclid . Starting from a photometric sample of galax-ies selected using the
Euclid magnitude limit and grouped us-ing the
Euclid colours and the corresponding spectroscopic sub-samples available in the Cosmological Evolution Survey (COS-MOS, Scoville et al. 2007) field, M15 estimated that a total col-lection of ∼ −
15 K spectra would be necessary in order tofill the galaxy colour space and cover the whole set of param-eters characterising the galaxy population that will be observedby
Euclid . About half of them are already available from variousspectroscopic surveys in the literature, whereas approximately5000 new redshifts should be observed in order to calibrate thecurrent photometric redshift techniques and meet the
Euclid re-
Article number, page 2 of 21. Guglielmo et al.: C3R2 VLT / KMOS quirements. Galaxies in these unexplored regions of colour spaceare generally fainter than i AB ∼
23 and lie at intermediate red-shift, 0 . < z < .
0; they correspond to a population of faint, bluegalaxies at intermediate redshift, which have not been targetedbecause they are near the magnitude limit of previous surveys.However, their abundance and unique colours make them animportant part of the galaxy population and crucial sources forweak lensing cosmology. Based on their spectral energy distribu-tions, we expect the objects targeted to be mostly low-metallicitygalaxies with strong emission lines. A minor number of cellscontain faint red galaxies that are either passively evolving ordust obscured, but these constitute only 10–20% of the unex-plored sample. Hence, M15 collected a large number of exist-ing spectroscopic measurements in the COSMOS field (Capaket al. 2007; Scoville et al. 2007; Lilly et al. 2007) to identify thetype (and number) of sources that require spectroscopic follow-up in order to accurately map the full colour-redshift relation ofgalaxies. The work has since then been extended to four addi-tional fields: the VIMOS VLT Deep Survey (VVDS) field, theSubaru / XMM-
Newton
Deep Survey (SXDF) field, the ExtendedGroth Strip field (EGS, within the All-Wavelength EGS Interna-tional Survey, AEGIS), and the Extended
Chandra
Deep Field-South (E-CDFS) field.
The Complete Calibration of the Color–Redshift Relation(C3R2; Masters et al. 2017; M17 hereafter) survey was designedto perform a systematic spectroscopic e ff ort by means of two ob-serving campaigns involving two telescope facilities. Part of thespectroscopic follow-up is conducted with the Keck telescopesusing a combination of the DEIMOS, LRIS, and MOSFIRE in-struments, with time allocated from all Keck partners (M17).The second part is overseen by the ESO Very Large Telescope(VLT) and its UT1 instruments FORS2 and KMOS.M17 presented the results of the first five nights of observa-tions using the Keck facilities during the 2016A semester, lead-ing to the release of 1283 high-confidence redshifts (Data Re-lease 1). A further 3171 new high-quality spectroscopic redshiftswere obtained during 2016B and 2017A semesters and are re-leased in (Masters et al. 2019, M19, Data Release 2). A thirdC3R2 (cid:64) Keck data release is in preparation (Stanford et al., inprep.). (cid:64)
VLT
In order to build a large sample of spectroscopic redshifts for thecalibration of the photometric redshifts of upcoming cosmologi-cal surveys we obtained a 200 h large programme (199.A-0732;PI F. J. Castander) in service mode over four semesters (Pe-riod P99: 1 st April 2017 – P102: 31 st March 2019 + carryover ).The large programme allocated 112 h to FORS2, a multi-objectoptical slit spectrograph and 88 .
3. Target selection and KMOS IFU settings
In order to reduce the impact of sample variance on the calibra-tion of photometric redshifts, the spectroscopic follow-up obser-vations are conducted in a number of extragalactic calibrationsand deep fields planned for the
Euclid mission. However, we ex-pect these commonly observed fields to also be the calibrationfields of other upcoming surveys such as LSST and WFIRST;this spectroscopic follow-up e ff ort will therefore be beneficialfor the wide field survey community at large.The major driving criterion in the choice of such fields is thepossibility of collecting a homogeneous and well-calibrated pho-tometric sample of galaxies observed in eight filters ( ugrizY JH ,seven colours) from the optical to the near-infrared domain downto the Euclid limiting magnitude but with five times highersignal-to-noise ratio. A combination of the Canada-France-Hawaii Telescope Legacy Survey (CFHTLS) deep fields in the ugriz optical magnitude and the VISTA or CFHT-WIRCAMDeep Survey (WIRDS) in the
Y HK near-infrared bands wasfound to meet these requirements. The finally targeted fieldsare COSMOS (from which the SOM was derived; RA = h m ,Dec =
2° 12 (cid:48) ), the VIMOS-VLT Deep Survey field centred atRA = h (VVDS-02h, VVDS hereafter; Le Fèvre et al. 2005;RA = h m Dec = −
4° 30 (cid:48) ), the Subaru / XMM-
Newton
Deep Sur-vey field (SXDF; Furusawa et al. 2008; RA = h m Dec = − Chandra
Deep Field-South Survey field(ECDFS; Lehmer et al. 2005; four fields centred at the fol-lowing coordinates: Field 1, RA = h m . s
61 Dec = −
27° 41 (cid:48) . (cid:48)(cid:48)
84; Field 2, RA = h m . s
43 Dec = −
27° 41 (cid:48) . (cid:48)(cid:48)
80; Field 3,RA = h m . s
94 Dec = −
27° 57 (cid:48) . (cid:48)(cid:48)
56; Field 4, RA = h m . s
93 Dec = −
27° 57 (cid:48) . (cid:48)(cid:48) = h m Dec =
52° 41 (cid:48) ), inaccessible to VLT facilities. We note that theSXDF and E-CDFS fields currently lack uniform photome-try in the full suite of the aforementioned optical and near-infrared filters at the required depth, but as they provide a con-siderable number of spectroscopic redshifts, they were includedafter applying a rough colour correction to convert into theCFHTLS + VISTA / WIRDS-like system (see M17).
C3R2 prioritises targets in regions of the SOM that lack spec-troscopic redshifts. High-priority targets have colours that arefrequent (i.e. fall in cells with high occupation) and are thereforeextremely valuable in calibrating the redshift-to-colour relation.The C3R2 prioritisation scheme (extensively described in M19)therefore gives higher weights to sources with common coloursin still uncharted cells. As observations are obtained and spec-troscopic redshifts determined, the target catalogue and priorityflags are updated.Spectroscopic redshift measurements are based on the iden-tification of emission lines in the observed galaxy spectra, withhigher priority given to the detection of the often prominent H α line ( λ .
61 Å ). The grisms selected for the KMOS obser- In order to operate at near-infrared wavelengths, the entire workingparts of the instrument are cooled to below − − α restframe = & A proofs: manuscript no. main
Fig. 1: Telluric absorption curve (black curve) in wavelengthrange covered by the KMOS H - and K -band gratings (red hor-izontal lines); the light grey spectrum in the bottom part of thepanel represents the emission lines produced by the OH radicalin the atmosphere between 0.61 µ m and 2.62 µ m. The red labelson the top horizontal axis indicate the redshift (1 . < z < . α emission line falls at the wavelength indi-cated by the position of the vertical red dashed lines.vations are H (1.456 – 1.846 µ m) and K (1.934 – 2.460 µ m); wethus target galaxies with a photometric redshift that positions theH α line within the observed wavelength range but avoids its con-tamination by atmospheric absorption windows as well as OHnight-sky emission lines, as shown in Fig. 1.We selected high-redshift star-forming galaxy candidateswith 1 . < z phot < . . < z phot < . H and K grisms, respectively, and divide them into twoclasses based on the prioritisation scheme defined in M19:- H -band, priority 1: 1 . ≤ z phot ≤ . i tot ≤ .
5, and thepriority flag computed in M19 ( P F ) ≥ ;- H -band, priority 2: 1 . ≤ z phot ≤ . i tot ≤ .
5, and 200 ≤ P F < K -band, priority 1: 2 . ≤ z phot ≤ . i tot ≤ .
7, and P F ≥ K -band, priority 2: 2 . ≤ z phot ≤ . i tot ≤ .
7, and 200 ≤ P F < H -band P F ≥
500 corresponds to the top 7 .
2% of KMOSselection list, P F ≥
200 corresponds to the top 18%. K -band pri-ority >
500 corresponds to the top 16% of the KMOS selectionlist, priority >
200 corresponds to the top 33%.A fraction of the COSMOS, SXDF, and E-CDFS fields havebeen extensively observed in the past with KMOS as part of theKMOS3D programme, one of the KMOS Guarantee Time Ob-servations programmes (Wilkinson et al. 2015) using the
Y J , H , and K gratings. We removed all sources already observed by theKMOS3D team from the present target selection. Their spec-troscopic redshifts (of exquisite precision) are available publicly(Wisnioski et al. 2019) and are going to be used for the cali-bration of the Euclid photometric redshifts (KMOS3D, ).
4. Observations and data reduction
In this section, we describe the acquisition and reduction of thedata. The P F parameter computed in M19 ranges from 0 up to 3750; 89%of the SOM cells have P F ≤ KMOS is a multiplexed near-infrared integral field system (IFS)with 24 deployable image slicers (commonly referred to as‘arms’), surveying a 7 (cid:48) . (cid:48)(cid:48) . × (cid:48)(cid:48) . ×
14 pixel IFS units)and a spatial resolution of 0 (cid:48)(cid:48) . / spaxel. The IFS units connectto three cryogenic grating spectrometers with 2k ×
2k Hawaii-2RG HgCdTe detectors. As previously mentioned, among thefive available KMOS gratings ( IZ , Y J , H , K , HK ), our obser-vations make use of the H - and K -bands (plus tentative Y J ),characterised by a typical spectral resolution of about 3500. Theobservations were prepared with the KMOS ARM Allocator(KARMA; Wegner & Muschielok 2008) software, and submit-ted through the Phase 2 Proposal Preparation (P2PP) tool. Here-after an individual KARMA setup (made of 24 arm allocations)is referred to as a ‘pointing’. Each pointing was observed fora total of 3600 s split into single exposures of 300 s each, us-ing an O-S-O-O-S-O pattern (i.e. a ‘sky’ exposure is observedevery two ‘object’ exposures). The sky exposures were o ff setwith respect to targets to the closest position uncontaminated bysources. Additional sub-pixel / pixel dithering shifts were also ap-plied at every exposure to minimise the impact of pixel-to-pixelvariation and bad pixels in the final science data cube. One ofthe 24 KMOS IFUs was allocated to a star (with an observedmagnitude of 15 . < H < .
5) during the science observations(with the exception of 7 /
36 pointings). The star allows us to trackvariations in the PSF and photometric conditions between theframes; the star is therefore referred to as the PSF star.The standard requirements of the KARMA software forpreparing a KMOS pointing are, firstly, the presence of a suf-ficient number of acquisition stars (with observed magnitudes13 . < H <
17) within the patrol field of a given KMOS point-ing and preferentially and equally distributed among the 24 armsand three spectrometers / detectors (these stars are used to alignKMOS). The second requirement is the absence of bright stars(which would create persistency) superposed with the path ofthe KMOS arms on the field of view. The final requirementis the presence of at least one bright guide star (with an ob-served magnitude 9 < R <
12) in the vicinity of the pointingto maintain telescope tracking. All the aforementioned stellarsources must have low proper motion. Specifically, we required | µ RA | and | µ Dec | <
20 mas yr − .The observations cover four distinct fields whose observabil-ity spreads adequately throughout the year. The number of hoursallocated per semester and per field is reported in Table 1. Thecorresponding number of pointings are indicated in parentheses,split between the H - and K -bands, with a slight preference of H -band over K -band to maximise the redshift measurement suc-cess rate. A detailed list of the pointings observed in P99–P103is reported in Table 2. Each observing block (OB) is composedof two pointings of 1 h on sky, which provides about 40 minuteson source. These pointings can either be observed during thesame night or on di ff erent nights. In the latter case, the obser-vations are reduced separately and then combined. Only duringthe last awarded period (P102) was the on-source time for K -band pointings doubled in order to increase the detectability ofthe targeted galaxies. The data-reduction procedure, described inthe next section, is applied to the single science and sky framesseparately, and the frames are combined at the end of the reduc-tion, after the whole pointing (two OBs) has been observed. Article number, page 4 of 21. Guglielmo et al.: C3R2 VLT / KMOS
The data were reduced with the Software Package for Astronom-ical Reduction with KMOS (SPARK; Davies et al. 2013) usingrecipes outlined in the SPARK instructional guide . The reduc-tion first applies a correction for detector e ff ects, including (1)the correction of the readout channel variations via the refer-ence pixels (pixels without photodiodes but with full electronicsreadout), and (2) the correction for the picture-frame e ff ects af-fecting IFUs at the edges of the detector, using median DARKframes. The reduction then proceeds through the standard cali-bration steps, namely flat fielding, illumination correction, wave-length calibration (the accuracy of the wavelength solution isto the order of 30 km s − ), reduction of the spectrophotometricstandards, and finally the data cube reconstruction. After thisstage, an additional custom processing was performed on thesereconstructed data cubes to further subtract the sky lines. Thecustom-made sky-line correction routine is an adaptation of theZurich Atmosphere Purge (ZAP; Soto et al. 2017) approach tothe KMOS data. The routine subtracts the closest sky frame tothe science frame in the O-S-O-O-S-O sequence and then fur-ther optimises the fitting to the OH sky-line residuals via a ZAPprincipal-component analysis (Wisnioski et al. 2019). The back-ground continuum is removed using o ff set sky frames withoutattempting to correct for short time scale background variations,and thus some residual continuum levels are still expected. Anillumination correction is then applied to flatten out the IFU spa-tial response. A heliocentric correction is finally performed be-fore the data cubes are combined.A further set of reduction steps is applied by means of a rou-tine developed by the KMOS GTO team in order to perform theflux calibration and a refined background subtraction (Wisnioskiet al. 2019). The flux calibration procedure can be summarisedin three operations: a) correction for the grism + detector wave-length response using a telluric star; b) application of the zeropoint to convert fluxes to units of 10 − W m − µ m − (to be fur-ther multiplied by 0.1 to obtain erg cm − s − Å − ); and c) fit ofthe PSF star in the science data with a Mo ff at function for themonitoring of the flux and estimation of the PSF from its averageFWHM across the frames, and measured again on the combineddata cubes for consistency checks. Individual frames are thenmedian-combined into final cubes using spatial shifts measuredfrom the average centre of the stars within the same pointings(when applicable) or using the information given in the headerof each frame. Variations in flux and seeing among the com-bined frames are typically 10% and 0 (cid:48)(cid:48) .
1, respectively. A detaileddescription of the data reduction for KMOS data cubes can befound in Wisnioski et al. (2019). ftp: // ftp.eso.org / pub / dfs / pipelines / kmos / kmos-pipeline-cookbook-0.9.pdf Article number, page 5 of 21 & A proofs: manuscript no. main
Table 1: List of the awarded time (in h) for KMOS observations. Below the number of hours, in parenthesis, the number of theobserved pointings is indicated, together with the selected filter, for example, 3 H + K means that three pointings have been observedin the H -band and two pointings have been observed in the K -band.Field P99 P100 P101 P102 TotalCOSMOS 7.6 10.8 0 10.8 29.2(2 H + Y J (cid:63) ) (3 H + K ) (5 H ) (10 H + K )ECDFS 0 0 2.2 0 2.2(1 H ) (1 H )SXDF 0 8.7 5.4 10.8 24.9(2 H + K ) (1 H + K ) (3 H (cid:63)(cid:63) + K ) (6 H + K )VVDS 6.5 10.8 6.5 8.7 32.5(2 H + K ) (3 H + K ) (2 H + K ) (2 H + K ) (9 H + K )Total 14.1 30.3 14.1 30.3 88.8 (cid:63) We had initially planned to target sources with 1 . < z phot < .
0, for which the O ii doublet is in the YJ -grating. The detection of O ii is challenging in high-redshift galaxies, and our firstobservations in P99 had a low success rate. We therefore decided to start in P100 to exclusively concentrate on the detection of H α in the H - and K -gratings. (cid:63)(cid:63) The observation of the last three H -band pointings in the SXDF field (see Table 2 for details) was carried over P103. Table 2: List of the observed pointings.Pointing ID RA cen
Dec cen
Exp_time Filter UT Date Success Rate(deg) (deg) (s) (yyyy.mm.dd) (3 ≤ Q ≤ / Q = / Observed)P99_COSMOS_HaHP1 149.8900 1.9003 2 × H / / H / / H − . × H / / − . H / / H − . K / / K H / / H × H / / × H / / K / / K K / / K − . × H / / − . × H / / − . K / / H − . K / / H − . H / / H − . H / / H − . K / / K − . K / / K − . H / / H − . H / / H − . H / / H − . × K / / K K − . × H / / − . × H / / Article number, page 6 of 21. Guglielmo et al.: C3R2 VLT / KMOS
P101_VVDS_haKP1 36.7296 − . × K / / (cid:63) − . K / / K (cid:63) − . K / / K − . H / / H − . × H / / H / / H H / / H × H / / H / / H × H / / − . K / / K K K − . H / / H − . H / / H − . H / / (cid:63) These pointings are replicated configurations of two K -band VVDS pointings with low success rates observed during P99 (P102_P99_VVDS_HaKP1) and P100(P102_P100_VVDS_HaKP1); the overall configuration is maintained, but new objects have been allocated to arms in which a good spectroscopic redshift was derived during the ear-lier observations (quality flag from three to four, which means that we replaced five to seven galaxies per pointing). Article number, page 7 of 21 & A proofs: manuscript no. main
5. Redshift assignment
The observational programme performed with KMOS (cid:64)
VLTaims to derive the spectroscopic redshift of 1 . (cid:46) z phot (cid:46) . α in the H - and K -band filters.Each observed spectrum was analysed by two co-authors toindependently determine the redshift and the quality flag. Theresults were then reconciled and discussed by the two people. Wedeveloped an interactive routine that we applied to the reducedand combined data cubes for the redshift assignment. There areseveral steps towards the application of the code: – when continuum is visible, find the position of the targetedsource in the spatial plane of the median image of the datacube, otherwise we use the nominal centre at the pixel withcoordinates ( x , y ) = (9 , – create two-dimensional (2D) vertical / horizontal spectracomputing the median flux at each wavelength of fourlines / columns around the central pixel; – identify the presence of an emission line either in the verticaland / or in the horizontal 2D spectrum and select a narrower(about 10 pixels) wavelength range to determine the pixelswhere the emission is detected; – plot the ( x , y ) spatial image of the cube at four pixels cor-responding to the wavelengths where the emission has thehighest intensity in order to identify both the wavelength (inpixel units) of the peak of the emission and the ( x , y ) coordi-nates of its centre; – plot the 1D spectrum of the selected central spaxel and the1D spectrum obtained by summing the flux in a number ofadjacent pixels to increase the signal to noise (the number ofpixels varies from a cross of five to a square of nine, depend-ing on the spatial extension of the source); – perform a Gaussian fit weighted by the noise spectrum on theidentified emission line; – choose the most appropriate-looking value of the emission-line centre, between the position of the mean of the fittedGaussian and the position of the peak pixel; – compute the redshift with the formula z spec = ( λ peak / Gaussian − λ H α ) /λ H α , (1)where λ peak / Gaussian is the wavelength (in µ m) correspondingto the pixel peak or to the centre of the fitted Gaussian, and λ H α is the H α vacuum wavelength expressed in µ m. Each redshift measurement is assigned a preliminary quality flagreproducing the flagging scheme presented in M17: – Q =
4: indicates a secure redshift measurement based on theidentification of more than one emission line. Specifically,the H α line is associated with the N ii doublet at λ λ ii doublet ( λ λ α line. (Detailson how the identification and fit of these groups of lines isperformed is given in Sect. 5.2); – Q = .
5: indicates a secure redshift measurement based on asingle emission line (usually H α ); – Q =
3: indicates a likely secure redshift determination, butwith a low probability of an incorrect identification or an un-certain redshift due to low signal-to-noise data or sky-linecontamination a ff ecting the Gaussian fit; – Q =
2: flag 2 indicates a reasonable but not secure enoughguess. The targets being assigned with this flag are discardedfrom the calibration sample, and not included in the releasedcatalogue.
KUBEVIZ
Maps of the emission-line fluxes were obtained from the reduceddata cube using the IDL routine
KUBEVIZ (Fossati et al. 2016).The code simultaneously fits groups of lines (defined as ‘linesets’, e.g. H α and the N ii λ λ iii λ λ < . KUBEVIZ on the KMOS reduced data cubes. Firstly, fitting the H α + N ii lineset improves the z spec measurement; starting from the H α emission map of the galaxy and its corresponding velocity ( v )map, we arbitrarily chose the centre ( v =
0) of the galaxy asthe spaxel that best compromises the peak of the H α emissionwith the centre of the galaxy signal / velocity map (if present),and we corrected the input z spec and the relative velocity of ev-ery spaxel accordingly. Furthermore, a successful KUBEVIZ fit oflow-quality spectroscopic candidates (those that were assigned a Q = z spec measurement, and thus their inclusion in the calibration sam-ple. Finally, the KUBEVIZ outputs constitute the groundwork formeasuring the total H α flux of the sources, which is described indetail in Sect. 7.2. We collected all available multi-wavelength photometry for thegalaxy sample observed during the KMOS programme frompublic data releases in the three fields . We start from the COSMOS2015 catalogue released in Laigleet al. (2016), which contains precise PSF-matched photome-try for more than half a million sources in the COSMOS field.Among the wide collection of photometric bands available in thedata release, we selected CFHT u (cid:48) and Subaru B , V , R , i + , z + and z ++ optical aperture magnitudes (3 (cid:48)(cid:48) ), Y JHK s near-infrared aper-ture magnitudes (3 (cid:48)(cid:48) ) from the UltraVISTA-DR2 survey, mid-infrared data from the
Spitzer
Large Area Survey with Hyper-Suprime-Cam (SPLASH) legacy programme (IRAC ch1, ch2,ch3, ch4 total magnitudes), and GALEX
NUV total magnitudes.We computed total magnitudes in the optical and near-infrared domain starting from the aperture magnitudes and thecorrection factors given in the released catalogue using Eq. (9) The multicolour photometry used here is optimised to measuredphysical parameters of galaxies of known spectroscopic redshift; otherchoices might be preferable when computing photometric redshifts (seeMasters et al. 2015).Article number, page 8 of 21. Guglielmo et al.: C3R2 VLT / KMOS in the Appendix of Laigle et al. (2016):
MAG _ TOTAL i , f = MAG _ APER3 i , f + o i − s f , (2)where i identifies the single objects, f the considered filter, MAG_APER3 is the magnitudes computed within a 3 (cid:48)(cid:48) radius aper-ture contained in the catalogue, o i is the photometric o ff set com-puted for scaling aperture magnitudes to total ones, and s f is thesystematic o ff set computed in the paper using spectroscopic red-shifts. Finally, all magnitudes should also be corrected for fore-ground Galactic extinction using the reddening values given inthe released catalogue for each object (Eq. 10 in the Appendix): MAG _ TOTAL i , f , extcorr = MAG _ TOTAL i , f − E ( B − V ) i × F f , (3)where F f is the extinction factor of any given filter.Besides the photometric information, we also kept the z phot and physical properties ( E ( B − V ), absolute magnitudes, medianstellar masses, and SFR from the maximum likelihood – ML –analysis of LePhare ) derived in Laigle et al. (2016) by means ofthe SED fitting code
LePhare (Arnouts et al. 1999; Ilbert et al.2006) run on the complete 30-band photometric data set.
We collected multi-band photometry in the SPLASH survey datarelease Mehta et al. (2018). We considered optical aperture mag-nitudes (3 (cid:48)(cid:48) ) from CFHT u filter and from the Hyper Suprime-Cam (HSC) UltraDeep layer in the griz filters; the near-infraredregime is fully covered by the VISTA Deep Extragalactic Ob-servations (VIDEO) Survey Y JHK s aperture magnitudes (3 (cid:48)(cid:48) ),and the mid-infrared takes advantage of the IRAC coverage (ch1,ch2, ch3, ch4) from SPLASH.Aperture magnitudes were corrected to total values using theo ff sets given in the released catalogue table ( OFFSET_MAG ) andall magnitudes were corrected for foreground extinction follow-ing the same procedure described in Sect. 5.3.1 for the COSMOSfield. Consistent with what was done in Laigle et al. (2016) forthe COSMOS field, Mehta et al. (2018) performed the SED fit-ting analysis of the SXDF photometric sample using
LePhare .We took advantage of the outputs of their analysis to collect thephysical properties of all our observed galaxies ( E ( B − V ), abso-lute magnitudes, best fit stellar masses, and SFRs). A complete and homogeneous collection of photometry in theVVDS-02h field is contained in the VIDEO Survey, which hasbeen merged with the CFHTLS Deep1 optical ( ugriz ) catalogue(Jarvis, M. & Häussler, B., priv. comm.). The catalogue con-tains aperture magnitudes within a 2 (cid:48)(cid:48) radius measured in a ho-mogeneous manner in all the optical and near-infrared filters.We computed the aperture to total magnitude o ff sets using the SExtractor
MAG_AUTO values given in the catalogues andthe photometric errors, according to Eq. (4) and (5) in Laigleet al. (2016): o = (cid:80) filters i w i × (cid:88) filters i ( MAG
AUTO − MAG
APER ) i × w i , (4)where w i = σ AUTO + σ APER ) i . (5) The o ff sets are computed for each object in the catalogue ( i ) us-ing all the bandpasses in the optical and near-infrared domain.We finally corrected total magnitudes for Milky Way foregroundextinction using the Schlegel et al. (1998) maps (consistent withwhat was used in Laigle et al. 2016) at the coordinates of eachobject and using the appropriate filter factors, as given in Eq. (3).In order to investigate and compare the properties of all theobserved galaxies with the spectroscopically confirmed ones,and to have consistent z phot measurements throughout the threeexplored fields, we ran LePhare on the whole set of collectedfilters and derived z phot and physical properties of all observedVVDS galaxies ( E ( B − V ), absolute magnitudes, median stellarmasses, and SFR from the ML analysis).
6. Results I: The success rate of the redshiftassignment
In light of the concepts outlined above, the success rate (SR)of the KMOS spectroscopic campaign in the context of theC3R2 survey must be evaluated in two ways: (1) as any spec-troscopic survey, as the ratio (or, equivalently, percentage) ofthe total number of high-quality z spec measured with respectto the number of targets observed; (2) as the total number ofempty / undersampled cells that are newly filled with spectroscop-ically confirmed galaxies. Needless to say, these two quantitiesshould be considered together: a large number of high-quality z spec assigned to a small number of cells is less valuable than asmaller number of high-quality z spec covering a larger number ofempty SOM cells.The total number of z phot targets observed with KMOS was805, 424 of which provided a secure redshift measurement ( Q ≥ H -band observations is twice that of the K -band observations,likely primarily due to the higher backgrounds at longer wave-length. Additional challenges are caused by the lowering of theprecision of currently available template fitting techniques asredshift increases, and also the lower brightness of the targetsthemselves. Doubling the exposure time of K -band pointings andrepeating the observation of two K -band pointings observed dur-ing P99 and P100, was not conclusive in this respect: the K -bandSR in P102 only slightly increased compared to previous peri-ods. Whether this result is mainly due to the limited accuracy of z phot -based target selection or to the necessity of longer exposuretimes to increase the SNR of the spectra is still unclear, but a de-tailed analysis of the spectroscopic failures is presented in Sect.6. Figure 2 presents a comparison between the photometric(individual and SOM-based) redshifts and high-quality ( Q ≥
3) KMOS spectroscopic redshifts. The dashed lines trace theboundaries outside which the photometric redshifts are consid-ered catastrophic outliers, | z phot − z spec | / (1 + z spec ) ≥ z phot redshift esti-mates with our z spec measurements: according to these quanti-ties, our sample contains one catastrophic outlier. This galaxy,observed in the H -band, has a z phot = . z spec = . Q = .
0. A detailed analysis of this target revealed a dis-crepancy between the individual (from template fitting) and theSOM-based z phot estimates ( z phot , SOM = . z spec - z phot plane. Furthermore, we notice that there is a target observed inthe H -band with z phot ≤ .
6, but validated at z spec ≥
2, thanksto the identification of the O iii ( λ Article number, page 9 of 21 & A proofs: manuscript no. main
Table 3: Success rate of KMOS observations.Period H -band K -bandP99 72 / (cid:63) (81.8%) 5 / (cid:63)(cid:63) (22.7%)P100 106 /
176 (60.2%) 46 /
132 (34.8%)P101 53 /
89 (59.6%) 13 /
44 (29.5%)P102 117 /
220 (53.2%) 12 /
51 (30.4%)Total 348 /
573 (60.7%) 76 /
232 (32.8%) (cid:63)
72 galaxies with accurate z spec estimate (Q ≥
3) over 88 observed targets. (cid:63)(cid:63)
Pointing re-observed during P102. Since 17 out of 22 galaxies were re-observed, the contribution to the total number of observed objects in the K -band from P99 is just five. Fig. 2:
Top left : comparison between z phot and z spec for high-quality ( Q ≥
3) redshift galaxies observed during the four periods ofthe KMOS Large Programme. Lower redshift targets are observed with the H -band grism, higher redshift ones with the K -band.The dashed lines define the region outside which the z phot is considered a ‘catastrophic failure’ (grey area in the plot), defined by aredshift error | z phot − z spec | / (1 + z spec ) ≥ Top right : histogram of the ( z phot − z spec ) / (1 + z spec ) of all high-quality redshift targets.A Gaussian with mean and sigma equal to the bias and σ NMAD , respectively, is overplotted with a red dashed line. Bottom left : sameas the top-left panel but comparing z phot , SOM and z spec . Bottom right : same as the top-right panel but with z phot , SOM .The bottom panels of Fig. 2 show the same statistical analysis tocompare the z spec with the redshift of the SOM cell each galaxybelongs to ( z phot , SOM ).We point out that the SOM is not intended to be used forindividual redshift estimates, and therefore one should not be surprised that its performance in terms of recovering individual z phot values is worse than for individual multi-band template fit-ting. However, comparing the distribution of z phot and z spec inindividual SOM cells is fundamental for a better understandingof cell occupation (e.g. in order to quantify the z phot dispersion of Article number, page 10 of 21. Guglielmo et al.: C3R2 VLT / KMOS galaxies occupying the same cell or to pinpoint multiple peaksin the distribution of galaxies) and for highlighting problematicregions in the SOM.The incidence of catastrophic outliers is significantly higherwhen z phot , SOM is considered. These 25 galaxies fall into 18 dif-ferent cells in the SOM, and have an individual z phot more in linewith the measured z spec ; furthermore, in case of multiple obser-vations within the same SOM cell, these galaxies have individualredshifts, which are in line with the other galaxies populating thecell. This result leads us to conclude that there is a misalignmentbetween the redshift of the cell and the redshift of the individualgalaxies that compose it. A better understanding of the distribu-tion of individual z phot of galaxies in the aforementioned SOMcells is given in Fig. 3. All galaxies in the C3R2 parent z phot sample are used to populate the cells, and the z phot , SOM is alsorepresented inside each panel with the dashed vertical line. Asis noticeable from the dispersion values of the histograms (hor-izontal errorbars centred on the mean z phot ), the z phot distribu-tion peaks close to the z phot , SOM value, but high dispersion and / ordouble peaks are present in many of the cells; multiple spectro-scopic redshift measurements occupy a narrow redshift range inthe panels, often separated from the z phot , SOM . Euclid galaxiesthat are assigned to these problematic cells need to be flagged,as their photometric redshift could be di ffi cult to calibrate.The mean value of the redshift di ff erencemean (cid:32) z phot − z spec + z spec (cid:33) (6)is represented as the mean value of the (red dashed) Gaussian inFig. 2. When comparing z spec with the individual z phot , the valueis − . − . H - and K -bands, respectively, further confirming the decreasing preci-sion of current photometric redshift estimates with increasingredshift. The redshift di ff erence increases to 0.027 when con-sidering the comparison between z spec and z phot , SOM , and 0.030,0.013 in the H - and K -bands, respectively. The higher H -bandbias reflects the increased number of catastrophic outliers, whichare all located at z spec ≤ . σ NMAD = . × median (cid:32) | z phot − z spec | + z spec (cid:33) , (7)is 0.0301 (3%) when individual z phot are considered, and 0.0443( (cid:38) z phot , SOM are used, pointing out that not only thenumber of catastrophic outliers increases, but also the dispersionof the data points in the white region of the (left-hand panels)scatter plots in Fig. 2. The values of the ∆ (cid:104) z (cid:105) and σ NMAD are inagreement with the results presented in M17 and M19.We computed the number of cells containing P1 / P2 targets(according to the priorities defined in Sect. 3.2) with a SOMphotometric redshift 1 . < z phot , SOM < . H -band targets)and 2 . < z phot , SOM < . K -band targets). The SOM hasa number of P1 and P2 cells in this redshift range of 283 and327, respectively. These numbers indicate the nominal goal ofC3R2 in the near-infrared, and will be used as a reference. Thenumber of P1 / P2 cells covered by all KMOS observations (i.e.by all targets placed in KMOS pointings from P99 until P103)is 274 and 162, respectively. Of the P1 cells occupied by theKMOS z phot candidates, 57% (156 / / Spectroscopic failures and uncalibrated cells
We next analysed the properties of galaxies that were observedbut for which we could not assign a spectroscopic redshift. Themain purpose of this analysis is to understand whether there arebiases in the data and where these failures are located in theSOM. To this end, we considered the physical parameters de-rived from SED fittings in Laigle et al. (2016) and Mehta et al.(2018) for the COSMOS and the SXDF field, respectively. Thereason for this choice is twofold. First, when trying to explorethe properties of non-spectroscopically validated galaxies, weare forced to rely on z phot − and z phot -based physical parameters,which are better determined when a broader photometric samplein terms of the number of available filters is used. Both Laigleet al. (2016) and Mehta et al. (2018) based their SED fitting anal-yses on a broad number of filters spanning the whole spectrum.Furthermore, the two are comparable as the same PSF homo-geneisation was adopted for the data, and the same template li-brary was used for photometric redshift calculation. Secondly,our LePhare setup is a close imitation of what was performed inthe two data releases, though limited to a restricted number offilters. In order to check that we did not introduce any bias, weran LePhare on the photometric samples with the same config-uration described in Sect. 7, but without fixing the redshift, andwe compared the results with those from Laigle et al. (2016) andMehta et al. (2018). In the COSMOS field, the average di ff erencebetween stellar masses is 0.090 with an rms of 0.17, and betweenthe (SED fitting based) SFRs it is 0.003 with an rms of 0.229. Inthe SXDF field, the average di ff erence between stellar masses is0.069 with a rms of 0.313 and between the (SED fitting based)SFR is 0.237 with a rms of 0.473. In light of the above, our setof physical parameters is compatible within the errors with theliterature but with larger uncertainties. Although all the conclu-sions discussed below do not change with our derivation, in thefollowing we always refer to the results from the literature.Figure 5 illustrates the distributions of the z phot , observed to-tal H magnitudes and SED-fitting star formation rates (SFRs),and stellar masses for all galaxies observed during our KMOSprogramme (green histograms), for the sub-samples of spectro-scopically confirmed targets (orange histograms) and for the tar-gets that could not be assigned a redshift (blue open histograms).The distributions of validated and non-validated targets presentsome di ff erences, with the former being slightly brighter witha higher star formation rate: the median value of H is 22.78 inthe validated sample and 22.84 in the non-validated one. Simi-larly, the median log (SFR / M (cid:12) yr − ) values are 1.41 and 1.21in the two samples, respectively. From the bottom right panel ofthe figure, we can finally notice that our spectroscopic complete-ness, in terms of number of galaxies validated with respect to thetotal number of galaxies observed, is a function of stellar mass.Specifically, at low stellar masses (log ( M (cid:63) / M (cid:12) ) < . . < log ( M (cid:63) / M (cid:12) ) <
10 and finallydecreases to the lowest values at higher stellar masses. A betterunderstanding of the reasons that prevented us from assigning ahigh-quality spectroscopic redshift to all galaxies can be reached
Article number, page 11 of 21 & A proofs: manuscript no. main
Fig. 3: Histogram of z phot of galaxies populating each cell falling in the grey region of the z phot , SOM - z spec plane (bottom left panel ofFig. 2). The distribution is normalised by dividing the number of galaxies in each z phot bin by the total number of z phot populatingthe considered cell; the number is indicated with the letter N in the top left panel of the figures, and written at the same positionin the others. Similarly, the cell number (CellID) and coordinates (CellX, CellY) are also given inside each panel. The z phot , SOM is represented by the dashed line, whereas dotted lines indicate z spec measured during our KMOS programme. The horizontal barcentred on the mean z phot is the rms of the histogram.by analysing the distribution of the validated and non-validatedtargets in the SOM.In the central panel of Fig. 6, validated cells are colour-codedaccording to the value of the assigned z spec . Cells populated withmultiple observations have been assigned a median z spec value.This panel again highlights a prevalence of low-redshift targetsas already discussed in Sect. 6, mainly concentrated at low val-ues of the X − indices, and spread along the whole Y − index range.In the right panel, we show the z phot of the observed targets forwhich we could not measure z spec , and we mask the spectroscopi-cally confirmed cells. The comparison between the z spec and z phot SOMs confirms that, despite the higher number of spectroscop-ically confirmed H -band targets, there is no systematic (photo-metric) redshift bias in the observed and non-validated targets:the SOM cells that were observed but could not be filled witha highly confident z spec have values ranging from the lowest H -band to the highest redshifts reachable with the K -band setup.However, if the lack of measurement is due to observational dif-ficulties in the K -band and lower accuracy in the SED fitting z phot determination used to select the observed targets, the cause of theconcentration of lower redshift ( H -band) galaxies present in thebottom region of the SOM (dark blue cells) must be investigatedmore thoroughly.We searched for the reason behind these spectroscopic fail-ures in the colours and star formation properties of galaxies.Figure 7 represents the rest-frame ( u − g ) colour, the best fit E ( B − V ) , the and SED fitting SFR of the non-validated sample.Again, the cells containing more than one target have been as-signed a median value. The peculiarity of the bottom part of theSOM stands out: the galaxies populating these cells are, on aver-age, redder and have lower star formation rates compared to the other empty cells. Moreover, as it noticeable from the E ( B − V )shown in the middle panel, they are not particularly dusty. Ourobserving strategy, and in particular the integration time, mayrequire modifications for obtaining the necessary SNR requiredto measure emission-line redshifts.
7. Results II: The physical properties of galaxies
The physical properties of galaxies were derived again for thespectroscopically confirmed targets, by taking advantage of theuse of z spec as a constraint to the fit. We applied the SED fit-ting code LePhare to the spectrophometric catalogues obtainedfrom merging the spectroscopic redshift measurement with themulti-band photometry collected from the parent surveys. A de-tailed list of the filters used in the three fields is reported in Table4, and the appropriate reference to the parent photometric cata-logues is given in the table caption. The code is provided withspectroscopic redshifts and total magnitudes as input, and we setthe priors on fitting parameters and galaxy libraries (based ona collection of di ff erent star formation histories, SFHs) takingadvantage of the knowledge of the average properties of our tar-get galaxies: these are high-redshift, star-forming galaxies, withconsistent H α emission. Out of the whole library of availablemodels, we selected a number of exponentially declining SFHs( τ models), of delayed SFH and of constant SFR, with sub-solar( Z = . Z = Z (cid:12) = .
02) metallicity. We used afine grid of E ( B − V ) ranging from 0 to 0.7, and two di ff erentextinction laws (Calzetti et al. 2000; Arnouts et al. 2013), arealso adopted. We obtain the stellar masses, absolute magnitudes,best fit E ( B − V ) values, and other physical parameters such as Article number, page 12 of 21. Guglielmo et al.: C3R2 VLT / KMOS
Fig. 4: Success rate in terms of number of cells filled with high-quality z spec . The observed targets are divided into high (P1) andlow (P2) priority targets according to the prioritisation schemedescribed in Sect. 3.2. Purple horizontal bars represent the to-tal number of undersampled cells requiring z spec measurements;orange histograms represent the number of cells targeted by allKMOS observations, and green histograms represent the numberof cells that provided accurate z spec measurements.the SFR as output. In the following, for stellar masses and SEDfitting SFRs, we use the median values computed from the MLanalysis of LePhare .The histogram of the resulting stellar masses from
LePhare in the three fields is shown in Fig. 8. The median stellarmass value in the total spectrophotometric sample of galaxiesobserved during the KMOS programme is log ( M (cid:63) / M (cid:12) ) = .
69, and the values in the three di ff erent fields are:log ( M (cid:63) / M (cid:12) ) COSMOS = .
73, log ( M (cid:63) / M (cid:12) ) SXDF = . ( M (cid:63) / M (cid:12) ) VVDS = . P ( z | C ), the properties of the galaxies observed by the C3R2 sur-vey is of unique importance and interest. Building a sample ofspectra spanning the whole redshift range up to z ∼ . α fluxes and stellar masses. In the following sections, we deter-mine and discuss the physical properties of the spectroscopicallyconfirmed galaxies in the COSMOS, VVDS, and SXDF fields,leaving aside the ECDFS field which contributes with only 12galaxies to the release. α fluxes The velocity and H α maps from KUBEVIZ allow the measure-ment of the total H α flux of the sources. Starting from the cen-tre coordinates, the final z spec and the velocity map, we esti-mate the H α flux in a fixed circular aperture of 1 (cid:48)(cid:48) . .2 radius. Thiscorresponds to about 10 kpc at redshifts 1 . (cid:46) z (cid:46) .
5. van Table 4: Summary of the photometry used in each field. Thecomplete filter set used in the COSMOS and SXDF data releaseis given in Table 1 of Laigle et al. (2016) and Table 1 of Mehtaet al. (2018).Field Instrument / Telescope Filter Central(Survey) λ (Å)COSMOS GALEX
NUV 2313.9MegaCam / CFHT u (cid:63) B / Subaru V r i + z ++ Y UD / VISTA J UD H UD K UDS / Spitzer ch1 35634.3(SPLASH) ch2 45110.1ch3 57593.4ch4 79594.9SXDF MegaCam / CFHT u (cid:63) g r i z y Y J H K S / Spitzer ch1 35573(SPLASH) ch2 45049ch3 57386ch4 79274VVDS MegaCam / CFHT u g r i z Y J H K S Hubble
Space Telescope)and CANDELS galaxies, as well as ACS / F814W (8073.43 Å),WFC3 / F125W (12501.04 Å), and WFC3 / F160W (15418.27 Å)filters for measuring sizes, estimated the evolution of the ef-fective radius ( R e ) of star-forming galaxies in various stellarmass and redshift bins. They estimated that massive star-forminggalaxies ( M (cid:63) ∼ M (cid:12) ) have R e ∼ M (cid:12) (Fig.8), we considered an aperture from the galaxy centre that dou-bles the R e estimated in van der Wel et al. (2014). This way,we sample our sources up to the outskirts and obtain the totalemission-line fluxes.A summary of the procedure followed for computing the H α aperture fluxes is shown, for a typical case of a galaxy with ve-locity field, in Fig. 9. We started from the velocity di ff erence Article number, page 13 of 21 & A proofs: manuscript no. main
Fig. 5:
Top Left : histogram of z phot of individual galaxies from the literature. Top right : histogram of the observed H total magnitudefor all observed targets (green filled), for those with high-quality spectroscopic redshifts (validated targets; orange filled) and forthose that could not be assigned a spectroscopic redshift (not validated targets; open blue line). Bottom left : histogram of the SFRderived from SED fitting for the same samples.
Bottom right : histogram of the stellar mass derived from SED fitting for the samesamples.with respect to the galaxy centre estimated with
KUBEVIZ andsaved it as output in the velocity map (top-left panel of the fig-ure). We also assigned a peculiar velocity to the spaxels enteringthe 1 (cid:48)(cid:48) . KUBEVIZ fit. This value is computed progressively asthe mean of the peculiar velocities of the neighbouring spaxels,starting from the most populated (i.e. with the highest number ofgood fit neighbouring spaxels) regions in the map. This method,leading to the smooth velocity map in the aperture (shown inthe bottom-right panel of the figure), assumes that the velocitycurves we are considering are smooth (see Wilman et al. 2020),which is not a strong assumption for discy star-forming galaxies. We then produced a total rest-frame 1D spectrum in the apertureby summing all the spaxels corrected for their relative velocity,as shown in Fig. 10 – where the same galaxy of Fig. 9 is used.Furthermore, we estimated the integrated flux by performing aweighted Gaussian fit to the total rest-frame H α emission line,which was weighted for the noise spectrum. We subtracted thecontinuum contribution in two di ff erent ways. Firstly, we gave arough estimate of the continuum of the spectrum as the mediansigma clipped counts in two windows of 300 pixels in widthblueward and redward of the emission line. Secondly, we con-sidered the continuum on the H α emission as it was estimatedby KUBEVIZ . The method outlined above for measuring the H α emission-line flux does not take into account the H α stellar ab- Article number, page 14 of 21. Guglielmo et al.: C3R2 VLT / KMOS
Fig. 6: Representation of SOM cells targeted by the KMOS programme.
Left : coloured cells are filled with high-quality spectro-scopic redshift measurements in the three fields targeted by our survey, whereas empty cells are occupied by observed and notspectroscopically confirmed targets. The high-quality spectroscopically assigned cells are colour-coded according to the occupationlevel, meaning the number of validated galaxies occupying the same colour cell.
Middle : the SOM cells filled with high-qualityspectroscopic redshift measurements are colour-coded according to the assigned z spec . Right : the observed but still empty SOMcells are colour-coded according to the z phot of the observed targets, whereas high-quality spectroscopic redshift measurements arecoloured in white.Fig. 7: Representation of SOM cells targeted by the KMOS programme. The cells filled with high-quality spectroscopic redshiftmeasurements are coloured in white. Left : cells are colour-coded according to the restframe ( u − g ) colour. Middle : the cells arecolour-coded according to the best fit E ( B − V ) resulting from SED fitting analysis on the photometric sample. Right : the cells arecolour-coded according to the best fit SFR resulting from SED fitting analysis on the photometric sample.sorption, but this is small and can be neglected. Using syntheticspectra representative of our galaxy population (same redshiftrange, delayed SFHs in agreement with the
LePhare best fitmodels), we estimate that the ratio between the equivalent width(EW) of the H α stellar absorption and the EW of the H α emis-sion line (as measured from the KMOS data) is lower than 5%. The H α flux is one of the primary SFR indicators, according tothe well-known Kennicutt (1998) relation, which sets a propor-tionality between H α flux and SFR, see Eq. 8 below. It is knownthat the extinction on the nebular emission is enhanced, on av-erage, with respect to the extinction towards the stellar compo-nent, and several methods and calibrations have been performedto derive it. (1) Observed spectra covering a broad enough wave-length range allow the direct estimate of the absorption throughthe computation of observed emission-line ratios and their com-parison to the theoretical value set by quantum physics, such asthe ratio of the Balmer nebular emission lines H α / H β . (2) Anumber of relations linking the absorption in the continuum tothat in the emission lines (Calzetti et al. 2000; Wuyts et al. 2013) have been studied at various redshift and in di ff erent wavelengthregimes over the last few years (3) Finally, the Kennicutt SFR–H α relation has also been calibrated by means of multiple SFRindicators to derive the best fit nebular extinction value a poste-riori , such as the work performed in Kashino et al. (2019).Considering the items above, the Kennicutt (1998) equation,for a Chabrier (2003) IMF, becomes: F H α [erg cm − s − ] = SFR [M (cid:12) yr − ]4 . × − · π d · − . A H α , (8)where d L is the luminosity distance, and A H α = K H α E ( B − V )f neb . (9) K H α = .
54 is the wavelength dependence of extinction accord-ing to Cardelli et al. (1989), E ( B − V ) is the reddening resultingfrom LePhare, and f neb = . ± .
01 is the enhancement of ex-tinction towards nebular lines calibrated in Kashino et al. (2019).The error associated with each object is 0.15 dex, and it is addedin quadrature to the typical error associated to the flux measure-ment (vertical error bar in Figure 11). We derived SFR using Eq.
Article number, page 15 of 21 & A proofs: manuscript no. main
Fig. 8: Histogram of stellar masses computed by
LePhare on thespectrophotometric catalogues ( z spec sample) built in the threefields. The fields are shown with separate histograms as indicatedby the legend.(8) with the H α aperture fluxes (Sect. 7.2) and the luminositydistance based on the spectroscopic redshift measurements.Figure 11 shows the resulting H α -based SFRs comparedwith those estimated from SED fitting with LePhare . Both dis-tributions peak at log (SFR / M (cid:12) yr − ) ∼ α fluxes (of the order of 0.05-0.1 dex in each of the threefields). We point out that the SFRs derived with LePhare areinstantaneous, in agreement with the definition of a H α -basedSFR. However, di ff erences may arise from (1) the necessary ap-proximations adopted in the SED-fitting procedure in order toderive SFRs as well as other physical parameters (e.g. the num-ber of input SED, the limited number of ages in the grid); (2) theuncertainties in the extinction values derived through the SEDfitting (see Laigle et al. (2019) for details); and (3) the uncer-tainties in the relation between continuum and line absorptionthat we had to adopt to derive the SFR from H α fluxes. Further-more, in light of the considerations previously performed on thesizes of our galaxy sample, this systematic shift is not likely tobe attributable to the di ff erent area considered in the photometrywith respect to the aperture considered for computing the totalH α flux. Indeed, as is noticeable from the stellar mass distribu-tion, these galaxies are less massive than those considered as areference for choosing the appropriate flux aperture. Moreover,SFRs derived from SED fitting are compatible with the scatterof the plot around the 1:1 line (approximately 0.5 dex).The distribution of the derived SFR and stellar masses in theSFR mass plane is shown in Fig. 12. The star-forming mainsequence (MS, black dashed line) parametrisation adopted isa broken power law defined in the stellar mass range 9 . ≤ log ( M (cid:63) / M (cid:12) ) ≤ . . ≤ z ≤ . H -band, the SFRmass relation is lower thanthat at higher redshift ( K -band). In particular, the distribution of Fig. 9: Summary of procedure followed to estimate the H α fluxwithin the 1 (cid:48)(cid:48) . top-left panel shows the velocity map from KUBEVIZ . The star at the centre of the image reprensents thepixel position from which the aperture is estimated. The bottom-left panel shows the distance matrix that defines the six-pixelradius corresponding to the aperture. The top-right panel showswhich spaxels from the original map are discarded because theyfall outside the aperture. Finally, the bottom-right panel showsthe corrected velocity field obtained following the procedure de-scribed in the main text for assigning a peculiar velocity to thespaxels flagged as bad in
KUBEVIZ .both the KMOS H − and K -band sources is systematically higherthan the star-forming main sequence. As already discussed in theSR analysis (Fig. 5, bottom-right panel), this trend indicates thatdue to the low stellar mass of the galaxies observed, the SR isbiased towards highly star-forming galaxies above the MS.In the figure, we also included the SED-fitting-based SFR ofnon-validated galaxies (grey crosses). As is noticeable, at 1 . ≤ z ≤ . H -band), the population of low star-forming galaxiespreviously identified in Sect. 6 emerges; the distribution of greycrosses at 2 . ≤ z ≤ . K -band) is not remarkably di ff erentfrom that of spectroscopically confirmed targets (grey circles),further confirming that spectroscopic failures in this regime aremore likely due to higher uncertainties in z phot .The KMOS (cid:64) C3R2 stellar mass distribution peaks atlog ( M (cid:63) / M (cid:12) ) ∼ .
5, which corresponds to the lower edge ofthe stellar mass distribution of the KMOS 3D galaxies (Wis-nioski et al. 2019). The integration between the two samples laysthe groundwork for building a high-redshift SFR mass relationthat is able to probe a wider stellar mass range, with the ultimategoal of determining the characteristic mass above which a flat-tening of the MS relation is expected to occur (Elbaz et al. 2007at z ∼
1; Daddi et al. 2007 at z ∼
8. Catalogue release
Following the methodology outlined above, we built a table con-taining the redshift assigned in each of the observed pointings,together with some relevant information regarding the observedtargets. The released catalogue collects all high-quality ( Q ≥ Article number, page 16 of 21. Guglielmo et al.: C3R2 VLT / KMOS
Fig. 10: One-dimensional spectrum estimated by summing up all the spaxel spectra in the 1 (cid:48)(cid:48) . .2 radius aperture, corrected for theirpeculiar velocity according to the aperture-corrected velocity map described in the main text (Sect. 7.2). The same galaxy as theone shown in Fig. 9 is used. The main panel shows a wavelength cut of the whole 1D sum spectrum around the H α and N ii lines,which are indicated with orange and black dashed lines, respectively. The inset panel is a zoom-in around the H α peak and showsthe integral of the line that is estimated for measuring the total flux (light blue area) weighted by the noise (red dashed line), and itis also continuum corrected.Fig. 11: Left : histogram of SFR derived from aperture H α fluxes, and that estimated from LePhare
SED fitting.
Right : comparisonbetween the H α and SED-fitting SFRs, colour-coded by galaxy stellar mass. The black dashed line is the one-to-one correlation.The plot also shows the typical error on the SFR from LePhare (horizontal black error bar, calculated using the SFR_INF andSFR_SUP released in the catalogue) and on the H α SFR (considering a typical uncertainty of 10% on the flux measurement, seeWisnioski et al. 2019).catalogue. The properties of a sub-sample of galaxies are givenin Table 5, while the total sample can be found at CDS.The columns indicate the following parameters:1. OBJ_ID: identification number for galaxies2. RA: right ascension (deg)3. Dec: declination (deg)4. Pointing: name of the KMOS OB in which the galaxy hasbeen observed (see Table 2) 5. Z_SPEC: redshift assigned and validated as described inSect. 56. Q_flag: quality flag of the redshift measurement, assignedaccording to the criteria described in Sect. 57. PHOTO-Z: photometric redshift from the galaxy parent sur-vey (details are given in Sect. 5.3)8. Priority (M17): observational priority of the target, accordingto the scheme described in M179. EBV_BEST: E ( B − V ) computed with LePhare
Article number, page 17 of 21 & A proofs: manuscript no. main
Fig. 12: (H α − based) SFR (grey circles) and (SED fitting based) SFR (grey crosses) vs stellar mass. The left panel shows the lowerredshift targets observed in H -band in the three surveys considered in the scientific analysis prensented here, and the right panelrepresents the same for higher redshift K -band targets. The black solid lines are the best fit to the star-forming main sequence (MS)in the same redshift range from Whitaker et al. (2014); the dashed and dotted lines show 4 × and 10 × above and below the MS andbracket the distribution of the data points of the 3D-HST galaxies (see Fig. 7 in Wisnioski et al. 2019).10. MASS_INF: sixteenth percentile of the galaxy stellar massfrom the maximum likelihood (ML) analysis of LePhare
11. MASS_MED: median value of the galaxy stellar mass fromthe ML analysis of
LePhare
12. MASS_SUP: eighty-fourth percentile of the galaxy stellarmass from the ML analysis of
LePhare
13. SFR_INF: sixteenth percentile of the SFR from the maxi-mum likelihood (ML) analysis of
LePhare
14. SFR_MED: median value of the SFR from the ML analysisof
LePhare
15. SFR_SUP: eighty-fourth percentile of the SFR from the MLanalysis of
LePhare
16. F H α, . : H α flux computed within an aperture of 1 (cid:48)(cid:48) .
9. Conclusions
In this work, we present the first results of a 200 h ESO LargeProgramme (199.A-0732; PI F.J. Castander) consisting of VLTspectroscopic observations, as part of the C3R2 survey. Themain goal of C3R2 is to acquire accurate spectroscopic redshiftsacross the relevant galaxy colour space in order to accuratelydetermine the colour-redshift relation for the
Euclid weak lens-ing cosmological survey. As a contribution to this challenginggoal, we release a spectrophotometric catalogue of high-redshiftstar-forming galaxies observed for 88 h with the near-infraredKMOS spectrograph. A total of 424 high-quality spectroscopicredshifts have been determined over five semesters in four ex-tragalactic fields (COSMOS, SXDF, ECDFS, and VVDS-02h),mainly measured as single emission-line redshifts (Q ≤ .
5) intwo near-infrared filters: the H (1.456 – 1.846 µ m) filter allows us to detect H α ( λ = .
61 Å) at 1 . ≤ z ≤ . , and the K (1.934 – 2.460 µ m) filter allows us to detect H α at 2 . ≤ z ≤ . The spectroscopic SR
A total number of 150 new redshifts were measured to galax-ies belonging to the COSMOS field, 81 redshifts to galaxies be-longing to the SXDF field, and 181 to galaxies in the VVDS-02hfield, with an overall SR of 60.7% for H -band observations and32.8% for K -band observations. We divided our target galaxiesinto two priority classes (P1 and P2). We were able to fill the57% of the observed P1 empty cells of the galaxy colour SOM,and 70% of the observed P2 empty cells. In Fig. 4, we notice thatless than 4% of P1 cells and about 50% of P2 cells in the near-infrared domain remain unexplored. However, 18 out of the total269 cells we filled presented some problems in terms of z phot distribution, so they need to be investigated further, and possiblyexcluded from the Euclid calibration sample. Considering ourspectroscopic failures, we found that they mainly include (1) K -band targets whose SR is lower due to observational di ffi cultiesand lower accuracy of the z phot estimate used at the sample se-lection stage, and (2) H -band galaxies with redder colours andlower SFR, which are more di ffi cult to detect with the 1 h inte-gration time adopted by our observations.A follow-up near-infrared observing programme is ongoingwith the Large Bincocular Telescope (LBT), making use of the Article number, page 18 of 21. Guglielmo et al.: C3R2 VLT / KMOS
Table 5: Sub-sample of ten galaxies in the catalogue with their properties. The full table can be found at CDS. The explanation ofthe di ff erent columns is given in Sect. 8. The column ‘ID’ is repeated at the beginning of each part of the table for the sake of clarity.We estimated that, due to the uncertainties in the spectrophotometric calibrations, the precision on the H α flux measurement is notbetter than 10%. OBJ_ID RA Dec Pointing Z_SPEC Q_flag373952 150.36320 2.46340 P100_COSMOS_HaHP1 1.7195 4.0399202 150.37578 2.51607 P100_COSMOS_HaHP1 1.5130 4.0399761 150.34360 2.51690 P100_COSMOS_HaHP1 1.3991 4.0388984 34.74180 − . − . − . − . − . − . − . − . H α, . log ( M (cid:63) / M (cid:12) ) log ( M (cid:63) / M (cid:12) ) log ( M (cid:63) / M (cid:12) ) log (M (cid:12) yr − ) log (M (cid:12) yr − ) log (M (cid:12) yr − ) 10 − erg cm − s − two multi-object spectrographs LUCI1 and LUCI2. Our observ-ing strategy is to simultaneously observe the same pointing using H − and K − band masks with LUCI1 and LUCI2, maintainingthe same integration time of KMOS observations (1 h). This al-lows us to observe many galaxies simultaneously in both filters,and helps us understand how much of the non-detection can beassessed with a broader wavelength range in the spectrum (e.g.in case of the more insecure photo-z estimates in K band targets). The physical properties of the released galaxies
We measured the physical properties of the spectroscopicallyconfirmed galaxies using their KMOS resolved spectra and theiroptical and near-infrared photometry from public data releasecatalogues in the three fields. We measured total H α fluxes in1 (cid:48)(cid:48) . ( M (cid:63) / M (cid:12) ) = .
69 and is similar within the error barsacross the three fields. We finally derived SFRs from the apertureH α flux following the Kennicutt (1998) prescription, taking intoaccount enhanced extinction towards nebular lines in the star- forming regions according to Kashino et al. 2019. We studied thedistribution of our galaxies in the SFR mass plane and comparedour data points with the best fit high-redshift main sequence fromWhitaker et al. (2014). Galaxies observed during our KMOSprogramme are located, on average, at higher SFRs with respectto the average population of similar stellar masses. This result isdue, especially at low stellar masses, to the limitations imposedby our observing strategy, of which the primary goal was to max-imise the number of spectroscopic redshifts measured. The pe-culiarity of our galaxy sample with respect to the literature, andin particular with respect to the KMOS-3D survey, is the stellarmass regime exploited. Our galaxies are, on average, less mas-sive than those observed in KMOS-3D, and could be used as astarting point for future studies aiming to probe the lower stellarmass regime of the high-redshift SFR mass relation. Acknowledgements.
The Euclid Consortium acknowledges the European SpaceAgency and the support of a number of agencies and institutes that have sup-ported the development of
Euclid . A detailed complete list is available on the
Euclid web site ( ). In particular the Academy ofFinland, the Agenzia Spaziale Italiana, the Belgian Science Policy, the CanadianEuclid Consortium, the Centre National d’Etudes Spatiales, the Deutsches Zen-trum für Luft- und Raumfahrt, the Danish Space Research Institute, the Fundaçãopara a Ciênca e a Tecnologia, the Ministerio de Economia y Competitividad,the National Aeronautics and Space Administration, the Netherlandse Onder-zoekschool Voor Astronomie, the Norvegian Space Center, the Romanian SpaceAgency, the State Secretariat for Education, Research and Innovation (SERI) at
Article number, page 19 of 21 & A proofs: manuscript no. main the Swiss Space O ffi ce (SSO), and the United Kingdom Space Agency. Basedon observations collected at the European Southern Observatory under ESO pro-gramme 199.A-0732 (B,D,F,H). VG, RS, AG and RB acknowledge support bythe Deutsches Zentrum f´’ur Luft- und Raumfahrt (DLR) grant 50 QE 1101. FJCacknowledges support from the Spanish Ministry of Science, Innovation andUniversities through grant ESP2017-89838-C3-1-R, and the H2020 programmeof the European Commission through grant 776247. AG acknowledges a Sin-ergia grant from the Swiss National Science Foundation. SA thank the supportPRIN MIUR 2015 “Cosmology and Fundamental Physics: Illuminating the DarkUniverse with Euclid". References
Aihara, H., Armstrong, R., Bickerton, S., et al. 2017, ArXiv e-prints,arXiv:1702.08449Amara, A. & Réfrégier, A. 2007, MNRAS, 381, 1018Arnouts, S., Cristiani, S., Moscardini, L., et al. 1999, MNRAS, 310, 540Arnouts, S., Le Floc’h, E., Chevallard, J., et al. 2013, A&A, 558, A67Calzetti, D., Armus, L., Bohlin, R. C., et al. 2000, ApJ, 533, 682Capak, P., Aussel, H., Ajiki, M., et al. 2007, ApJS, 172, 99Cardelli, J. A., Clayton, G. C., & Mathis, J. S. 1989, ApJ, 345, 245Chabrier, G. 2003, PASP, 115, 763Daddi, E., Dickinson, M., Morrison, G., et al. 2007, ApJ, 670, 156Dahlen, T., Mobasher, B., Faber, S. M., et al. 2013, ApJ, 775, 93Dark Energy Survey Collaboration, Abbott, T., Abdalla, F. B., et al. 2016, MN-RAS, 460, 1270Davies, R. I., Agudo Berbel, A., Wiezorrek, E., et al. 2013, A&A, 558, A56de Jong, J. T. A., Kuijken, K., Applegate, D., et al. 2013, The Messenger, 154,44Elbaz, D., Daddi, E., Le Borgne, D., et al. 2007, A&A, 468, 33Fossati, M., Fumagalli, M., Boselli, A., et al. 2016, MNRAS, 455, 2028Furusawa, H., Kosugi, G., Akiyama, M., et al. 2008, ApJS, 176, 1Hildebrandt, H., Arnouts, S., Capak, P., et al. 2010, A&A, 523, A31Ilbert, O., Arnouts, S., McCracken, H. J., et al. 2006, A&A, 457, 841Ilbert, O., Capak, P., Salvato, M., et al. 2009, ApJ, 690, 1236Kashino, D., Silverman, J. D., Sanders, D., et al. 2019, ApJS, 241, 10Kennicutt, Robert C., J. 1998, ARA&A, 36, 189Kitching, T. D., Taylor, P. L., Capak, P., Masters, D., & Hoekstra, H. 2019, arXive-prints, arXiv:1901.06495Kohonen, T. 2001, Self-Organizing Maps (Springer)Laigle, C., Davidzon, I., Ilbert, O., et al. 2019, MNRAS, 486, 5104Laigle, C., McCracken, H. J., Ilbert, O., et al. 2016, ApJS, 224, 24Laureijs, R., Amiaux, J., Arduini, S., et al. 2011, arXiv e-prints, arXiv:1110.3193Le Fèvre, O., Vettolani, G., Garilli, B., et al. 2005, A&A, 439, 845Lehmer, B. D., Brandt, W. N., Alexander, D. M., et al. 2005, ApJS, 161, 21Lilly, S. J., Le Fèvre, O., Renzini, A., et al. 2007, ApJS, 172, 70LSST Science Collaboration, Abell, P. A., Allison, J., et al. 2009, arXiv e-prints,arXiv:0912.0201Ma, Z., Hu, W., & Huterer, D. 2006, ApJ, 636, 21Masters, D., Capak, P., Stern, D., et al. 2015, ApJ, 813, 53Masters, D. C., Stern, D. K., Cohen, J. G., et al. 2017, ApJ, 841, 111Masters, D. C., Stern, D. K., Cohen, J. G., et al. 2019, ApJ, 877, 81Mehta, V., Scarlata, C., Capak, P., et al. 2018, ApJS, 235, 36Schlegel, D. J., Finkbeiner, D. P., & Davis, M. 1998, ApJ, 500, 525Scoville, N., Aussel, H., Brusa, M., et al. 2007, ApJS, 172, 1Soto, E., de Mello, D. F., Rafelski, M., et al. 2017, ApJ, 837, 6Spergel, D., Gehrels, N., Baltay, C., et al. 2015, arXiv e-prints, arXiv:1503.03757van der Wel, A., Franx, M., van Dokkum, P. G., et al. 2014, ApJ, 788, 28Wegner, M. & Muschielok, B. 2008, Society of Photo-Optical InstrumentationEngineers (SPIE) Conference Series, Vol. 7019, KARMA: the observationpreparation tool for KMOS, 70190TWhitaker, K. E., Franx, M., Leja, J., et al. 2014, ApJ, 795, 104Wilkinson, D. M., Maraston, C., Thomas, D., et al. 2015, MNRAS, 449, 328Wilman, D. J., Fossati, M., Mendel, J. T., et al. 2020, ApJ, 892, 1Wisnioski, E., Förster Schreiber, N. M., Fossati, M., et al. 2019, ApJ, 886, 124Wright, A. H., Hildebrandt, H., van den Busch, J. L., & Heymans, C. 2019, arXive-prints, arXiv:1909.09632Wuyts, S., Förster Schreiber, N. M., Nelson, E. J., et al. 2013, ApJ, 779, 135 Max Planck Institute for Extraterrestrial Physics, Giessenbachstr. 1,D-85748 Garching, Germany Universitäts-Sternwarte München, Fakultät für Physik, Ludwig-Maximilians-Universität München, Scheinerstrasse 1, 81679 München,Germany Institute of Space Sciences (ICE, CSIC), Campus UAB, Carrer deCan Magrans, s / n, 08193 Barcelona, Spain Institut d’Estudis Espacials de Catalunya (IEEC), 08034 Barcelona,Spain Department of Astronomy, University of Geneva, ch. d’Écogia 16,CH-1290 Versoix, Switzerland INAF-Osservatorio di Astrofisica e Scienza dello Spazio di Bologna,Via Piero Gobetti 93 /
3, I-40129 Bologna, Italy California institute of Technology, 1200 E California Blvd, Pasadena,CA 91125, USA Cosmic Dawn Center (DAWN), Niels Bohr Institute, Universityof Copenhagen, Vibenshuset, Lyngbyvej 2, DK-2100 Copenhagen,Denmark Aix-Marseille Univ, CNRS, CNES, LAM, Marseille, France Infrared Processing and Analysis Center, California Institute ofTechnology, Pasadena, CA 91125, USA Jet Propulsion Laboratory, California Institute of Technology, 4800Oak Grove Drive, Pasadena, CA, 91109, USA INAF-Osservatorio Astronomico di Brera, Via Brera 28, I-20122Milano, Italy Universidad de la Laguna, E-38206, San Cristóbal de La Laguna,Tenerife, Spain Instituto de Astrofísica de Canarias. Calle Vía Làctea s / n, 38204,San Cristóbal de la Laguna, Tenerife, Spain Dipartimento di Fisica e Astronomia, Universitá di Bologna, ViaGobetti 93 /
2, I-40129 Bologna, Italy INFN-Sezione di Bologna, Viale Berti Pichat 6 /
2, I-40127 Bologna,Italy IFPU, Institute for Fundamental Physics of the Universe, via Beirut2, 34151 Trieste, Italy INAF-Osservatorio Astronomico di Trieste, Via G. B. Tiepolo 11,I-34131 Trieste, Italy INAF-Osservatorio Astrofisico di Torino, Via Osservatorio 20,I-10025 Pino Torinese (TO), Italy Department of Mathematics and Physics, Roma Tre University, Viadella Vasca Navale 84, I-00146 Rome, Italy Institut de Recherche en Astrophysique et Planétologie (IRAP),Université de Toulouse, CNRS, UPS, CNES, 14 Av. Edouard Belin,F-31400 Toulouse, France INAF-Osservatorio Astronomico di Capodimonte, Via Moiariello16, I-80131 Napoli, Italy INFN-Bologna, Via Irnerio 46, I-40126 Bologna, Italy Dipartimento di Fisica e Scienze della Terra, Universitá degli Studidi Ferrara, Via Giuseppe Saragat 1, I-44122 Ferrara, Italy INAF, Istituto di Radioastronomia, Via Piero Gobetti 101, I-40129Bologna, Italy INFN-Sezione di Torino, Via P. Giuria 1, I-10125 Torino, Italy Dipartimento di Fisica, Universitá degli Studi di Torino, Via P. Giuria1, I-10125 Torino, Italy Université Côte d’Azur, Observatoire de la Côte d’Azur, CNRS,Laboratoire Lagrange, France INAF-IASF Milano, Via Alfonso Corti 12, I-20133 Milano, Italy Institut de Física d’Altes Energies IFAE, 08193 Bellaterra,Barcelona, Spain Instituto de Astrofísica e Ciências do Espaço, Faculdade de Ciências,Universidade de Lisboa, Tapada da Ajuda, PT-1349-018 Lisboa,Portugal AIM, CEA, CNRS, Université Paris-Saclay, Université Paris Diderot,Sorbonne Paris Cité, F-91191 Gif-sur-Yvette, France INAF-Osservatorio Astronomico di Roma, Via Frascati 33, I-00078Monteporzio Catone, Italy Observatoire de Sauverny, Ecole Polytechnique Fédérale de Lau-sanne, CH-1290 Versoix, Switzerland Department of Physics "E. Pancini", University Federico II, ViaArticle number, page 20 of 21. Guglielmo et al.: C3R2 VLT / KMOSCinthia 6, I-80126, Napoli, Italy INFN section of Naples, Via Cinthia 6, I-80126, Napoli, Italy INAF-Osservatorio Astrofisico di Arcetri, Largo E. Fermi 5, I-50125,Firenze, Italy Centre National d’Etudes Spatiales, Toulouse, France Institute for Astronomy, University of Edinburgh, Royal Observatory,Blackford Hill, Edinburgh EH9 3HJ, UK University of Nottingham, University Park, Nottingham NG7 2RD,UK European Space Agency / ESRIN, Largo Galileo Galilei 1, 00044Frascati, Roma, Italy ESAC / ESA, Camino Bajo del Castillo, s / n., Urb. Villafranca delCastillo, 28692 Villanueva de la Cañada, Madrid, Spain Univ Lyon, Univ Claude Bernard Lyon 1, CNRS / IN2P3, IP2I Lyon,UMR 5822, F-69622, Villeurbanne, France University of Lyon, UCB Lyon 1, CNRS / IN2P3, IUF, IP2I Lyon,France Mullard Space Science Laboratory, University College London,Holmbury St Mary, Dorking, Surrey RH5 6NT, UK Departamento de Física, Faculdade de Ciências, Universidade deLisboa, Edifício C8, Campo Grande, PT1749-016 Lisboa, Portugal Instituto de Astrofísica e Ciências do Espaço, Faculdade de Ciências,Universidade de Lisboa, Campo Grande, PT-1749-016 Lisboa, Portugal Department of Physics, Oxford University, Keble Road, Oxford OX13RH, UK INFN-Padova, Via Marzolo 8, I-35131 Padova, Italy School of Physics, HH Wills Physics Laboratory, University ofBristol, Tyndall Avenue, Bristol, BS8 1TL, UK Department of Physics, P.O. Box 64, 00014 University of Helsinki,Finland Department of Physics and Helsinki Institute of Physics, GustafHällströmin katu 2, 00014 University of Helsinki, Finland Dipartimento di Fisica "Aldo Pontremoli", Universitá degli Studi diMilano, Via Celoria 16, I-20133 Milano, Italy INFN-Sezione di Milano, Via Celoria 16, I-20133 Milano, Italy Astronomisches Institut, Ruhr-Universität Bochum, Universitätsstr.150, 44801 Bochum, Germany Leiden Observatory, Leiden University, Niels Bohrweg 2, 2333 CALeiden, The Netherlands von Hoerner & Sulger GmbH, SchloßPlatz 8, D-68723 Schwetzin-gen, Germany Max-Planck-Institut für Astronomie, Königstuhl 17, D-69117Heidelberg, Germany Aix-Marseille Univ, CNRS / IN2P3, CPPM, Marseille, France Institut d’Astrophysique de Paris, 98bis Boulevard Arago, F-75014,Paris, France Institut de Physique Nucléaire de Lyon, 4, rue Enrico Fermi, 69622,Villeurbanne cedex, France Université de Genève, Département de Physique Théorique andCentre for Astroparticle Physics, 24 quai Ernest-Ansermet, CH-1211Genève 4, Switzerland European Space Agency / ESTEC, Keplerlaan 1, 2201 AZ Noordwijk,The Netherlands Institute of Theoretical Astrophysics, University of Oslo, P.O. Box1029 Blindern, N-0315 Oslo, Norway NOVA optical infrared instrumentation group at ASTRON, OudeHoogeveensedijk 4, 7991PD, Dwingeloo, The Netherlands Institute of Cosmology and Gravitation, University of Portsmouth,Portsmouth PO1 3FX, UK Argelander-Institut für Astronomie, Universität Bonn, Auf demHügel 71, 53121 Bonn, Germany Centre for Extragalactic Astronomy, Department of Physics, DurhamUniversity, South Road, Durham, DH1 3LE, UK Université Côte d’Azur, Observatoire de la Côte d’Azur, CNRS,Laboratoire Lagrange, Bd de l’Observatoire, CS 34229, 06304 Nicecedex 4, France Université de Paris, F-75013, Paris, France, LERMA, Observatoirede Paris, PSL Research University, CNRS, Sorbonne Université,F-75014 Paris, France Istituto Nazionale di Astrofisica (INAF) - Osservatorio di Astrofisica e Scienza dello Spazio (OAS), Via Gobetti 93 /
3, I-40127 Bologna, Italy INAF-IASF Bologna, Via Piero Gobetti 101, I-40129 Bologna, Italy INFN, Sezione di Lecce, Via per Arnesano, CP-193, I-73100, Lecce,Italy Department of Mathematics and Physics E. De Giorgi, University ofSalento, Via per Arnesano, CP-I93, I-73100, Lecce, Italy Dipartimento di Fisica e Astronomia “G.Galilei", Universitá diPadova, Via Marzolo 8, I-35131 Padova, Italy Departamento de Física, FCFM, Universidad de Chile, BlancoEncalada 2008, Santiago, Chile Centro de Investigaciones Energéticas, Medioambientales y Tec-nológicas (CIEMAT), Avenida Complutense 40, 28040 Madrid, Spain Universidad Politécnica de Cartagena, Departamento de Electrónicay Tecnología de Computadoras, 30202 Cartagena, Spain Jodrell Bank Centre for Astrophysics, School of Physics andAstronomy, University of Manchester, Oxford Road, Manchester M139PL, UK80