SN 2000cx and SN 2013bh: They Are Twins

SN 2000cx and SN 2013bh: Extremely Rare, Nearly Twin Type Ia Supernovae

Jeffrey M. Silverman, Jozsef Vinko, Mansi M. Kasliwal, Ori D. Fox, Yi Cao, Joel Johansson, Daniel A. Perley, David Tal, J. Craig Wheeler, Rahman Amanullah, Iair Arcavi, Joshua S. Bloom, Avishay Gal-Yam, Ariel Goobar, Shrinivas R. Kulkarni, Russ Laher, William H. Lee, G. H. Marion, Peter E. Nugent, Isaac Shivvers
Department of Astronomy, University of Texas at Austin, Austin, TX 78712, USA
NSF Astronomy and Astrophysics Postdoctoral Fellow
Department of Optics and Quantum Electronics, University of Szeged, Dóm tér 9, 6720 Szeged, Hungary
Observatories of the Carnegie Institution of Science, Pasadena, CA 91101, USA
Department of Astronomy, University of California, Berkeley, CA 94720-3411, USA
Cahill Center for Astrophysics, California Institute of Technology, Pasadena, CA 91125, USA
The Oskar Klein Centre, Department of Physics, AlbaNova, Stockholm University, SE-106 91 Stockholm, Sweden
Benoziyo Center for Astrophysics, The Weizmann Institute of Science, Rehovot 76100, Israel
Spitzer Science Center, California Institute of Technology, MC 314-6, Pasadena, CA 91125, USA
Instituto de Astronomía, Universidad Nacional Autónoma de México, Apartado Postal 70-264, 04510 México, D.F., México
Harvard-Smithsonian Center for Astrophysics, 60 Garden St., Cambridge, MA 02138, USA
Lawrence Berkeley National Laboratory, Berkeley, CA 94720, USA
Accepted . Received ; in original form

The Type Ia supernova (SN Ia) SN 2000cx was one of the most peculiar transients ever discovered, with a rise to maximum brightness typical of a SN Ia, but a slower decline and a higher photospheric temperature. Thirteen years later SN 2013bh (aka iPTF13abc), a near identical twin, was discovered and we obtained optical and near-IR photometry and low-resolution optical spectroscopy from discovery until about 1 month past -band maximum brightness. The spectra of both objects show iron-group elements (CoII, NiII, FeII, FeIII, and high-velocity features [HVFs] of TiII), intermediate-mass elements (SiII, SiIII, and SII), and separate normal velocity features ( 12000 km s) and HVFs ( 24000 km s) of CaII. Persistent absorption from FeIII and SiIII, along with the colour evolution, imply high blackbody temperatures for SNe 2013bh and 2000cx ( 12000 K). Both objects lack narrow NaI D absorption and exploded in the outskirts of their hosts, indicating that the SN environments were relatively free of interstellar or circumstellar material and may imply that the progenitors came from a relatively old and low-metallicity stellar population. Models of SN 2000cx, seemingly applicable to SN 2013bh, imply the production of up to  1 M of Ni and (4.3–5.5) M of fast-moving Ca ejecta.

supernovae: general – supernovae: individual (SN 2000cx, SN 2013bh)
pagerange: SN 2000cx and SN 2013bh: Extremely Rare, Nearly Twin Type Ia SupernovaeReferencespubyear: 2013

1 Introduction

Resulting from the thermonuclear explosion of C/O white dwarfs (WDs; e.g., Nugent et al., 2011; Bloom et al., 2012), Type Ia supernovae (SNe Ia) provided the first clear indication that the expansion of the Universe is accelerating (Riess et al., 1998; Perlmutter et al., 1999) and have been used as precise distance indicators to accurately measure cosmological parameters (e.g., Conley et al., 2011; Sullivan et al., 2011; Suzuki et al., 2012). Despite their utility, the specifics of SN Ia progenitor systems and explosion mechanisms are still unclear (see Howell, 2011, for further information). In general, the two leading progenitor scenarios are the single-degenerate (SD) channel, when the WD accretes matter from a non-degenerate companion star (e.g., Whelan & Iben, 1973), and the double-degenerate (DD) channel, which is the result of the merger of two WDs (e.g., Iben & Tutukov, 1984; Webbink, 1984).

The ability to determine cosmological distances using SNe Ia lies mainly in the fact that they follow a light-curve decline rate versus peak luminosity correlation (i.e., the “Phillips relation”; Phillips, 1993); however, the scatter in this relation is at least partially caused by the inclusion of various peculiar SNe Ia that nominally follow the correlation. Some of these objects, which are definitively classified as SNe Ia, can show extreme peculiarities. One of the most well-known and infamous of these objects is SN 2000cx.

SN 2000cx was discovered using the 0.76 m Katzman Automatic Imaging Telescope (KAIT; Richmond et al. 1993) on 2000 Jul. 17.5 (Yu et al., 2000, UT dates are used throughout), as part of the Lick Observatory Supernova Search (LOSS; Filippenko et al. 2001). It occurred in the outskirts of the nearby () S0 galaxy NGC 524 and was intensely observed by multiple groups (e.g. Li et al., 2001; Candia et al., 2003). It was soon discovered that SN 2000cx was peculiar in many ways. Its rise to maximum brightness in the band was typical for a SN Ia, but it declined much slower than normal. Thus standard light curve fitting algorithms could not be reliably used on SN 2000cx (Li et al., 2001). Furthermore, the optical colours of the SN were redder than nearly all other SNe Ia near maximum brightness, but quickly became bluer than other SNe Ia by 15 d after maximum (Candia et al., 2003).

Spectroscopically, SN 2000cx showed strong FeIII and weak SiII and SII features, and thus resembled the overluminous SN 1991T (Filippenko et al., 1992; Phillips et al., 1992). Unlike SN 1991T (and pretty much all other SNe Ia ever observed), SN 2000cx continued to show strong FeIII features (indicative of a high photospheric temperature) through 20 d past maximum brightness (Li et al., 2001). In addition, high-velocity features (HVFs) of CaII were observed at velocities  km s, separated from a second set of CaII features at more typical velocities ( 12000 km s, Thomas et al., 2004; Branch et al., 2004), the latter of which we will refer to as the photospheric velocity features (PVFs). These extreme observables indicate that SN 2000cx likely produced  M of Ni (Sollerman et al., 2004) and possibly up to  1 M of Ni (Li et al., 2001). SN 2000cx remained unique–until now.

Herein we present optical and near-IR photometry and low-resolution optical spectra of a near twin of SN 2000cx, SN 2013bh (aka iPTF13abc), during the first month after its discovery. We discuss the discovery of this object in §2 and present our observations and describe our data reduction in §3. Our analysis of the data, as well as a discussion of the host galaxy and possible progenitor system of SN 2013bh, can be found in §4. We summarize our conclusions in §5.

2 Discovery

iPTF13abc was discovered on 2013 Mar. 23.5 with  mag as part of the intermediate Palomar Transient Factory (iPTF; Law et al., 2009; Rau et al., 2009; Kulkarni, 2013). It was found at J2000.0 coordinates , in the star-forming galaxy SDSS J150214.17+103843.6 (, Ahn et al., 2012). The field of iPTF13abc is shown in Figure 1 with the SN marked with the green cross-hairs and a bright offset star marked with the red square. Up is north and left is east; the host galaxy of iPTF13abc is east and slightly south of the SN.

Figure 1: The field of iPTF13abc with the SN marked with the green cross-hairs and a bright offset star marked with the red square. Up is north and left is east. The host galaxy of iPTF13abc is east and slightly south of the SN.

About 10 days later, this SN was independently discovered by the Catalina Real-Time Transient Survey (CRTS; Drake et al., 2009) as CSS130403:150213+103846, and upon public announcement of the discovery was christened SN 2013bh (Morales-Garoffolo et al., 2013). On 2013 Apr. 1.3, we obtained a spectrum of SN 2013bh and found that it was similar to the aforementioned SN 2000cx a few days before -band maximum brightness. Three days later, the Public ESO Spectroscopic Survey of Transient Objects (PESSTO; e.g., Maund et al., 2013)111 collaboration obtained a spectrum of this object and also found SN 2013bh to be spectroscopically similar to SN 2000cx (Drake et al., 2013).

3 Observations and Data Reduction

3.1 Photometry

SN 2013bh was discovered using the 48 in Samuel Oschin Telescope at Palomar Observatory (P48). Every night since 15 May 2013, P48 observed the field of SN 2013bh three times in the -band,222All photometric data presented herein are in AB mags, except the - and -band data which are in Vega mags. weather-permitting. Adjacent visits were separated by at least 45 min. The P48 images of SN 2013bh were processed with the iPTF LBL pipeline (Nugent et al. in preparation). Aperture photometry was then performed on each of these images with an aperture radius equal to the seeing of each image. Aperture correction coefficients were calculated by measuring unsaturated stars that have signal-to-noise ratio (S/N) with two apertures, one of which had a radius of seeing while the other had a radius of three times seeing. Relative photometry was done among all P48 images while the absolute photometry was calibrated to -band data of the Sloan Digital Sky Survey (SDSS) Data Release 9 (DR9) Catalogue (Ahn et al., 2012). The P48 -band data of SN 2013bh is presented in Table 1.

JD-2456000 (mag)
366.88 21.22
370.02 21.46
375.02 19.70 (0.14)
376.03 19.43 (0.10)
377.01 19.22 (0.11)
386.98 18.35 (0.24)
392.95 18.49 (0.11)
393.78 18.50 (0.11)
396.94 18.80 (0.13)
1 uncertainties are in parentheses.
Table 1: P48 Photometry of SN 2013bh

After discovery, multiple telescopes were employed in order to follow SN 2013bh photometrically. As part of regular iPTF operations (Law et al., 2009), robotic multi-band () follow-up photometry was obtained with the Palomar 60 in (P60). First, the images were reduced using the automated P60 pipeline (Cenko et al., 2006). Next, calibration star magnitudes were measured on the best seeing image. Finally, aperture photometry on SN 2013bh was undertaken (with the radius equal to the seeing and the background annulus equal to 3 and 5 times the seeing) and calibrated relative to these stars. Given the offset from the host galaxy, image subtraction was not required. Table 2 shows our P60 photometry of SN 2013bh.

JD-2456000 (mag) (mag) (mag)
375.40 19.42 (0.05)
388.44 18.38 (0.02) 18.31 (0.03) 18.94 (0.06)
389.34 18.41 (0.03) 18.31 (0.03) 18.93 (0.05)
390.25 18.44 (0.05) 18.34 (0.04) 19.14 (0.11)
392.21 18.53 (0.06) 18.36 (0.05) 19.10 (0.12)
393.21 18.46 (0.02) 19.08 (0.05)
406.17 19.39 (0.14)
409.19 20.11 (0.42) 19.33 (0.11)
412.42 20.18 (0.08) 19.50 (0.05) 19.89 (0.09)
413.41 19.85 (0.07)
414.31 20.32 (0.06)
415.32 19.91 (0.08)
416.22 20.54 (0.06) 19.60 (0.08)
417.36 19.95 (0.15)
422.38 21.06 (0.08) 20.17 (0.06) 20.39 (0.11)
423.36 21.11 (0.10) 20.20 (0.06) 20.38 (0.10)
426.22 21.13 (0.10) 20.32 (0.06) 20.51 (0.11)
427.21 21.26 (0.15) 20.43 (0.08) 20.62 (0.15)
1 uncertainties are in parentheses.
Table 2: P60 Photometry of SN 2013bh

Using ALFOSC at the Nordic Optical Telescope (NOT), La Palma, we obtained photometry of SN 2013bh. All data have been reduced with standard IRAF333IRAF: The Image Reduction and Analysis Facility is distributed by the National Optical Astronomy Observatory, which is operated by the Association of Universities for Research in Astronomy (AURA) under cooperative agreement with the National Science Foundation (NSF). routines, using the QUBA pipeline (see Valenti et al., 2011, for more information). The magnitudes are measured with aperture photometry (with the radius equal to the 2 times the seeing and the background annulus equal to 3 and 5 times the seeing) and calibrated to the Landolt system through observations of standard stars PG1047+003 and PG1525-071. In order to match the other SN 2013bh photometry, the data from NOT were converted to magnitudes using the conversions presented by Jordi et al. (2006). These data are presented in Table 3.

JD-2456000 (mag) (mag) (mag) (mag)
396.65 18.98 (0.03) 18.48 (0.03) 18.78 (0.03) 19.77 (0.03)
407.45 19.69 (0.09) 19.27 (0.09)
1 uncertainties are in parentheses. magnitudes have been converted from data using the conversions presented by Jordi et al. (2006).
Table 3: NOT Photometry of SN 2013bh

Optical and near-IR photometry () were obtained with the multi-channel Reionization And Transients InfraRed camera (RATIR; Butler et al., 2012) mounted on the 1.5 m Johnson telescope at the Mexican Observatorio Astronoḿico Nacional on Sierra San Pedro Mártir in Baja California, México (Watson et al., 2012). Typical observations include a series of 60 s exposures in , with dithering between exposures. The fixed IR filters of RATIR cover half of their respective detectors, automatically providing off-target IR sky exposures while the target is observed in the neighboring filter. Master IR sky frames are created from a median stack of off-target images in each IR filter. No off-target sky frames were obtained on the optical CCDs, but the small galaxy size and sufficient dithering allowed for a sky frame to be created from a median stack of all the images in each filter. Flat-field frames consist of evening sky exposures. Given the lack of a cold shutter in the RATIR design, IR darks are not available. Laboratory testing, however, confirms that dark current is negligible in both IR detectors (Fox et al., 2012).

The RATIR data were reduced, co-added, and analysed using standard CCD and IR processing techniques in IDL and Python, utilizing online astrometry programs SExtractor and SWarp444SExtractor and SWarp can be accessed from .. Calibration was performed using field stars with reported fluxes in both 2MASS (Skrutskie et al., 2006) and the SDSS DR9 Catalogue (Ahn et al., 2012). Table 4 lists our RATIR photometry of SN 2013bh.

JD-2456000 (mag) (mag) (mag) (mag) (mag) (mag)
386.00 18.34 (0.01) 18.88 (0.01) 19.15 (0.05) 18.50
389.00 18.26 (0.01) 18.93 (0.01) 19.39 (0.05) 18.50
397.00 18.61 (0.03) 19.66 (0.03)
398.00 18.64 (0.01) 19.56 (0.02) 19.95 (0.11) 18.50
399.00 18.77 (0.01) 19.73 (0.02) 19.98 (0.07) 19.93 (0.12) 19.40 18.50
410.00 19.88 (0.02) 20.45 (0.10) 19.73 (0.06) 19.90 18.50
423.00 20.30 (0.02) 20.25 (0.03) 20.65 (0.12) 20.02 (0.11) 19.30 18.50
1 uncertainties are in parentheses.
Table 4: RATIR Photometry of SN 2013bh

In addition to obtaining our own data described above, we also downloaded publicly available -band data from CRTS (Table 5).555

JD-2456000 (mag)
385.50 18.44 (0.07)
392.50 18.48 (0.17)
402.50 19.16 (0.96)
411.50 19.19 (0.77)
1 uncertainties are in parentheses.
Table 5: CRTS Photometry of SN 2013bh

3.2 Spectroscopy

A spectroscopic time series of SN 2013bh was also obtained soon after discovery. Low-resolution optical spectra were obtained mainly using the Marcario Low-Resolution Spectrograph (LRS; Hill et al., 1998) on the 9.2 m Hobby-Eberly Telescope (HET) at McDonald Observatory, though spectral data were also acquired with the Low Resolution Imaging Spectrometer (LRIS; Oke et al., 1995) on the 10 m Keck I telescope and the Double Beam Spectrograph (DBSP; Oke & Gunn, 1982) on the Palomar 200 in telescope. We also downloaded, via WISeREP (the Weizmann Interactive Supernova data REPository; Yaron & Gal-Yam, 2012),666 a publicly available spectrum of SN 2013bh obtained by the PESSTO collaboration using the ESO Faint Object Spectrograph and Camera v.2 (EFOSC2; Buzzoni et al., 1984) on the ESO New Technology Telescope (NTT). Table 6 summarizes the spectral data of SN 2013bh presented here and upon publication all of these spectra will be available in electronic format on WISeREP.

UT Date Epoch Instrument Range (Å) Res. (Å) Exposure (s)
2013 Apr. 1.31 LRS 4192–10204 15.6 2000
2013 Apr. 5.40 0.2 EFOSC2 3764–9283 18 1500
2013 Apr. 6.46 1.2 LRS 4192–10128 15.4 2000
2013 Apr. 9.56 4.1 LRIS 3764–10204 4.5/6 480
2013 Apr. 11.44 5.8 LRS 4192–10204 15.5 2000
2013 Apr. 13.48 7.7 DBSP 3624–9872 3/4 300
2013 Apr. 16.43 10.5 LRS 4192–10204 15.2 2000
2013 Apr. 22.26 15.9 LRS 4204–10204 15.6 2000
2013 Apr. 27.23 20.5 LRS 4288–10200 15.9 2500
Rest-frame days relative to the date of -band maximum (2013 Apr. 5.2).
LRS = Low-Resolution Spectrograph on the 9.2 m Hobby-Eberly Telescope at McDonald Observatory; EFOSC2 = ESO Faint Object Spectrograph and Camera v.2 on the NTT; LRIS = Low Resolution Imaging Spectrometer on the 10 m Keck I telescope; DBSP = Double Spectrograph on the Palomar 200 in telescope.
Approximate full width at half-maximum intensity (FWHM) resolution. If two numbers are listed, they represent the blue- and red-side resolutions, respectively.
Table 6: Journal of Spectroscopic Observations of SN 2013bh

All of our spectra were reduced using standard techniques (e.g., Silverman et al., 2012a). Routine CCD processing and spectrum extraction were completed with IRAF. We obtained the wavelength scale from low-order polynomial fits to calibration-lamp spectra. Small wavelength shifts were then applied to the data after cross-correlating a template sky to the night-sky lines that were extracted with the SN. Using our own IDL routines, we fit spectrophotometric standard-star spectra to the data in order to flux calibrate our spectra and to remove telluric lines (Wade & Horne, 1988; Matheson et al., 2000).

4 Analysis and Results

4.1 Light Curves

We present our light curves of SN 2013bh in Figure 2 (points), compared with those of SN 2000cx (solid lines, Li et al., 2001; Candia et al., 2003). Upper limits are shown as downward-pointing arrows. In the figure, the SN 2000cx data has been shifted in time to match the -band peak of SN 2013bh. It has also been shifted in apparent magnitude to the distance of SN 2013bh, however no other vertical shift has been applied. Neither of the objects have been dereddened. The photometric data of SN 2000cx matches SN 2013bh in all bands for which there are observations of both objects. The observed brightness of the two objects is consistent in and , while SN 2013bh appears to be  mag brighter than SN 2000cx in and . The upper limits on SN 2013bh in the and bands are also consistent with the observed magnitudes of SN 2000cx in these bandpasses.

Figure 2: light curves of SN 2013bh (points), compared with those of SN 2000cx (solid lines, Li et al., 2001; Candia et al., 2003). The SN 2000cx data has been shifted in time to match the -band peak of SN 2013bh and shifted in apparent magnitude to the distance of SN 2013bh. Neither of the objects have been dereddened.

From low-order polynomial fits, we find that SN 2013bh reached a peak -band magnitude of on 2013 Apr.  and a peak -band magnitude of on 2013 Apr. . The peak values for the band are somewhat uncertain due to the fact that our earliest data for this bandpass are very close to maximum brightness. SN light-curve declines are often described by , i.e., the decrease in magnitudes in a given bandpass from peak to 15 d past peak. SN 2013bh has  mag and  mag, which are both somewhat smaller than the corresponding values for SN 2000cx (0.94 and 1.06 mag, respectively; Li et al., 2001), implying that the light curve of SN 2013bh is slightly broader than that of SN 2000cx.

Given  mag (Schlegel et al., 1998), no reddening from the host galaxy of SN 2013bh (see Section 4.3), and  mag (see Section 4.6), the peak absolute magnitudes of SN 2013bh in our best observed bands are  mag and  mag. These values are in agreement with those of SN 2000cx, after correcting for  mag and no host reddening ( and , respectively; Li et al., 2001).

4.1.1 The and Bands

While we do not have data on the rising portion of the -band light curve of SN 2013bh, we note that SN 2000cx was found to have a similar rise time to that of the normal Type Ia SN 1994D (Li et al., 2001). Based on two data points, SN 2013bh has a slightly slower decline than SN 2000cx in the band (much like the and bands, see below), but has a very similar -band light curve. SN 2000cx was found to have a normal -band decline rate until 6 d after -band maximum brightness, after which time the decline slowed (Li et al., 2001).

4.1.2 The Band

The best observed bandpass of SN 2013bh is the band, and in Figure 3 we present its -band light curve, along with data from various comparison objects: SN 2000cx (solid; Li et al., 2001), the overluminous type Ia SN 1991T (dotted; Lira et al., 1998), and the normal type Ia SN 2011fe (dashed; Vinkó et al., 2012). All comparison objects have had their published -band data converted to -band magnitudes using the conversions in Jordi et al. (2006) and have been deredshifted and dereddened by values given in their respective references. The top panel of Figure 3 shows absolute -band magnitudes while the bottom panel is the same data, but shifted such that all objects have the same peak magnitude.

Figure 3: The -band light curve of SN 2013bh with some comparison objects: SN 2000cx (solid; Li et al., 2001), the overluminous type Ia SN 1991T (dotted; Lira et al., 1998), and the normal type Ia SN 2011fe (dashed; Vinkó et al., 2012). All -band photometry has been converted to -band magnitudes using conversions found in Jordi et al. (2006). All data have been deredshifted and dereddened by values given in their respective references. Absolute -band magnitudes are shown (top) as well a shifted version of the data such that all objects have the same peak magnitude (bottom).

Once again, the light curve of SN 2013bh matches well to that of SN 2000cx and both are slightly more luminous at peak (in the -band) than SN 2011fe, but less luminous than SN 1991T. The -band is the only band for which we have pre-maximum data of SN 2013bh. Based on these observations, it appears that SN 2013bh has a relatively normal rise time, similar to that of SN 2000cx.

The band of SN 2013bh shows a plateau that begins between 12 and 17 d after -band maximum brightness, similar to SN 2000cx for which the plateau begins at  15 d after maximum, which is later than more normal SNe Ia ( 12 d for SN 1994D and  14 d for SN 2011fe, Li et al., 2001; Vinkó et al., 2012, respectively). Since the plateau begins later and the secondary -band maximum is weaker in SNe 2013bh and 2000cx that in other SNe Ia, the amount these objects have faded shortly after maximum brightness is larger than that of the comparison objects shown in Figure 3. As mentioned above, SN 2013bh has a slightly slower -band decline rate after maximum brightness than SN 2000cx, but it is clear that both are faster than the other objects in the Figure. Finally, the -band plateau in SN 2013bh ends between 23 and 27 d past maximum, again similar to SN 2000cx.

4.1.3 The Bands

The -band data of SN 2000cx was normal until about 7 d after -band maximum brightness at which time it began to decline faster than normal SNe Ia. At 14 d past maximum it had faded  1.1 mag below peak after which it re-brightened somewhat (Li et al., 2001). It is likely that SN 2013bh also behaved this way, but unfortunately, we do not have -band data at this epoch. We do see evidence, however, that SN 2013bh (like SN 2000cx) has a weak secondary maximum, as compared to more normal SNe Ia, in the band (just like the band). These secondary maxima are usually associated with FeIII recombining to form FeII as the SN ejecta cools (Pinto & Eastman, 2000a). Weak secondary maxima indicate that the SN photosphere remains relatively hot through these epochs, which is consistent with the blackbody temperatures and strong FeIII absorptions seen in SN 2013bh at these epochs (see Section 4.4).

Moving further into the IR, the and data of SN 2013bh also show weak secondary maxima as compared to other SNe Ia, which is similar to the weak secondary maxima in data of SN 2000cx (Candia et al., 2003). SN 2000cx, and many other SNe Ia, have been found to have  mag at 10 d after -band maximum (Candia et al., 2003), which is consistent with our observed upper limit of  mag at this epoch. Furthermore, Candia et al. (2003) observed a deep dip in the -band light curve of SN 2000cx at about 13 d past maximum brightness and this is certainly consistent with our upper limits for SN 2013bh.

4.2 Colour Curves

In Figure 4 we plot the colour evolution of SN 2013bh, along with the same comparison objects that were shown in Figure 3. As above, all objects have been corrected for Galactic extinction using the dust maps of Schlegel et al. (1998) and information in their respective references.

Figure 4: Colour curves of SN 2013bh and the same comparison objects used in Figure 3. All data have been deredshifted and corrected for extinction using the reddening values provided in their respective references.

The colour of SN 2013bh is redder than all of the comparison objects at  8 d after -band maximum brightness, but then becomes nearly as blue as SN 2000cx by  18 d after maximum. The colour of SN 2000cx was also seen to be quite red at early times, becoming extremely blue by 2 weeks after maximum. This was due to a plateau in the colour curve at  mag for  d (Li et al., 2001; Candia et al., 2003). Even though SN 2013bh data are sparse, the observations are consistent with a plateau at  mag at similar epochs as the one seen in SN 2000cx, though perhaps a bit later.

The single observation of SN 2013bh from  8 d after maximum brightness is slightly redder than the colour of SN 2000cx at the same epoch. SN 2000cx (and presumably SN 2013bh as well) is bluer than normal and overluminous SNe Ia at all epochs studied herein and shows a deep blue dip at  12 d after maximum (as seen in the upper-right panel of Figure 4 and Li et al., 2001). In it was also seen that SN 2000cx was bluer than other SNe Ia and again showed a deep blue dip, though in this colour the dip is a few days earlier than in . We see a similar behavior in our one data point for SN 2013bh, though it is significantly ( 0.3 mag) bluer than even SN 2000cx.

The colours of SN 2013bh, for which we have a handful of data points, are found to match SN 2000cx near -band maximum brightness. Soon after maximum, however, SN 2013bh evolves to be bluer than SN 2000cx in . The minimum in the colour curve of SN 2013bh occurs at  9 d after maximum and is about 0.2 mag bluer than that of SN 2000cx. At later epochs, SN 2013bh remains  0.1 mag bluer than SN 2000cx. Note that the majority of the difference in of SN 2013bh at about 1 month past maximum as measured from P60 and RATIR data comes from the -band photometry, though it is still consistent at about the 2 level. Furthermore, no instrumental response corrections have been applied to our data and, as mentioned above, no host-galaxy subtraction was performed.

In summary, based on limited data, the colours of SN 2013bh tend to follow those of SN 2000cx, and certainly more closely than either SN 1991T or SN 2011fe. However, SN 2013bh appears to have more extreme colours than SN 2000cx. In general SN 2013bh seems to be redder than SN 2000cx in the bluest colours, but bluer than SN 2000cx in the reddest bands. This implies an optical continuum that peaks redder than SN 2000cx, thus SN 2013bh seems to have a lower blackbody temperature which is consistent with our spectral models (see Section 4.4).

4.3 Spectra

All of our spectra of SN 2013bh are shown in black in Figure 5 and are labelled with their age relative to -band maximum brightness. The over-plotted red, dashed curves are spectra of SN 2000cx at similar epochs to our observations of SN 2013bh, also labelled by their age relative to -band maximum brightness (converted from the date of -band maximum brightness as determined by Li et al., 2001). The SN 2000cx comparison data come from the in-depth study of that object by Li et al. (2001) and the large CfA SN Ia spectral dataset presented by Matheson et al. (2008). Both objects have been deredshifted and dereddened. The spectral match between SN 2013bh and SN 2000cx at all epochs studied in this work is remarkable. We do note, however, that SN 2000cx seems to have a slightly bluer continuum than SN 2013bh, which is indicative of a higher blackbody temperature, once again consistent with our spectral models (see Section 4.4).

Figure 5: Spectra of SN 2013bh (solid black) and SN 2000cx (dashed red, taken from Li et al., 2001; Matheson et al., 2008), both labelled with their age relative to -band maximum brightness. Telluric absorption in the maximum light spectrum is marked by the Earth symbol () and both objects have been deredshifted and dereddened.

We assumed above no reddening from the host galaxy of SN 2013bh. This is due mainly to the facts that we see no evidence of absorption from NaI D and because it was found in the outskirts of its host (much like SN 2000cx). In our highest S/N spectrum from 4 d after -band maximum brightness, we find a 2 upper limit of 0.2 Å for the equivalent width (EW) of NaI D absorption. Using the empirical relation of Poznanski et al. (2011), this converts to  mag, which is negligible (especially given the uncertainties in the conversion, i.e.  0.3 mag). Also supporting no host reddening is the location of SN 2013bh in its host galaxy. It went off  24 kpc (projected linear distance) from the centre of SDSS J150214.17+103843.6, which is equivalent to  3 Petrosian radii (Petrosian, 1976). See Section 4.6 for more information on the host galaxy of SN 2013bh.

In Figure 6 we compare our highest S/N spectrum of SN 2013bh (from  4 d past maximum brightness) with various other SNe Ia at similar epochs. Shown in the Figure are its near twin, SN 2000cx (Li et al., 2001), as well as the overluminous SN 1991T (Filippenko et al., 1992) and the extremely normal SN 2011fe (Parrent et al., 2012). Important spectral features are labelled. SN 2011fe stands out from the other three objects with overall much stronger absorption features and a much broader and more blended CaII near-IR triplet. Further differences are seen in the blue end of the spectrum where SN 2011fe is dominated by absorption from blends of FeII and MgII lines, while the complex profiles in the other objects are due in large part to FeIII and TiII (see §4.4 for further information).

Figure 6: A spectrum of SN 2013bh along with comparisons to other SNe Ia: SN 1991T (Filippenko et al., 1992), SN 2000cx (Li et al., 2001), and SN 2011fe (Parrent et al., 2012). Each spectrum is labelled with its age relative to -band maximum brightness and the data have all been deredshifted and dereddened. Significant spectral features have been identified.

Even though SNe 2013bh and 2000cx bear some resemblance to SN 1991T, there are significant differences as well. The OI line is more prominent in SN 1991T than in either SNe 2013bh and 2000cx, where it is almost non-existent. This (along with a lack of C absorption) may be an indication of more complete WD burning in SNe 2013bh and 2000cx, which is consistent with the relatively large amount of Ni inferred for these objects (see § 4.6). SNe 2013bh and 2000cx both also show significant absorption from HVFs of TiII while SN 1991T lacks these features in the blue end of the spectrum. In addition, just blueward of the CaII near-IR triplet in these latter two objects, a second absorption component from HVFs of CaII is easily seen and is conspicuously absent from SN 1991T.

4.4 Spectral Fits

The spectrum-synthesis code SYNAPPS (Thomas et al., 2011) was employed to help identify the species present in our spectra of SN 2013bh. SYNAPPS (and its modeling kernel SYN++) is derived from SYNOW (Fisher et al., 1997), which computes synthetic spectra of SNe in the photospheric phase using the Sobolev approximation (Sobolev, 1960; Castor, 1970; Jeffery, 1989). SYNAPPS is capable of varying a large number of parameters automatically, thus finding an optimum fit via -minimization.

In SYNAPPS models the spectral lines are assumed to form via resonance scattering above a sharp photosphere. The location of the photosphere is expressed in velocity coordinates as (in km s) taking into account the homologous expansion of the SN ejecta. Consequently, for a particular ion, the minimum and maximum velocity coordinates of the line-forming region are denoted with and (both in km s), respectively. If , then the line forming region is considered “detached” from the photosphere.

The optical depths for each transition of a given species are controlled with two additional parameters: the optical depth of a “reference line,” (which is usually the strongest line in the optical band, defined internally within the code), and the e-folding width, (in km s), of the optical depth profile above the photosphere (assumed to be an exponential function in SYNAPPS). For other features, the optical depth is computed relative to the reference line assuming Boltzmann-excitation (i.e. local thermodynamic equilibrium) using an excitation temperature (in K). Non-LTE effects are mimicked partly by allowing different values for each species, all of which can be different from the photospheric temperature . The latter is used only in computing the blackbody radiation emitted by the photosphere. Since the number of species appearing in a typical SN spectrum is usually between 5 and 10, and each ion has 5 tunable parameters, the number of adjustable parameters in a SYNAPPS session can be substantial.

The spectral sequence of SN 2013bh in Figure 5 suggests that the optical spectra shortly before and after maximum brightness are formed by the same set of ions, and the spectral evolution can be modeled by simply tuning the optical depth and other parameters of the same species. This is supported by the similar photospheric temperature of the pre-maximum and post-maximum spectra, which suggests that the excitation of ions should not change drastically near maximum.

A model containing the usual composition of a SN Ia was constructed to fit the spectra of SN 2013bh and is based on the model for SN 2000cx by Branch et al. (2004) and Thomas et al. (2004). The model was initially fit to the 4.1 d spectrum, then it was optimized for the and +6.3 d spectra. SYNAPPS fits to these three SN 2013bh spectra are shown in Figure 7. The data are shown as solid black, while the models are shown as dashed red. Major spectral features have been labelled. The parameters of our SYNAPPS fit to the 4.1 d spectrum are listed in Table 7.

Figure 7: Observed spectra of SN 2013bh (solid black) and SYNAPPS fits to the data (dashed red). Each spectrum is labelled with its age relative to -band maximum brightness and the data have all been deredshifted and dereddened. Major spectral features are identified.
( km s) ( km s) ( km s) ( K)
SiII 1.57 12.0 40.0 1.0 8.8
SiIII 1.34 16.0 40.0 2.0 13.0
SII 12.0 40.0 1.0 5.2
CaII (PVF) 0.80 12.0 40.0 4.8 12.0
CaII (HVF) 2.90 25.0 40.0 3.0 12.0
TiII 2.09 21.0 40.0 2.0 5.5
FeII 0.94 21.2 40.0 4.0 5.0
FeIII 0.03 11.0 40.0 2.0 12.0
CoII 0.60 11.0 40.0 2.0 6.3
NiII 1.00 11.0 40.0 2.0 5.1
 km s,  K.
Table 7: SYNAPPS Model Parameters for the 4.1 d Spectrum

The model includes PVFs of CaII, SiII, SII, FeIII, CoII, and NiII, in addition to detached, HVFs of CaII, SiIII, TiII, and FeII. The photospheric velocity is  km sfor all modeled spectra and the detached, HVFs span a range between 20000 and 24000 km s. These modeled velocities are fully consistent with what is calculated from direct measurements of the spectral features themselves (§ 4.5), including a SiII velocity plateau seen in our SN Ia-like SYNAPPS fits. From Figure 7, it is clear that most of the spectral features between 3000 and 6300 Å can be fit successfully with the models. As also seen in other, more normal SNe Ia at similar epochs (e.g., Silverman et al., 2012a), redward of  6300 Å, the continuum is significantly depressed relative to the blackbody continuum, which cannot be modeled self-consistently within the framework of SYNAPPS. Therefore, this region has been omitted from the automatic fitting process.

We identify CaII via the near-IR triplet (at  8000 Å) as well as CaII H&K (at  3650 Å). The strong, unblended feature near 6150 Å is due to SiII 6355 and the feature near 5480 Å is from SII. The model is able to fit the observed feature near 4000 Å, which was identified in SN 2000cx as a blended HVF from TiII by Branch et al. (2004) and could also be due to SiII 4130, despite the presence of these ions in our model. Note that the feature at  4500 Å, suspected to be a HVF of H in SN 2000cx by Branch et al. (2004), can be fit adequately as a blend of iron-group elements.

As mentioned above, our SYNAPPS model of SN 2013bh is extremely similar to models of SN 2000cx. Li et al. (2001) found that SiII 6355 strengthened with time in SN 2000cx while FeIII remained strong through maximum brightness, which is identical to what we find for SN 2013bh. SII was found to be weak in both objects, but vanished quicker in SN 2013bh than SN 2000cx. HVFs of FeII occurred in both objects, but we find evidence of only PVFs for FeIII and SiII while SN 2000cx showed evidence for HVFs of these ions (Li et al., 2001; Thomas et al., 2004). Evidence for OI and MgII absorption in SN 2000cx was presented by Thomas et al. (2004) and Branch et al. (2004), respectively, though our SYNAPPS fits do not support the same claim for SN 2013bh. Models in both of those studies do indicate CaII velocities in SN 2000cx that match what we find for SN 2013bh quite well. In addition, Branch et al. (2004) note significant absorption due to HVFs of TiII in their models of SN 2000cx and we find similar results for our models of SN 2013bh.

For epochs studied herein, the photospheric temperatures inferred from our SYNAPPS fits to SN 2013bh are  11000 K. This is higher than more normal SNe Ia, though slightly cooler than what was found for SN 2000cx (12000 K; Thomas et al., 2004). Note, however, that at slightly earlier epochs (about 1 week before maximum brightness), a much higher blackbody temperature for SN 2000cx was inferred from near-IR spectra (20000-25000 K; Rudy et al., 2002).

4.5 Expansion Velocities

For a few of the spectral features in SN 2013bh (and SN 2000cx) which are easily-identifiable and relatively well-separated, the expansion velocities are measured. Here we investigate the SiII 6355 feature, the SII “W” feature,777The two broad absorptions that make up the SII “W” are fit using a single spline, but we calculate the expansion velocity of the absorption complex using the minimum of the bluer of the two features relative to its rest wavelength. the CaII H&K feature, and the CaII near-IR triplet. For the SiII and SII features, the method of velocity determination is the same as that used in Silverman et al. (2012), but, briefly, consists of: defining two endpoints on either side of the feature of interest, fitting a linear pseudo-continuum between those endpoints, fitting a spline function to the data between the endpoints, and using the minimum of the spline fit to calculate an expansion velocity for the feature.

For the CaII features, however, we employed a slightly different fitting method. Similar to the method outlined above, we begin by choosing endpoints for the spectral feature in question and, using these endpoints, define and then subtract from the data a linear pseudo-continuum. Then a non-linear least squares fitter is used to simultaneously fit multiple Gaussian components to each component of the feature. The CaII H&K feature (CaII near-IR triplet) consists of 2 (3) distinct components. Given that the relative fluxes (set by the weights) and rest wavelengths of each component are known and since we require that all components in a given fit have the same Gaussian width, each CaII fit contains only 3 free parameters: the Gaussian height, width, and centroid of the strongest component. By-eye initial estimates of these parameters for each fit are used as inputs to the fitting routine.

For the CaII H&K profile, we also allow for the existence of absorption due to SiII 3858 at a similar velocity to SiII 6355 (e.g., Foley, 2012). We find that SiII 3858 is extremely weak (or non-existent) in our spectra of SN 2013bh that cover this wavelength range. We do, however, find evidence for SiII 3858 absorption in SN 2000cx at velocities consistent with SiII 6355.

All of the measured velocities for SN 2013bh are consistent with our SYNAPPS fits and are plotted in Figure 8 (filled black), relative to -band maximum brightness. We also plot (open grey) our re-measured velocities of SN 2000cx (also relative to -band maximum) using spectra presented in Li et al. (2001) and Matheson et al. (2008). Note that the velocities for SN 2000cx calculated herein are consistent with previous work (Li et al., 2001; Thomas et al., 2004; Branch et al., 2004).

Figure 8: The temporal evolution of the expansion velocities (relative to -band maximum brightness) of SN 2013bh (filled black) and SN 2000cx (open grey). The top panel shows velocities of the SiII 6355 (circles) and SII “W” features (stars). The bottom panel shows velocities of PVFs and HVFs of CaII H&K (stars and circles, respectively) and PVFs and HVFs of the CaII near-IR triplet (squares and triangles, respectively). The light grey swath in the top panel is the 1 region around the average SiII 6355 velocity of normal SNe Ia as determined by the entire BSNIP dataset. Note that the HVFs and PVFs of the CaII near-IR triplet in SN 2013bh become blended by 16 d after maximum brightness and thus are shown as a single (black filled square) data point from then on.

The upper panel of Figure 8 shows the velocities of the SiII 6355 and SII “W” features for SNe 2013bh and 2000cx, along with the 1 region (light grey) around the average SiII 6355 velocity of normal SNe Ia as determined by the entire Berkeley SN Ia Program sample (BSNIP; Silverman et al., 2012). The SiII 6355 velocity of SN 2013bh is initially similar to other, more normal SNe Ia, but shows a plateau until  10 d after maximum brightness, thus becoming larger than the typical SN Ia SiII 6355 velocity. After this epoch, the velocity decreases some, eventually matching the more normal SNe Ia. SN 2000cx also shows a plateau at nearly the same velocity ( 11500-12000 km s) though it is much longer lived, lasting through all epochs for which we measure a SiII 6355 velocity in this work. In fact, Li et al. (2001) found that this plateau continues to d. Such long-lasting SiII velocity plateaus are not unheard of, though all other examples of them are found in extremely peculiar SNe Ia (e.g., Scalzo et al., 2010; Childress et al., 2013).

The width of the SiII 6355 feature in SN 2013bh is larger than that of SN 2000cx, but we do not see any convincing evidence for HVFs from SiII, even though this has been seen in other SNe Ia (e.g., Silverman et al., 2012b; Childress et al., 2013; Marion et al., 2013). In both SNe 2000cx and 2013bh, the SiII 6355 and SII “W” features both strengthen with time, though while the former stays strong through all epochs considered here, the latter begins to weaken after maximum brightness and completely disappears by a few days after maximum.

As seen in Figure 5, the SII “W” feature is somewhat weaker in SN 2013bh as compared to SN 2000cx near maximum brightness. This spectral feature is usually found to have velocities 1000-2000 km s lower than the SiII 6355 feature (e.g., Silverman et al., 2012), though in SNe 2013bh and 2000cx both features have similar velocities before maximum. At this point the velocity plummets, likely due to the feature getting progressively weaker and becoming blended with other ions (notably SiIII, see § 4.4).

The spectra of SN 2000cx blueward of the CaII near-IR triplet were initially found to be quite complex and confusing (Li et al., 2001), but it was eventually realized, through spectral modeling, that this object showed HVFs of CaII at velocities  km s (Thomas et al., 2004; Branch et al., 2004). This was in addition to a second set of CaII features at more typical velocities ( 12000 km s). HVFs from CaII have been seen in many individual SNe Ia (e.g., Silverman et al., 2012b; Childress et al., 2013; Marion et al., 2013) and might be common in the early-time spectra of SNe Ia (e.g., Mazzali et al. 2005; Childress et al. 2013; Silverman et al. in preparation). Unsurprisingly, SN 2013bh once again matches SN 2000cx and shows strong HVFs from CaII at early times.

The bottom panel of Figure 8 shows the velocities for SN 2013bh (filled black) and SN 2000cx (open grey) of PVFs and HVFs of CaII H&K (stars and circles, respectively) and PVFs and HVFs of the CaII near-IR triplet (squares and triangles, respectively). In both objects, through  10 d past maximum, we measure distinct PVFs and HVFs for both the CaII H&K feature and the CaII near-IR triplet. The PVF velocities are similar to more normal SNe Ia at these epochs, while the HVF velocities of both objects are significantly larger than most SNe Ia (Childress et al. 2013; Silverman et al. in preparation). By 16 d past maximum, we are only able to measure a single CaII near-IR triplet feature in SN 2013bh. Since the measured velocity at this epoch is nearly the average of the HVF and PVF velocities from the previous spectrum, it appears that the two components have become blended and thus are shown in the Figure as a single (black filled square) data point from then on.

The measured velocities of the PVFs for both CaII features in SN 2000cx are nearly identical, while the HVF velocities of this objects also match each other quite well. This is a useful sanity check for our velocity determinations since one might expect all CaII features in a given spectrum to have nearly equal velocities. The velocities of the PVFs and HVFs are quite self-consistent for SN 2013bh as well. Also note the relatively small scatter in all PVF velocity measurements (both objects and both spectral features), as well as the consistency of all of the HVF velocity measurements. This once again emphasizes the similarity between these two objects.

To highlight the separation between the HVFs and the PVFs, as well as our multiple Gaussian component fitting technique, we plot the full CaII near-IR triplet profile for SNe 2013bh (top) and 2000cx (bottom) in Figure 9. Both spectra were obtained  4 d after maximum and have had their linear pseudo-continua (grey dotted line) removed. The data are the black curves, the combined Gaussian fits are the grey solid curves, and the individual Gaussian components are the grey short-dashed curves. The separation between the HVFs (left-hand triplet of Gaussians) and the PVFs (right-hand triplet of Gaussians) can easily be seen in both objects. While we fully admit that the fits are not perfect and fitting a set of Gaussians to the full profile is an oversimplification, they do appear to capture the overall shapes of both profiles, as well as at least some of the more subtle details. The fits tend to have reduced values of 2–3.

Figure 9: The CaII near-IR triplet profile for SNe 2013bh (top) and 2000cx (bottom), after removing the linear pseudo-continuum (grey dotted line). The rest-frame age of each spectrum is listed next to the data (black). Over-plotted are the combined Gaussian fits (solid grey) and the individual Gaussian components (short-dashed grey). The HVFs are the left-side Gaussians while the PVFs are the right-side Gaussians. The CaII near-IR triplet velocity of SN 2013bh (SN 2000cx) is shown as the vertical solid (long-dashed) line.

The vertical solid line is the CaII near-IR triplet velocity for SN 2013bh, while the vertical long-dashed line is the velocity for SN 2000cx. The PVFs in these two objects have consistent expansion velocities at this epoch, however the HVFs of SN 2013bh are found to be at larger velocities than those of SN 2000cx (which can also be seen in Figure 8). Perhaps this indicates a higher CaII optical depth in SN 2013bh. A more detailed analysis of HVFs is beyond the scope of this paper, but an in-depth study of HVFs in a large set of more normal SN Ia spectra will be undertaken in the near future (Silverman et al. in preparation).

4.6 Host Galaxy and Progenitor System

SN 2013bh lies in the outskirts of the star-forming galaxy SDSS J150214.17+103843.6 at a redshift of (Ahn et al., 2012), which, assuming a CDM cosmology with  km s Mpc, implies a distance of  Mpc and a distance modulus of  mag. It is located west and north of the centre of its host, a projected distance of  25.3 kpc (or  3 Petrosian radii, Petrosian, 1976). Correlation with a library of galaxy spectra using the “SuperNova IDentification” code (SNID; Blondin & Tonry, 2007) indicates that SDSS J150214.17+103843.6 is a spiral (Sa/Sb) galaxy with , quite close to the SDSS DR9 value of (Ahn et al., 2012). While this is not an atypical type of host galaxy for a SN Ia (Li et al., 2011), very few (if any) SNe Ia are found further away from the centre of their host (e.g., Wang et al., 2013).

While inner parts of the host galaxy of SN 2013bh may be currently forming stars, the outer edges (where the SN is located) show no obvious -band emission (Ahn et al., 2012), indicating a lack of star formation ( M yr; e.g., Moustakas06). In addition, typical metallicity gradients in spiral galaxies (e.g., Henry & Worthey, 1999) indicate that SN 2013bh went off in a region that likely has relatively low-metallicity.

SN 2000cx also exploded in the periphery of its host galaxy, the nearby ( 38 Mpc) S0 galaxy NGC 524. It was found 18 kpc (linear projected distance) from the centre of its host, in a low-metallicity region (Li et al., 2001). High-resolution spectra of SN 2000cx presented by Patat et al. (2007) do not show evidence of NaI D absorption to very tight limits and they infer NaI cm and (for solar abundances) H cm. As mentioned above, our highest S/N spectrum of SN 2013bh also did not show obvious absorption from NaI D (with a 2 upper limit of 0.2 Å) or evidence for star formation from narrow emission lines.

Both SN 2013bh and SN 2000cx lack NaI D absorption which indicates low host-galaxy reddening and a relatively “clean” circumstellar environment (i.e., free of expelled material from the companion star). Their locations on the periphery of their host galaxies, along with no obvious associated recent star formation, point to old stellar population progenitors with relatively low-metallicity for both objects. All of these observations are consistent with many DD models (e.g., Iben & Tutukov, 1984), though we cannot definitively rule out all SD models (e.g., Whelan & Iben, 1973).

Li et al. (2001) and Candia et al. (2003) found that none of the standard SN Ia light-curve fitting algorithms could match the observations of SN 2000cx, and thus the same can be said for SN 2013bh; however, the delayed-detonation DD3 model (Woosley & Weaver, 1994; Pinto & Eastman, 2000b) did match many of the observations of SN 2000cx. This model produced higher kinetic energies and more nuclear burning than models of more normal SNe Ia (including the overluminous SN 1991T). As a result, the DD3 model produced a relatively large amount of Ni ( 1 M), higher-than-normal blackbody temperatures, and high expansion velocities. The latter two observables were seen in both SN 2000cx (Li et al., 2001; Thomas et al., 2004; Branch et al., 2004) and SN 2013bh (Sections 4.4 and 4.5). Furthermore, this model also predicted peak magnitudes and values consistent with those observed for both SN 2000cx and SN 2013bh.

An alternative explanation for the HVFs observed in these two objects comes from a 3D model presented by Thomas et al. (2004). They explain the HVFs of CaII as coming from clumpy ejecta such that the high-velocity material only partially covers the SN photosphere. This idea is supported by Leonard et al. (2000), who found significant polarization intrinsic SN 2000cx. Thomas et al. (2004) also derive (4.3–5.5) M for the high-velocity ejecta mass in Ca, which they claim can be explained by primordial material alone. This conclusion is consistent with the “clean” environment determined for SNe 2000cx and 2013bh by the lack of NaI D or narrow CaII H&K absorption from circumstellar material.

5 Conclusions

In this work we have presented optical and near-IR photometry of SN 2013bh (aka iPTF13abc), a near twin of SN 2000cx, from discovery through  40 d past -band maximum brightness and low-resolution optical spectroscopy through  20 d past maximum. SN 2013dh reached a peak absolute magnitude of  mag and had a decline rate of  mag, matching nearly identically to SN 2000cx. The colours of SN 2013bh mostly follow those of SN 2000cx, but show evidence of a slightly “flatter” optical continuum that indicates a slightly lower blackbody temperature than SN 2000cx. This is consistent with the relative spectral shapes of these two objects. The colours of SNe 2000cx and 2013bh also indicate photospheric temperatures that are higher than more normal SNe Ia. This is supported by the presence of strong absorption features from FeIII and SiIII that persist well past maximum brightness (which is not seen in any other SN Ia). Our spectral models from SYNAPPS and direct measurements of the strongest spectral features also indicate the presence of other iron-group elements (CoII, NiII, FeII, and HVFs of TiII) and intermediate-mass elements (SiII and SII). Furthermore, in all of our spectra we observe separate normal velocity features ( 12000 km s) and HVFs ( 24000 km s) of CaII.

The environments of both SN 2013bh and SN 2000cx appear to be relatively “clean,” with negligible amounts of CSM, consistent with their locations in the outskirts of their host galaxies and the lack of obvious associated recent star formation. This points to the progenitors of these objects being relatively low-metallicity, old stars, consistent with the DD scenario (though the SD scenario cannot be completely disproved). The delayed-detonation DD3 model of Woosley & Weaver (1994); Pinto & Eastman (2000b) was successfully applied to SN 2000cx and given the observational similarities with SN 2013bh, it is reasonable to use it for this object as well. This model implies a relatively large amount of Ni produced ( 1 M), higher-than-normal blackbody temperatures, and high expansion velocities. The latter two of these were directly observed in SN 2013bh. The model also predicts peak magnitudes and values consistent with those observed for both SN 2000cx and SN 2013bh. A 3D model presented by Thomas et al. (2004) explains the HVFs of CaII as coming from (4.3–5.5) M of clumpy ejecta, which can be explained by primordial material alone.

Of the  2300 SNe Ia given IAU designations from the beginning of 2000 through the end of 2012, SN 2000cx was the only one of its kind discovered. PTF, which ran for 4 years starting in 2009, discovered 1250 SNe Ia with no objects resembling SN 2000cx. iPTF, which began in Jan 2013, discovered SN 2013bh (aka iPTF13abc) and through the end of Jul 2013 has found 111 SNe Ia. Very roughly, it seems that an object like SN 2000cx or SN 2013bh is found in every  2000 SNe Ia. In other words, these objects are approximately 0.05% of the total SN Ia rate. The extreme rarity of of these objects and how they relate to more normal SNe Ia and overluminous SN 1991T-like objects is a challenge to SN progenitor models. As large-scale transient surveys (e.g., iPTF, Pan-STARRS, LSST) continue to (and will in the future) find many new objects, more objects similar to SNe 2000cx and 2013bh will likely be discovered.


We would like to thank M. Ganeshalingam, P. Kelly, and E. Ofek for helpful discussions, J. Caldwell, S. Odewahn, and S. Rostopchin for their assistance with some of the observations, as well as the PESSTO and CRTS collaborations for making some of their data on SN 2013bh publicly available. The HET is a joint project of the University of Texas at Austin, the Pennsylvania State University, Stanford University, Ludwig-Maximilians-Universität München, and Georg-August-Universität Göttingen. The HET is named in honor of its principal benefactors, William P. Hobby and Robert E. Eberly. The Marcario Low Resolution Spectrograph is named for Mike Marcario of High Lonesome Optics who fabricated several optics for the instrument but died before its completion. The LRS is a joint project of the HET partnership and the Instituto de Astronomía de la Universidad Nacional Autónoma de México. Some of the data presented herein were obtained at the W. M. Keck Observatory, which is operated as a scientific partnership among the California Institute of Technology, the University of California, and the National Aeronautics and Space Administration (NASA); the observatory was made possible by the generous financial support of the W. M. Keck Foundation. The authors wish to recognize and acknowledge the very significant cultural role and reverence that the summit of Mauna Kea has always had within the indigenous Hawaiian community; we are most fortunate to have the opportunity to conduct observations from this mountain. This work is partially based on observations made with the Nordic Optical Telescope, operated on the island of La Palma jointly by Denmark, Finland, Iceland, Norway, and Sweden, in the Spanish Observatorio del Roque de los Muchachos of the Instituto de Astrofisica de Canarias. We thank the RATIR instrument team and the staff of the Observatorio Astronómico Nacional on Sierra San Pedro Mártir. RATIR is a collaboration between the University of California, the Universidad Nacional Autonóma de México, NASA Goddard Space Flight Center, and Arizona State University, benefiting from the loan of an H2RG detector from Teledyne Scientific and Imaging. RATIR, the automation of the Harold L. Johnson Telescope of the Observatorio Astronómico Nacional on Sierra San Pedro Mártir, and the operation of both are funded by the partner institutions and through NASA grants NNX09AH71G, NNX09AT02G, NNX10AI27G, and NNX12AE66G, CONACyT grants INFR-2009-01-122785, UNAM PAPIIT grant IN113810, and a UC MEXUS-CONACyT grant. The National Energy Research Scientific Computing Center, supported by the Office of Science of the U.S. Department of Energy, provided staff, computational resources, and data storage for this project. This research has made use of the NASA/IPAC Extragalactic Database (NED) which is operated by the Jet Propulsion Laboratory, California Institute of Technology, under contract with NASA. Funding for SDSS-III has been provided by the Alfred P. Sloan Foundation, the Participating Institutions, the NSF, and the U.S. Department of Energy Office of Science. The SDSS-III web site is . JMS is supported by an NSF Astronomy and Astrophysics Postdoctoral Fellowship under award AST-1302771. JV is supported by Hungarian OTKA Grant NN 107637. MMK acknowledges generous support from the Hubble Fellowship and Carnegie-Princeton Fellowship. JCW’s supernova group at UT Austin is supported by NSF Grant AST 11-09801. Some work on this paper by JCW was done in the hospitable clime of the Aspen Center for Physics that is supported by NSF Grant PHY-1066293. JSB acknowledges the generous support of a CDI grant (#0941742) from the National Science Foundation.


  • Ahn et al. (2012) Ahn C. P., et al., 2012, ApJS, 203, 21
  • Blondin & Tonry (2007) Blondin S., Tonry J. L., 2007, ApJ, 666, 1024
  • Bloom et al. (2012) Bloom J. S., et al., 2012, ApJ, 744, L17
  • Branch et al. (2004) Branch D., et al., 2004, ApJ, 606, 413
  • Butler et al. (2012) Butler N., et al., 2012, in Society of Photo-Optical Instrumentation Engineers (SPIE) Conference Series Vol. 8446 of Society of Photo-Optical Instrumentation Engineers (SPIE) Conference Series, First Light with RATIR: An Automated 6-band Optical/NIR Imaging Camera. p. 34
  • Buzzoni et al. (1984) Buzzoni B., et al., 1984, The Messenger, 38, 9
  • Candia et al. (2003) Candia P., et al., 2003, PASP, 115, 277
  • Castor (1970) Castor J. I., 1970, MNRAS, 149, 111
  • Cenko et al. (2006) Cenko S. B., et al., 2006, PASP, 118, 1396
  • Childress et al. (2013) Childress M. J., et al., 2013, ApJ, 770, 29
  • Childress et al. (2013) Childress M. J., Filippenko A. V., Ganeshalingam M., Schmidt B. P., 2013, MNRAS, submitted (arXiv:1307.0563)
  • Conley et al. (2011) Conley A., et al., 2011, ApJS, 192, 1
  • Drake et al. (2009) Drake A. J., et al., 2009, ApJ, 696, 870
  • Drake et al. (2013) Drake A. J., et al., 2013, Central Bureau Electronic Telegrams, 3480, 1
  • Filippenko et al. (1992) Filippenko A. V., et al., 1992, ApJ, 384, L15
  • Filippenko et al. (2001) Filippenko A. V., Li W. D., Treffers R. R., Modjaz M., 2001, in Paczyński B., Chen W. P., Lemme C., eds, Small-Telescope Astronomy on Global Scales. Vol. 246, The Lick Observatory Supernova Search with the Katzman Automatic Imaging Telescope. Astron. Soc. Pac., San Francisco, p. 121
  • Fisher et al. (1997) Fisher A., Branch D., Nugent P., Baron E., 1997, ApJ, 481, L89
  • Foley (2012) Foley R. J., 2012, MNRAS, submitted (arXiv:1212.6261)
  • Foley et al. (2003) Foley R. J., et al., 2003, PASP, 115, 1220
  • Fox et al. (2012) Fox O. D., et al., 2012, in Society of Photo-Optical Instrumentation Engineers (SPIE) Conference Series Vol. 8453 of Society of Photo-Optical Instrumentation Engineers (SPIE) Conference Series, Performance and calibration of H2RG detectors and SIDECAR ASICs for the RATIR camera. p. 59
  • Fox & Filippenko (2013) Fox O. D., Filippenko A. V., 2013, ApJ, in press (arXiv:1304.4934)
  • Henry & Worthey (1999) Henry R. B. C., Worthey G., 1999, PASP, 111, 919
  • Hill et al. (1998) Hill G. J., et al., 1998, in D’Odorico S., ed., Society of Photo-Optical Instrumentation Engineers (SPIE) Conference Series Vol. 3355 of Society of Photo-Optical Instrumentation Engineers (SPIE) Conference Series, Hobby-Eberly Telescope low-resolution spectrograph: mechanical design. pp 433–443
  • Howell (2011) Howell D. A., 2011, Nature Communications, 2, 350
  • Howell et al. (2006) Howell D. A., et al., 2006, Nature, 443, 308
  • Iben & Tutukov (1984) Iben Jr. I., Tutukov A. V., 1984, ApJS, 54, 335
  • Jeffery (1989) Jeffery D. J., 1989, ApJS, 71, 951
  • Jordi et al. (2006) Jordi K., Grebel E. K., Ammon K., 2006, A&A, 460, 339
  • Kasliwal et al. (2012) Kasliwal M. M., et al., 2012, ApJ, 755, 161
  • Kulkarni (2013) Kulkarni S. R., 2013, The Astronomer’s Telegram, 4807, 1
  • Law et al. (2009) Law N. M., et al., 2009, PASP, 121, 1395
  • Leonard et al. (2000) Leonard D. C., Filippenko A. V., Chornock R., Li W. D., 2000, IAU Circ., 7471, 3
  • Li et al. (2001) Li W., et al., 2001, PASP, 113, 1178
  • Li et al. (2011) Li W., et al., 2011, MNRAS, 412, 1473
  • Lira et al. (1998) Lira P., et al., 1998, AJ, 115, 234
  • Marion et al. (2013) Marion G. H., et al., 2013, ApJ, submitted (arXiv:1302.3537)
  • Matheson et al. (2008) Matheson T., et al., 2008, AJ, 135, 1598
  • Matheson et al. (2000) Matheson T., Filippenko A. V., Ho L. C., Barth A. J., Leonard D. C., 2000, AJ, 120, 1499
  • Maund et al. (2013) Maund J. R., et al., 2013, MNRAS, 431, L102
  • Mazzali et al. (2005) Mazzali P. A., et al., 2005, ApJ, 623, L37
  • Morales-Garoffolo et al. (2013) Morales-Garoffolo A., et al., 2013, The Astronomer’s Telegram, 4955, 1
  • Nugent et al. (2011) Nugent P. E., et al., 2011, Nature, 480, 344
  • Oke et al. (1995) Oke J. B., Cohen J. G., Carr M., et al., 1995, PASP, 107, 375
  • Oke & Gunn (1982) Oke J. B., Gunn J. E., 1982, PASP, 94, 586
  • Parrent et al. (2012) Parrent J. T., et al., 2012, ApJ, 752, L26
  • Patat et al. (2007) Patat F., et al., 2007, Science, 317, 924
  • Perlmutter et al. (1999) Perlmutter S., et al., 1999, ApJ, 517, 565
  • Petrosian (1976) Petrosian V., 1976, ApJ, 209, L1
  • Phillips (1993) Phillips M. M., 1993, ApJ, 413, L105
  • Phillips et al. (1992) Phillips M. M., Wells L. A., Suntzeff N. B., Hamuy M., Leibundgut B., Kirshner R. P., Foltz C. B., 1992, AJ, 103, 1632
  • Pinto & Eastman (2000a) Pinto P. A., Eastman R. G., 2000a, ApJ, 530, 757
  • Pinto & Eastman (2000b) Pinto P. A., Eastman R. G., 2000b, ApJ, submitted (arXiv:000.6171)
  • Poznanski et al. (2011) Poznanski D., Ganeshalingam M., Silverman J. M., Filippenko A. V., 2011, MNRAS, 415, L81
  • Rau et al. (2009) Rau A., et al., 2009, PASP, 121, 1334
  • Richmond et al. (1993) Richmond M., Treffers R. R., Filippenko A. V., 1993, PASP, 105, 1164
  • Riess et al. (1998) Riess A. G., et al., 1998, AJ, 116, 1009
  • Rudy et al. (2002) Rudy R. J., Lynch D. K., Mazuk S., Venturini C. C., Puetter R. C., Höflich P., 2002, ApJ, 565, 413
  • Scalzo et al. (2010) Scalzo R. A., et al., 2010, ApJ, 713, 1073
  • Schlegel et al. (1998) Schlegel D. J., Finkbeiner D. P., Davis M., 1998, ApJ, 500, 525
  • Silverman et al. (2012a) Silverman J. M., et al., 2012a, MNRAS, 425, 1789
  • Silverman et al. (2012b) Silverman J. M., et al., 2012b, ApJ, 756, L7
  • Silverman et al. (2013) Silverman J. M., et al., 2013, ApJS, accepted (arXiv:1303.0763)
  • Silverman et al. (2011) Silverman J. M., Ganeshalingam M., Li W., Filippenko A. V., Miller A. A., Poznanski D., 2011, MNRAS, 410, 585
  • Silverman et al. (2012) Silverman J. M., Kong J. J., Filippenko A. V., 2012, MNRAS, 425, 1819
  • Skrutskie et al. (2006) Skrutskie M. F., et al., 2006, AJ, 131, 1163
  • Sobolev (1960) Sobolev V. V., 1960, Moving envelopes of stars. Cambridge: Harvard University Press
  • Sollerman et al. (2004) Sollerman J., et al., 2004, A&A, 428, 555
  • Sullivan et al. (2011) Sullivan M., et al., 2011, ApJ, 737, 102
  • Suzuki et al. (2012) Suzuki N., et al., 2012, ApJ, 746, 85
  • Thomas et al. (2004) Thomas R. C., Branch D., Baron E., Nomoto K., Li W., Filippenko A. V., 2004, ApJ, 601, 1019
  • Thomas et al. (2011) Thomas R. C., Nugent P. E., Meza J. C., 2011, PASP, 123, 237
  • Valenti et al. (2011) Valenti S., et al., 2011, MNRAS, 416, 3138
  • Vinkó et al. (2012) Vinkó J., et al., 2012, A&A, 546, A12
  • Wade & Horne (1988) Wade R. A., Horne K., 1988, ApJ, 324, 411
  • Wang et al. (2013) Wang X., Wang L., Filippenko A. V., Zhang T., Zhao X., 2013, Science, 340, 170
  • Watson et al. (2012) Watson A. M., et al., 2012, in Society of Photo-Optical Instrumentation Engineers (SPIE) Conference Series Vol. 8444 of Society of Photo-Optical Instrumentation Engineers (SPIE) Conference Series, Automation of the OAN/SPM 1.5-meter Johnson telescope for operations with RATIR
  • Webbink (1984) Webbink R. F., 1984, ApJ, 277, 355
  • Whelan & Iben (1973) Whelan J., Iben Jr. I., 1973, ApJ, 186, 1007
  • Woosley & Weaver (1994) Woosley S. E., Weaver T. A., 1994, ApJ, 423, 371
  • Yaron & Gal-Yam (2012) Yaron O., Gal-Yam A., 2012, PASP, 124, 668
  • Yu et al. (2000) Yu C., Modjaz M., Li W. D., 2000, IAU Circ., 7458, 1
Comments 0
Request Comment
You are adding the first comment!
How to quickly get a good reply:
  • Give credit where it’s due by listing out the positive aspects of a paper before getting into which changes should be made.
  • Be specific in your critique, and provide supporting evidence with appropriate references to substantiate general statements.
  • Your comment should inspire ideas to flow and help the author improves the paper.

The better we are at sharing our knowledge with each other, the faster we move forward.
The feedback must be of minimum 40 characters and the title a minimum of 5 characters
Add comment
Loading ...
This is a comment super asjknd jkasnjk adsnkj
The feedback must be of minumum 40 characters
The feedback must be of minumum 40 characters

You are asking your first question!
How to quickly get a good answer:
  • Keep your question short and to the point
  • Check for grammar or spelling errors.
  • Phrase it like a question
Test description