DRAFT VERSION MAY 2, 2018
Typeset using L
A
T
E
X default style in AASTeX61
THE DISCOVERY OF A GRAVITATIONALLY LENSED SUPERNOVA IA AT REDSHIFT 2.22
D. RUBIN,
1, 2, 3
B. HAYDEN,
1, 3, 4
X. HUANG,
5
G. ALDERING,
3
R. AMANULLAH,
6
K. BARBARY,
3
K. BOONE,
3, 4
M. BRODWIN,
7
S. E. DEUSTUA,
2
S. DIXON,
3, 4
P. EISENHARDT,
8
A. S. FRUCHTER,
2
A. H. GONZALEZ,
9
A. GOOBAR,
6
R. R. GUPTA,
3
I. HOOK,
10
M. J. JEE,
11
A. G. KIM,
3
M. KOWALSKI,
12
C. E. LIDMAN,
13
E. LINDER,
3
K. LUTHER,
3, 4
J. NORDIN,
12
R. PAIN,
14
S. PERLMUTTER,
3, 4
Z. RAHA,
5, 3
M. RIGAULT,
12
P. RUIZ-LAPUENTE,
15, 16
C. M. SAUNDERS,
3, 4, 14
C. SOFIATTI,
3, 4
A. L. SPADAFORA,
3
S. A. STANFORD,
17
D. STERN,
8
N. SUZUKI,
18
AND S. C. WILLIAMS
10
(THE SUPERNOVA COSMOLOGY PROJECT)
1
D. Rubin and B. Hayden were equal coauthors.
2
Space Telescope Science Institute, 3700 San Martin Drive, Baltimore, MD 21218
3
E.O. Lawrence Berkeley National Lab, 1 Cyclotron Rd., Berkeley, CA, 94720
4
Department of Physics, University of California Berkeley, Berkeley, CA 94720
5
Department of Physics and Astronomy, University of San Francisco, San Francisco, CA 94117-1080
6
The Oskar Klein Centre, Department of Physics, AlbaNova, Stockholm University, SE-106 91 Stockholm, Sweden
7
Department of Physics and Astronomy, University of Missouri-Kansas City, Kansas City, MO 64110
8
Jet Propulsion Laboratory, California Institute of Technology, Pasadena, CA, 91109
9
Department of Astronomy, University of Florida, Gainesville, FL 32611
10
Physics Department, Lancaster University, Lancaster LA1 4YB, United Kingdom
11
Department of Astronomy and Center for Galaxy Evolution Research, Yonsei University, 50 Yonsei-ro, Seoul 03772, Korea
12
Institut f
¨
ur Physik, Newtonstr. 15, 12489 Berlin, Humboldt-Universit
¨
at zu Berlin, Germany
13
Australian Astronomical Observatory, PO Box 296, Epping, NSW 1710, Australia
14
Laboratoire de Physique Nucleaire et de Hautes Energies, Universite Pierre et Marie Curie, 75252 Paris Cedec 05, France
15
Institute of Cosmos Sciences, University of Barcelona, E-08028 Barcelona, Spain
16
Instituto de Fisica Fundamental, CSIC, E-28006 Madrid, Spain
17
Department of Physics, University of California Davis, One Shields Avenue, Davis, CA 95616
18
Kavli Institute for the Physics and Mathematics of the Universe, University of Tokyo, Kashiwa, 277-8583, Japan
Submitted to ApJ
ABSTRACT
We present the discovery and measurements of a gravitationally lensed supernova (SN) behind the galaxy cluster
MOO J1014+0038. Based on multi-band Hubble Space Telescope and Very Large Telescope (VLT) photometry of the su-
pernova, and VLT spectroscopy of the host galaxy, we find a 97.5% probability that this SN is a SN Ia, and a 2.5% chance of
a CC SN. Our typing algorithm combines the shape and color of the light curve with the expected rates of each SN type in the
host galaxy. With a redshift of 2.2216, this is the highest redshift SN Ia discovered with a spectroscopic host-galaxy redshift.
A further distinguishing feature is that the lensing cluster, at redshift 1.23, is the most distant to date to have an amplified SN.
The SN lies in the middle of the color and light-curve shape distributions found at lower redshift, disfavoring strong evolution to
z = 2.22. We estimate an amplification due to gravitational lensing of 2.8
+0.6
−0.5
(1.10 ± 0.23 mag)—compatible with the value
estimated from the weak-lensing-derived mass and the mass-concentration relation from ΛCDM simulations—making it the most
amplified SN Ia discovered behind a galaxy cluster.
Keywords: cosmology: observations — gravitational lensing — galaxies: clusters: individual (MOO J1014+0038)
supernovae: general
Corresponding author: David Rubin
drubin@stsci.edu
arXiv:1707.04606v2 [astro-ph.GA] 1 May 2018
2 RUBIN AND HAYDEN ET AL.
1. INTRODUCTION
Gravitational lensing by massive galaxy clusters offers an amplified and magnified view of the high-redshift universe. Several
examples of supernovae (SNe) lensed by foreground clusters have been found in recent years (Goobar et al. 2009; Amanullah
et al. 2011; Nordin et al. 2014; Patel et al. 2014; Kelly et al. 2015; Rodney et al. 2015a; Petrushevska et al. 2016). Four of
these cluster-lensed SNe have been of Type Ia (SNe Ia), from which the amplification due to lensing has been determined. (Two
additional SNe Ia that were lensed by field galaxies have also been found, Quimby et al. 2014; Goobar et al. 2017; we summarize
all referenced SNe in Table 1).
The uniquely precise standardization possible with SNe Ia provides amplification information, breaking the so-called mass-
sheet degeneracy that is problematic for most shear-based lensing models, and permits independent direct tests of cluster mass
models obtained from galaxy lensing (Nordin et al. 2014; Patel et al. 2014; Rodney et al. 2015a). Lenses that produce multiple
SN images provide additional model constraints (Kelly et al. 2016a). To date the redshifts of gravitationally lensed background
SNe Ia have been at z < 1.39, a redshift range where even without lensing the complete (normal) SN Ia population can be detected
using single-orbit HST visits. In cases of stronger amplification of sources beyond the redshift reach of normal observations, tests
of SN Ia properties and rates over a larger look-back time become possible. Here we report the discovery of the most amplified
SN Ia behind a galaxy cluster ever found, with a spectroscopic host-galaxy redshift making it also the most distant.
2. DISCOVERY
The Supernova Cosmology Project’s “See Change” program (PI: Perlmutter) monitored twelve massive galaxy clusters in the
redshift range 1.13 to 1.75 using the Hubble Space Telescope (HST) Wide Field Camera 3 (WFC3) with the goals of greatly
expanding the range of the high-redshift SN Ia Hubble diagram and accurately determining high-redshift cluster masses via
weak lensing. We observed each cluster every 5 weeks for one orbit split between the UVIS F 814W , the IR F105W , and
the IR F 140W filters. The supernova survey was very deep; our 50% completeness for simulated SN detection was AB 26.6
in F 105W + F 140W (Hayden et al. in prep.); this is ∼ 1 magnitude fainter than a z = 1.75 SN Ia at maximum.
1
When a
promising Type Ia supernova candidate was found, at least one extra visit was triggered and executed within 2–3 observer frame
weeks after the initial discovery. These extra visits provided better light-curve sampling, usually close to maximum light, and
frequently extended the wavelength range with the IR F 160W filter.
On 2016 Feb 29 UTC, we searched images of the WISE-selected (Wright et al. 2010) massive galaxy cluster MOO J1014+0038
(z = 1.23, Decker et al. in prep.; M
200
= (5.6 ± 0.6) × 10
14
M
, Brodwin et al. 2015), from the Massive and Distant Clusters
of WISE Survey (MaDCoWS; Gettings et al. 2012, Stanford et al. 2014, Gonzalez et al. 2015). We detected a red (WFC3
F 140W = 25.1, F 105W − F 140W = 1.4 AB mag; upper limit only in F 814W ) supernova at α = 153.
◦
52655, δ = +0.
◦
64041
(J2000, aligned to USNO-B1). This SN was internally designated as SN SCP16C03,
2
as it was the third SN found in this
cluster field (alphabetically labeled cluster “C”) in 2016. As shown in Figure 1, the SN lies on a red galaxy having a color of
F 105W − F 140W = 1.5 AB mag and is located 0.
00
7 from the core of this galaxy (α = 153.
◦
52643, δ = +0.
◦
64025). We
conclude that this galaxy must be the host galaxy; there are no other galaxies nearby, the light curve is incompatible with those
possible from an intra-cluster SN within MOO J1014+0038 (as discussed in Section 6), and the probability of a chance projection
is negligible (∼ 0.2%, as discussed in Appendix B).
3. SPECTROSCOPIC AND PHOTOMETRIC FOLLOW-UP
We activated ToO spectroscopic observations (PI: Hook) using X-shooter, the multi-wavelength medium resolution spectro-
graph on the Very Large Telescope (Vernet et al. 2011). Five “Observation Blocks” were taken between 2016 Mar 4-5 (UT),
yielding a total integration time of 3.8, 4.4, and 5.0 hours in the X-shooter UVB, VIS, and NIR arms, respectively. Although
the slit orientation captured both the host galaxy and the SN (shown in Figure 2), the galaxy was ∼ 4 magnitudes brighter so
only a host-galaxy spectrum could be extracted. The data were reduced to wavelength- and flux-calibrated, sky-subtracted, 2D
spectra using the Reflex software (Freudling et al. 2013). Optimal 1D-extractions were determined using a combination of cus-
tom and IRAF (Tody 1993) routines.
3
The extractions used a Gaussian profile in the cross-dispersion direction, the width of
which was determined by fitting the trace in each (binned) 2d spectrum individually. These extractions were combined using the
weighted mean to create a final 1D spectrum. We corrected for telluric absorption using a model atmosphere computed by the
Line-By-Line Radiative Transfer Model (Clough et al. 1992; Clough et al. 2005) retrieved through Telfit (Gullikson et al. 2014).
Our observations were obtained at a low mean airmass of ∼ 1.15 with a NIR slit width of 0.
00
9 (matching the mean seeing). The
1
When stacking all the epochs together, we reach 5σ point-source depths of 28.0 AB mag in F 105W and 27.9 AB mag in F 140W .
2
Nicknamed “Joseph.”
3
IRAF is distributed by the National Optical Astronomy Observatory, which is operated by the Association of Universities for Research in Astronomy, Inc.,
under cooperative agreement with the National Science Foundation.
LENSED z = 2.22 SN IA BEHIND MOO J1014+0038 3
Table 1. Other high-redshift or gravitationally lensed SNe.
SN Redshift Redshift Type Lens Lens Redshift Amplification Type Reference
CAND-ISAAC 0.64 Spec Cluster 0.18 3.6 IIP Goobar et al. (2009); Stanishev et al. (2009)
PS1-10afx 1.39 Spec Field 1.12 30
a
Ia Quimby et al. (2014)
A1, CLA11Tib 1.14 Spec Cluster 0.19 1.4 Ia Nordin et al. (2014); Patel et al. (2014)
H1, CLN12Did 0.85 Spec Cluster 0.35 1.3 Ia Nordin et al. (2014); Patel et al. (2014)
L2, CLO12Car 1.28 Spec Cluster 0.39 1.7 Ia Nordin et al. (2014); Patel et al. (2014)
Refsdal 1.49 Spec Cluster 0.54 ∼ 100
a
II Kelly et al. (2015, 2016b)
HFF14Tom 1.35 Spec Cluster 0.31 1.7 Ia Rodney et al. (2015a)
GND13Sto 1.80 ± 0.02 Phot · · · · · · · · · Ia Rodney et al. (2015b)
GND12Col 2.26
+0.02
−0.10
Phot · · · · · · · · · Ia Rodney et al. (2015b)
CAND-669 0.67 Spec Cluster 0.18 1.3 IIL Petrushevska et al. (2016)
CAND-821 1.70 Spec Cluster 0.18 4.3 IIn Amanullah et al. (2011); Petrushevska et al. (2016)
CAND-1392 0.94 Spec Cluster 0.18 2.7 IIP Petrushevska et al. (2016)
CAND-10658 0.94
+0.07
−0.27
Spec Cluster 0.18 1.7 IIn Petrushevska et al. (2016)
CAND-10662 1.03
+0.20
−0.17
Spec Cluster 0.18 2.6 IIP Petrushevska et al. (2016)
iPTF16geu 0.41 Spec Field 0.22 52
a
Ia Goobar et al. (2017)
SCP16C03 (“Joseph”) 2.22 Spec Cluster 1.23 2.8 Ia This Work
a
Multiple images; amplification is sum over all images.
4 RUBIN AND HAYDEN ET AL.
10''
0.500
0.750
1.250
2.000
E
N
SN SCP16C03
Figure 1. Color image of the central part of MOO J1014+0038 made with an F 814W stack and the F 105W and F 160W data from our first
triggered HST imaging near maximum light. The inset panel uses different scaling, and incorporates only IR data (F 105W , F 125W , and
F 160W ). The compass arrows are 1
00
in linear scale. Both panels use hyperbolic arcsine intensity scaling. In green, we show contours of our
computed lensing amplification model (in magnitudes), described in Section 8.
X-shooter NIR arm is aligned with the optical at 13,100
˚
A
4
; the atmospheric differential refraction over the NIR wavelength
range for this airmass range is < 0.
00
1. For a Gaussian PSF (with FWHM 0.
00
9) shifted by 0.
00
1, the differential slit loss would be
2%. Thus, atmospheric differential refraction has a negligible impact on the calibration of our NIR spectrum. The host-galaxy
spectrum is shown in Figure 3. Balmer Hδ and Hβ absorption lines are clearly detected, as are the Ca H&K absorption lines.
Mg b is likely detected as well. These lines provide the main contribution to the weighted cross-correlation, as shown in the lower
panel of Figure 3, yielding a redshift of 2.2216 ± 0.0002 for the host galaxy. No emission lines were detected (see Section 5.3
for further details).
4
X-shooter user manual, Section 2.2.1.7 http://www.eso.org/sci/facilities/paranal/instruments/xshooter/doc/VLT-MAN-ESO-14650-4942 v87.pdf
LENSED z = 2.22 SN IA BEHIND MOO J1014+0038 5
Slit PA = 27.4
E
N
1''
Figure 2. Illustration of the X-shooter slit orientation and location on the SN and host galaxy.
12000 14000 16000 18000 20000 22000
Observer-Frame Wavelength (
Å
)
0
2
4
6
8
10
12
14
16
f
λ
(10
−
19
ergs
/
cm
2
/
s
/
Å
)
H
δ
HK
H
α
H
β
Mg b
[NII]
Fe-
E
2
[SII]
H
δ
HK
H
α
H
β
Mg b
[NII]
Fe-
E
2
[SII]
SCP16C03 host galaxy X-shooter
best-fit FAST BC03 z=2.2216
strong atmospheric absorption regions
12000 14000 16000 18000 20000 22000
Observer-Frame Wavelength (
Å
)
4
2
0
2
4
6
8
WeightedCC (Arbitrary Units)
3500 4000 4500 5000 5500 6000 6500 7000
Rest-Frame Wavelength (
Å
)
Figure 3. Top: Flux-calibrated, telluric corrected, de-lensed (by a factor of 2.8; see Section 7) and binned (14
˚
A observer-frame) spectrum of
the host galaxy (red line). Gray shaded regions indicate strong atmospheric absorption. The best-fit FAST model (fit to both the photometry and
spectrum) is shown in black, convolved by the 305 km/s velocity dispersion estimated by ppxf. Prominent features identified in both spectra
are highlighted with light blue shading, and labeled; the Hα line is labeled for reference, but no prominent contribution to the cross-correlation
is detected from it. Bottom: The contribution to the weighted cross-correlation of each wavelength element of the spectrum. Regions of heavy
atmospheric absorption are properly de-weighted, and the prominent features easily identified by eye provide a confident redshift determination
of 2.2216 ± 0.0002.
This SN was 15
00
away from the center of a dense concentration of cluster galaxies. Our initial flux amplification estimate
was ∼ 2.2, calculated by assuming that the optical and halo center were the same, and employing a Navarro-Frenk-White dark
matter profile (Navarro et al. 1997). Based on the likelihood that this would turn out to be a highly magnified, high-redshift
SN Ia, we applied for Director’s Discretionary time on HST and were granted a three-orbit disruptive ToO to ensure a good
measurement of the SN SED. We triggered F 105W , F 125W , and F 160W imaging from our “See Change” allocation. We did
not request HST grism spectroscopy, as the roll angle range available would not have allowed for a clean separation of the SN
and its host. In addition to our triggered followup, we obtained additional WFC3 F814W , F 105W, and F 140W imaging at two
more previously scheduled orbits (the cadenced search visits for discovering SNe). We also were granted Director’s Discretionary
time on VLT for imaging in the K
s
band (PI: Nordin) with HAWK-I (Kissler-Patig et al. 2008) in order to extend the wavelength
range redder than is possible with HST.
![Figure 4. In blue, we show cutouts of the spectral regions for [OII] (left panel) and Hα (right panel). No emission lines are detected in either region. In black, we show the best-fit FAST template, convolved to the ppxf velocity dispersion of 305 km/s. In red, we show the best-fit [OII] (left panel) and Hα (right panel) contribution.](/figures/figure-4-in-blue-we-show-cutouts-of-the-spectral-regions-for-3e9vtfn7.webp)
