J/A+A/602/A86 Blazar candidates among Fermi/LAT 3FGL catalog (Lefaucheur+, 2017)
Research and characterisation of blazar candidates among the Fermi/LAT 3FGL
catalogue using multivariate classifications.
Lefaucheur J., Pita S.
<Astron. Astrophys. 602, A86 (2017)>
=2017A&A...602A..86L 2017A&A...602A..86L (SIMBAD/NED BibCode)
ADC_Keywords: Active gal. nuclei; BL Lac objects ; Gamma rays
Keywords: gamma rays; galaxies - galaxies: active -
BL Lacertae objects: general - methods: statistical - catalogs
Abstract:
We present a study to search for, and characterise blazar candidates
among the Fermi/LAT 3FGL catalogue using machine-learning
classification methods. Classifiers are based on the exploitation of
statistical differences imprinted in the 3FGL Fermi/LAT catalogue,
such as variability and spectral shape, between different populations
of sources.
Description:
We obtained a sample of 595 blazar candidates among the unassociated
sources of the 3FGL catalogue. Performance metrics are derived
separately for high (|b|>10-degrees) and low (|b|≤10-degrees)
latitude sources, and according to the existence of a 3FGL caution
flag. The number of candidates are sumarised below:
- 345 high-latitude blazar candidates with no flag (estimated number
of false positives ∼4.8)
- 80 high-latitude blazar candidates with a flag (estimated number of
false positives ∼4.5)
- 72 low-latitude blazar candidates with no flag (estimated number of
false positives ∼8.8)
- 98 low-latitude blazar candidates with a flag (estimated number of
false positives ∼54.0)
We also propose an assignation, BL Lac or FSRQ, for the candidates we
proposed and the blazar candidates from the 3FGL catalogue, labelled
as BCUs. Only sources with no flag where considered. In total, we
obtained a sample of 509 BL Lacs and 295 FSRQs with a number of false
positives respectively estimated to ∼29 and ∼70.
File Summary:
--------------------------------------------------------------------------------
FileName Lrecl Records Explanations
--------------------------------------------------------------------------------
ReadMe 80 . This file
table5.dat 255 595 Sample of blazar candidates from unassociated
3FGL sources
table7.dat 216 903 Type assignation of blazar candidates
--------------------------------------------------------------------------------
See also:
J/ApJ/753/83 : Associations to 1FGL sources (Ackermann+, 2012)
J/MNRAS/424/L64 : AGN/pulsar distinction for 2FGL sources (Mirabal+, 2012)
J/ApJS/206/13 : Blazars with γ-ray counterparts. II. (Massaro+, 2013)
J/ApJS/207/4 : Unidentified γ-ray sources. III. Radio (Massaro+, 2013)
J/ApJ/782/41 : 231 AGN candidates from the 2FGL catalog (Doert+, 2014)
J/ApJ/820/8 : 3FGL sources statistical classifications
(Saz Parkinson+, 2016)
Byte-by-byte Description of file: table5.dat
--------------------------------------------------------------------------------
Bytes Format Units Label Explanations
--------------------------------------------------------------------------------
1- 12 A12 --- 3FGL 3FGL source name, JHHMM.m+DDMM (Source_Name)
14- 34 F21.17 deg GLON Galactic longitude
36- 57 F22.18 deg GLAT Galactic latitude
59- 61 I3 --- Flags Flags (1)
63- 82 F20.17 [-] logsigmac Discriminant parameter normalised curvature
(NormalizedCurvature) (2)
84-102 F19.17 [-] logTS Discriminant parameter normalised variability
(NormalizedVariability) (G2)
104-125 E22.18 --- HR23 Discriminant parameter HR23
hardness ratio (G3)
127-148 E22.19 --- HR34 Discriminant parameter HR34
hardness ratio (G3)
150-171 E22.19 --- HR23-HR34 Discriminant parameter curvature, HR23-HR34
(Curvature)
173-190 F18.16 --- lambda Discriminant parameter lambda (Lambda) (3)
192-210 F19.17 --- zetaBDT Output parameter zeta from BDT decision
(zeta_BDT) (G5)
212-230 F19.17 --- zetaMLP Output parameter zeta from MLP decision
(zeta_MLP) (G5)
232-235 A4 --- Type Blazar type (BlazarType) (G6)
237-239 A3 --- SP16 [yes/no ] In Saz Parkinson et al.
(2016, Cat. J/ApJ/820/8) (SazParkinson2016)
241-243 A3 --- Mi12 [yes/no ] In Mirabal et al.
(2012, Cat. J/MNRAS/424/L64) (Mirabal_2012)
245-247 A3 --- A12 [yes/no ] In Ackermann et al.
(2012, Cat. J/ApJ/753/83) (Ackermann_2012)
249-251 A3 --- Ma13 [yes/no ] In Massaro et al.
(2013, Cat. J/ApJS/207/4 and J/ApJS/206/13)
(Massaro_2013)
253-255 A3 --- D14 [yes/no ] In Doert & Errando
(2014, Cat. J/ApJ/782/41) (Doert_2014)
--------------------------------------------------------------------------------
Note (1): Flags (from table3 of Acero et al. (2015ApJS..218...23A 2015ApJS..218...23A) as follows:
1 = Source with TS≥35, which went to TS≤25 when changing the diffuse model
or the analysis method.
Sources with TS≲35 are not flagged with this bit because normal
statistical fluctuations can push them to TS<25
2 = Not used
3 = Flux (>1GeV) or energy flux (>100MeV) changed by more than 3σ when
changing the diffuse model or the analysis method. Requires also that the
flux change by more than 35% (to not flag strong sources)
4 = Source-to-background ratio less than 10% in highest band in which TS>25
Background is integrated over πr682 or 1 square degree, whichever
is smaller
5 = Closer than θref from a brighter neighbor. θref is defined
in the highest band in which source TS≥25, or the band with highest TS
if all are <25. θref is set to 2.17° (FWHM) below 300MeV,
1.38° between 300MeV and 1GeV, 0.87° between 1 and 3GeV,
0.67° between 3 and 10GeV, and 0.45° above 10GeV (2r68)
6 = On top of an interstellar gas clump or small-scale defect in the model of
diffuse emission
7 = Unstable position determination; result from GTFINDSRC outside the 95%
ellipse from pointlike
8 = Not used
9 = Localization Quality >8 in pointlike or long axis of 95%
ellipse >0.25°
10 = Spectral Fit Quality >16.3
11 = Possibly due to the Sun
12 = Highly curved spectrum; LogParabola β fixed to 1 or PLExpCutoff
Spectral_Index fixed to 0.5
Note (2): Normalised curvature, defined as σc/σ where σc is
the significance of the curvature and and σ is the detection significance
(Doert & Errando, 2014, Cat. J/ApJ/782/41).
Note (3): ratio between the spectral index of the preferred hypothesis and the
spectral index of the power law hypothesis, called γ.
--------------------------------------------------------------------------------
Byte-by-byte Description of file: table7.dat
--------------------------------------------------------------------------------
Bytes Format Units Label Explanations
--------------------------------------------------------------------------------
1- 12 A12 --- 3FGL 3FGL source name, JHHMM.m+DDMM (Source_Name)
14- 33 F20.16 deg GLON Galactic longitude
35- 56 F22.18 deg GLAT Galactic latitude
58- 61 A4 --- Class [BCU UnId] Classification (CLASS) (1)
63- 80 F18.16 --- gamma Spectral index of the power law hypothesis
(PowerLaw_Index)
82- 99 F18.16 [-] logEp Discriminant parameter pivot energy (which is
somewhat correlated to the position of the
high energy peak) (Pivot_Energy)
101-119 F19.17 [-] logTS Discriminant parameter normalised variability
(NormalizedVariability) (G2)
121-142 E22.19 --- HR23 Discriminant parameter HR23 hardness ratio (G3)
144-165 E22.19 --- HR34 Discriminant parameter HR34 hardness ratio (G3)
167-188 F22.19 --- zetaBDT Output parameter zeta from BDT decision
(zeta_BDT) (G5)
190-211 E22.19 --- zetaMLP Output parameter zeta from MLP decision
(zeta_MLP) (G5)
213-216 A4 --- Type Blazar type (BlazarType) (G6)
--------------------------------------------------------------------------------
Note (1): Classification as follows:
BCU = blazar candidates of uncertain type
UnId = unidentified
--------------------------------------------------------------------------------
Global notes:
Note (G2): Normalised variability, given by the ratio between the index
variability TS and the detection significance σ (Doert & Errando 2014,
Cat. J/ApJ/782/41).
Note (G3): We use the definition of hardness ratio given in Ackermann et al.
(2012, Cat. J/ApJ/753/83) which is HRij=(Fj<Ej>-Fi<Ei>)/(Fi<Ei>+Fj<Ej>),
where Fi is the integrated flux in the energy band i and < Ei> is the mean
energy of the band.
Note (G5): optimal cutoff value with boosted decision trees (BDT) and a
multilayer perceptron (MLP) neural network.
Note (G6): Blazar type as follows:
bll = BL Lac object
fsrq = FSRQ (Flat Spectrum Radio Quasar)
unc = uncertain
NULL = not analyzed
--------------------------------------------------------------------------------
Acknowledgements:
Julien Lefaucheur, julien.lefaucheur(at)obspm.fr
(End) J. Lefaucheur [Obs. Paris/Meudon - LUTH], P. Vannier [CDS] 13-Mar-2017