J/ApJ/905/94 Classif. for PS1-MDS SNe with SuperRAENN (Villar+, 2020)
SuperRAENN: a semisupervised supernova photometric classification pipeline
trained on Pan-STARRS1 Medium-Deep Survey supernovae.
Villar V.A., Hosseinzadeh G., Berger E., Ntampaka M., Jones D.O.,
Challis P., Chornock R., Drout M.R., Foley R.J., Kirshner R.P., Lunnan R.,
Margutti R., Milisavljevic D., Sanders N., Pan Y.-C., Rest A.,
Scolnic D.M., Magnier E., Metcalfe N., Wainscoat R., Waters C.
<Astrophys. J., 905, 94 (2020)>
=2020ApJ...905...94V 2020ApJ...905...94V
ADC_Keywords: Supernovae; Redshifts; Spectral types; Photometry;
Spectra, optical
Keywords: Supernovae ; Astrostatistics ; Light curve classification
Abstract:
Automated classification of supernovae (SNe) based on optical
photometric light-curve information is essential in the upcoming era
of wide-field time domain surveys, such as the Legacy Survey of Space
and Time (LSST) conducted by the Rubin Observatory. Photometric
classification can enable real-time identification of interesting
events for extended multiwavelength follow-up, as well as archival
population studies. Here we present the complete sample of
5243 "SN-like" light curves (in gP1rP1iP1zP1) from the Pan-STARRS1
Medium-Deep Survey (PS1-MDS). The PS1-MDS is similar to the planned
LSST Wide-Fast-Deep survey in terms of cadence, filters, and depth,
making this a useful training set for the community. Using this data
set, we train a novel semisupervised machine learning algorithm to
photometrically classify 2315 new SN-like light curves with host
galaxy spectroscopic redshifts. Our algorithm consists of an RF
supervised classification step and a novel unsupervised step in which
we introduce a recurrent autoencoder neural network (RAENN). Our final
pipeline, dubbed SuperRAENN, has an accuracy of 87% across five SN
classes (Type Ia, Ibc, II, IIn, SLSN-I) and macro-averaged purity and
completeness of 66% and 69%, respectively. We find the highest
accuracy rates for SNe Ia and SLSNe and the lowest for SNe Ibc. Our
complete spectroscopically and photometrically classified samples
break down into 62.0% Type Ia (1839 objects), 19.8% Type II
(553 objects), 4.8% Type IIn (136 objects), 11.7% Type Ibc
(291 objects), and 1.6% Type I SLSNe (54 objects).
Description:
PS1 is a wide-field survey telescope located near the summit of
Haleakala, Maui, with a 1.8m diameter primary mirror and a
1.4 gigapixel camera (GPC1). The Pan-STARRS1 Medium-Deep Survey
(PS1-MDS), one of several PS1 surveys (Chambers+ 2016, II/349), was
conducted between 2009 and 2014 July. It consisted of
10 single-pointing fields, each of approximately 7.1deg2, with a
pixel scale of 0.25". During its 5yr survey, PS1-MDS discovered
5243 SN-like objects.
Of the 5243 SN-like objects, 4090 host galaxies were targeted through
a concerted observational effort. The majority (3321 objects) were
observed using the Hectospec multifiber instrument on MMT. Additional
host redshifts were obtained with the Anglo-Australian Telescope (AAT;
290 objects), the WIYN telescope (217 objects), and the Apache Point
Observatory 3.5m telescope (APO; 5 objects). Additional host redshifts
were obtained from archival survey data. See Section 2.
File Summary:
--------------------------------------------------------------------------------
FileName Lrecl Records Explanations
--------------------------------------------------------------------------------
ReadMe 80 . This file
table1.dat 171 5243 Supernovae (SNe) properties
table2.dat 88 2885 SNe classification
table3.dat 74 15 Rare transients classification
--------------------------------------------------------------------------------
See also:
III/250 : The VIMOS VLT deep survey (VVDS-DEEP) (Le Fevre+ 2005)
VII/259 : 6dF galaxy survey final redshift release (Jones+, 2009)
II/349 : The Pan-STARRS release 1 (PS1) Survey - DR1 (Chambers+, 2016)
J/ApJS/78/1 : Galaxies in 7 clusters 0.35<z<0.55 (Dressler+, 1992)
J/ApJS/155/271 : Chandra Deep Field-South: Opt. spectroscopy (Szokoly+, 2004)
J/MNRAS/372/425 : 2dF-SDSS Luminous Red Galaxy Survey, 2SLAQ (Cannon+, 2006)
J/AJ/132/2409 : Deep ATLAS radio observations of CDFS (Norris+, 2006)
J/A+A/474/473 : XMM-LSS survey: AGN classifications (Garcet+, 2007)
J/ApJS/172/70 : zCOSMOS-bright catalog (Lilly+, 2007)
J/A+A/467/73 : 3σ hard sample of XMDS survey (Tajer+, 2007)
J/A+A/477/717 : Spectroscopy of Type Ia supernovae (Bronder+, 2008)
J/MNRAS/387/1323 : SDSS photometry of luminous red galaxies (Ross+, 2008)
J/A+A/495/53 : Physical properties of VVDS galaxies (Lamareille+, 2009)
J/ApJS/184/218 : The zCOSMOS 10k-bright spectroscopic sample (Lilly+, 2009)
J/ApJS/182/625 : WIYN spectroscopy in the deep SWIRE field (Owen+, 2009)
J/ApJ/696/1195 : COSMOS AGN spectroscopic survey. I. (Trump+, 2009)
J/A+A/512/A12 : VLT/VIMOS spectroscopy in GOODS-S field (Balestra+, 2010)
J/ApJ/711/928 : Low-redshift Lyα galaxies from GALEX (Cowie+, 2010)
J/MNRAS/401/1429 : WiggleZ dark energy survey (DR1) (Drinkwater+, 2010)
J/MNRAS/405/2302 : Improved redshifts for SDSS quasar spectra (Hewett+, 2010)
J/MNRAS/401/294 : Optical identification of XMM-LSS sources (Stalin+, 2010)
J/A+A/529/A135 : AGN Opt/IR properties in Lockman Hole (Rovilos+, 2011)
J/ApJ/738/162 : SN Ia candidates from the SDSS-II SN Survey (Sako+, 2011)
J/MNRAS/422/25 : SDSS DR7 groups, clusters and filaments (Smith+, 2012)
J/ApJ/750/99 : The Pan-STARRS1 photometric system (Tonry+, 2012)
J/ApJ/769/39 : SN Ibn PS1-12sk optical & NIR light curves (Sanders+, 2013)
J/ApJ/794/23 : Pan-STARRS1 transients optical photometry (Drout+, 2014)
J/MNRAS/441/1802 : Low-redshift QSOs in SDSS Stripe 82 (Karhunen+, 2014)
J/ApJ/795/44 : PS1 SNe Ia (0.02<z<0.7) griz light curves (Rest+, 2014)
J/ApJ/799/208 : Type IIP supernovae from Pan-STARRS1 (Sanders+, 2015)
J/ApJ/807/178 : Rich galaxy clusters identified in SDSS-DR12 (Wen+, 2015)
J/ApJ/837/120 : Lick Observatory SN Search (LOSS) revisited (Graur+, 2017)
J/ApJ/857/51 : Dark energy properties with PS1 SNe. II. (Jones+, 2018)
J/ApJ/852/81 : 17 PS1 superluminous SNe LCs + classif. sp. (Lunnan+, 2018)
J/A+A/611/A97 : Phot. quasar candidates in Stripe 82 (Pasquet-Itam+, 2018)
J/ApJ/859/101 : The supernovae Ia Pantheon sample (Scolnic+, 2018)
J/ApJ/895/32 : Zwicky Transient Facility BTS. I. (Fremling+, 2020)
J/ApJ/905/93 : Classif. of PS1-MDS SNe with Superphot (Hosseinzadeh+, 2020)
http://outerspace.stsci.edu/display/PANSTARRS/ : Pan-STARRS1 home page
Byte-by-byte Description of file: table1.dat
--------------------------------------------------------------------------------
Bytes Format Units Label Explanations
--------------------------------------------------------------------------------
1- 11 A11 --- Name Object Name
13- 21 A9 --- PScID Pan-STARRS1 Medium-Deep Survey (PS1-MDS)
transient identifier
23- 31 A9 --- IAU IAU Name
33- 71 A39 --- CBET CBET ADS bibcode(s)
73- 80 F8.4 deg RAdeg [34.3/354] SN Right Ascension (J2000)
82- 89 F8.4 deg DEdeg [-29.4/59.7] SN Declination (J2000)
91- 96 F6.4 mag E(B-V) [0.0034/0.14] Milky Way reddening
98-110 A13 --- Type SN Type
112-120 F9.7 --- z [0/2.77]? Redshift estimate from either the
transient spectra or host galaxy spectra
122-129 F8.4 deg RAHdeg ? Host Right Ascension, decimal degree (J2000)
131-138 F8.4 deg DEHdeg ? Host Declination, decimal degree (J2000)
140-142 I3 --- Npt [0/206] Number of >5σ datapoints
in light curve
144-167 A24 --- Tel Telescope or source (ADS bibcode) used to
acquire galaxy or SN spectra (1)
169-169 A1 --- uSup Included in unsupervised training set
171-171 A1 --- Sup Included in supervised training set
--------------------------------------------------------------------------------
Note (1): Host galaxies observations with telescopes as below:
MMT = the Hectospec multifiber instrument on MMT (3321 occurrences)
AAT = Anglo-Australian Telescope (291 occurrences)
WIYN = the WIYN telescope (208 occurrences)
APO = the Apache Point Observatory 3.5m telescope (5 occurrences)
--------------------------------------------------------------------------------
Byte-by-byte Description of file: table2.dat
--------------------------------------------------------------------------------
Bytes Format Units Label Explanations
--------------------------------------------------------------------------------
1- 9 A9 --- PScID PS1-MDS Transient Identifier
11- 14 F4.2 --- pSLSN [0/1] Probability, SLSN
16- 20 F5.2 --- e_pSLSN [0/0.22] Lower uncertainty on pSLSN
22- 25 F4.2 --- E_pSLSN [0/0.3] Lower uncertainty on pSLSN
27- 30 F4.2 --- pII [0/1] Probability, Type II
32- 36 F5.2 --- e_pII [0/0.24] Lower uncertainty on pII
38- 41 F4.2 --- E_pII [0/0.21] Lower uncertainty on pII
43- 46 F4.2 --- pIIn [0/1] Probability, Type IIn
48- 52 F5.2 --- e_pIIn [0/0.26] Lower uncertainty on pIIn
54- 57 F4.2 --- E_pIIn [0/0.3] Lower uncertainty on pIIn
59- 62 F4.2 --- pIa [0/1] Probability, Type Ia SN
64- 67 F4.2 --- e_pIa [0/0.26] Lower uncertainty on pIa
69- 72 F4.2 --- E_pIa [0/0.23] Lower uncertainty on pIa
74- 77 F4.2 --- pIbc [0/1] Probability, Type Ibc
79- 83 F5.2 --- e_pIbc [0/0.22] Lower uncertainty on pIbc
85- 88 F4.2 --- E_pIbc [0/0.23] Lower uncertainty on pIbc
--------------------------------------------------------------------------------
Byte-by-byte Description of file: table3.dat
--------------------------------------------------------------------------------
Bytes Format Units Label Explanations
--------------------------------------------------------------------------------
1- 9 A9 --- PScID PS1-MDS Transient Identifier
11- 19 A9 --- Name Object name
21- 29 A9 --- Class Spec. Class (FELT=fast-evolving luminous
transient, TDE=tidal disruption event)
31- 49 A19 --- r_Class Reference for Class
51- 54 F4.2 --- pSLSN [0/0.88] Probability, Type SLSN
56- 59 F4.2 --- pII [0/0.76] Probability, Type II
61- 64 F4.2 --- pIIn [0/0.79] Probability, Type IIn
66- 69 F4.2 --- pIa [0.02/0.95] Probability, Type Ia
71- 74 F4.2 --- pIbc [0.01/0.42] Probability, Type Ibc
--------------------------------------------------------------------------------
History:
From electronic version of the journal
(End) Prepared by [AAS], Emmanuelle Perret [CDS] 27-Jul-2022