CLASH: Discovery of a Bright z~6.2 Dwarf Galaxy Quadruply Lensed by MACS J0329.6-0211

2.1 MB·English

by  A. Zitrin

Submitted to the Astrophysical Journal Letters
PreprinttypesetusingLATEXstyleemulateapjv.8/13/10
CLASH:DISCOVERYOFABRIGHTz (cid:39)6.2DWARFGALAXYQUADRUPLYLENSEDBYMACSJ0329.6-0211
A. Zitrin1, J. Moustakas2, L. Bradley3, D. Coe3, L.A. Moustakas4, M. Postman3, X. Shu5, W. Zheng6, N.
Ben´ıtez7, R. Bouwens8, T. Broadhurst9,10, H. Ford6, O. Host11, S. Jouvel11, A. Koekemoer3, M. Meneghetti12,
P. Rosati13, M. Donahue14, C. Grillo15, D. Kelson16, D. Lemze6, E. Medezinski6, A. Molino7, M. Nonino17, and S.
Ogaz3
Submitted to the Astrophysical Journal Letters
ABSTRACT
We report the discovery of a z =6.18+0.05 (95% confidence level) dwarf galaxy, lensed into four
phot −0.07
2
images by the galaxy cluster MACS J0329.6-0211 (z = 0.45). The galaxy is observed as a high-
1 l
redshift dropout in HST/ACS/WFC3 CLASH and Spitzer/IRAC imaging. Its redshift is securely
0
determined due to a clear detection of the Lyman-break in the 18-band photometry, making this
2
galaxy one of the highest-redshift multiply-lensed objects known to date with an observed magnitude

n ofF125W=24.00±0.04ABmagforitsmostmagnifiedimage. Wealsopresentthefirststrong-lensing
a analysis of this cluster uncovering 15 additional multiply-imaged candidates of five lower-redshift
J sourcesspanningtherangez (cid:39)2−4. Themassmodelindependentlysupportsthehighphotometric
s
8 redshiftandrevealsmagnificationsof11.6+8.9,17.6+6.2,3.9+3.0,and3.7+1.3,respectively,forthefour
−4.1 −3.9 −1.7 −0.2
2 images of the high-redshift galaxy. By delensing the most magnified image we construct an image of
the source with a physical resolution of ∼ 200 pc when the universe was ∼ 0.9 Gyr old, where the
] z (cid:39)6.2galaxyoccupiesasource-planeareaofapproximately2.2kpc2. Modelingtheobservedspectral
O
energy distribution (SED) using population synthesis models, we find a demagnified stellar mass of
C ∼109 M , subsolar metallicity (Z/Z ∼0.5), low dust content (A ∼0.1 mag), a demagnified SFR
(cid:12) (cid:12) V
h. of∼3.2M(cid:12) yr−1,andaspecificSFRof∼3.4Gyr−1,allconsistentwiththepropertiesoflocaldwarf
p galaxies.
- Subject headings: darkmatter,galaxies: clusters: individuals: MACSJ0329.6-0211,galaxies: clusters:
o
general, galaxies: high-redshift, gravitational lensing: strong
r
t
s
a
[ 1. INTRODUCTION highest-redshift lensedgalaxy is at z =7.6 inthe fieldof
Abell1689(Bradleyetal.2008),butamongthehighest-
By gravitationally lensing distant background sources,
3 redshift multiply-lensed galaxies known are a galaxy at
galaxy clusters act as natural magnifying lenses in the
v z =6.03inthefieldofAbell383(comprisingtwoimages;
sky, providing a unique window into the early Uni-
6 Richardetal.2011;Zitrinetal.2011c), az ∼6.5galaxy
verse. Severalhigh-redshiftgalaxieshavebeendiscovered
0 inthefieldofAbell2218(comprisingthreeimages;Kneib
through the magnification power of galaxy clusters (e.g.
0 et al. 2004; Egami et al. 2005), and a z ∼7 candidate in
Franxetal.1997;Fryeetal.2002;Bouwensetal.2009b;
5 the “bullet cluster” (comprising two images; Hall et al.
Zheng et al. 2009; Bradley et al. 2011). The current
.
1 2011). In addition, the lens model can be used to map
1 the lensed images back into the source plane, while the
1Institut fu¨r Theoretische Astrophysik, Universit¨at Heidel-
1 high magnification enables their spatially resolved inter-
berg;[email protected]
1 2Center for Astrophysics and Space Sciences, University of nalstructuralpropertiestobemeasured. Suchmeasure-
: California,SanDiego ments are not possible without the aid of the lensing
v 3SpaceTelescopeScienceInstitute,Baltimore power of the cluster (e.g. Zitrin et al. 2011b).
i 4Jet Propulsion Laboratory, California Institute of Technol-
X The Lyman-alpha forest produces a sharp drop in flux
ogy
r 5UniversityofScienceandTechnologyofChina,Hefei,Anhui below 1216 ˚A. The expansion of the universe moves
a 6DepartmentofPhysicsandAstronomy,TheJohnsHopkins this spectral break to longer wavelengths, allowing high-
University
7InstitutodeAstrof´ısicadeAndaluc´ıa(CSIC) redshift galaxies to be identified as “dropouts” in the
8LeidenObservatory,UniversityofLeiden observed-frame optical and near-IR (see Madau 1995;
9Department of Theoretical Physics, University of Basque Franx et al. 1997). As these young, high-redshift galax-
Country ies are usually actively forming stars, their rest-frame
10IKERBASQUE,BasqueFoundationforScience
UV spectra should be relatively blue. This combination
11Department of Physics & Astronomy, University College
London of properties allows the redshift of the galaxy to be es-
12INAF, Osservatorio Astronomico di Bologna; INFN, timated accurately, even without spectroscopy: as low
SezionediBologna as ∼ 1% uncertainty on the redshift is obtained here,
13EuropeanSouthernObservatory
for 95% confidence levels. The study of high-redshift
14Physics and Astronomy Department, Michigan State
University galaxies enables important constraints to be placed on
15Excellence Cluster Universe, Technische Universita¨t galaxy evolution and structure formation. Particularly,
Mu¨nchen observinghigh-redshiftgalaxiesprovidesdirectmeasure-
16Observatories of the Carnegie Institution of Washington,
ments of the high-z luminosity-function, early SFR, and
Pasadena
17INAF-OsservatorioAstronomicodiTrieste the epoch of the intergalactic medium reionization (see
2 Zitrin et al.
Fig. 1.— Galaxy cluster MACS0329 (z = 0.45) imaged with HST/ACS/WFC3. We number the multiple-images uncovered by our
model,wherethetwocandidatesystemsaremarkedin“c”. Allsystemsareaccuratelyreproducedbyourmodel,withanaverageimage-
planereproductionuncertaintyof2.07(cid:48)(cid:48) perimage, andimage-planermsof2.32(cid:48)(cid:48), excludingcandidatesystems. Thewhitecriticalcurve
correspondstosystem1atzs=6.18,andthebluecriticalcurvecorrespondstosystem2atzs=2.17. Thecompositionofthiscolorimage
isRed=F105W+F110W+F125W+F140W+F160W,Green=F606W+F625W+F775W+F814W+F850LP,andBlue=F435W+F475W.The
upper-right insetshowsablow-upofthez=6.18system,belowwhichweinsetthesource-galaxyasreproducedbyourmodelbydelensing
the most magnified image 1.2 (µ ∼ 17.6) to the source plane, yielding a resolution of ∼ 200 pc per pixel. Overplotted therein are bars
showing the source-planesize of 0.2(cid:48)(cid:48) ((cid:39)1.2 kpc). Noisecleaningand colormanipulation procedures were performed for abetterview of
theinternaldetailsofthesource.
Bradley et al. 2011, and references therein). the central field of MACS0329 (Fig. 1) so that its mass
We report the discovery of one of the highest-redshift distributionandinnerprofilecanbewell-constrained,al-
(z = 6.18+0.05) multiply-lensed galaxies known to lowing us to deduce the source-plane properties of the
phot −0.07
date, lensed into four images by the galaxy cluster high-redshift galaxy presented here.
MACS J0329.6-0211 (z=0.45; MACS0329 hereafter). The paper is organized as follows: In §2 we describe
MACS0329 is an X-ray selected system found by the theobservations,andin§3wedetailtheSLmodel. In§4
Massive Cluster Survey, MACS (Ebeling et al. 2001, we report the results of the SL analysis and the physical
2010). Maughan et al. (2008) classified MACSJ0329 properties of the high-redshift galaxy based on detailed
as having evidence for substructure in its X-ray surface SED modeling. The results and conclusions are summa-
brightness, although Schmidt & Allen (2007) classified rized in §5. We adopt a concordance ΛCDM cosmology
it as relaxed. We found no record of previous strong with Ωm0 =0.3, ΩΛ0 =0.7, and h=0.7. With these pa-
lensing (SL) analysis of this cluster, and only 1-band rameters, one arcsecond corresponds to a physical scale
shallow HST/WFPC2 previous imaging. The 16 HST of 5.76 kpc at the cluster redshift (z =0.45; Allen et al.
bandschosenfortheClusterLensingandSupernovasur- 2008), and 5.62 kpc at z =6.18.
vey with Hubble (CLASH; Postman et al. 2011) enable
us to conduct the first SL study of this cluster, and to 2. OBSERVATIONSANDREDSHIFTS
obtain accurate photometric redshifts for the multiply- MACS0329 was observed as part of CLASH with HST
lensed background galaxies. We use these data in con- from 2011 August to 2011 October. This is one of 25
junction with a parametric SL modeling method (e.g., clusters to be observed to a total depth of 20 HST or-
Broadhurstetal.2005;Zitrinetal.2009,2011a,b;Merten bits in 16 filters with the Wide Field Camera 3 (WFC3)
etal.2011),tofindseveralmultipleimagefamiliesacross UVIS and IR cameras, and the Advanced Camera for
Surveys (ACS) WFC. The images are reduced and mo-
z (cid:39)6.2 Quadruply-Lensed Dwarf Galaxy 3
saiced (0(cid:48).(cid:48)065 pixel−1) using standard techniques imple- exampleseeFig. 4),andthemodelissuccessivelyrefined
mentedintheMosaicDrizzlepipeline(Koekemoeretal. as additional sets of multiple images are incorporated.
2002, 2011). HST photometry for the arcs is obtained
within the apertures shown in Fig. 2, and photometric 4. RESULTS
redshifts are estimated for all the galaxies in the field
We uncovered in HST/CLASH and Spitzer/IRAC
using the full 16-band UVIS/ACS/WFC3-IR aperture-
imaging of MACS0329 one of the highest-redshift
matched (e.g., Fig 2) photometry via both the BPZ
multiply-lensed galaxies known to date. We use the in-
(Ben´ıtez 2000; Ben´ıtez et al. 2004; Coe et al. 2006) and
dependent photometric-redshift distributions of its four
LePhare (Arnouts et al. 1999; Ilbert et al. 2006) pro-
lensed images to obtain a source redshift (and 2σ lim-
grams. See Postman et al. (2011) for additional details.
its, see Table 1) of z = 6.18+0.05. Using our mass
We supplement these HST observations with Infrared phot −0.07
model and the extensive imaging and resulting photo-
Array Camera (IRAC) imaging obtained in 2011 Octo-
metric redshifts, we physically matched 15 additional
ber as part of the Spitzer Cycle 8 warm mission (PI:
new multiple images and candidates of five background,
R. Bouwens; see also Postman et al. 2011). The Warm
lower-z sources (see Figure 1), which are used in turn to
Spitzer IRAC observations in Channels 1 and 2 (3.6 and
refinethemassmodel. Explicitly, inadditiontothefour
4.5µm) have a total integration time of 13,290s, and
images of the high-z galaxy, we use as additional con-
clearlydetectthreeofthefourimagesofthebackground
straintsthesevenmultiple-imagesofsystems2and3, at
source we report here (1.1 through 1.3), with a formal
photometric redshifts (and 2σ limits) of z = 2.17+0.12
∼ 2σ detection for arc 1.4. Because of crowding in the s −0.03
Spitzer images, we adopt a circular 2.4(cid:48)(cid:48) diameter aper- andzs =2.89+−00..0055,respectively,andverifythatallother
turefortheIRACphotometry. Backgroundlevelsarees- systems and candidates are plausible in the context of
timated by sampling non-crowded sections of the IRAC the resulting mass model.
mosaic while avoiding nearby bright sources. The ac- The magnification by the cluster is coupled to the
curacy of this (manual) photometry is verified by sub- surface-densitygradientineachpoint(i.e.,themasspro-
tracting possible contaminating sources with GALFIT1, file), so multiple-systems with redshift information are
especially for arc 1.2 which has a bright star nearby. We important to assess the true magnification across the
use a flux aperture correction factor of 1.9×, based on cluster field. For the four images of the zphot = 6.18
a curve of growth analysis of unresolved sources in the galaxy, we obtain magnifications (and 1σ uncertainties)
Post Basic Calibrated Data (PBCD) mosaic. The rele- of11.6+8.9,17.6+6.2,3.9+3.0,and3.7+1.3,respectively,so
−4.1 −3.9 −1.7 −0.2
vant multi-wavelength photometry is given in Table 1. that in total the galaxy is magnified by a factor of (cid:39)37.
We leverage the magnification boost of the most mag-
3. STRONGLENSINGMODELINGANDANALYSIS nified image, arc 1.2, to construct a high-resolution im-
Thelensmodelingmethodusedhere(e.g.,Broadhurst age of the background galaxy, which occupies a source-
etal.2005;Zitrinetal.2009)beginswiththeassumption plane area of ∼2.2 kpc2 (see Figure 1). To further con-
thatmassapproximatelytraceslight. Wemodelthedis- strain the physical properties of this galaxy, we model
tribution of cluster mass by assigning a power-law mass the observed SED of arc 1.2 using the Bayesian SED-
profile to each red-sequence cluster galaxy, scaled by its fitting code iSEDfit (see Moustakas et al. 2011) cou-
(relative) brightness. The sum of all galaxy contribu- pled to the flexible stellar population synthesis models
tions represents the lumpy galaxy component, which is of Conroy et al. (2009). We adopt the Chabrier (2003)
then smoothed using 2D spline interpolation to obtain a initial mass function from 0.1 − 100 M , assume the
(cid:12)
smooth-component representing the dark matter (DM) Calzetti et al. (2000) dust attenuation law, and adopt
distribution. The polynomial degree of smoothing and uniformpriors(basedonMonteCarlodraws)onthestel-
the index of the power-law are the most important free lar metallicity (0.0002<Z <0.03), V-band attenuation
parameters determining the mass profile. (0 < A < 2 mag), and galaxy age (0.005 − 1 Gyr).
V
A worthwhile improvement in fitting the location of For reference, the age of the universe at z = 6.18 is
the lensed images is generally found by introducing an 0.9 Gyr. We parameterize the star formation history
external shear describing the overall matter ellipticity. ψ(t) as an exponentially declining function of time, t,
The direction of the shear and its amplitude are free given by ψ(t)∝exp(−t/τ), where τ is the characteristic
parameters, allowing for some flexibility in the relation time for star formation. We draw τ from a uniform dis-
between the distribution of DM and the distribution of tribution between 0.01−5 Gyr, which spans the range
galaxies. Theweightofthelumpycomponentrelativeto from passively evolving to continuous star formation.
the DM, and the overall normalization, bring the total In Figure 3 we show the observed and rest-frame
number of free parameters in our modeling to six (see SED of arc 1.2, the maximum likelihood model fit, and
Zitrin et al. 2009). the posterior distributions on all the model parameters.
The best fit is assessed by a χ2 minimization in the Adoptingthemedianofeachposteriordistributionasour
imageplane, whichispreferredbecauseitgenerallydoes best estimate of the properties of the galaxy, we find a
not produce solutions that are biased towards shallow demagnified stellar mass of ∼109 M , low dust content
(cid:12)
mass profiles (and high magnifications), as is the case (A ∼0.1 mag), a demagnified SFR of ∼3.2 M yr−1,
V (cid:12)
withsource-planeminimization. Thepositionsandmor- and a SFR-weighted age of ∼ 180 Myr. These results
phologies of the multiply-imaged arcs are accurately re- imply a specific SFR, sSFR≡SFR/M, of ∼ 3.4 Gyr−1,
produced by the best-fit mass model in their measured corresponding to a mass-doubling time of just 600 Myr
photometric redshifts (for a morphological comparison (assuming a 50% return fraction). The stellar metal-
licity and τ parameter (not shown) are not particularly
1http://users.obs.carnegiescience.edu/peng/work/galfit/galfit.html wellconstrained,althoughsolutionswithsubsolarmetal-
4 Zitrin et al.
Fig. 2.— Thumbnails (5(cid:48)(cid:48)×5(cid:48)(cid:48)) of the four arcs of the z = 6.18 source in the different bands, with tailored apertures used for the
photometryoverlaid. Asisclear,thesearenotseenbelowtheF814Wband,inagreementwiththephotometricredshiftestimate. SeeFig.
3andTable1fortheSEDandphotometry,respectively.
TABLE 1
Multiple-images data and multiband photometry for the four images of the z=6.18 galaxy
ARC RA DEC BPZzphot LPZzphot Magnification,µ
ID (J2000.0) (J2000.0) (best)[95%C.L.] (best)[95%C.L.]
1.1 03:29:40.18 -02:11:45.60 6.185[6.079–6.282] 6.155[5.885–6.374] 11.6+8.9
−4.1
1.2 03:29:40.06 -02:11:51.72 6.137[6.024–6.204] 6.194[5.982–6.347] 17.6+6.2
−3.9
1.3 03:29:41.28 -02:12:05.04 6.152[6.055–6.257] 6.253[5.828–6.573] 3.9+3.0
−1.7
1.4 03:29:43.16 -02:11:17.30 6.246[6.073–6.395] 5.965[5.482–6.327] 3.7+1.3
−0.2
Multi-bandphotometry
ARC F775W F814W F850LP F105W F110W F125W F140W F160W 3.6µm 4.5µm
1.1 26.24±0.32 25.69±0.12 24.73±0.11 24.41±0.05 24.39±0.04 24.46±0.05 24.52±0.04 24.62±0.05 24.4±0.2 24.7±0.3
1.2 >26.52(2σ) 25.32±0.10 24.29±0.08 23.94±0.04 23.90±0.03 24.00±0.04 24.04±0.03 24.11±0.04 23.6±0.2 23.5±0.1
1.3 26.52±0.33 26.30±0.17 25.52±0.17 24.83±0.06 24.93±0.05 24.87±0.06 24.91±0.05 25.01±0.06 24.9±0.6 24.5±0.3
1.4 >29.94(2σ) >27.76(2σ) 25.28±0.14 26.17±0.13 26.25±0.10 26.54±0.17 26.42±0.13 26.28±0.13 24.8±0.2 >24.9(2σ)
Note. — Data for the multiple images of the z = 6.18 galaxy. Top: Columns are: arc ID; RA and DEC; photo-z using BPZ and LePhare
(hereafterLPZ),respectively;µ,magnificationpredictedbythemassmodel. Bottom: Multibandphotometryforthedropoutband(F775W)and
all redder bandpasses, for the four images of the z = 6.18 galaxy, in AB mags. Lower (2σ) limits are given for non-detections; we find no 2σ
detections in any of the 8 bluer bands. Note also that the (relative) magnitudes between the different images of this system agree overall with
the(relative)magnificationsderivedbyourmassmodel(see§4),withsomedeviationforthelessmagnifiedimages,1.3(1.4),whichseemslightly
less(more)magnifiedthanourmodelimplies.
licity and τ (cid:29) 0 Gyr are generally favored; the median ferent centers, we find this offset may introduce uncer-
of the posterior distributions imply subsolar metallicity, tainty of ∼0.3 mag in the IRAC photometry. The effect
Z/Z ∼0.5, and τ ∼2.4 Gyr. of possible contamination from bright neighbors was re-
(cid:12)
We verified that performing our SED modeling on the examinedforbothemission-centers,andwasfoundtobe
other (less magnified) images yields overall similar re- typically0.4mag. However,weimportantlyverifiedthat
sults. ThedemagnifiedSEDofarc1.1isnearlyidentical these higher uncertainties have only a negligible (< 0.1
to that of arc 1.2, and therefore the posterior distribu- dex)effectontheresultingphysicalproperties, asthefit
tionsofthephysicalquantitieswederiveareverysimilar is governed by the HST photometry.
to those shown in Figure 3; the median quantities all Finally,weuseourlensmodeltoconstrainthephysical
agree to well within 0.1 dex (±50%). The demagnified characteristics of MACS0329. For the z = 6.18 source,
s
F125W magnitude of arc 1.3(1.4), on the other hand, is thecriticalcurvesenclosearelativelylargearea,withan
0.8magbrighter(fainter)thanarcs1.1and1.2,although effective Einstein radius of r =33(cid:48).(cid:48)9±3(cid:48)(cid:48) ((cid:39)195 kpc at
E
this discrepancy is well within the statistical uncertain- z =0.45), and a projected mass of 1.89+0.10×1014 M
l −0.06 (cid:12)
tiesonthemagnifications. Nevertheless,atfacevaluethe
(Figure 1). For the lower redshift of system 2, z =
s
fainter two arcs imply an 0.3-0.4 dex larger stellar mass 2.17,theEinsteinradiusis(cid:39)27(cid:48).(cid:48)7,andthecriticalcurve
andSFR,0.2-0.6magmoredustattenuation,andavery encloses a projected mass of 1.40+0.09 ×1014 M . For
similar age, 200 Myr. However, we emphasize that the −0.05 (cid:12)
completeness, we measure the (total) mass profile slope,
posteriordistributionsonthesequantitiesoverlapsignif-
dlogΣ/dlogr (cid:39) −0.61+0.05 (in the range 1(cid:48)(cid:48) −52(cid:48)(cid:48), or
icantlywiththoseofarcs1.1and1.2,andthereforethese −0.1
differencesdonotaffectourconclusionsaboutthenature 6 (cid:46) r (cid:46) 300 kpc; about twice the Einstein radius for
of this object. zs ∼2), typical of relaxed and well-concentrated lensing
Note also that the center of emission in the 3.6µm im- clusters (e.g. Broadhurst et al. 2005; Zitrin et al. 2009),
age is slightly different than that of the 4.5µm image andinagreementwiththefairlycircularly-symmetricX-
(Figure 2). Repeating the photometry on the two dif- ray contours centered on the BCG (see Mann & Ebeling
z (cid:39)6.2 Quadruply-Lensed Dwarf Galaxy 5
Fig. 4.—Reproduction ofsystems1 byourmodel(lower row),
comparedwiththerealimages(upperrow). Wedelensimage1.1to
thesourceplaneandrelensitbacktotheimage-planetoreproduce
theotherimagesofthissystem. Ascanbeseen,ourmodelrepro-
duceswelltheimagesofthissystem. Notealso,ourmodelrequires
largerdeflectionanglesforthez=6.18source,relativetolower-z
systems such as systems 2 and 3, independently strengthening its
detectionasahigh-redshiftsource.
a source-plane area of ∼2.2 kpc2, similar to previously
deduced sizes of high-z lensed galaxies (e.g. Zitrin et al.
2011b). Due to the hierarchical growth of structure,
galaxies are expected to be small at high redshifts, with
dwarfgalaxiesconstitutingthebuildingmaterialoflarger
structures. Our source-plane reconstruction shows at
leastthree(possiblystar-forming)knots,consistentwith
several other reports of high-redshift galaxies with mul-
Fig. 3.— SED modeling results of arc 1.2, the most magnified
image of the z (cid:39)6.18 galaxy. The upper panel shows the ob- tiple components (Franx et al. 1997; Bradley et al. 2008,
phot
servedandrest-frameSEDandthemaximumlikelihoodmodelfit. 2011; Zheng et al. 2009; Oesch et al. 2010; Zitrin et al.
Thefilledredpointsindicatebandsinwhichtheobjectiswellde- 2011b), possibly as the result of merging. In addition,
tected,whiletheopenbluetrianglesindicate2σ upperlimits;the we measure an overall half-light radius of ∼ 0.12(cid:48)(cid:48), con-
open squares show the photometry of the best-fitting model con-
volvedwiththeWFC3,ACS,andIRACfilterresponsefunctions. sistent with that found in Bouwens et al. (2004, 2006)
Theopenredcircleindicatesaspurious,marginallysignificantde- and (Oesch et al. 2010).
tectionofarc1.2intheF606Wband;however,weemphasizethat (4) The SED fits to the multiband photometry of the
noneoftheotherimagesaredetectedinthisband. Wealsoshowan
source suggest a demagnified stellar mass of ∼109 M ,
insetwiththephotometricredshiftdistribution,P(z),obtainedby (cid:12)
combiningtheindependentphotometricredshiftmeasurementsof a SFR-weighted age of ∼ 180 Myr, subsolar metallicity
thefourarcs,yieldingasourceredshiftofz=6.18+0.05 withBPZ, (Z/Z ∼ 0.5), low dust content (A ∼ 0.1 mag), and a
−0.07 (cid:12) V
and6.18+0.11 withLPZ(95%confidencelevels). Thelowerpanels demagnified SFR of ∼ 3.2 M yr−1. The specific SFR
−0.13 (cid:12)
show the posterior distributions on the demagnified stellar mass, of ∼3.4 Gyr−1, which is slightly higher than that found
SFR weighted age, stellar metallicity (relative to Z(cid:12) = 0.019), byotherrecentstudies(Gonzalezetal.2011;Starketal.
restV-banddustattenuation,demagnifiedSFR,andspecificSFR,
2009; Labb´e et al. 2010; McLure et al. 2011), implies a
sSFR≡SFR/M. Theverticaldashedlinesshowthemedianofeach
posteriordistribution. Wesummarizetheresultsin§4. mass-doubling time of just 600 Myr and therefore vigor-
ous ongoing star formation considering its low mass.
2011). (5) The UV continuum is blue, with a UV-slope β =
−2.5±0.06, consistent with measurements of other faint
5. DISCUSSIONANDCONCLUSIONS z ∼6galaxiesandsuggeststhatthesesourcesarelargely
The discovery of a high-redshift galaxy in the field of dust free (Bouwens et al. 2009a, 2011; Finkelstein et al.
MACS0329 adds to several known high-reshift galaxies 2011; Vanzella et al. 2011).
lensedbygalaxyclusters(e.g.,Egamietal.2005;Bradley Thediscoveryofthegalaxypresentedhereshowsonce
etal.2008,2011;Zhengetal.2009;Richardetal.2011). more the novelty and tremendous potential of galaxy
Here we summarize the properties of this unique source. clusters for observationally accessing the faint early uni-
(1) It is one of the highest-redshift multiply lensed ob- verse.
jects known to date, and lensed into four separate im-
ages. Theangularseparationbetweenarcs1.2and1.4is ACKNOWLEDGMENTS
∼ 1(cid:48), considerably larger than previously reported cases The authors thank Saurabh Jha for useful discussions,
(Egami et al. 2005; Richard et al. 2011). and the anonymous referee for valuable comments. The
(2)Thesourceisoneofthebrightestatz >6: itsJ CLASH Multi-Cycle Treasury Program (GO-12065) is
125
magnitude is 24.0 AB, making it a viable candidate for based on observations made with the NASA/ESA Hub-
follow-up spectroscopy. ble Space Telescope. The Space Telescope Science In-
(3)Thegalaxyisconsistentwithbeingadwarfgalaxy. stitute is operated by the Association of Universities for
Its intrinsic (delensed) magnitude of J = 27.1 AB ResearchinAstronomy, Inc. underNASAcontractNAS
125
makes it a sub-L galaxy at this redshift. It occupies 5-26555. This work is based in part on observations
∗
6 Zitrin et al.
made with the Spitzer Space Telescope, which is oper- award issued by JPL/Caltech. We thank the hospital-
ated by the Jet Propulsion Laboratory, California insti- ity of Institut fu¨r Theoretische Astrophysik, Universit¨at
tute of Technology under a contract with NASA. Sup- Heidelberg, where part of this work took place, and the
port for this work was provided by NASA through an Baden-Wu¨erttemberg Stiftung.
REFERENCES
Allen,S.W.,Rapetti,D.A.,Schmidt,R.W.,Ebeling,H., Ilbert,O.,etal.2006,A&A,457,841
Morris,R.G.,&Fabian,A.C.2008,MNRAS,383,879 Kneib,J.-P.,Ellis,R.S.,Santos,M.R.,&Richard,J.2004,ApJ,
Arnouts,S.,Cristiani,S.,Moscardini,L.,Matarrese,S.,Lucchin, 607,697
F.,Fontana,A.,&Giallongo,E.1999,MNRAS,310,540 Koekemoer,A.M.,Fruchter,A.S.,Hook,R.N.,&Hack,W.
Ben´ıtez,N.2000,ApJ,536,571 2002,inThe2002HSTCalibrationWorkshop:Hubbleafter
Ben´ıtez,N.,etal.2004,ApJS,150,1 theInstallationoftheACSandtheNICMOSCoolingSystem,
Bouwens,R.J.,Illingworth,G.D.,Blakeslee,J.P.,Broadhurst, ed.S.Arribas,A.Koekemoer,&B.Whitmore,337
T.J.,&Franx,M.2004,ApJ,611,L1 Koekemoer,A.M.,etal.2011,ApJS,197,36
Bouwens,R.J.,Illingworth,G.D.,Blakeslee,J.P.,&Franx,M. Labb´e,I.,etal.2010,ApJ,716,L103
2006,ApJ,653,53 Madau,P.1995,ApJ,441,18
Bouwens,R.J.,etal.2009a,ApJ,705,936 Mann,A.W.,&Ebeling,H.2011,arXiv,1111.2396
—.2009b,ApJ,690,1764 Maughan,B.J.,Jones,C.,Forman,W.,&VanSpeybroeck,L.
—.2011,arXiv,1109.0994 2008,ApJS,174,117
Bradley,L.D.,etal.2008,ApJ,678,647 McLure,R.J.,etal.2011,MNRAS,1787
—.2011,arXiv,1104.2035 Merten,J.,etal.2011,arXiv,1103.2772
Broadhurst,T.,etal.2005,ApJ,621,53 Moustakas,J.,etal.2011,ApJsubmitted
Calzetti,D.,Armus,L.,Bohlin,R.C.,Kinney,A.L.,Koornneef, Oesch,P.A.,etal.2010,ApJ,709,L21
J.,&Storchi-Bergmann,T.2000,ApJ,533,682 Postman,M.,etal.2011,ApJSsubmitted,arXiv,1106.3328
Chabrier,G.2003,PASP,115,763 Richard,J.,Kneib,J.-P.,Ebeling,H.,Stark,D.P.,Egami,E.,&
Coe,D.,Ben´ıtez,N.,S´anchez,S.F.,Jee,M.,Bouwens,R.,& Fiedler,A.K.2011,MNRAS,414,L31
Ford,H.2006,AJ,132,926 Schmidt,R.W.,&Allen,S.W.2007,MNRAS,379,209
Conroy,C.,Gunn,J.E.,&White,M.2009,ApJ,699,486 Stark,D.P.,Ellis,R.S.,Bunker,A.,Bundy,K.,Targett,T.,
Ebeling,H.,Edge,A.C.,&Henry,J.P.2001,ApJ,553,668 Benson,A.,&Lacy,M.2009,ApJ,697,1493
Ebeling,H.,Edge,A.C.,Mantz,A.,Barrett,E.,Henry,J.P., Vanzella,E.,etal.2011,ApJ,730,L35
Ma,C.J.,&vanSpeybroeck,L.2010,MNRAS,407,83 Zheng,W.,etal.2009,ApJ,697,1907
Egami,E.,etal.2005,ApJ,618,L5 Zitrin,A.,Broadhurst,T.,Barkana,R.,Rephaeli,Y.,&Ben´ıtez,
Finkelstein,S.L.,etal.2011,arXiv,1110.3785 N.2011a,MNRAS,410,1939
Franx,M.,Illingworth,G.D.,Kelson,D.D.,vanDokkum,P.G., Zitrin,A.,Broadhurst,T.,Coe,D.,Liesenborgs,J.,Ben´ıtez,N.,
&Tran,K.-V.1997,ApJ,486,L75 Rephaeli,Y.,Ford,H.,&Umetsu,K.2011b,MNRAS,237
Frye,B.,Broadhurst,T.,&Ben´ıtez,N.2002,ApJ,568,558 Zitrin,A.,etal.2009,MNRAS,396,1985
Gonzalez,V.,Bouwens,R.,Labbe,I.,Illingworth,G.,Oesch,P., —.2011c,arXiv,1103.5618
Franx,M.,&Magee,D.2011,arXiv,1110.6441
Hall,N.,etal.2011,arXiv,1101.4677