Hostname: page-component-7c8c6479df-nwzlb Total loading time: 0 Render date: 2024-03-19T10:21:36.682Z Has data issue: false hasContentIssue false

The ANU WiFeS SuperNovA Programme (AWSNAP)

Published online by Cambridge University Press:  08 November 2016

Michael J. Childress*
Affiliation:
Research School of Astronomy and Astrophysics, Australian National University, Canberra, ACT 2611, Australia ARC Centre of Excellence for All-sky Astrophysics (CAASTRO), Canberra, ACT, Australia School of Physics and Astronomy, University of Southampton, Southampton, SO17 1BJ, UK
Brad E. Tucker
Affiliation:
Research School of Astronomy and Astrophysics, Australian National University, Canberra, ACT 2611, Australia ARC Centre of Excellence for All-sky Astrophysics (CAASTRO), Canberra, ACT, Australia
Fang Yuan
Affiliation:
Research School of Astronomy and Astrophysics, Australian National University, Canberra, ACT 2611, Australia ARC Centre of Excellence for All-sky Astrophysics (CAASTRO), Canberra, ACT, Australia
Richard Scalzo
Affiliation:
Research School of Astronomy and Astrophysics, Australian National University, Canberra, ACT 2611, Australia
Ashley Ruiter
Affiliation:
Research School of Astronomy and Astrophysics, Australian National University, Canberra, ACT 2611, Australia ARC Centre of Excellence for All-sky Astrophysics (CAASTRO), Canberra, ACT, Australia
Ivo Seitenzahl
Affiliation:
Research School of Astronomy and Astrophysics, Australian National University, Canberra, ACT 2611, Australia ARC Centre of Excellence for All-sky Astrophysics (CAASTRO), Canberra, ACT, Australia
Bonnie Zhang
Affiliation:
Research School of Astronomy and Astrophysics, Australian National University, Canberra, ACT 2611, Australia
Brian Schmidt
Affiliation:
Research School of Astronomy and Astrophysics, Australian National University, Canberra, ACT 2611, Australia
Borja Anguiano
Affiliation:
Department of Physics and Astronomy, Macquarie University, NSW 2109, Australia
Suryashree Aniyan
Affiliation:
Research School of Astronomy and Astrophysics, Australian National University, Canberra, ACT 2611, Australia
Daniel D. R. Bayliss
Affiliation:
Research School of Astronomy and Astrophysics, Australian National University, Canberra, ACT 2611, Australia Observatoire Astronomique de l’Université de Genève, 51 ch. des Maillettes, 1290 Versoix, Switzerland
Joao Bento
Affiliation:
Research School of Astronomy and Astrophysics, Australian National University, Canberra, ACT 2611, Australia
Michael Bessell
Affiliation:
Research School of Astronomy and Astrophysics, Australian National University, Canberra, ACT 2611, Australia
Fuyan Bian
Affiliation:
Research School of Astronomy and Astrophysics, Australian National University, Canberra, ACT 2611, Australia
Rebecca Davies
Affiliation:
Research School of Astronomy and Astrophysics, Australian National University, Canberra, ACT 2611, Australia
Michael Dopita
Affiliation:
Research School of Astronomy and Astrophysics, Australian National University, Canberra, ACT 2611, Australia
Lisa Fogarty
Affiliation:
Sydney Institute for Astronomy (SIfA), School of Physics, The University of Sydney, NSW 2006, Australia
Amelia Fraser-McKelvie
Affiliation:
School of Physics and Astronomy, Monash University, Clayton, Victoria 3800, Australia Monash Centre for Astrophysics (MoCA), Monash University, Clayton, Victoria 3800, Australia
Ken Freeman
Affiliation:
Research School of Astronomy and Astrophysics, Australian National University, Canberra, ACT 2611, Australia
Rajika Kuruwita
Affiliation:
Research School of Astronomy and Astrophysics, Australian National University, Canberra, ACT 2611, Australia
Anne M. Medling
Affiliation:
Research School of Astronomy and Astrophysics, Australian National University, Canberra, ACT 2611, Australia
Simon J. Murphy
Affiliation:
Research School of Astronomy and Astrophysics, Australian National University, Canberra, ACT 2611, Australia
Simon J. Murphy
Affiliation:
Sydney Institute for Astronomy (SIfA), School of Physics, The University of Sydney, NSW 2006, Australia Stellar Astrophysics Centre, Department of Physics and Astronomy, Aarhus University, 8000 Aarhus C, Denmark
Matthew Owers
Affiliation:
Department of Physics and Astronomy, Macquarie University, NSW 2109, Australia Australian Astronomical Observatory, PO Box 915, North Ryde, NSW 1670, Australia
Fiona Panther
Affiliation:
Research School of Astronomy and Astrophysics, Australian National University, Canberra, ACT 2611, Australia ARC Centre of Excellence for All-sky Astrophysics (CAASTRO), Canberra, ACT, Australia
Sarah M. Sweet
Affiliation:
Research School of Astronomy and Astrophysics, Australian National University, Canberra, ACT 2611, Australia
Adam D. Thomas
Affiliation:
Research School of Astronomy and Astrophysics, Australian National University, Canberra, ACT 2611, Australia
George Zhou
Affiliation:
Research School of Astronomy and Astrophysics, Australian National University, Canberra, ACT 2611, Australia Harvard-Smithsonian Center for Astrophysics, 60 Garden St, Cambridge, MA 02138, USA
Rights & Permissions [Opens in a new window]

Abstract

This paper presents the first major data release and survey description for the ANU WiFeS SuperNovA Programme. ANU WiFeS SuperNovA Programme is an ongoing supernova spectroscopy campaign utilising the Wide Field Spectrograph on the Australian National University 2.3-m telescope. The first and primary data release of this programme (AWSNAP-DR1) releases 357 spectra of 175 unique objects collected over 82 equivalent full nights of observing from 2012 July to 2015 August. These spectra have been made publicly available via the WISEREP supernova spectroscopy repository.

We analyse the ANU WiFeS SuperNovA Programme sample of Type Ia supernova spectra, including measurements of narrow sodium absorption features afforded by the high spectral resolution of the Wide Field Spectrograph instrument. In some cases, we were able to use the integral-field nature of the Wide Field Spectrograph instrument to measure the rotation velocity of the SN host galaxy near the SN location in order to obtain precision sodium absorption velocities. We also present an extensive time series of SN 2012dn, including a near-nebular spectrum which both confirms its ‘super-Chandrasekhar’ status and enables measurement of the sub-solar host metallicity at the SN site.

Type
Research Article
Copyright
Copyright © Astronomical Society of Australia 2016 

1 INTRODUCTION

In the last decade, wide-field extragalactic transient surveys—such as the Palomar Transient Factory (PTF; Rau et al. Reference Rau2009; Law et al. Reference Law2009), the Panoramic Survey Telescope and Rapid Response System (PanSTARRS; Kaiser et al. Reference Kaiser, Stepp, Gilmozzi and Hall2010), the Catalina Real-time Transient Survey (CRTS; Drake et al. Reference Drake2009), the Texas Supernova Search (Quimby Reference Quimby2006; Yuan Reference Yuan2010), and the All-Sky Automated Survey for Supernovae (ASAS-SN; Shappee et al. Reference Shappee2014; Holoien et al. Reference Holoien2016)—have revolutionised our understanding of the myriad ways in which stars explode through the discovery of new classes of exotic transients. Simultaneously, these surveys have discovered hundreds of supernovae (SNe) of ‘traditional’ types (see Filippenko Reference Filippenko1997, for a review), enabling statistical analyses of the properties of these SNe.

Whilst imaging surveys have provided discovery and light curves for this wealth of new transients, complementary spectroscopy surveys have provided the critical insight into the physical origins of these events. Numerous supernova spectroscopy surveys have released thousands of high-quality spectra of nearby SNe into the public domain (Matheson et al. Reference Matheson2008; Blondin et al. Reference Blondin2012; Silverman et al. Reference Silverman2012c; Folatelli et al. Reference Folatelli2013; Modjaz et al. Reference Modjaz2014). These surveys have frequently been dedicated to the spectroscopic follow-up of Type Ia supernovae (SNe Ia) which, due to their rates and luminosities, dominate any magnitude-limited imaging survey. Such surveys have revealed that photometrically similar SNe can still exhibit diversity of spectroscopic behaviour, indicating spectra remain a critical tool for revealing the full nature of the supernova progenitors (particularly for SNe Ia). Additionally, spectra remain critical for supernova classification—particularly at early phases when the full photometric evolution has yet to be revealed. Such early classifications then inform the use of additional SN follow-up facilities, including those operating outside the optical window.

Recently, the Public ESO Spectroscopic Survey for Transient Objects (PESSTO; Smartt et al. Reference Smartt2015) began a multi-year programme on the NTT 3.6-m telescope in Chile, with the goal of obtaining high-quality spectral time series for roughly 100 transients (of all kinds) to be released to the public. This survey has already released hundreds of spectra in its first two annual data releases, and continues to release all SN classification spectra within typically 1 d from observation. Other ongoing SN spectroscopy programmes, such as the Asiago Supernova Programme (Tomasella et al. Reference Tomasella2014), also make important contributions to the transient community through timely SN classification and spectroscopy releases.

Here, we describe our ongoing spectroscopy programme AWSNAP—the ANU WiFeS SuperNovA Programme—which uses the Wide Field Spectrograph (WiFeS; Dopita et al. Reference Dopita, Hart, McGregor, Oates, Bloxham and Jones2007, Reference Dopita2010) on the Australian National University (ANU) 2.3-m telescope at Siding Spring Observatory in Australia. In this paper, we describe the data processing procedures for this ongoing programme, and describe the first AWSNAP data release (AWSNAP DR1) comprising 357 spectra of 175 supernova of various types obtained during 82 classically scheduled observing nights over a 3-yr period from 2012 July to 2015 August. Most of these spectra have been released publicly via the Weizmann Interactive Supernova data REPository (WISeREPFootnote 1 — Yaron & Gal-Yam Reference Yaron and Gal-Yam2012), with the remainder set to be released within the next year as part of forthcoming PESSTO papers. This programme will continue to observe SNe of interest and classify SN discoveries from transient searches such as the new SkyMapper Transients Survey (Keller et al. Reference Keller2007). We aim to release future SN classification spectra from AWSNAP publicly via WISeREP in parallel with any classification announcements.

This paper is organised as follows. Section 2 describes the WiFeS data processing and SN spectrum extraction procedures. Section 3 presents general properties of our SN sample and compares AWSNAP DR1 to other public SN spectra releases. In Section 4, we present some analysis of the properties of the SNe Ia in our sample, including measurement of narrow sodium absorption features afforded by the intermediate resolution of the WiFeS spectrograph. Some concluding remarks follow in Section 5.

2 OBSERVATIONS AND DATA DESCRIPTION

Observations for AWSNAP were conducted with the WiFeS—(Dopita et al. Reference Dopita, Hart, McGregor, Oates, Bloxham and Jones2007, Reference Dopita2010) on the ANU 2.3-m telescope at Siding Spring Observatory in northern New South Wales, Australia. Observing nights were classically scheduled with a single night of observing every 8–15 d. On some occasions, special objects of interest were observed during non-AWSNAP nights. A full list of the AWSNAP transient spectra is presented in Table A3 in Appendix A. In the sections that follow, we describe the processing of the WiFeS data, then characterise both the long-term performance of the WiFeS instrument and observing conditions at Siding Spring.

2.1. Data reduction and supernova spectrum extraction

The WiFeS instrument is an image-slicing integral field spectrograph with a wide 25 arcsec × 38 ′′arcsec field of view. For AWSNAP, this frequently provided simultaneous integral field observations of SNe and their host galaxies. The WiFeS image slicer breaks the field of view into 25 ‘slitlets’ of width 1 arcsec, which then pass through a dichroic beamsplitter and volume phase holographic (VPH) gratings before arriving at 4k × 4k CCD detectors. AWSNAP observations were always conducted with a CCD binning of 2 in the vertical direction—this sets the vertical spatial scale of the detector to be 1 arcsec, yielding final integral field elements (or ‘spaxels’) of size 1 arcsec × 1 arcsec. Typically seeing at Siding Spring is roughly 2 arcsec (see Section 2.3).

The VPH gratings utilised by WiFeS provide a higher wavelength resolution than traditional glass gratings. The low- and high-resolution gratings provide resolutions of R = 3000 and R = 7000, respectively, yielding velocity resolutions of up to σv ~ 45 km s−1 which is ideal for observing nebular emission lines from ionised regions in galaxies. For SNe, this can reveal narrow absorption features (see Section 4.2) from circumstellar material (CSM) which are typically smeared out by lower resolution spectrographs. AWSNAP observations were generally conducted with the lower resolution B3000 and R3000 gratings for the blue and red arms of the spectrograph, respectively, with the RT560 dichroic beamsplitter. Occasionally, the R7000 grating was deployed on the red arm to provide higher resolution observations of sodium absorption features. In Table 1, we provide the wavelength range, spectroscopic pixel size, and wavelength resolution (determined as the FWHM of calibration lamp emission lines) for the three gratings used for AWSNAP observations.

Table 1. Details of WiFeS gratings.

Data for AWSNAP observations were reduced with version 0.7.0 of the PyWiFeS pipeline (Childress et al. Reference Childress, Vogt, Nielsen and Sharp2014a). PyWiFeS performs standard image pre-processing such as overscan and bias subtraction, as well as cosmic ray rejection using a version of LACosmic (van Dokkum Reference van Dokkum2001) tailored for WiFeS data. The wavelength solution for WiFeS is derived using an optical model of the spectrograph which achieves an accuracy of 0.05 to 0.10Å (for R = 7000 and R = 3000, respectively) across the entire detector. Spectral flatfielding (i.e. correction of pixel-to-pixel quantum efficiency variations) is achieved with an internal quartz lamp, whilst spatial flat-fielding across the full instrument field of view is facilitated by twilight sky flats. Once the data has been pre-processed and flat-fielded, it is resampled onto a rectilinear three-dimensional (x, y, λ) grid (a ‘data cube’). We then flux calibrate the data cubes with the use of spectrophotometric standard stars (from, e.g. Oke Reference Oke1990; Bessell Reference Bessell1999; Stritzinger et al. Reference Stritzinger, Suntzeff, Hamuy, Challis, Demarco, Germany and Soderberg2005) and the Siding Spring Observatory extinction curve from Bessell (Reference Bessell1999), whilst telluric features are removed using observations of smooth spectrum stars.

To further facilitate data reduction for the AWSNAP data release, we developed an SQL database for WiFeS observations using the Python Django framework. We created modified versions of the PyWiFeS reduction scripts that allow the user to request that all data from a specific night using a specific grating be fully reduced. The reduction scripts then query the database for all science observations and calibration data from that night and perform the required reduction procedures. For nights where a full suite of calibrations was not available (e.g. due to guest observations on a non-AWSNAP night), calibration solutions from the closest (in time) AWSNAP night were automatically identified via the database and employed in the reduction. Both the wavelength solution and spatial flat-fielding of the instrument are incredibly stable on long (~year) timescales (see Section 2.2), thus validating the choice to use calibrations from different nights where necessary. This new database-driven data processing mode for PyWiFeS will be an important component of the upcoming effort to develop fully robotic queue-based observing capabilities for the ANU 2.3-m telescope.

The output of the PyWiFeS pipeline is a flux calibrated data cube that contains signal from the target supernova, the night sky, and occasionally the SN host galaxy. Extraction of the SN spectrum requires isolation of the spaxels containing supernova signal and subtraction of the underlying sky (and possibly galaxy) background. To achieve this, we constructed a custom Python-based GUI which allows the user to manually select spaxels containing the target SN (‘object’ spaxels) and spaxels used to determine the background (‘sky’ spaxels, which may contain some galaxy signal). The background spectrum is determined as the median spectrum across all ‘sky’ spaxels, and this median spectrum is subtracted from each ‘object’ spaxel. The final SN spectrum is then the sum of the sky-subtracted object spaxels, and the variance spectrum is the sum of the variance from all object spaxels (with no subtraction of sky variance).

This SN spectrum extraction technique produces excellent quality sky subtraction due to the robust spatial flat-fielding achieved with PyWiFeS. However, some obvious sky subtraction residual features are evident in the redder wavelengths of most R3000 spectra where the night sky exhibits sharp emission features from rotational transitions of atmospheric OH molecules. The intrinsic line width of these emission lines is below the resolution of the spectrograph, meaning the observed line width is that of the spectrograph—which in this case is only slightly larger than the detector pixel size. The natural wavelength solution of the spectrograph shifts spaxel-to-spaxel, so when all spaxels are resampled onto the same wavelength grid this means sky lines experience pixel-wise resampling that is not uniform across the full instrument field of view. Thus, when a median sky spectrum is subtracted from a specific spaxel, some residual features arise due to this resampling effect. A more robust technique for correcting this would be to model the intrinsic sky spectrum using the multiple samplings achieved across all spaxels and resample it to the wavelength solution of each spaxel before subtracting it (as was demonstrated for two-dimensional spectroscopic data by Kelson Reference Kelson2003). Such a technique is beyond the scope of the current data release, but is being prioritised for future AWSNAP releases.

2.2. Long-term behaviour of the WiFeS instrument

One key advantage of having a long-running observing programme is the ability to characterise the long-term behaviour of the WiFeS instrument. Below we analyse the stability of the wavelength solution and instrument throughput on multi-year timescales. WiFeS did undergo a major change in early 2013 when the detectors for both arms of the spectrograph were replaced with higher throughput E2V CCDs. Thus, we restrict our analysis to dates from late 2013 May until the end of the current data release in 2015 August.

We collected the wavelength solution fits for the B3000 (R3000) grating from 48 (45) distinct epochs following the WiFeS detector upgrade. In Figure 1, we plot the difference between the wavelength solution for each individual epoch and the mean wavelength solution across all epochs. Epochs are colour-coded from earliest (red) to latest (purple) to illustrate potential coherent long-term shifts in the wavelength solution. The top panels present the B3000 grating, whilst the bottom panels present the R3000 gratings. The left and middle columns represent the WiFeS slitlets at the top and middle of the detectors, respectively. In the right hand panels of this Figure, we show the average deviation of the wavelength solution for a given epoch from the global mean wavelength solution, as a function of epoch. This allows us to track any coherent evolution of the detector wavelength solution with time.

Figure 1. Evolution of the WiFeS wavelength solution over a 2-yr period spanning 2013 May to 2015 April. In the left two panels of each row, we plot the deviation of individual wavelength solutions from the mean solution (averaged over the full 2-yr period) for the top and middle slitlets of the instrument (left and middle columns, respectively) for the B3000 (top row) and R3000 (bottom row) gratings. Wavelength solution residuals are colour-coded by date from earliest (red) to latest (purple), with the pixel size (dashed black lines) and wavelength solution residual RMS (solid black lines) displayed for comparison. In the right panels of each row, we show the average deviation (across all wavelengths) from the mean wavelength solution as a function of epoch, to trace the temporal evolution of the instrument solution (points for individual epochs obey the same colour scheme as the left panels).

From these plots, we see clear demonstration of the remarkable stability of the WiFeS instrument. The RMS variation of wavelength solution is smaller than a single pixel for both gratings (i.e. both detectors) for nearly all wavelengths, with the exception of the lower throughput regions near the dichroic boundary. There is some evidence of a coherent shift of the blue detector wavelength solution over the 2-yr time period probed here, but this is still less than two pixels shift. For the red detector, we can say confidently that the wavelength solution has shifted by less than a pixel over a timescale of 2 yrs. This remarkable stability achieved (by design) by WiFeS means the use of the wavelength solution from an adjacent night yields negligible changes in wavelength, and thus small instrumental velocity shifts—highly suitable for supernova analyses.

We then collected illumination corrections (derived from twilight flat observations) from 46 (45) distinct epochs for the B3000 (R3000) grating. In Figure 2, we show the mean illumination correction (left panels) and RMS of the illumination correction (right panels—presented in fractional form, i.e. the RMS divided by the mean) for both the B3000 (top) and R3000 (bottom) gratings. We find the mean variation of the illumination correction to be 1.4 and 1.2% for the B3000 and R3000 gratings, respectively, for the full WiFeS field of view. The nature of the instrument optics are such that the outer regions of instrument field of view have slightly lower throughput, and thus a higher fractional RMS than the inner most region. For the innermost 8 arcsec × 8 arcsec region typically used for SN observation in AWSNAP, the RMS of the illumination correction is 0.8 and 0.3% for the blue and red detectors (i.e. B3000 and R3000 gratings). Thus, the throughput of the WiFeS instrument has remarkable spatial uniformity on long (multi-year) timescales.

Figure 2. The normalised mean (left column) and RMS (right column) of the illumination corrections for the B3000 (top row) and R3000 (bottom row) gratings. The RMS images are shown as fractional values of the mean illumination correction.

2.3. Observing conditions at siding spring observatory

In addition to the long-term behaviour of the WiFeS instrument, our observing programme allows us to monitor the observing conditions at Siding Spring Observatory. Below we briefly discuss the atmospheric throughput and seeing conditions experienced during AWSNAP observing.

We collected the flux calibration solutions from 59(54) epochs for the B3000 (R3000) grating from 2013 May to 2015 August, and these are plotted in Figure 3 after being normalised in the wavelength range with highest throughput (45 00–5 400 Å and 6 500–8 000 Å for B3000 and R3000, respectively). These curves represent the normalised throughput (in magnitudes) of the instrument and atmosphere as measured with spectrophotometric standard stars (typically from Oke Reference Oke1990; Bessell Reference Bessell1999; Stritzinger et al. Reference Stritzinger, Suntzeff, Hamuy, Challis, Demarco, Germany and Soderberg2005) whose flux has already been corrected using the nominal Siding Spring extinction curve from Bessell (Reference Bessell1999). The effects of the instrument throughput and atmospheric transmission are degenerate here, as WiFeS instrument throughput cannot be independently measured using local (i.e. terrestrial) calibration sources.

Figure 3. Flux calibration solutions for the B3000 (top) and R3000 (bottom) gratings. As in Figure 1, these are colour-coded by date from earliest (red) to latest (purple), with the mean flux calibration solution shown as the solid black line.

From the curves in Figure 3, we see that the total combined throughput of the instrument plus atmosphere is relatively stable. We calculated the RMS colour variation in the throughput curves and find variations of σ(UB) = 0.09 mag for B3000 (note the V band runs into the wavelength range where the dichroic splits light between the blue and red channels) and σ(ri) = 0.04 mag for R3000. We note these are calculated from the mean flux calibration solution for each night, where no attempt has been made to derive a unique extinction curve for a given night. These colour variations are relatively small, and comparable in size to the colour dispersion found when comparing spectrophotometry to imaging photometry for other SN spectroscopy samples (e.g. Silverman et al. Reference Silverman2012c; Blondin et al. Reference Blondin2012; Modjaz et al. Reference Modjaz2014).

We also measured the atmospheric seeing during our observing programme, using both guide star FWHM measurements recorded from the WiFeS guider camera and low airmass (secz ⩽ 1.5) WiFeS data cubes convolved with the B- and R-band filter curves (for the blue and red cameras, respectively). We plot the observed distribution of seeing values in Figure 4. The seeing distribution peaks slightly above 1.5 arcsec (for all measurements) with a tail predominantly filled to 2.5 arcsec, with little or no sub-arcsecond observations and a small number of observations with incredibly poor seeing (3.0 arcsec or greater).

Figure 4. Seeing measurements at Siding Spring Observatory as measured during AWSNAP observations. These include measurements from the WiFeS guider camera (green dash-dot histogram), and measurements of low airmass standard star WiFeS datacubes convolved with B-band (blue solid histogram—from the blue detector) and R-band (red dashed histogram—from the red detector) filter curves.

3 THE AWSNAP SUPERNOVA SAMPLE AND SPECTROSCOPIC DATA RELEASE

In this Section, we briefly describe the global characteristics of the sample of SNe comprising the first data release (DR1) for AWSNAP. This consists of observations made between 2012 July18 and 2015 August17, a total of 357 epochs of 175 total SNe. These spectra have all been uploaded to WISeREP (Yaron & Gal-Yam Reference Yaron and Gal-Yam2012), with most made publicly available and the small remainder set to be made public with the associated PESSTO publication within the next year. Some additional spectra taken after 2015 August 17 have been processed and released via WISeREP, and we expect the future release of AWSNAP spectra to proceed in a continuous fashion via the same procedures outlined in this work.

AWSNAP originated as the spectroscopic observation programme for the supernova group at the ANU. The programme was intended to cover a broad range of science topics driven by transient discoveries from contemporaneous photometric surveys, with a large emphasis on timely classification of newly discovered transients. The observing and target selection strategy for AWSNAP was heavily influenced by the scheduling of our observing time, which typically consisted of one single classically scheduled full night of observing every 8–15 d throughout the entirety of the calendar year. The most significant implication of this scheduling was that dense spectroscopic sampling (i.e. 2–3 d cadence) was generally not viable for our preferred targets (though on rare occasions we requested time exchange with other observers for targets of particular importance). Furthermore, additional targets were always needed to fill an entire night of observing. Thus, we frequently chose to make complementary observations of targets being observed through the PESSTO programme (Smartt et al. Reference Smartt2015), or chose targets whose spectroscopic data would have legacy value for future analyses. As a result, we observed a diverse range of targets that covered the entire Southern sky. The full list of targets and their classification information in presented in Table A1 in Appendix A, and their distribution on the sky is shown in Figure 5.

Figure 5. On-sky distribution (in equatorial coordinates) of all extragalactic targets in AWSNAP DR1 (white diamonds) plotted over the 857 GHz all-sky map from the Planck satellite (Planck Collaboration et al. 2011) which reveals emission by Milky Way dust. The physical pointing limit of the ANU 2.3-m telescope (~+40° declination) is shown as the solid gray bar.

The AWSNAP spectra have relatively high signal-to-noise (S/N) by design so that they might be useful for precision spectral feature measurement. Our sample’s median S/N is 16 per Å, with 1σ high and low S/N values of 40 and 6 per Å, respectively. We show some example AWSNAP spectra in Figure 6 exhibiting low (− 1σ), medium (median), and high (+ 1σ) S/N.

Figure 6. Example spectra spanning the typical range of signal-to-noise (S/N) in the AWSNAP sample. Shown here are spectra of SN 2012dt (a SN II, with a ‘low’ S/N of 5 per Å), ASASSN-15ba (a SN Ia, with a ‘medium’ S/N of 13 per Å), and SN 2012dj (a SN Ib, with a ‘high’ S/N of 33 per Å).

In Figure 7, we present a histogram of the number of spectroscopic observations per target for the AWSNAP sample compared to three major SN Ia spectroscopic data releases: the Berkeley SN Ia Programme (BSNIP; Silverman et al. Reference Silverman2012c), the Harvard Center for Astrophysics (CfA) Supernova Programme (Matheson et al. Reference Matheson2008; Blondin et al. Reference Blondin2012), and the Carnegie Supernova Project (CSP; Folatelli et al. Reference Folatelli2013). Our sample has a high fraction of singly observed targets compared to the other programmes. This is in part due to the fact that our data release includes all our SN classification spectra, as well as the persistent need for us to fill our classically scheduled observing queue. We thus also show in the inset of Figure 7 the normalised histogram of spectroscopic epochs for multiply observed SNe (i.e. targets with N obs > 2), which shows a much more similar distribution to the other programmes.

Figure 7. Normalised histogram of the number of spectroscopic epochs per target for the full sample of AWSNAP DR1 targets. For comparison, we also show the same histogram for previous SN spectroscopy surveys: BSNIP (Silverman et al. Reference Silverman2012c), CfA (Matheson et al. Reference Matheson2008; Blondin et al. Reference Blondin2012), and CSP (Folatelli et al. Reference Folatelli2013). The inset shows the (re-)normalised histograms of the number of spectroscopic epochs for multiply observed targets (i.e. those with N obs > 1) in the same surveys.

The types of SNe observed in AWSNAP DR1 is summarised in Table 2 and presented graphically in Figure 8. We show both the number of targets and number of spectra for AWSNAP, and for reference we compare this to the magnitude-limited rates for SNe in the local universe calculated by Li et al. (Reference Li2011a). For this figure, we have grouped the SNe by type into four main categories:

  • SNe Ia: includes the standard (Branch-normal) SNe Ia, sub-luminous (SN 1991bg-like Filippenko et al. Reference Filippenko1992a), over-luminous (SN 1991T- and SN 1999aa-like Filippenko et al. Reference Filippenko1992b), candidate ‘super-Chandrasekhar’ SNe Ia (Scalzo et al. Reference Scalzo2010, Reference Scalzo2012), and SNe Iax (Foley et al. Reference Foley2013).

    Figure 8. Total number of spectra (outer ring) and SN targets (middle annulus) for AWSNAP broken down by SN type, compared to the volume-limited SN rates (inner circle) for the LOSS survey (Li et al. Reference Li2011a, —note this galaxy-targeted survey did not find any SLSNe). In each ring, the regions are colour-coded by (broad) SN type (see text for discussion): SNe Ia (blue), SNe II(red), SNe Ib/Ic (green), and SLSNe (purple).

    Table 2. AWSNAP objects and spectra by SN type.

  • SNe Ibc: SNe Ic, SNe Ib, and SNe IIb—the standard classes of ‘stripped-envelope’ SNe (e.g. Bianco et al. Reference Bianco2014; Modjaz et al. Reference Modjaz2014; Graur et al. Reference Graur, Bianco, Modjaz, Maoz, Shivvers, Filippenko and Li2015; Liu et al. Reference Liu, Modjaz, Bianco and Graur2016).

  • SNe II: SNe IIP and SNe IIL—which we note cannot be distinguished spectroscopically—and SNe IIn.

  • SLSNe: ‘superluminous’ SNe—for our sample, this consists entirely of the ‘SLSN-Ic’ type (see, e.g. Inserra et al. Reference Inserra2013) which show blue continua with weak absorption features and no hydrogen signatures.

We note these groupings are made strictly to provide broad perspective on the sample statistics, but we reiterate that the SN sub-types within each group have their own unique physical mechanisms.

From Figure 8, we clearly see that, as expected, SNe Ia comprise the majority of objects and spectra in the AWSNAP sample, but comprise a smaller fraction than might be expected from a pure magnitude limited sample of SNe such as that of the Lick Observatory Supernova Search (LOSS, whose relative SN rates are presented in Li et al. Reference Li2011a). This may be due in part to the fact that many of the SNe in AWSNAP were discovered by untargeted supernova searches such as the ASAS-SN (Shappee et al. Reference Shappee2014). These surveys find SNe in low-mass galaxies that are missed by targeted surveys such as LOSS. Due to the increased star-formation intensity in low-mass galaxies (e.g. Salim et al. Reference Salim2007), this means the relative rate of core-collapse SNe will be higher and thus CCSNe will comprise a higher fraction of the sample. Additionally, we explicitly targeted a higher fraction of CCSN discoveries owing to the comparative paucity of CCSN spectroscopic samples compared to SNe Ia (again arising from the magnitude-limited rates). Thus, our higher representation of CCSNe than the expectation from the magnitude-limited rates of Li et al. (Reference Li2011a) is likely a combination of the higher CCSN fraction for our untargeted feeder surveys and our explicit emphasis on preferential targeting of CCSNe.

4 TYPE IA SUPERNOVA SPECTRA FROM AWSNAP

As a first demonstration of the usefulness of the AWSNAP dataset, we analyse basic spectroscopic features of the SNe Ia in our sample—this comprises 180 total spectra of 101 objects. In Figure 9, we plot a histogram of the phasesFootnote 2 of our SN Ia spectra, again compared to previous large SN Ia spectroscopy surveys (CfA, BSNIP, CSP). Our sample shows a similar phase coverage as previous surveys, with perhaps a slight increase in earlier phases due to improved SN discovery efficiency from nearby SN searches. In the sections that follow, we further analyse features of interest from the AWSNAP SN Ia spectroscopy sample.

Figure 9. Histograms of spectroscopic phases (with respect to B-band maximum light) for SN Ia spectra in AWSNAP and other SN Ia spectroscopy samples (the same as in Figure 7). Note the AWSNAP phases are based on the spectroscopic-based phase reported with the SN classification, which may have an associated uncertainty of 3–5 d.

4.1. Spectral features in SNe Ia at maximum light

We begin by inspecting the spectra of SNe Ia close to the epoch of peak brightness. These maximum light spectra have a long history of providing key insights into the diversity of SN Ia explosions through the study of ‘spectral indicators’ (Nugent et al. Reference Nugent, Phillips, Baron, Branch and Hauschildt1995; Hatano et al. Reference Hatano, Branch, Lentz, Baron, Filippenko and Garnavich2000; Benetti et al. Reference Benetti2005; Bongard et al. Reference Bongard, Baron, Smadja, Branch and Hauschildt2006, Reference Bongard, Baron, Smadja, Branch and Hauschildt2008; Hachinger, Mazzali, & Benetti Reference Hachinger, Mazzali and Benetti2006; Hachinger et al. Reference Hachinger, Mazzali, Tanaka, Hillebrandt and Benetti2008; Branch et al. Reference Branch2006; Branch, Dang, & Baron Reference Branch, Dang and Baron2009; Bronder et al. Reference Bronder2008; Silverman, Kong, & Filippenko Reference Silverman, Kong and Filippenko2012b), as well as potential avenues for improving the cosmological standardisation of SNe Ia (Wang et al. Reference Wang2009; Bailey et al. Reference Bailey2009; Blondin, Mandel, & Kirshner Reference Blondin, Mandel and Kirshner2011; Silverman et al. Reference Silverman, Ganeshalingam, Li and Filippenko2012a). Thus, we actively targeted many SNe Ia (which had already been classified) on an epoch close to maximum light to obtain high-quality spectra facilitating such studies.

We isolated the sample SNe Ia with a spectrum within 5 d of estimated peak brightness. It is important to reiterate here that the phases for our SN Ia sample are extrapolated from the reported spectroscopic classification phase and date. It has been demonstrated previously (e.g. Blondin et al. Reference Blondin2012) that the phase determined for SNe Ia via spectroscopic matching codes such as SNID (Blondin & Tonry Reference Blondin and Tonry2007) typically has an uncertainty of at least 3–5 d. Thus, we may have included SN Ia spectra as much as 10 d removed from maximum light. Our analysis here is intended to be illustrative of the utility of our published spectra, and more detailed quantitative spectroscopic analyses should always be coupled to a robust photometric data set.

For the analysis that follows, we measure two key quantities of interest: the silicon absorption ratio R Si, and the strength of high-velocity features (HVFs) R HVF. For this work, we define R Si as the ratio of the pseudo equivalent width (pEW) of the 5962 Å feature to the pEW of the 6355 Å feature. The pEW for each feature is measured by fitting a linear pseudo-continuum across narrow regions redward and blueward of the given feature (similar to methods employed in Silverman et al. Reference Silverman, Kong and Filippenko2012b), then integrating the flux in the normalised absorption feature.

R Si is known to correlate strongly with the SN Ia light curve decline rate (Nugent et al. Reference Nugent, Phillips, Baron, Branch and Hauschildt1995). We quantified the relationship between our definition of R Si (the pEW ratio) and light curve decline rate Δm 15 by fitting a linear relationship between these quantities for a sample of 342 SNe Ia collated from the CfA, BSNIP, and CSP samples. The fit to these data is presented in Appendix A2, shown graphically in Figure A1 with a best fit trend given by Equation (A1). Below we will use this relation to display the corresponding range of Δm 15 spanned by our observed values of R Si.

We measure the HVF strength for our SN Ia sample using the same techniques employed in Childress et al. (Reference Childress, Filippenko, Ganeshalingam and Schmidt2014b), and similarly define R HVF as the ratio of the pEW of the HVF to the pEW of the photospheric velocity feature (PVF). Briefly, we first normalise the Ca NIR feature using a linear pseudo-continuum fit. We then fit the Ca NIR feature with two Gaussians in velocity space with velocity centres, widths, and absorption depths fitted with a Python mpfit routine. Each component in velocity space corresponds to a triplet in wavelength space, and we set the absorption depth of each component equal to each other (i.e. the optically thick limit). As in Childress et al. (Reference Childress, Filippenko, Ganeshalingam and Schmidt2014b), we require the photospheric velocity component to have a velocity centre within 20% of the value derived for the Si 6355 Å feature, though in most cases the results are the same if this requirement is removed.

In Figure 10, we plot the measured HVF strength (R HVF) versus the silicon absorption strength ratio R Si. For illustrative purposes, we also show the scale of light curve decline rate Δm 15 corresponding to the plotted range of R Si using the above relation. A representative error bar for Δm 15 from the above relation and a typical R HVF error-bar are shown in the figure.

Figure 10. Strength of the high-velocity features (HVFs) in the Ca II NIR triplet—using the quantity R HVF as defined by Childress et al. (Reference Childress, Filippenko, Ganeshalingam and Schmidt2014b)—plotted against the Si II absorption strength ratio R Si defined by Nugent et al. (Reference Nugent, Phillips, Baron, Branch and Hauschildt1995). On the top axis, we show the rough equivalent light curve decline rate Δm 15 values corresponding to the range of R Si values. The crosshair in the upper right represents the characteristic errors in measurement of R HVF (typically 20%, plotted here for the larger values of R HVF) and the error in Δm 15 when converted from R Si.

Figure 10 demonstrates a tendency for SNe Ia with strong HVFs to also have low values of R Si and thus broad light curves (low Δm 15). This correlation of HVFs with SN Ia light curve width was first observed by Maguire et al. (Reference Maguire2012) in composite high-redshift SN Ia spectra, and subsequently confirmed by numerous studies of low-redshift SNe Ia (Childress et al. Reference Childress, Filippenko, Ganeshalingam and Schmidt2014b; Maguire et al. Reference Maguire2014; Pan et al. Reference Pan, Sullivan, Maguire, Gal-Yam, Hook, Howell, Nugent and Mazzali2015; Silverman et al. Reference Silverman, Vinkó, Marion, Wheeler, Barna, Szalai, Mulligan and Filippenko2015; Zhao et al. Reference Zhao2015). Our results support these studies with spectra alone.

4.2. Narrow sodium absorption features in SN Ia spectra

A great advantage of the higher resolution of WiFeS (compared to many spectrographs deployed for other SN spectroscopy surveys) is the ability to detect narrow absorption features, particularly the Na I doublet at λλ5890/5896 Å. This feature has a long history of being used to infer the presence of foreground dust in SNe Ia (Barbon et al. Reference Barbon, Benetti, Rosino, Cappellaro and Turatto1990; Turatto, Benetti, & Cappellaro Reference Turatto, Benetti, Cappellaro, Hillebrandt and Leibundgut2003; Poznanski et al. Reference Poznanski, Ganeshalingam, Silverman and Filippenko2011; Poznanski, Prochaska, & Bloom Reference Poznanski, Prochaska and Bloom2012; Phillips et al. Reference Phillips2013). Earlier works attempted to derive correlations between SN Ia colours and sodium absorption strength (i.e. the absorption equivalent width), but Poznanski et al. (Reference Poznanski, Ganeshalingam, Silverman and Filippenko2011) showed sodium to be a poor indicator of SN Ia reddening. Phillips et al. (Reference Phillips2013) refined this result by showing that SN Ia reddening exhibits a strong correlation with absorption strength of the diffuse interstellar bands (found exclusively in the interstellar medium) but some SNe Ia have an excess of sodium absorption that does not coincide with associated reddening of the SN. Thus, sodium features in SNe Ia remain instructive but should be considered with a measure of caution.

Recently, velocities of sodium absorption features in SN Ia spectra have been used as a diagnostic of CSM, including cases of some SNe Ia (Patat et al. Reference Patat2007; Simon et al. Reference Simon2009) where the sodium absorption features exhibit variability (though most do not—see Sternberg et al. Reference Sternberg2014), indicating SN-CSM interaction. More recently, a statistical analysis of the velocity distribution of sodium features in SNe Ia by Sternberg et al. (Reference Sternberg2011) found an excess of SNe Ia with blueshifted sodium features, indicating that a fraction of SNe Ia explode inside an expanding shell of material that was presumably shed by the SN Ia progenitor system prior to explosion. Maguire et al. (Reference Maguire2013) extended this work to show that the excess of SNe Ia with blueshifted sodium absorption was populated predominantly by the more luminous SNe Ia with slow declining light curves (i.e. high ‘stretch’).

We thus searched for the presence of sodium absorption features in our sample of SN Ia spectra. For most SNe Ia (13), this was done with the lower resolution (R3000) observations of the SN at maximum light. For six SNe Ia, we also had a higher resolution (R7000) observation of the SN at maximum light. In a few instances, this observation was triggered by the presence of strong reddening being reported in the SN classification announcement. For the other instances, we obtained both a low-resolution (R3000) and high-resolution (R7000) spectrum in the red whilst obtaining a longer exposure blue (B3000) spectrum during nights near full moon when the blue sky background was exceptionally high.

We fitted the sodium doublet absorption profile as follows. First, the spectrum was normalised to the local continuum by fitting a quadratic to the SN flux between 10–25 Å redward or blueward of the doublet centre (5893 Å). The absorption profile was fitted as two Gaussians with rest wavelengths set by the doublet wavelengths but with unknown (common) velocity shift and velocity width and (independent) absorption depths. This was done with an mpfit routine in Python, which accounts for variable covariances and returns the appropriate fit values and uncertainties. We show representative examples for fits to data from both gratings in Figure 11. The outcomes for our sodium profile fits are presented in Table A4 in Appendix A, and comprise six successful fits for targets with R7000 spectra, and 13 for R3000 spectra.

Figure 11. Sodium absorption fit examples for both the R7000 (top) and R3000 (bottom) gratings. Data (which have been normalised to the local continuum fit) are shown as blue diamonds whilst the best fit absorption profile is shown as the solid red curve. For reference, we also mark the continuum level (horizontal black line at value 1.00) and the rest wavelengths of the sodium doublet (vertical dotted gray lines).

The primary quantity we wanted to measure from the sodium absorption feature was its velocity with respect to the local standard of rest at the SN site. The default local rest velocity was initially set to be the systemic velocity of the SN host galaxy. In some cases, we detected clear nebular emission lines at the SN site present in the SN spectrum—for these cases, the local rest velocity (i.e. the rotational velocity of the host galaxy at the site of the SN) was measured from the Hα emission line.

In two cases (SN 2014ao and ASASSN-14jg), no local Hα emission was present in the SN spectrum, and the SN sodium velocity differed from the host systemic velocity by more than 100 km s−1. To obtain the true local rest velocities for these two SNe Ia, we took advantage of the integral-field data provided by WiFeS, which allows us to measure host galaxy properties (such as velocity) over a broad field of view.

We illustrate this for SN 2014ao in Figure 12: The WiFeS field of view extends from the host galaxy core to the outer edge of its spiral arms in the direction of the SN. We were able to extract the velocity of an H ii region along the SN-host axis and found its velocity differed significantly from that of the host core (Δv = 174 ± 10 km s−1). As galaxy velocity curves tend to flatten at large radii, we use the H ii region velocity as the local rest velocity of SN 2014ao—a value much closer to the measured sodium absorption velocity.

Figure 12. Determination of the local velocity for SN 2014ao in NGC 2615 with WiFeS. Left: SDSS (York et al. Reference York2000) gri colour composite—created with SWARP (Bertin et al. Reference Bertin, Mellier, Radovich, Missonnier, Didelon, Morin, Bohlender, Durand and Handley2002) and STIFF (Bertin Reference Bertin, Ballester, Egret and Lorente2012)—with the WiFeS field of view (red rectangle) and SN location (red dot) highlighted. Middle: Image of the SN 2014ao WiFeS data cube in the isolated wavelength range within ± 6 Å (i.e. ± 300 km s−1) of the wavelength of Hα at the published redshift of NGC 2615, with host core (purple square) and nearby H ii region (brown square) highlighted—the SN is the bright object near the centre. Right: Extracted WiFeS spectra of the nearby H ii region (top) and host core (bottom) near the Hα+NII emission line group, with the expected location of those lines at the published redshift of NGC 2615 (z = 0.014083, Theureau et al. Reference Theureau, Bottinelli, Coudreau-Durand, Gouguenheim, Hallet, Loulergue, Paturel and Teerikorpi1998a) shown as the vertical gray lines.

With the final sodium absorption velocities for our sample, we can inspect the relationship between sodium velocity and other spectroscopic properties of our SN Ia sample. In Figure 13, we plot the silicon absorption ratio (R Si), the HVF absorption strength (R HVF), and the absorption strength (i.e. equivalent width) of sodium itself against the velocity of the sodium absorption feature. Based on results of Maguire et al. (Reference Maguire2013), we would expect SNe Ia with low R Si and high R HVF values (the high stretch, slow declining SNe Ia) to have a slight excess of blueshifted sodium absorption. Our sample size here is too small (and not well-selected) to make any robust statement about such preferences. However, we note that this analysis was not the explicit objective of our observations, but instead was a supplemental outcome facilitated by the nature of the WiFeS data.

Figure 13. Top: Silicon absorption ratio R Si plotted against velocity centre of the narrow sodium absorption feature (as in Figure 10 we show the corresponding values of Δm 15, though note the smaller range). Middle: HVF strength (R HVF) plotted against sodium absorption velocity. Bottom: Absorption equivalent width of the combined D1+D2 sodium lines plotted against sodium absorption velocity. On the right axis of this panel, we use the relation of Poznanski et al. (Reference Poznanski, Prochaska and Bloom2012) to show the reddening values E(BV) corresponding to the measured sodium equivalent widths if the absorption arises solely from the ISM—though this is unlikely to be true for all SNe Ia (Poznanski et al. Reference Poznanski, Ganeshalingam, Silverman and Filippenko2011; Phillips et al. Reference Phillips2013—see discussion in text). In all panels, higher resolution observations with the R7000 grating are displayed as green squares, whilst lower resolution R3000 observations are shown as blue circles.

Thus, WiFeS is an excellent instrument for measuring sodium absorption in SNe Ia, a key observable for investigating SN Ia progenitor systems. This is particularly true for observations taken with the R7000 grating, which provides sodium velocity uncertainties of order a few km s−1, thus enabling a robust classification of the SN as being ‘blueshifted’ or ‘redshifted’. More importantly, perhaps, is the capability of WiFeS to observe a wide field of view around the SN. This enables a measurement of the local systemic velocity at the SN location, even in cases where emission from the host galaxy is weak at the SN location itself.

4.3. Spectroscopic evolution of SN 2012dn

The SN with the greatest number of spectroscopic epochs in AWSNAP is SN 2012dn, and we show its AWSNAP spectroscopic time series in Figure 14. SN 2012dn was a spectroscopically peculiar SN Ia whose photometric and spectroscopic evolution was studied extensively by Chakradhari et al. (Reference Chakradhari, Sahu, Srivastav and Anupama2014)—they found it to be similar to the ‘candidate super-Chandrasekhar’ SN Ia SN 2006gz (Hicken et al. Reference Hicken, Garnavich, Prieto, Blondin, DePoy, Kirshner and Parrent2007). For brevity, we will refer to this class of objects as ‘super-Chandra’, though we caution the reader that the origin of the extreme luminosity of these objects is still under debate (see, e.g. Taubenberger et al. Reference Taubenberger2011; Hachinger et al. Reference Hachinger, Mazzali, Taubenberger, Fink, Pakmor, Hillebrandt and Seitenzahl2012). We comment on only a few additional outcomes for SN 2012dn from the AWSNAP data, but refer readers to Chakradhari et al. (Reference Chakradhari, Sahu, Srivastav and Anupama2014) and Parrent et al. (Reference Parrent2016) for a thorough discussion of this interesting object.

Figure 14. AWSNAP time series of SN 2012dn, labelled by phase with respect to the date of maximum light (2012 July 24, as determined by Chakradhari et al. Reference Chakradhari, Sahu, Srivastav and Anupama2014). Note these observations come from the first semester of AWSNAP when observing time was allocated in multi-night blocks separated sometimes by a month or more.

We obtained a very high signal-to-noise spectrum of SN 2012dn at phase +91 d (with respect to the date of maximum light 2012 July 24, as determined by Chakradhari et al. Reference Chakradhari, Sahu, Srivastav and Anupama2014). At this epoch, the SN is beginning to enter the nebular phase when the ejecta become optically thin, revealing emission from the iron group elements (IGEs) near the centre of the SN. Spectra at these epochs provide an excellent diagnostic of the nucleosynthetic products of the SN explosion. In Figure 15, we present our +91 d spectrum of SN 2012dn compared to very late spectra of other candidate super-Chandra SNe Ia SN 2007if (Scalzo et al. Reference Scalzo2010; Yuan et al. Reference Yuan2010; Taubenberger et al. Reference Taubenberger2013) and SN 2009dc (Silverman et al. Reference Silverman, Ganeshalingam, Li, Filippenko, Miller and Poznanski2011; Taubenberger et al. Reference Taubenberger2011; Yamanaka et al. Reference Yamanaka2009; Tanaka et al. Reference Tanaka2010; Hachinger et al. Reference Hachinger, Mazzali, Taubenberger, Fink, Pakmor, Hillebrandt and Seitenzahl2012; Kamiya et al. Reference Kamiya, Tanaka, Nomoto, Blinnikov, Sorokina and Suzuki2012; Taubenberger et al. Reference Taubenberger2013), as well as the gold standard normal SN Ia SN 2011fe (Nugent et al. Reference Nugent2011; Li et al. Reference Li2011b; Parrent et al. Reference Parrent2012; Pereira et al. Reference Pereira2013).

Figure 15. SN 2012dn at its latest AWSNAP epoch (+91 d on 2012 October 23) compared to other candidate super-Chandra SNe Ia SN 2007if at +98 d (top panel, from Silverman et al. Reference Silverman, Ganeshalingam, Li, Filippenko, Miller and Poznanski2011) and SN 2009dc at +97 d (middle panel, from Taubenberger et al. Reference Taubenberger2011), as well as the normal SN 2011fe (bottom panel, from Pereira et al. Reference Pereira2013).

This comparison clearly reveals a strong spectroscopic similarity between SN 2012dn and the candidate super-Chandrasekhar SNe Ia, and a distinct dissimilarity with SN 2011fe. Perhaps, most prominent is the weaker Fe III line complex at ~ 4700 Å for the super-Chandra SNe Ia compared to SN 2011fe. This discrepancy is also evident in fully nebular spectra of super-Chandra SNe Ia at ~ 1 yr past maximum light, as discussed by Taubenberger et al. (Reference Taubenberger2013). This indicates that the ionisation state of the super-Chandra SNe Ia at these phases is different from normal SNe Ia—whether the diminished Fe iii emission arises from a higher or lower average ionisation state remains uncertain.

Additionally, the velocity profile of the emission features in the super-Chandra SNe Ia exhibits a marked difference to that of SN 2011fe (and other normal SNe Ia). The normal SN Ia profile appears very Gaussian (and indeed is generally well fit by a Gaussian profile—Childress et al. Reference Childress2015), whilst the super-Chandra velocity profile appears sharper at the centre. This is particularly evident for the line features at ~ 5 900 and ~ 6 300 Å, which at these epochs are dominated by Co iii. Further spectral modelling of this and other late-phase super-Chandra spectra may reveal important insights into the structure and composition of the super-Chandra ejecta.

Finally, the high resolution of WiFeS reveals narrow emission lines from the host galaxy of SN 2012dn. We isolated the narrow host galaxy lines using the longest exposure of SN 2012dn, the late-phase observation of 2012 October 23 (+91 d). We fit a simple linear ‘continuum’ near each emission line by fitting a line to the SN spectrum between 5 and 15 Å away from the line centre on both the blue and red sides of the line. We illustrate this technique and present the SN-subtracted emission features in Figure 16.

Figure 16. Extraction of host galaxy emission line flux (green) from late SN 2012dn spectrum (blue) using simple linear continuum fits (red).

In Table 3, we present the measured emission fluxes and errors for the major galaxy emission lines, all of which have been scaled by the observed flux in the Hβ line. By comparing the observed ratio of the Hα and Hβ lines to its expected value of 2.87 (the Balmer decrement Osterbrock & Ferland Reference Osterbrock and Ferland2006), we can determine the amount of reddening in the H ii regions giving rise to the host emission lines. We use the Cardelli, Clayton, & Mathis (Reference Cardelli, Clayton and Mathis1989) to find a host reddening of E(BV) = 0.17 ± 0.10, a value remarkably similar to the SN reddening of E(BV) = 0.18 (Milky Way plus host) determined by Chakradhari et al. (Reference Chakradhari, Sahu, Srivastav and Anupama2014). We correct the host emission line fluxes for the value E(BV) = 0.17 and report the corrected values (which we also re-scale to the de-reddened Hβ flux) in Table 3.

Table 3. SN 2012dn local emission line fluxes.

a Observer frame fluxes, scaled to Hβ.

b De-reddened using Balmer decrement reddening of E(BV) = 0.17 so that F(Hβ) = F(Hα)/2.87 with a CCM reddening law, and scaled to the de-reddened flux of Hβ.

With the de-reddened host galaxy emission line fluxes, we calculate a gas-phase metallicity at the site of SN 2012dn. The N2 method of Pettini & Pagel (Reference Pettini and Pagel2004, hereafter PP04) yields 12 + log(O/H) = 8.29 ± 0.04, whilst the O3N2 method of PP04 yields 12 + log(O/H) = 8.35 ± 0.03. The former value is remarkably close to the estimated site metallicity for SN 2006gz of 12 + log(O/H) = 8.26 calculated by Khan et al. (Reference Khan, Stanek, Stoll and Prieto2011) using the same metallicity method. If we convert the O3N2 value to the Tremonti et al. (Reference Tremonti2004) scale using the formulae of Kewley & Ellison (Reference Kewley and Ellison2008) as was done in Childress et al. (Reference Childress2011), we measure 12 + log(O/H) T04 = 8.51 ± 0.04.

Comparing the above values to the solar oxygen abundance of 12 + log(O/H) = 8.69 (Asplund et al. Reference Asplund, Grevesse, Sauval and Scott2009), we find the site metallicity for SN 2012dn is in the range 40–65% solar, depending on the chosen metallicity calibration. This sub-solar metallicity value is consistent with the previously reported trend for super-Chandra SNe Ia to prefer low metallicity environments (Taubenberger et al. Reference Taubenberger2011; Childress et al. Reference Childress2011; Khan et al. Reference Khan, Stanek, Stoll and Prieto2011).

5 CONCLUSIONS

This work marks the primary data release for the AWSNAP, comprising 357 distinct spectra of 175 unique SNe. These data were collected using the WiFeS on the ANU 2.3-m telescope during 82 nights of observing over a 3-yr period from mid-2012 to mid-2015.

The AWSNAP spectroscopy sample is comparable in size to other SN spectra data releases, and its composition of SN types is roughly in line with expectations for a magnitude-limited SN search. The phase coverage of the AWSNAP SN Ia sample is comparable to other published SN Ia spectroscopy datasets (for SNe Ia with multiple epochs of observation), with the inclusion of more SNe Ia with a single observation (i.e. classification spectra only).

We presented some analyses of the AWSNAP SN Ia sample, including some results uniquely enabled by the fine wavelength resolution available with WiFeS. We measured broad absorption features in SN Ia spectra at maximum light, including the ratio of silicon absorption features R Si and the strength of HVFs R HVF. Additionally, we measured the strength and velocity of narrow sodium absorption features, including some cases where the integral-field nature of the instrument allowed us to measure the local systemic velocity within the SN host galaxy. Some expected feature trends, such as a correlation between R HVF and R Si, were recovered in our data set. The nature of sodium absorption in our sample was limited by small number statistics. Finally, we presented our observations of the candidate super-Chandrasekhar SN Ia SN 2012dn, and used narrow host galaxy emission features to show the SN site exhibits sub-solar metallicity.

The WiFeS instrument presents several unique advantages for the study of transients, particularly owing to its comparatively narrow velocity resolution (σv ~ 45 km s−1). It has previously been employed in the study of SNe with strong narrow emission features such as SN 2009ip (Fraser et al. Reference Fraser2013, Reference Fraser2015), SN 2012ca (Inserra et al. Reference Inserra2014, Reference Inserra2016), and SN 2013fc (Kangas et al. Reference Kangas2016). Here, we also demonstrated that the fine velocity resolution allows for the measurement of narrow absorption features, particularly sodium absorption in SNe Ia. The higher resolution of WiFeS also frequently revealed narrow host galaxy emission features at the site of the SN, which can at times be used to determine a SN site metallicity (as we showed for SN 2012dn). Finally, the integral field nature of WiFeS allowed us to measure the local host galaxy rotational velocity at the site of several SNe, even when there was no emission directly at the SN location. Thus, WiFeS is an instrument capable of not only standard SN spectroscopic observations, but also a unique suite of capabilities not commonly found in transient follow-up instruments.

ACKNOWLEDGEMENTS

We are very grateful for the excellent technical support staff for the ANU 2.3-m telescope and the WiFeS instrument: Peter Verwayen, Ian Adams, Peter Small, Ian Price, Peter Young, and Jon Nielsen. We thank the ANU telescope time allocation committee who continue to support the observations presented herein. We also thank Julie Banfield, Michael Ireland, Stefan Keller, Lisa Kewley, Jeremy Mould, Chris Owen, Aaron Rizzuto, Dary Ruiz, and Tian-Tian Yuan for additional observations. We also thank the anonymous referee for very helpful comments.

This research was conducted by the Australian Research Council Centre of Excellence for All-sky Astrophysics (CAASTRO), through project number CE110001020. IRS was supported by Australian Research Council Laureate Grant FL0992131. 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 the National Aeronautics and Space Administration. This research has made use of NASA’s Astrophysics Data System (ADS).

APPENDIX A1: TARGET AND OBSERVATION DETAILS

In this Appendix, we present the observational metadata for the AWSNAP data release, as well as sodium fit results and host (or local) redshifts used in the sodium fitting.

AWSNAP targets: Table A1 presents the discovery and classification details for the full list of 175 objects featured in the AWSNAP DR1 data release. We list the discovery location and date, as well as discovery reference and group that discovered the transient. Note the discovery references are typically either Astronomer’s Telegrams (ATel’s, denoted as, e.g. ‘A1234’) or telegrams from the Central Bureau for Electronic Telegrams (CBET’s, denoted as, e.g. ‘C1234’), though occasionally a separate discovery notice was not issued—in these cases, typically the webpage of the discovering group was cited in the classification reference. Classification references are given with similar notation, listed along with classification date, quoted supernova type and phase from the classification, as well as the redshift listed in the classification.

Table A1. All AWSNAP Targets.

In Table A1, the discovery and classification groups are listed either as acronyms of the group name, or listed as the first author in the discovery/classification notice. The acronyms used in the table for discovery groups are as follows:

The acronyms used in the table for classification groups are as follows:

Some transients in Table A1 are known by other aliases or are listed in shorthand form (due to a lengthy full name). These transients and their aliases (or full names) are noted in Table A2.

Table A2. Alternate and/or full designations for SNe in the AWSNAP sample.

Figure A1. Literature data used to fit the trend of Δm 15 versus R Si. The best fit is the thick solid line, and the thin dashed line represent the trend ± 1σ (where σ is the dispersion of the data about the trend).

AWSNAP spectra: The full list of AWSNAP spectra released here is given in Table A3, along with pertinent observation details such as grating and exposure time.

SN Ia sodium fits: In Table A4, we present the best fit parameters (and uncertainties) for our fits to narrow sodium absorption in our sample of SN Ia maximum light spectra. Table A5 lists the redshifts used to establish the local rest velocity of the SN—typically this was the redshift of the host galaxy but occasionally is obtained from galaxy emission features at the location of the SN (denoted as ‘Local’).

APPENDIX A2: DERIVATION OF THE SN Ia R Si–Δm 15 RELATION FROM LITERATURE DATA

Our fit to the relation between Δm 15 and R Si is presented here. Our fit to the data is shown in Figure A1, and from these data we derive a best fit relation of

Table A3. All AWSNAP spectra.

Table A4. Fits of Na line in SNe Ia.

Table A5. Host galaxy redshifts for SN Ia Na sample.

(A1) $$\begin{equation} \Delta m_{15} = 2.04 R_{{\rm Si}} + 0.808. \end{equation}$$

The scatter about this best fit curve is 0.190 mag.

Footnotes

2 Based on the spectroscopically estimated phase reported at the time of classification.

a Notation: A1234 ≡ ATel # 1234; C1234 ≡ CBET # 1234; CTOCP ≡ CBET Transient Object Confirmation Page; objects newly classified here denoted as (NEW); all others refer to transient announcement webpages for stated discovery group(s).

b Discovery group acronyms (where appropriate—see text for details) or lead author of the discovery alert.

c Classifcation group acronyms (where appropriate—see text for details) or lead author of the classification alert.

d Phase with respect to maximum light (where appropriate) for the best spectroscopic match, as reported by the classifying group. SNe IIn do not typically have a clearly identifiable spectroscopic phase so are listed as ‘(unk)’ (i.e. unknown). Very young SNe II (with very blue featureless continua) similarly have ambiguous phases so are listed as ‘(young)’.

a Via NED. ‘Local’ denotes local velocity measured with WiFeS, either from narrow lines in the SN spectrum or from galaxy rotation characteristics (see text for details).

References

REFERENCES

Aldering, G., et al. 2002, in SPIE Conf. Ser., Vol. 4836, Survey and Other Telescope Technologies and Discoveries, eds. Tyson, J. A. & Wolff, S. (Waikoloa: SPIE), 61 http://dx.doi.org/10.1117/12.458107, doi:10.1117/12.458107 Google Scholar
Asplund, M., Grevesse, N., Sauval, A. J., & Scott, P. 2009, ARA&A, 47, 481 10.1146/annurev.astro.46.060407.145222 2009ARA&A..47..481A Google Scholar
Bailey, S., et al. 2009, A&A, 500, L17 10.1051/0004-6361/200911973 2009A&A...500L..17B Google Scholar
Baltay, C., et al. 2013, PASP, 125, 683 10.1086/671198 2013PASP..125..683B CrossRefGoogle Scholar
Barbon, R., Benetti, S., Rosino, L., Cappellaro, E., & Turatto, M. 1990, A&A, 237, 79 1990A&A...237...79B Google Scholar
Benetti, S., et al. 2005, ApJ, 623, 1011 10.1086/428608 2005ApJ...623.1011B CrossRefGoogle Scholar
Bertin, E. 2012, in ASP Conf. Ser., Vol. 461, Astronomical Data Analysis Software and Systems XXI, eds. Ballester, P., Egret, D., & Lorente, N. P. F. (San Francisco: ASP), 263 Google Scholar
Bertin, E., Mellier, Y., Radovich, M., Missonnier, G., Didelon, P., & Morin, B. 2002, in ASP Conf. Ser., Vol. 281, Astronomical Data Analysis Software and Systems XI, eds. Bohlender, D. A., Durand, D., & Handley, T. H. (San Francisco: ASP), 228 Google Scholar
Bessell, M. S. 1999, PASP, 111, 1426 10.1086/316454 1999PASP..111.1426B CrossRefGoogle Scholar
Bianco, F. B., et al. 2014, ApJS, 213, 19 10.1088/0067-0049/213/2/19 2014ApJS..213...19B CrossRefGoogle Scholar
Blagorodnova, N., Koposov, S. E., Wyrzykowski, Ł., Irwin, M., & Walton, N. A. 2014, MNRAS, 442, 327 10.1093/mnras/stu837 2014MNRAS.442..327B CrossRefGoogle Scholar
Blagorodnova, N., Van Velzen, S., Harrison, D. L., Koposov, S., Mattila, S., Campbell, H., Walton, N. A., & Wyrzykowski, Ł. 2016, MNRAS, 455, 603 10.1093/mnras/stv2308 2016MNRAS.455..603B CrossRefGoogle Scholar
Blondin, S., Mandel, K. S., & Kirshner, R. P. 2011, A&A, 526, A81 10.1051/0004-6361/201015792 2011A&A...526A..81B Google Scholar
Blondin, S., & Tonry, J. L. 2007, ApJ, 666, 1024 10.1086/520494 2007ApJ...666.1024B CrossRefGoogle Scholar
Blondin, S., et al. 2012, AJ, 143, 126 10.1088/0004-6256/143/5/126 2012AJ....143..126B CrossRefGoogle Scholar
Bongard, S., Baron, E., Smadja, G., Branch, D., & Hauschildt, P. H. 2006, ApJ, 647, 513 10.1086/505322 2006ApJ...647..513B CrossRefGoogle Scholar
Bongard, S., Baron, E., Smadja, G., Branch, D., & Hauschildt, P. H. 2008, ApJ, 687, 456 10.1086/590107 2008ApJ...687..456B CrossRefGoogle Scholar
Branch, D., Dang, L. C., & Baron, E. 2009, PASP, 121, 238 10.1086/597788 2009PASP..121..238B CrossRefGoogle Scholar
Branch, D., et al. 2006, PASP, 118, 560 10.1086/502778 2006PASP..118..560B CrossRefGoogle Scholar
Bronder, T. J., et al. 2008, A&A, 477, 717 10.1051/0004-6361:20077655 2008A&A...477..717B Google Scholar
Brown, T. M., et al. 2013, PASP, 125, 1031 10.1086/673168 2013PASP..125.1031B CrossRefGoogle Scholar
Cardelli, J. A., Clayton, G. C., & Mathis, J. S. 1989, ApJ, 345, 245 10.1086/167900 1989ApJ...345..245C CrossRefGoogle Scholar
Chakradhari, N. K., Sahu, D. K., Srivastav, S., & Anupama, G. C. 2014, MNRAS, 443, 1663 10.1093/mnras/stu1258 2014MNRAS.443.1663C CrossRefGoogle Scholar
Childress, M., et al. 2011, ApJ, 733, 3 10.1088/0004-637X/733/1/3 2011ApJ...733....3C CrossRefGoogle Scholar
Childress, M. J., Filippenko, A. V., Ganeshalingam, M., & Schmidt, B. P. 2014b, MNRAS, 437, 338 10.1093/mnras/stt1892 2014MNRAS.437..338C CrossRefGoogle Scholar
Childress, M. J., Vogt, F. P. A., Nielsen, J., & Sharp, R. G. 2014a, Ap&SS, 349, 617 10.1007/s10509-013-1682-0 2014Ap&SS.349..617C Google Scholar
Childress, M. J., et al. 2015, MNRAS, 454, 3816 10.1093/mnras/stv2173 2015MNRAS.454.3816C CrossRefGoogle Scholar
Cortese, L., et al. 2008, MNRAS, 383, 1519 10.1111/j.1365-2966.2007.12664.x 2008MNRAS.383.1519C CrossRefGoogle Scholar
da Costa, L. N., Pellegrini, P. S., Davis, M., Meiksin, A., Sargent, W. L. W., & Tonry, J. L. 1991, ApJS, 75, 935 10.1086/191555 1991ApJS...75..935D CrossRefGoogle Scholar
Donzelli, C. J., & Pastoriza, M. G. 2000, AJ, 120, 189 10.1086/301420 2000AJ....120..189D CrossRefGoogle Scholar
Dopita, M., Hart, J., McGregor, P., Oates, P., Bloxham, G., & Jones, D. 2007, Ap&SS, 310, 255 10.1007/s10509-007-9510-z 2007Ap&SS.310..255D Google Scholar
Dopita, M., et al. 2010, Ap&SS, 327, 245 10.1007/s10509-010-0335-9 2010Ap&SS.327..245D Google Scholar
Drake, A. J., et al. 2009, ApJ, 696, 870 10.1088/0004-637X/696/1/870 2009ApJ...696..870D CrossRefGoogle Scholar
Filippenko, A. V. 1997, ARA&A, 35, 309 10.1146/annurev.astro.35.1.309 1997ARA&A..35..309F Google Scholar
Filippenko, A. V., et al. 1992a, AJ, 104, 1543 10.1086/116339 1992AJ....104.1543F CrossRefGoogle Scholar
Filippenko, A. V., et al. 1992b, ApJL, 384, L15 10.1086/186252 1992ApJ...384L..15F CrossRefGoogle Scholar
Folatelli, G., et al. 2013, ApJ, 773, 53 10.1088/0004-637X/773/1/53 2013ApJ...773...53F CrossRefGoogle Scholar
Foley, R. J., et al. 2013, ApJ, 767, 57 10.1088/0004-637X/767/1/57 2013ApJ...767...57F CrossRefGoogle Scholar
Fraser, M., et al. 2013, MNRAS, 433, 1312 10.1093/mnras/stt813 2013MNRAS.433.1312F CrossRefGoogle Scholar
Fraser, M., et al. 2015, MNRAS, 453, 3886 10.1093/mnras/stv1919 2015MNRAS.453.3886F CrossRefGoogle Scholar
Ganeshalingam, M., et al. 2010, ApJS, 190, 418 10.1088/0067-0049/190/2/418 2010ApJS..190..418G CrossRefGoogle Scholar
Graur, O., Bianco, F. B., Modjaz, M., Maoz, D., Shivvers, I., Filippenko, A. V., & Li, W. 2015, preprint (arXiv:1509.08432)2015arXiv150908432G Google Scholar
Grogin, N. A., Geller, M. J., & Huchra, J. P. 1998, ApJS, 119, 277 10.1086/313164 1998ApJS..119..277G CrossRefGoogle Scholar
Hachinger, S., Mazzali, P. A., & Benetti, S. 2006, MNRAS, 370, 299 10.1111/j.1365-2966.2006.10468.x 2006MNRAS.370..299H CrossRefGoogle Scholar
Hachinger, S., Mazzali, P. A., Tanaka, M., Hillebrandt, W., & Benetti, S. 2008, MNRAS, 389, 1087 10.1111/j.1365-2966.2008.13645.x 2008MNRAS.389.1087H CrossRefGoogle Scholar
Hachinger, S., Mazzali, P. A., Taubenberger, S., Fink, M., Pakmor, R., Hillebrandt, W., & Seitenzahl, I. R. 2012, MNRAS, 427, 2057 10.1111/j.1365-2966.2012.22068.x 2012MNRAS.427.2057H CrossRefGoogle Scholar
Hatano, K., Branch, D., Lentz, E. J., Baron, E., Filippenko, A. V., & Garnavich, P. M. 2000, ApJL, 543, L49 10.1086/318169 2000ApJ...543L..49H CrossRefGoogle Scholar
Hicken, M., Garnavich, P. M., Prieto, J. L., Blondin, S., DePoy, D. L., Kirshner, R. P., & Parrent, J. 2007, ApJL, 669, L17 10.1086/523301 2007ApJ...669L..17H CrossRefGoogle Scholar
Holoien, T. W.-S., et al. 2016, MNRAS 10.1093/mnras/stw2273 2016MNRAS.tmp.1390H Google Scholar
Inserra, C., et al. 2013, ApJ, 770, 128 10.1088/0004-637X/770/2/128 2013ApJ...770..128I CrossRefGoogle Scholar
Inserra, C., et al. 2014, MNRAS, 437, L51 10.1093/mnrasl/slt138 2014MNRAS.437L..51I CrossRefGoogle Scholar
Inserra, C., et al. 2016, MNRAS, 459, 2721 10.1093/mnras/stw825 2016MNRAS.459.2721I CrossRefGoogle Scholar
Kaiser, N., et al. 2010, in Ground-based and Airborne Telescopes III, SPIE Proc. 7733, eds. Stepp, L.M., Gilmozzi, R., & Hall, H.J. (San Diego: SPIE), 77330E CrossRefGoogle Scholar
Kamiya, Y., Tanaka, M., Nomoto, K., Blinnikov, S. I., Sorokina, E. I., & Suzuki, T. 2012, ApJ, 756, 191 10.1088/0004-637X/756/2/191 2012ApJ...756..191K CrossRefGoogle Scholar
Kangas, T., et al. 2016, MNRAS, 456, 323 10.1093/mnras/stv2567 2016MNRAS.456..323K CrossRefGoogle Scholar
Keller, S. C., et al. 2007, PASA, 24, 1 10.1071/AS07001 2007PASA...24....1K CrossRefGoogle Scholar
Kelson, D. D. 2003, PASP, 115, 688 10.1086/375502 2003PASP..115..688K CrossRefGoogle Scholar
Kewley, L. J., & Ellison, S. L. 2008, ApJ, 681, 1183 10.1086/587500 2008ApJ...681.1183K CrossRefGoogle Scholar
Khan, R., Stanek, K. Z., Stoll, R., & Prieto, J. L. 2011, ApJL, 737, 24 10.1088/2041-8205/737/1/L24 CrossRefGoogle Scholar
Klotz, A., Boer, M., Atteia, J.-L., Gendre, B., Le Borgne, J.-F., Frappa, E., Vachier, F., & Berthier, J. 2013, Msngr, 151, 6 2013Msngr.151....6K Google Scholar
Law, N. M., et al. 2009, PASP, 121, 1395 10.1086/648598 2009PASP..121.1395L CrossRefGoogle Scholar
Li, W., et al. 2011a, MNRAS, 412, 1441 10.1111/j.1365-2966.2011.18160.x 2011MNRAS.412.1441L CrossRefGoogle Scholar
Li, W., et al. 2011b, Nature, 480, 348 10.1038/nature10646 2011Natur.480..348L CrossRefGoogle Scholar
Lipunov, V. M., et al. 2004, AN, 325, 580 10.1002/asna.200410284 2004AN....325..580L Google Scholar
Liu, Y.-Q., Modjaz, M., Bianco, F. B., & Graur, O. 2016, ApJ, 827, 90 10.3847/0004-637X/827/2/90 CrossRefGoogle Scholar
Maguire, K., et al. 2012, MNRAS, 426, 2359 10.1111/j.1365-2966.2012.21909.x 2012MNRAS.426.2359M CrossRefGoogle Scholar
Maguire, K., et al. 2013, MNRAS, 436, 222 10.1093/mnras/stt1586 2013MNRAS.436..222M CrossRefGoogle Scholar
Maguire, K., et al. 2014, MNRAS, 444, 3258 10.1093/mnras/stu1607 2014MNRAS.444.3258M CrossRefGoogle Scholar
Matheson, T., et al. 2008, AJ, 135, 1598 10.1088/0004-6256/135/4/1598 2008AJ....135.1598M CrossRefGoogle Scholar
Modjaz, M., et al. 2014, AJ, 147, 99 10.1088/0004-6256/147/5/99 2014AJ....147...99M CrossRefGoogle Scholar
Nugent, P., Phillips, M., Baron, E., Branch, D., & Hauschildt, P. 1995, ApJL, 455, L147 10.1086/309846 1995ApJ...455L.147N CrossRefGoogle Scholar
Nugent, P. E., et al. 2011, Nature, 480, 344 10.1038/nature10644 2011Natur.480..344N CrossRefGoogle Scholar
Ogando, R. L. C., Maia, M. A. G., Pellegrini, P. S., & da Costa, L. N. 2008, AJ, 135, 2424 10.1088/0004-6256/135/6/2424 2008AJ....135.2424O CrossRefGoogle Scholar
Oke, J. B. 1990, AJ, 99, 1621 10.1086/115444 1990AJ.....99.1621O CrossRefGoogle Scholar
Osterbrock, D. E., & Ferland, G. J. 2006, Astrophysics of gaseous nebulae and active galactic nuclei (Sausalito: University Science Books)Google Scholar
Pan, Y.-C., Sullivan, M., Maguire, K., Gal-Yam, A., Hook, I. M., Howell, D. A., Nugent, P. E., & Mazzali, P. A. 2015, MNRAS, 446, 354 10.1093/mnras/stu2121 2015MNRAS.446..354P CrossRefGoogle Scholar
Parrent, J. T., et al. 2012, ApJL, 752, L26 10.1088/2041-8205/752/2/L26 2012ApJ...752L..26P CrossRefGoogle Scholar
Parrent, J. T., et al. 2016, MNRAS, 457, 3702 10.1093/mnras/stw239 2016MNRAS.457.3702P CrossRefGoogle Scholar
Patat, F., et al. 2007, Science, 317, 924 10.1126/science.1143005 2007Sci...317..924P CrossRefGoogle Scholar
Pereira, R., et al. 2013, A&A, 554, A27 10.1051/0004-6361/201221008 2013A&A...554A..27P Google Scholar
Pettini, M., & Pagel, B. E. J. 2004, MNRAS, 348, L59 10.1111/j.1365-2966.2004.07591.x 2004MNRAS.348L..59P CrossRefGoogle Scholar
Phillips, M. M., et al. 2013, ApJ, 779, 38 10.1088/0004-637X/779/1/38 2013ApJ...779...38P CrossRefGoogle Scholar
Pignata, G., et al. 2009, in AIP Conf. Ser., Vol. 1111, eds. Giobbi, G., Tornambe, A., Raimondo, G., Limongi, M., Antonelli, L. A., Menci, N., & Brocato, E. (Sicily: AIP), 551 10.1063/1.3141608 Google Scholar
Planck Collaboration et al. 2011, A&A, 536, A1 10.1051/0004-6361/201116464 2011A&A...536A...1P Google Scholar
Poznanski, D., Ganeshalingam, M., Silverman, J. M., & Filippenko, A. V. 2011, MNRAS, 415, L81 2011MNRAS.415L..81P 10.1111/j.1745-3933.2011.01084.x CrossRefGoogle Scholar
Poznanski, D., Prochaska, J. X., & Bloom, J. S. 2012, MNRAS, 426, 1465 10.1111/j.1365-2966.2012.21796.x 2012MNRAS.426.1465P CrossRefGoogle Scholar
Quimby, R. M. 2006, PhD thesis, The University of Texas at AustinGoogle Scholar
Rau, A., et al. 2009, PASP, 121, 1334 10.1086/605911 2009PASP..121.1334R CrossRefGoogle Scholar
Salim, S., et al. 2007, ApJS, 173, 267 10.1086/519218 2007ApJS..173..267S CrossRefGoogle Scholar
Sandage, A., & Tammann, G. A. 1981, A revised Shapley-Ames Catalog of bright galaxies, Carnegie Institution of Washington Pub. 635 (Washington: Carnegie Institution)Google Scholar
Scalzo, R., et al. 2012, ApJ, 757, 12 10.1088/0004-637X/757/1/12 2012ApJ...757...12S CrossRefGoogle Scholar
Scalzo, R. A., et al. 2010, ApJ, 713, 1073 10.1088/0004-637X/713/2/1073 2010ApJ...713.1073S CrossRefGoogle Scholar
Shappee, B. J., et al. 2014, ApJ, 788, 48 10.1088/0004-637X/788/1/48 2014ApJ...788...48S CrossRefGoogle Scholar
Silverman, J. M., Ganeshalingam, M., Li, W., Filippenko, A. V., Miller, A. A., & Poznanski, D. 2011, MNRAS, 410, 585 10.1111/j.1365-2966.2010.17474.x 2011MNRAS.410..585S CrossRefGoogle Scholar
Silverman, J. M., Ganeshalingam, M., Li, W., & Filippenko, A. V. 2012a, MNRAS, 425, 1889 10.1111/j.1365-2966.2012.21526.x 2012MNRAS.425.1889S CrossRefGoogle Scholar
Silverman, J. M., Kong, J. J., & Filippenko, A. V. 2012b, MNRAS, 425, 1819 10.1111/j.1365-2966.2012.21269.x 2012MNRAS.425.1819S CrossRefGoogle Scholar
Silverman, J. M., Vinkó, J., Marion, G. H., Wheeler, J. C., Barna, B., Szalai, T., Mulligan, B. W., & Filippenko, A. V. 2015, MNRAS, 451, 1973 10.1093/mnras/stv1011 2015MNRAS.451.1973S CrossRefGoogle Scholar
Silverman, J. M., et al. 2012c, MNRAS, 425, 1789 10.1111/j.1365-2966.2012.21270.x 2012MNRAS.425.1789S CrossRefGoogle Scholar
Simon, J. D., et al. 2009, ApJ, 702, 1157 10.1088/0004-637X/702/2/1157 2009ApJ...702.1157S CrossRefGoogle Scholar
Smartt, S. J., et al. 2015, A&A, 579, A40 10.1051/0004-6361/201425237 2015A&A...579A..40S Google Scholar
Sternberg, A., et al. 2011, Science, 333, 856 10.1126/science.1203836 2011Sci...333..856S CrossRefGoogle Scholar
Sternberg, A., et al. 2014, MNRAS, 443, 1849 10.1093/mnras/stu1202 2014MNRAS.443.1849S CrossRefGoogle Scholar
Stritzinger, M., Suntzeff, N. B., Hamuy, M., Challis, P., Demarco, R., Germany, L., & Soderberg, A. M. 2005, PASP, 117, 810 10.1086/431468 2005PASP..117..810S CrossRefGoogle Scholar
Tanaka, M., et al. 2010, ApJ, 714, 1209 10.1088/0004-637X/714/2/1209 2010ApJ...714.1209T CrossRefGoogle Scholar
Taubenberger, S., et al. 2011, MNRAS, 412, 2735 10.1111/j.1365-2966.2010.18107.x 2011MNRAS.412.2735T CrossRefGoogle Scholar
Taubenberger, S., et al. 2013, MNRAS, 432, 3117 10.1093/mnras/stt668 2013MNRAS.432.3117T CrossRefGoogle Scholar
Theureau, G., Bottinelli, L., Coudreau-Durand, N., Gouguenheim, L., Hallet, N., Loulergue, M., Paturel, G., & Teerikorpi, P. 1998a, A&AS, 130, 333 10.1051/aas:1998416 1998A&AS..130..333T Google Scholar
Theureau, G., Bottinelli, L., Coudreau-Durand, N., Gouguenheim, L., Hallet, N., Loulergue, M., Paturel, G., & Teerikorpi, P. 1998b, A&AS, 130, 333 10.1051/aas:1998416 1998A&AS..130..333T Google Scholar
Tomasella, L., et al. 2014, AN, 335, 841 10.1002/asna.201412068 2014AN....335..841T Google Scholar
Tremonti, C. A., et al. 2004, ApJ, 613, 898 10.1086/423264 2004ApJ...613..898T CrossRefGoogle Scholar
Turatto, M., Benetti, S., & Cappellaro, E. 2003, in From Twilight to Highlight: The Physics of Supernovae, Proc. of the ESO/MPA/MPE Workshop, eds. Hillebrandt, W. & Leibundgut, B. (Garching: ESO), 200 10.1007/10828549_26 CrossRefGoogle Scholar
van Dokkum, P. G. 2001, PASP, 113, 1420 10.1086/323894 2001PASP..113.1420V CrossRefGoogle Scholar
Wang, X., et al. 2009, ApJL, 699, L139 10.1088/0004-637X/699/2/L139 2009ApJ...699L.139W CrossRefGoogle Scholar
Wyrzykowski, L., Hodgkin, S., Blogorodnova, N., Koposov, S., & Burgon, R. 2012, preprint (arXiv:1210.5007)2012arXiv1210.5007W Google Scholar
Wyrzykowski, Ł., et al. 2014, ACTA, 64, 197 2014AcA....64..197W Google Scholar
Yamanaka, M., et al. 2009, ApJL, 707, L118 10.1088/0004-637X/707/2/L118 2009ApJ...707L.118Y CrossRefGoogle Scholar
Yaron, O., & Gal-Yam, A. 2012, PASP, 124, 668 10.1086/666656 2012PASP..124..668Y CrossRefGoogle Scholar
York, D. G., et al. 2000, AJ, 120, 1579 10.1086/301513 2000AJ....120.1579Y CrossRefGoogle Scholar
Yuan, F. 2010, PhD thesis, University of MichiganGoogle Scholar
Yuan, F., et al. 2010, ApJ, 715, 1338 10.1088/0004-637X/715/2/1338 2010ApJ...715.1338Y CrossRefGoogle Scholar
Zhao, X., et al. 2015, ApJS, 220, 20 10.1088/0067-0049/220/1/20 2015ApJS..220...20Z CrossRefGoogle Scholar
Figure 0

Table 1. Details of WiFeS gratings.

Figure 1

Figure 1. Evolution of the WiFeS wavelength solution over a 2-yr period spanning 2013 May to 2015 April. In the left two panels of each row, we plot the deviation of individual wavelength solutions from the mean solution (averaged over the full 2-yr period) for the top and middle slitlets of the instrument (left and middle columns, respectively) for the B3000 (top row) and R3000 (bottom row) gratings. Wavelength solution residuals are colour-coded by date from earliest (red) to latest (purple), with the pixel size (dashed black lines) and wavelength solution residual RMS (solid black lines) displayed for comparison. In the right panels of each row, we show the average deviation (across all wavelengths) from the mean wavelength solution as a function of epoch, to trace the temporal evolution of the instrument solution (points for individual epochs obey the same colour scheme as the left panels).

Figure 2

Figure 2. The normalised mean (left column) and RMS (right column) of the illumination corrections for the B3000 (top row) and R3000 (bottom row) gratings. The RMS images are shown as fractional values of the mean illumination correction.

Figure 3

Figure 3. Flux calibration solutions for the B3000 (top) and R3000 (bottom) gratings. As in Figure 1, these are colour-coded by date from earliest (red) to latest (purple), with the mean flux calibration solution shown as the solid black line.

Figure 4

Figure 4. Seeing measurements at Siding Spring Observatory as measured during AWSNAP observations. These include measurements from the WiFeS guider camera (green dash-dot histogram), and measurements of low airmass standard star WiFeS datacubes convolved with B-band (blue solid histogram—from the blue detector) and R-band (red dashed histogram—from the red detector) filter curves.

Figure 5

Figure 5. On-sky distribution (in equatorial coordinates) of all extragalactic targets in AWSNAP DR1 (white diamonds) plotted over the 857 GHz all-sky map from the Planck satellite (Planck Collaboration et al. 2011) which reveals emission by Milky Way dust. The physical pointing limit of the ANU 2.3-m telescope (~+40° declination) is shown as the solid gray bar.

Figure 6

Figure 6. Example spectra spanning the typical range of signal-to-noise (S/N) in the AWSNAP sample. Shown here are spectra of SN 2012dt (a SN II, with a ‘low’ S/N of 5 per Å), ASASSN-15ba (a SN Ia, with a ‘medium’ S/N of 13 per Å), and SN 2012dj (a SN Ib, with a ‘high’ S/N of 33 per Å).

Figure 7

Figure 7. Normalised histogram of the number of spectroscopic epochs per target for the full sample of AWSNAP DR1 targets. For comparison, we also show the same histogram for previous SN spectroscopy surveys: BSNIP (Silverman et al. 2012c), CfA (Matheson et al. 2008; Blondin et al. 2012), and CSP (Folatelli et al. 2013). The inset shows the (re-)normalised histograms of the number of spectroscopic epochs for multiply observed targets (i.e. those with Nobs > 1) in the same surveys.

Figure 8

Figure 8. Total number of spectra (outer ring) and SN targets (middle annulus) for AWSNAP broken down by SN type, compared to the volume-limited SN rates (inner circle) for the LOSS survey (Li et al. 2011a, —note this galaxy-targeted survey did not find any SLSNe). In each ring, the regions are colour-coded by (broad) SN type (see text for discussion): SNe Ia (blue), SNe II(red), SNe Ib/Ic (green), and SLSNe (purple).

Figure 9

Table 2. AWSNAP objects and spectra by SN type.

Figure 10

Figure 9. Histograms of spectroscopic phases (with respect to B-band maximum light) for SN Ia spectra in AWSNAP and other SN Ia spectroscopy samples (the same as in Figure 7). Note the AWSNAP phases are based on the spectroscopic-based phase reported with the SN classification, which may have an associated uncertainty of 3–5 d.

Figure 11

Figure 10. Strength of the high-velocity features (HVFs) in the Ca II NIR triplet—using the quantity RHVF as defined by Childress et al. (2014b)—plotted against the Si II absorption strength ratio RSi defined by Nugent et al. (1995). On the top axis, we show the rough equivalent light curve decline rate Δm15 values corresponding to the range of RSi values. The crosshair in the upper right represents the characteristic errors in measurement of RHVF (typically 20%, plotted here for the larger values of RHVF) and the error in Δm15 when converted from RSi.

Figure 12

Figure 11. Sodium absorption fit examples for both the R7000 (top) and R3000 (bottom) gratings. Data (which have been normalised to the local continuum fit) are shown as blue diamonds whilst the best fit absorption profile is shown as the solid red curve. For reference, we also mark the continuum level (horizontal black line at value 1.00) and the rest wavelengths of the sodium doublet (vertical dotted gray lines).

Figure 13

Figure 12. Determination of the local velocity for SN 2014ao in NGC 2615 with WiFeS. Left: SDSS (York et al. 2000) gri colour composite—created with SWARP (Bertin et al. 2002) and STIFF (Bertin 2012)—with the WiFeS field of view (red rectangle) and SN location (red dot) highlighted. Middle: Image of the SN 2014ao WiFeS data cube in the isolated wavelength range within ± 6 Å (i.e. ± 300 km s−1) of the wavelength of Hα at the published redshift of NGC 2615, with host core (purple square) and nearby H ii region (brown square) highlighted—the SN is the bright object near the centre. Right: Extracted WiFeS spectra of the nearby H ii region (top) and host core (bottom) near the Hα+NII emission line group, with the expected location of those lines at the published redshift of NGC 2615 (z = 0.014083, Theureau et al. 1998a) shown as the vertical gray lines.

Figure 14

Figure 13. Top: Silicon absorption ratio RSi plotted against velocity centre of the narrow sodium absorption feature (as in Figure 10 we show the corresponding values of Δm15, though note the smaller range). Middle: HVF strength (RHVF) plotted against sodium absorption velocity. Bottom: Absorption equivalent width of the combined D1+D2 sodium lines plotted against sodium absorption velocity. On the right axis of this panel, we use the relation of Poznanski et al. (2012) to show the reddening values E(BV) corresponding to the measured sodium equivalent widths if the absorption arises solely from the ISM—though this is unlikely to be true for all SNe Ia (Poznanski et al. 2011; Phillips et al. 2013—see discussion in text). In all panels, higher resolution observations with the R7000 grating are displayed as green squares, whilst lower resolution R3000 observations are shown as blue circles.

Figure 15

Figure 14. AWSNAP time series of SN 2012dn, labelled by phase with respect to the date of maximum light (2012 July 24, as determined by Chakradhari et al. 2014). Note these observations come from the first semester of AWSNAP when observing time was allocated in multi-night blocks separated sometimes by a month or more.

Figure 16

Figure 15. SN 2012dn at its latest AWSNAP epoch (+91 d on 2012 October 23) compared to other candidate super-Chandra SNe Ia SN 2007if at +98 d (top panel, from Silverman et al. 2011) and SN 2009dc at +97 d (middle panel, from Taubenberger et al. 2011), as well as the normal SN 2011fe (bottom panel, from Pereira et al. 2013).

Figure 17

Figure 16. Extraction of host galaxy emission line flux (green) from late SN 2012dn spectrum (blue) using simple linear continuum fits (red).

Figure 18

Table 3. SN 2012dn local emission line fluxes.

Figure 19

Table A1. All AWSNAP Targets.

Figure 20

Table A2. Alternate and/or full designations for SNe in the AWSNAP sample.

Figure 21

Figure A1. Literature data used to fit the trend of Δm15 versus RSi. The best fit is the thick solid line, and the thin dashed line represent the trend ± 1σ (where σ is the dispersion of the data about the trend).

Figure 22

Table A3. All AWSNAP spectra.

Figure 23

Table A4. Fits of Na line in SNe Ia.

Figure 24

Table A5. Host galaxy redshifts for SN Ia Na sample.