Radio Weak BL Lac objects in the Fermi era

1.9 MB·

by  F. Massaro

pre-proof version
PreprinttypesetusingLATEXstyleemulateapjv.11/10/09
RADIO WEAK BL LAC OBJECTS IN THE Fermi ERA
F. Massaro1,2,3, E. J. Marchesini1,4,5, R. D’Abrusco6, N. Masetti7,8, I. Andruchow4,5 & Howard A. Smith6
DipartimentodiFisica,Universita`degliStudidiTorino(UniTO),viaPietroGiuria1,I-10125Torino,Italy.
IstitutoNazionalediFisicaNucleare,SezionediTorino,viaPietroGiuria1,I-10125Torino,Italy.
INAF–OsservatorioAstrofisicodiTorino,viaOsservatorio20,10025,PinoTorinese,Italy.
FacultaddeCienciasAstrono´micasyGeof´ısicas,UniversidadNacionaldeLaPlata,PaseodelBosque,B1900FWA,LaPlata,Argentina.
InstitutodeAstrof´ısicadeLaPlata,CONICET–UNLP,CCTLaPlata,PaseodelBosque,B1900FWA,LaPlata,Argentina.
SmithsonianAstrophysicalObservatory,60GardenStreet,02138Cambridge(MA),UnitedStates.
INAF–IstitutodiAstrofisicaSpazialeeFisicaCosmicadiBologna,viaGobetti101,I-40129,Bologna,Italia.and
DepartamentodeCienciasF´ısicas,UniversidadAndr`esBello,Fern`andezConcha700,LasCondes,Santiago,Chile
7 pre-proof version
1
ABSTRACT
0
2 The existence of “radio weak BL Lac objects” (RWBLs) has been an open question, still unsolved,
since the discovery that quasars could be radio-quiet or radio-loud. Recently several groups iden-
n
tified RWBL candidates, mostly found while searching for low energy counterparts of the unidenti-
a
J fied/unassociatedgamma-raysourceslistedintheFermicatalogs. ConfirmingRWBLsisachallenging
tasksincetheycouldbeconfusedwithwhitedwarfsorweakemissionlinequasarswhentherearenot
1
sufficient data to precisely draw their broad band spectral energy distribution and their classification
2
is mainly based on a featureless optical spectra. Motivated by the recent discovery that Fermi BL

Lacs appear to have very peculiar mid-IR emission, we show that it is possible to distinguish be-
]
E tween WDs, WELQs and BL Lacs using the [3.4]-[4.6]-[12]µm color-color plot built using the WISE
H magnitudes when the optical spectrum is available. On the basis of this analysis, we identify WISE
J064459.38+603131 and WISE J141046.00+740511.2 as the first two genuine RWBLs, both poten-
.
h tially associated with Fermi sources. Finally, to strengthen our identification of these objects as true
p RWBLs, we present multifrequency observations for these two candidates to show that their spectral
- behavior is indeed consistent with those of the BL Lac population.
o
r Subject headings: galaxies: active - galaxies: BL Lacertae objects - quasars: emission lines - quasars:
t general - radiation mechanisms: non-thermal
s
a
[
1. INTRODUCTION Over the entire electromagnetic spectrum both
1 BL Lacs and FSRQs feature a typical double-
BL Lac objects constitute the most elusive
v bumped spectral energy distribution (SEDs) with
class of radio-loud active galactic nuclei (AGNs;
7 the first component peaking between the IR and
Urry & Padovani 1995). Hosted in elliptical galaxies
6
the X-rays and the second one in the gamma-rays
0 (see e.g., Falomo et al. 2014), their emission extends
(see e.g., Padovani & Giommi 1995; Paggi et al. 2009;
6 from radio frequencies up to the highest TeV energies
Abdo et al. 2010b). Mid-IR colors clearly helps to
0 and is characterized by rapid and large-amplitude
determine the energy peak location of the low en-
. flux variability, flat radio spectra, even below ∼1GHz
1 ergy component (Massaro et al. 2011a). BL Lac SEDs
(see e.g., Massaro et al. 2013a; Massaro et al. 2013b),
0 and their broad-band features were described in terms
peculiar infrared (IR) colors (Massaro et al. 2011a;
7 of emission arising from particles accelerated in rel-
D’Abrusco et al. 2012) and significant and vari-
1 ativistic jets directly pointing towards the observer
able optical polarization (i.e., P ∼ 10%) (see e.g.,
: (Blandford & Rees 1978).
v Andruchow et al. 2005; Smith et al. 2007, and refer-
i ences therein). BL Lacs are traditionally discovered using ra-
X dio, X-ray surveys and/or their combination (see
BL Lacs show featureless optical spectra, sometimes
r havingweakemissionlinesofrest-frameequivalentwidth e.g., Stickel et al. 1991; Sambruna et al. 1996;
a Giommi et al. 2005), since in the optical energy
(EW ) lower than 5˚A or absorption lines due to their
r range the lack of spectral features combined with their
host galaxies, both superimposed to a blue continuum
flux variability makes a color-color selection extremely
(see e.g., Laurent-Muehleisen et al. 1999). BL Lac ob-
challenging (Jannuzi et al. 1993; Londish et al. 2002;
jectsareoneofthetwosubclassesoftheradio-loudAGNs
Collinge et al. 2005; Plotkin et al. 2008). However, re-
known as blazars (see e.g., Massaro et al. 2009, for a re-
centobservationstosearchforopticalpolarizationinBL
centreview). Theothersubclassisconstitutedbytheflat
LacsselectedusingdeepopticalsurveyastheSloanDig-
spectrumradioquasars(FSRQs),characterizedbyprop-
ital Sky Survey (see e.g., SDSS DR12 Alam et al. 2015,
ertiessimilartothoseofBLLacsbuthavingaquasar-like
and references therein) and the Two-Degree Field
optical spectra (i.e., with broad emission lines; see e.g.,
Quasar Redshift Survey (2QZ Smith et al. 2005, and
V´eron-Cetty & V´eron 2000) and radio spectral indices1
references therein) validated this strategy of investi-
lower than 0.5 between ∼300 MHz and a few GHz (see
gating proper motion and broad-band color properties
Massaro et al. 2014a, and references therein).
to minimize contamination by featureless Galactic
objects while searching for quasi-featureless spectra
1 Spectralindices,α,aredefinedbyfluxdensity,Sν ∝ν−α
2 F. Massaro, E. Marchesini, R. D’Abrusco, N. Masetti, I. Andruchow & Howard A. Smith 2016
(Smith et al. 2007). ments (see e.g., Liuzzo et al. 2013; Nulsen et al. 2013;
At the current moment, the most efficient survey Fabian 2012, and references therein). The exis-
for discovering new BL Lacs is the Fermi gamma- tence of RWBLs could be crucial to address differ-
ray one (Abdo et al. 2010b; Ackermann et al. 2011a; ent phases of jet formation in cases when relativis-
Nolan et al. 2012), in particular when combining tic jets are not able to form extended radio struc-
gamma-ray observations with mid-IR colors anal- tures (Blandford & K¨onigl 1979; Begelman et al. 1984),
yses (Massaro et al. 2012a; Massaro et al. 2012b) as it seems from low radio frequency (i.e., below
and follow up optical spectroscopic obser- ∼1 GHz) spectral studies (see Massaro et al. 2013b;
vations (Paggi et al. 2013; Shaw et al. 2013; Nori et al. 2014; Giroletti et al. 2016, and references
Massaro et al. 2015a; Massaro et al. 2015b). This therein).
occurs because BL Lacs are the largest known popula- There are two major problems to solve before claim-
tion emitting at MeV-GeV energies, constituting more ing the existence of RWBLs, even when featureless
than 30% of the associated sources in the latest release optical spectra are available. They can be con-
of the Fermi catalog (The Fermi - Large Area Tele- fused with or misinterpreted as hot white dwarfs
scope Third Source Catalog, 3FGL Acero et al. 2015). (WDs; see e.g. Kepler et al. 2015) and/or weak emis-
Moreover, recent optical spectroscopic campaigns sion line quasars (WELQs; Diamond-Stanic et al. 2009;
aimed to confirm the nature of blazar candidates Plotkin et al. 2010, andreferencestherein)whenoptical
of uncertain type listed in the 3FGL showed that spectroscopic data are available. Optical polarimet-
a significant fraction of those observed to date are ric observations could shed light on their nature. How-
also BL Lacs (see e.g. A´lvarez-Crespo et al. 2016a; ever, unfortunately, optical polarization in BL Lacs can
A´lvarez-Crespo et al. 2016b; be variable and it is not always observed thus not repre-
sentingthemostdefinitivestamptoconfirmtheirnature
A´lvarez-Crespo et al. 2016c).
(see e.g., Londish et al. 2004).
However,nowadays,alltheknownFermiBLLacshave
In the present paper we explore the use mid-IR col-
radiocounterpartsandradiosurveysarethereforecrucial
ors as a diagnostic tool to disentangle between BL Lacs,
toassociateγ-rayobjectswiththeircounterparts. Thus,
WDs and WELQs. This new tool could potentially
one of the major open issues for this rare AGN class
open a new window to search for RWBLs. This in-
is proving the existence of radio-weak BL Lac objects
vestigation is motivated by the recent discovery that
(RWBLs). We indicate them as radio-weak sources in-
Fermi BL Lacs show peculiar mid-IR colors (see also,
stead of using the term “radio-quiet”, generally adopted
Massaro & D’Abrusco 2016, and references therein), re-
for the AGN unification scenario. This designation is
sult made possible by observations performed with the
more useful because radio follow up observations could
WISE all-sky survey (Wright et al. 2010), covering a re-
reveal some radio emission. However, the current lack of
gion well isolated from that of other Galactic and extra-
counterpartsinthemajorradiosurveysavailabletodate
galactic objects. The main explanation underlying this
place them in the tail of the distribution of the IR-to-
distinction is the BL Lac non-thermal IR emission (see
radio flux ratios for the whole BL Lac population.
e.g., D’Abrusco et al. 2012; D’Abrusco et al. 2013).
At the current moment only a handful of sources
This manuscript is organized as follows. In § 2 we
have been claimed to be RWBLs on the ba-
presentallthesamplesusedtoperformourmid-IRcolor-
sis of their broad band properties as feature-
color analysis. In § 3 we compare the mid-IR colors of
less optical spectra combined with X-ray emission
BL Lacs with those of WELQs and WDs, while in § 4
and and their lying within the positional uncer-
we search for RWBLs. § 5 is dedicated to the analysis of
tainty regions of unidentified/unassociated gamma-
the IR variability for the best candidates selected on the
ray sources (UGSs; Paggi et al. 2013; Ricci et al. 2015;
basisofthemid-IRcolorsadtheopticalspectra. Thenin
Landoni et al. 2016).
§6wealsoinvestigatetheSWIFTobservationsavailable
Theirexistencewillhaveacrucialimpact(i)onthees-
forthesecandidateswhilein§7wediscussontheSEDs.
timates of their source counts and luminosity functions
Summary and conclusions are given in § 8
(Ajello et al. 2012; Ajello et al. 2014), a task already
Unless stated otherwise, for our numerical results, we
complicatedbythechallengeofmeasuringtheirredshifts
use cgs units. WISE magnitudes used here are in the
(Falomo et al. 2014), (ii) on the gamma-ray association
Vega system and are not corrected for the Galactic ex-
methods all currently based on radio surveys (see e.g,
tinction. As shown in our previous analyses, such cor-
Ackermann et al. 2015,andreferencestherein),ifproved
rections mostly affect the magnitude at 3.4µ, for sources
that RWBLs could be associated to Fermi sources,
lyingatlowGalacticlatitudesanditrangesonlybetween
as well as (iii) on the unification scenario of AGNs
2% and 5% (see e.g. D’Abrusco et al. 2014), thus being
(Urry & Padovani 1995),tonameafewimplications. An
negligible. In addition this allows us to compare plots in
accurate accounting is crucial to compute correctly their
thepresentanalysiswiththoseofpreviousinvestigations
contributiontotheextragalacticgamma-raybackground
(Massaro et al. 2013c).
for which, together with FSRQs (Ajello et al. 2012)
and radio galaxies (Inoue 2011; Di Mauro et al. 2014;
Massaro & Ajello 2011b) they provide the most signif- 2. SAMPLESELECTION
icant ones. Additional implications are related to
To perform our analysis we built three main sam-
jet physics. Radio emission in extragalactic radio-
ples. The first two composed of WDs and WELQs,
loud AGNs permits us to investigate the “large-
respectively; both used in comparison with the third
scale” (i.e. from pc to Mpc size) evolution of rel-
one of Fermi blazars. We searched for mid-IR coun-
ativistic jets and their interactions with the environ-
terparts of all samples in the latest release of the
Radio Weak BL Lac objects in the Fermi era 3
WISE catalog2. All our final samples of WDs, WELQs We also computed the chance probability of having spu-
and gamma-ray blazars only include sources having rious associations within 3(cid:48)(cid:48).3 when crossmatching the
a WISE counterpart within 3(cid:48)(cid:48).3 that is detected at AllWISE source catalog with the SDSSDR7WD; the re-
3.4µm, 4.6µm and 12µm so allowing us to build sult is less than 4% (see e.g., D’Abrusco et al. 2013, for
at least one of the mid-IR color-color plots used in moredetailsonthemethodadoptedforthepchancecal-
our previous blazar analyses (D’Abrusco et al. 2014; culation).
Massaro & D’Abrusco 2016). The choice of 3(cid:48)(cid:48).3 asso- To verify the reliability of our choice for the WD
ciation radius is the same adopted in our previous inves- sample we also used the latest release of the McCook
tigations (D’Abrusco et al. 2013) and it was computed & Sion White Dwarf Catalog5 available to date (see
on the basis of Montecarlo procedure corresponding to a McCook & Sion 1999, for details). This catalog has
probability of having spurious associations of the order been also used to search for dusty disks around WDs
of ∼1% (see also D’Abrusco et al. 2014, for additional (Hoard et al. 2013) using WISE observations. However
details). in this sample the number of sources detected at 3.4µm,
4.6µm and 12µm for which is possible to built the mid-
2.1. BL Lac objects and Flat Spectrum Radio Quasars IR colors is only 91. The chance probability of spurious
The first sample used in our analysis lists all the associations between the McCook and Sion WD catalog
blazars, both BL Lacs and FSRQs, included in the and the AllWISE survey is lower than 1%.
Roma-BZCAT3 (Massaro et al. 2015c) and having a
2.3. Weak emission line quasars
counterpart in the latest release of the Fermi catalog
(Acero et al. 2015). This sample has been recently used In the last two decades an intriguing and also elu-
toperformtheanalysisontheIR–gamma-rayconnection sive subclass of radio-quiet quasars was discovered: the
(Massaro & D’Abrusco 2016). According to the nomen- weak emission line quasars (WELQs), featuring ex-
clatureproposedintheRoma-BZCAThereinafterwewill ceptionally weak or completely missing broad emission
refer to BL Lac objects as BZBs and to the FSRQs as lines in the ultraviolet (UV) rest-frame energy range
BZQs. We choose this sample because it has the largest (see e.g, Collinge et al. 2005; Shemmer et al. 2006;
fraction of BZBs with a mid-IR counterpart in WISE Diamond-Stanic et al. 2009).
corresponding to 603 out of 610 (i.e., ∼99%) within our TheoriginalcriteriontodefineWELQswithrespectto
association radius; this fraction is similar to that occur- othermembersofthesameclass(i.e.,normalquasars)is
ring for the BZQs with 419 out of 426 (i.e., ∼98%). based on the EWr distribution of the Lyα λ1216+N V
λ1240 blend for a sample of quasars in the Sloan Digital
2.2. White Dwarfs Sky Survey (SDSS; ) at redshift z >3, being lower than
Hot WDs could be misclassified as BZBs because they 15˚A as occurs for the >3σ weak tail of the EW [Lyα+
r
could show almost featureless optical spectra and X- NV] distribution. Using this definition, a first sample of
ray radiation (see e.g., Heise 1985; Bilk´ov´a et al. 2010). ∼70 WELQs was defined (Diamond-Stanic et al. 2009).
Thus in the present analysis we want to show that the Sinceatlowz itisnotalwayspossibletoapplytheorigi-
mid-IRcolorsoftheWISEcounterpartsofWDsisdiffer- nalcriterionontheEWr[Lyα+NV]becausetheseemis-
ent from that of BZBs. We expect that for WDs mid-IR sion lines lie outside the optical energy range covered by
emission arises from the presence of a companion brown the SDSS spectrograph an equivalent criterion was de-
dwarf or the presence of a dusty disk4, both expected to fined selecting WELQs on the basis of the EW of Mg
have thermal IR colors (Debes et al. 2011) while those II λ2800, C III] λ1909, and/ or C IV λ1549 (see e.g.,
of BZBs are dominated by the non-thermal synchrotron Meusinger & Balafkan 2014).
radiation of particles accelerated in their jets. When WELQs were discovered the possibility of be-
To achieve this goal we considered Sloan Digital ing classified BZBs was explored and discarded (see e.g.
Sky Survey DR7 White Dwarf Catalog (SDSSDR7WD) Plotkin et al. 2015, forarecentreview), howeveracom-
(Kleiman et al. 2013). This new catalog of spectroscop- parison of their mid-IR properties with those of con-
icallyconfirmedWDsfromtheSloanDigitalSkySurvey firmed BZBs has not yet been completely performed.
(SDSS) Data Release 7 (DR7) listing more than 19000 Here we selected a sample of WELQs to show that
stars. CrossmatchingtheSDSSDR7WDcatalogwiththe their mid-IR colors are generally similar to those of
latest release of the WISE catalog we found 4125 have a normal quasars but not to BZBs. Additional analyses
unique mid-IR potential counterpart within 3(cid:48)(cid:48).3. How- based on the SEDs and/or broadband spectral indices
ever, onlyfor255outof4125sourcestheirmid-IRcoun- (i.e., radio-to-optical αro and optical-to-Xray αox) were
terparts were detected by WISE in all three of its short- also used to distinguish between WELQs and BL Lacs
wavelength bands, and so we only do a comparison with (Plotkin et al. 2010; Lane et al. 2011; Wu et al. 2012)
this subset. when multifrequency observations are available. Thus
Weassumethatmid-IRcorrespondencesfoundforthe in the following sections we also investigate the broad
SDSSDR7WD are the real counterparts of the WDs. band SEDs of RWBL candidates selected on the basis
However in the worst scenario if none of the WDs listed of the combination of the mid-IR colors and the optical
in the SDSSDR7WD has mid-IR emission this will au- spectra.
tomatically confirm that their mid-IR spectral behav- We considered two catalogs of WELQs, spectro-
ior is different from that of Fermi BZBs known to date. scopically selected from the SDSS. The first catalog
lists 73 WELQs in the redshift range between 3.03
2 http://wise2.ipac.caltech.edu/docs/release/allwise/ and 5.09 (Diamond-Stanic et al. 2009, hereinafter DS9
3 http://www.asdc.asi.it/bzcat/
4 http://www.stsci.edu/∼debes/wired.html 5 http://www.astronomy.villanova.edu/WDCatalog/index.html
4 F. Massaro, E. Marchesini, R. D’Abrusco, N. Masetti, I. Andruchow & Howard A. Smith 2016
subsample) while the second lists 46 sources with
0.614< z <3.351 (Meusinger & Balafkan 2014, here-
inafter M14 subsample). We excluded 6 sources out of
these 119 WELQs, namely: SDSS J074451.36+292006.0 2 generic mid-IR sources
and SDSS J092145.37+233548.1 from the M14 sub- g-ray BZBs
g-ray BZQs
sample and SDSS J141318.86+450522.9 from DS9
TBLs
subsample because they are all listed in the Roma-
BZCAT and SDSS J104831.29+211552.2, SDSS
J115326.70+361726.3, SDSS J144204.03+132916.0 g) (2.59,1.06)
a
again from the M14 subsample since they show a flat m1
radio spectrum that is widely interpreted as jet signa- 6] (
4. (3.01,0.77)
tcuonretinsuoumsugmgeasrtkisngnotnh-athtertmheairl sqyunacshir-ofetarotunreelmesisssioopnt.ical 4] - [ (1.00,0.46)
3.
Only 90 out of the remaining 113 sources arising from [
(1.78,0.27)
the combination of the DS9 and the M14 subsamples
have a counterpart in the AllWISE catalog. These 90 0
sources thus constitute our WELQ reference sample to
carry out the comparison with BZB mid-IR colors. We
estimatedthechanceprobabilityofspuriousassociations 0 1 2 3 4 5
betweentheSDSSDR7WDcatalogandtheWISEsurvey [4.6] - [12] (mag)
as being lower than 1%.
3. BZBS,WDSANDWELQSATMID-IRWAVELENGTHS Fig. 1.—Themid-IRcolor-colorplotbuiltwiththeWISEmag-
nitudes at 3.4µm, 4.6µm and 12µm. BZBs and BZQs are dis-
In Fig. 1 we show the mid-IR colors built with the in
playedascyancirclesandorangesquaresrespectively,thuscorre-
the 3.4µm, 4.6µm and 12µm magnitudes of the BZBs spondingtothe“WISEGamma-rayStrip”(Massaroetal. 2011a).
and the BZQs in our Fermi blazar sample in com- Generic mid-IR sources (grey circles) of about 3000 sources se-
parison with generic infrared sources selected at high lected at high Galactic latitudes are also show. The separa-
tion between the mid-IR colors of the Fermi blazars and that of
Galactic latitudes (|b| >50◦), to highlight their pecu-
other sources randomly chosen appears evident. The presence of
liar spectral behavior. Since we know that BZQs gen- BZBs in the region of the “WISE Gamma-ray Strip” occupied
erally lie in the region occupied by “normal” quasars, at by the BZQs is only limited to a handful of sources (see e.g.,
the top end of the so called “WISE Gamma-ray Strip” D’Abruscoetal. 2013; D’Abruscoetal. 2014, for more details).
ThedashedpolygonmarkstheregionusedtoselectpotentialRW-
(Massaro et al. 2011a), to highlight the mid-IR color re-
BLs.
gion where BZBs reside, we will only consider RWBL
candidatesthoseWISEsourceslyinginthedashedpoly- inFig.2. Therearealsointhiscaseonlytwoobjectsthat
gonshowninFig.1wherethecontaminationoftheBZB could be confused as BZBs on the basis of their mid-IR
region by BZQ is lower than 5%, lacking a radio coun- colors, namely WISE J085506.13+063904.1 and WISE
terpart. It is worth noting that the selected region also J115642.21+130603.9, both lying in the footprint of the
includes TeV BL Lacs. SDSS. Thus in both cases we checked the optical images
Fig.2showsthecomparisonbetweenWISEsourcesas- usingtheSDSSNavigatetool6. Wefoundthatabout6(cid:48)(cid:48)6˙
sociated with WDs and our blazar sample. It is evident from the location of WISE J085506.13+063904.1 there
that only 4 out of the 255 sources in the SDSSDR7WD is a clear WD, namely SDSS J085506.62+063904.7, for
cataloghavemid-IRcolorsconsistentwiththemainBZB which an optical spectrum is available being most likely
region of the “WISE Gamma-ray Strip”, while the re- the star listed in the McCook & Sion WD catalog and
maining ∼99% of the WDs lie well separated from this lacking a WISE counterpart. This mis-association is
region. Threeoutoffoursourcesareextragalactic,while due to the positional uncertainty of the sources in the
WISEJ122859.87+104032.8hasaclearWDopticalspec- McCook & Sion WD catalog. Thus we conclude that
trum, making it the only WD contaminant of the BZB WISEJ085506.13+063904.1isasimplecontaminantun-
region in the mid-IR color-color diagram. Moreover related to the WD selection. Then for the latter, WISE
WISE J122859.87+104032.8 has a negative value of the J115642.21+130603.9 the lack of its optical spectrum
u−r a situation that never occurs for the known BZBs combined with the uncertainties of the SDSS photomet-
(Massaro et al. 2012c; Massaro et al. 2014b). WISE ric observations does not allow us to verify its nature.
J172633.50+530300.3 also shows a featureless optical However, this object is classified as galaxy thus unlikely
spectrum with a peculiar continuum shape; this could to be a WD that can contaminate our RWBL selection.
be due to the presence of a nearby source clearly visible The comparison between mid-IR colors of BZBs and
in the SDSS image but not in the WISE one. This im- WDsprovetheusefulnessofthemid-IRcolor-diagramto
pliesarelativelysmallcontaminationofthemid-IRcolor distinguish the two classes while searching for RWBLs,
selection. once optical spectra are available.
Both WISE J001736.91+145101.9 and WISE Fig. 3 shows the mid-IR comparison between the
J121348.83+642520.0 are BL Lacs with a radio WELQs and the BZBs. Here too is clear the neat sepa-
counterpart in addition to WISE J121348.83+642520.0 ration between these two classes. It is also worth noting
listed in the latest release of the Roma-BZCAT catalog. that sources in the DS9 subsample appear to lie in a dif-
To complete our comparison between WDs and BZBs ferent region of the color parameter space with respect
we also show the mid-IR color-color plot with the 91
sources selected out of the McCook & Sion WD catalog 6 http://skyserver.sdss.org/dr12/en/tools/chart/navi.aspx
Radio Weak BL Lac objects in the Fermi era 5
2 2
generic mid-IR sources generic mid-IR sources
g-ray BZBs g-ray BZBs
g-ray BZQs g-ray BZQs
WDs WDs
1.5 1.5
g) g)
a a
m 1 m 1
6] ( 6] (
4. 4.
4] - [ 0.5 4] - [ 0.5
3. 3.
[ [
0 0
-0.5 -0.5
-0.5 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 -0.5 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
[4.6] - [12] (mag) [4.6] - [12] (mag)
Fig. 2.— The comparison between the mid-IR colors of the BZBs and the BZQs as shown in Fig. 1 and those of the WDs (black and
blue circles) from the SDSSDR7WD (left panel) and the McCook & Sion WD (right panel) catalogs, respectively. In both comparisons
themid-IRcolorsofblazarsappearwelldistinctfromthoseofWDs,withtheonlyfewexceptionsdiscussedin§3 andmarkedbythered
arrows. Genericmid-IRsourcesarealsoshowninthebackgroundinbothpanels. Uncertaintiesonthemid-IRcolorsofBZBsandBZQs
arenotreportedhereasinFig.1forsakeofsimplicitysincetheyareshownhereforcomparisononly.
In this case we highlight the fact that two WELQs lie
belowthethresholdmarkingtheBZBregionofthecolor-
color plot and additional three objects are close to it.
2 generic mid-IR sources These two sources are WISE J001514.87-103043.4 and
g-ray BZBs WISE J150427.69+543902.2; both showing MgIIλ2798
g-ray BZQs
emissionlineofrest-frameequivalentwidthEW consis-
WELQs (M14) r
1.5 WELQs (DS9) tent with the 5˚A threshold, usually adopted to classify
BL Lacs, within 3σ, thus being “border line cases”. The
ag) remaining three sources, belonging to the M14 subsam-
m 1
6] ( phlaeve(saeeraFdigio. 4c,ouwnhteerrepatrhteyata1re.4shGoHwzniinntmheagFeanitnat),Imal-l
4.
4] - [ 0.5 aHgeelsfanofdtehtealR.a2d0i1o5)SkcyataaltogT,wbeunttythceenlatcimkeotferad(dFiItRioSnTa;l
3.
[ radio data did not allow us to verify the flatness of their
radiospectra,althoughtheyaredefinitivelynotRWBLs.
0
Thelocationofallfivesourceswithrespecttothatofthe
BZBs in the mid-IR color-color plot is shown in Fig. 4.
Finally,weconcludethatbothWDsandWELQshave
-0.5
-0.5 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 distinct mid-IR colors with respect to the BL Lac pop-
[4.6] - [12] (mag) ulation and [3.4]-[4.6]-[12]µm color-color diagram can be
usedasdiagnostictooltoconfirmthesourcenaturehav-
ing featureless optical spectra. Sources that appear to
Fig. 3.— Thecomparisonbetweenthemid-IRcolorsoftheBZBs
be contaminants of the mid-IR selection procedure are
and the BZQs as shown in Fig. 1 and those of the WELQs. The
WELQs belonging to the M14 subsample are shown in magenta ruled out when WISE data are combined with optical
triangles(up)whilethoseoftheDS9subsampleinblack(down). In spectroscopic observations.
bothcomparisonsthemid-IRcolorsofblazarsappearwelldistinct
from those of WELQs, with the only few exceptions discussed in 4. RADIOWEAKBLLACS
§3 andmarkedherebytheblackarrows. Genericmid-IRsources
are also shown in the background in both panels. Uncertainties In the literature there are five sources tentative classi-
on the mid-IR colors of BZBs and BZQs are not reported here fied as RWBLs.
as in Fig. 1 for sake of simplicity since they are shown here for
comparisononly. 1. 2QZ J215454.3-305654 (Londish et al. 2002) for
which several multifrequency follow up obser-
totheM14subsample. Thereasonunderlyingthiseffect
vations have been performed to date was the
is due to their different redshift distributions. We note
first. However its nature is still uncertain since
that there is also a marginal difference between the mid-
it could be a RWBL or a “lineless” quasar
IR colors of the WELQs in the M14 sample and that
(Londish et al. 2004).
of the BZQs but it is less useful to search for RWBLs
hidden among this sample. 2. Paggi et al. (2014), carried out optical spec-
6 F. Massaro, E. Marchesini, R. D’Abrusco, N. Masetti, I. Andruchow & Howard A. Smith 2016
troscopic observations of blazar candidates, po- latest releases of the Fermi catalogs is typically grater
tential UGS counterparts and discovered WISE than 99.5% (Ackermann et al. 2015).
J064459.38+603131. This mid-IR sources lies Applying this mid-IR color-based analysis, in com-
within the positional uncertainty region of 2FGL bination with optical spectra, we conclude that WISE
J0644.6+6034.7 (3FGL J0644.6+6035). This ob- J064459.38+603131andWISEJ141046.00+740511.2,ly-
jectshowsasingleopticalemissionlineofEWcon- ing within the positional uncertainty regions of two dis-
sistentwiththe5˚Athreshold, arbitrarilysetinthe tinct UGSs listed in the 3FGL, are the first two RW-
literaturefortheBLLacclassdefinition,similarto BLsdetectedsofar. Bothofthemcanalsobeassociable
the two cases discussed at the end of § 3. with1FHLsourcesthusstrengtheningourinterpretation,
since the high energy sky above 10 GeV is mostly domi-
3. In 2015 WISE J173052.85-035247.2 was discov- nated by BZBs. In the following we will also show that
ered during the optical spectroscopic follow up additional multifrequency observations can strengthen
observations of UGSs (Ricci et al. 2015). This our results.
source could be the potential counterpart of 2FGL
J1730.6-0353 (a.k.a. 3FGL J1730.6-0357). 5. MID-IRVARIABILITY
BL Lac objects are known to be among the most
4. More recently Landoni et al. (2016) claimed that variable extragalactic sources at all frequencies, showing
alsoSDSSJ004054.65-0915268couldbeatentative variationseitherinspectralshapeorinflux/luminosities.
RWBL. Thispeculiarbehaviorcouldinprincipleminethemid-IR
color selection proposed here and in our previous works.
5. Finally, Marchesini et al. (2016) discovered However this situation does not seem to occur.
the last known candidate of RWBL, WISE BL Lacs appear to be variable in the mid-IR band
J141046.00+740511.2. This source was also (see e.g., D’Abrusco et al. 2012) although their color
found while carrying out spectroscopic obser- variations still reside within the boundaries of the
vations of UGS listed in the First Fermi- “WISE Gamma-ray Strip” (see Buson et al. 2016, for
LAT Catalog of Sources above 10 GeV (1FHL; more details). Here we show an example of two BZBs:
Ackermann et al. 2013) selected on the basis of 5BZBJ1542+6129 and 5BZBJ1959+6508, the latest also
theSWIFTX-rayobservations(Landi et al. 2015). detected at TeV energies (Aharonian et al. 2003), for
It has been indicated as potential counterpart of which more than 50 WISE observations are available,
1FHL J1410.4+7408 (a.k.a. 3FGL J1410.9+7406). when both sources are detected at least in the first three
mid-IR filters at 3.4µm, 4.6µm and 12µm. In Fig. 5 it
In Fig. 4 we show the positions of the five RW- is clear how both BZBs during their single-epoch WISE
BLs overlay to the “WISE Gamma-ray Strip” subre- observations move in the mid-IR color-color plot in the
gion of BZBs. SDSS J004054.65-0915268 (a.k.a WISE same region occupied by the BL Lac objects. In addi-
J004054.63-091526.8) has an upper limit at 12µm and tion their colors are also consistent with the boundaries
beingahighz sourceitseemstobemoreconsistentwith chosen for our RWBL selection procedure.
thelocationoftheDS9subsampleofWELQsratherthan AccordingtotheWISEcatalogoneofourbestRWBL
the BZB population. A similar situation occurs for 2QZ candidate: the WISE J141046.00+740511.2 is flagged as
J215454.3-305654(a.k.a. J215454.32-305653.2)thatonly variable in the mid-IR. This is in agreement with the
a follow up observations at 12µm could potentially re- fact that such emission it is unlikely due to dust emis-
veal if the source is consistent with the mid-IR colors of sion, as occurs in normal quasars or in white dwarfs
the BZBs. WISE J064459.38+603131.7, being selected (Debes et al. 2011). Fig. 6 shows the WISE magni-
on the basis of its mid-IR colors is definitively consis- tude at 3.4µm, 4.6µm and 12µm as function of time
tent with the BZB population on the 3.4µm - 4.6µm for two different epochs; mid-IR variability, not only
- 12µm diagram, while WISE J173052.85-035247.2 ap- between the two observing periods, but also in each
pears to have such colors more similar to the WELQs single set of observations is found. We also present
listed in the M14 subsample. the lightcurve for the second RWBL candidate WISE
The remaining case of WISE J141046.00+740511.2 J064459.38+603131 that clearly shows infrared variabil-
is the most intriguing one. This object is potentially ity being detected more times at 12µm during the
associated to a 1FHL object and its potential WISE WISE single-epoch observations with respect to WISE
counterpart lies in the region of the “WISE Gamma- J141046.00+740511.2. In Fig. 7 we also show their mid-
ray Strip” where all known TeV BL Lacs are located IR color variability as function of time. In the same fig-
(Massaro et al. 2013d; Arsioli et al. 2015). The lack of ure we also present the single epoch observations when
any radio counterpart in both the NRAO VLA Sky Sur- both sources are detected in the first three WISE fil-
vey (NVSS; Condon et al. 1998) and the FIRST cata- ters overlaid to the “WISE Gamma-ray Strip”. As
log (Helfand et al. 2015) that cover the field of WISE for the other two BZBs mentioned above these varia-
J141046.00+740511.2makesthissourcethemostpromis- tions are clearly consistent with the region chosen to se-
ing candidate to be the first, genuine, RWBL. Neverthe- lect RWBLs for WISE J141046.00+740511.2 while for
less is it worth emphasizing that the distance between WISE J064459.38+603131 there is partial agreement.
thismid-IRsourceandthegamma-raypositionof3FGL It is worth mentioning that the location of integrated
J1410.9+7406 is ∼90(cid:48)(cid:48)well below the average of the as- mid-IR colors for both WISE J064459.38+603131 and
sociatedFermiblazarslistedintheRoma-BZCATcorre- WISE J141046.00+740511.2 could appear different from
sponding to ∼130(cid:48)(cid:48). At the angular separation of ∼90(cid:48)(cid:48), that where they exhibit variability in the mid-IR color
the association probabilities of counterparts listed in the diagram. The underlying reason is that there are
Radio Weak BL Lac objects in the Fermi era 7
1.6 1.6
g-ray BZBs g-ray BZBs
g-ray BZQs g-ray BZQs
1.4 1.4 RWBLs known
J172633.50+530300.3 J173052.85-035247.2
1.2 1.2
g) J001736.91+145101.9 g)
a 1 a 1 J064459.38+603131.7
m m
4] - [4.6] (00..68 J121348.83+642520.0 J15J00402175.6194+.8574-31900320.423.4 4] - [4.6] (00..68 J141046.00+740511.2 J215454.32-305653.2
3. 3.
[ J122859.87+104032.8 [
0.4 0.4
J004054.63-091526.8
0.2 0.2
0 0
0.8 1.2 1.6 2 2.4 2.8 3.2 3.6 4 4.4 0.8 1.2 1.6 2 2.4 2.8 3.2 3.6 4 4.4
[4.6] - [12] (mag) [4.6] - [12] (mag)
Fig. 4.— Left panel) The mid-IR color-color plot where the positions of the blazars recognized in the WD and the WELQ samples
are highlighted with respect to the Fermi blazars. WISE J122859.87+104032.8 and WISE J172633.50+530300.3 are marked with a red
arrow since additional optical information allow us to discard them as potential RWBLs. The former has the unusual negative value of
theu−r neverseenforBLLacswhilethemid-IRcolorsofthelatterarecontaminatedbythepresenceofanearbysourcevisibleinthe
optical image (see 3 for more details). Right panel) The positions of the candidates RWBLs in the mid-IR color-color plot are shown in
comparisontothatofFermiblazarsbelongingtothe“WISEGamma-rayStrip”andthoseofgenericmid-IRsources(samecolorcodeof
previousfigures). Sourcedetailsfortheobjectsindicatedinbothpanelsaregivenin§3. Uncertaintiesonthemid-IRcolorsofBZBsand
BZQsarenotreportedhereasinFig.1forsakeofsimplicitysincetheyareshownhereforcomparisononly.
ing that for the two RWBL candidates the uncertainties
on the WISE magnitudes are greater than for the two
bright BZBs selected above as example.
1.5 g-ray BZBs On the basis of this investigation of mid-IR variabil-
5BZB J1542+6129 (a.k.a.3FGL J1542.9+6129) ity and spectral behavior we conclude that the evidence
5BZB J1959+6508 (a.k.a. 3FGL J2000.0+6509)
1.25 supports the interpretation of these sources as RWBLs.
(2.59,1.06)
6. SWIFTOBSERVATIONS
ag) 1 Several analyses were carried out to study the nature
m
4.6] (0.75 (3.01,0.77) oWfuWeEtLaQl.s2i0n12t)h.e XT-hreasyeso(bseseerev.agt.i,oSnhsewmemreeraelstoalu.se2d00t9o;
4] - [ bSuEiDlds t(hseeeire.bgr.o,aPdlobtkanindetspael.ct2r0a1l0i)ndtoicevseraifnyd/thoreirthpeoir-
[3. 0.5 tential similarities with BL Lac objects. Since the two
(1.00,0.46)
sourcesidentifiedhereasRWBLswerediscoveredduring
0.25 a campaign aimed to search for UGS counterparts and
(1.78,0.27)
all the Fermi UGSs are in a SWIFT observational pro-
gram (Stroh et al. 2013)7, we reduced the SWIFT data
0
0.8 1.2 1.6 2 2.4 2.8 3.2 3.6 4 available to date to build the RWBL SEDs.
[4.6] - [12] (mag) WISE J064459.38+603131 was observed by SWIFT
both with the X-Ray Telescope (XRT), for ∼ 3 ksec,
and with the Ultraviolet/Optical Telescope (UVOT) in-
Fig. 5.— The mid-IR color-color plot built with the WISE
struments on board but for the latter the only ultravi-
magnitudes at 3.4µm, 4.6µm and 12µm. BZBs are displayed as
cyan circles and correspond to the “WISE Gamma-ray Strip” olet filter used was M2. On the other hand, for WISE
(Massaroetal. 2011a). We show two cluster of points rel- J141046.00+740511.2, there are optical data in the U
ative to the WISE single epoch observations of the BZBs: bandandalltheUVfiltersavailable,togetherwithabout
5BZBJ1542+6129 (black circles) and 5BZBJ1959+6508 (red cir-
11 ksec of X-ray data.
cles). Itappearsevidentthatevenmid-IRcolorvariationstatesof
these two BL Lac objects are always consistent with the bound- Here adopted the standard XRT and UVOT data re-
aries of the “WISE Gamma-ray Strip”. The two crosses point to ductionprocedure,extensivelyusedinourpreviousanal-
thelocationof5BZBJ1542+6129(magenta)and5BZBJ1959+6508 yses (see e.g. Massaro et al. 2008a; Maselli et al. 2013;
(orange) on the mid-IR color diagram when integrated values for
Maselli et al. 2016, and references therein) while only
theentireWISEall-skysurveyareused.
the very basic details are described below. XRT data
manyWISEsingle-epochobservationswhensources(see were processed with the XRTDAS software package
Fig. 6) are not detected that contribute to the color
change for the integrated values. It is also worth not- 7 http://www.swift.psu.edu/unassociated/
8 F. Massaro, E. Marchesini, R. D’Abrusco, N. Masetti, I. Andruchow & Howard A. Smith 2016
WISE J064459.38+603131.7 WISE J141046.00+740511.2
e14 e15
d d
u u
nit13 nit14
ag12 3.4 m m ag 3.4 m m
m 4.6 m m m13 4.6 m m
IR 11 12 m m IR 12 12 m m
mid-10 mid-11
9
55276 55277 55278 55279 55280 55314 55314 55315 55316 55316 55316
Modified Julian Date Modified Julian Date
14.5 16
e e
d d
nitu 14 nitu15.5
ag 3.4 m m ag 15
m13.5 4.6 m m m
R R 14.5
mid-I 13 mid-I 14 34..46 mm mm
12.5 13.5
55467 55468 55469 55470 55471 55472 55506 55506 55506 55507 55508
Modified Julian Date Modified Julian Date
Fig. 6.— The light curve of WISE J064459.38+603131 (left panel) and WISE J141046.00+740511.2 (right panel), candidate RWBLs,
tentativelyassociatedto3FGLand1FHLFermisources. TheseareWISEsingle-epochobservationsfortwodifferenttimeperiods. Mid-IR
flux density is expressed using the WISE magnitude (not corrected for absorption), as function of time in Modified Julian Date. Each
epoch(topandlowerpanel)correspondtoapproximately2days. Infraredvariabilityisclearlynotableforbothsources. Trianglesareused
tomarkWISEobservationsthatcorrespondstoupperlimitsatagivenwavelength.
[3.4]-[4.6] mag
[4.6]-[12] mag
3.5 1.5
g-ray BZBs
3 WISE J064459.38+603131.7
or WISE J141046.00+740511.2
ol2.5 1.25
c
R 2 WISE J064459.38+603131.7
I (2.59,1.06)
d-1.5
mi g) 1
1 a
m
05.55276 55277 Modifie5d5 J2u7l8ian Date 55279 55280 4.6] (0.75 (3.01,0.77)
3.45 4] - [
olor2.53 [3. 0.5 (1.00,0.46)
c
R 2
I WISE J141046.00+740511.2
d-1.5 0.25 (1.78,0.27)
mi 1
0.5
0 0
55314 55315 55316 0.8 1.2 1.6 2 2.4 2.8 3.2 3.6 4
Modified Julian Date [4.6] - [12] (mag)
Fig. 7.—Leftpanel)Mid-IRcolorsreportedasfunctionoftimeinModifiedJulianDateforbothRWBLselectedinouranalysis: WISE
J064459.38+603131(top)andWISEJ141046.00+740511.2(bottom). ForWISEJ064459.38+603131weonlyconsideredWISEobservations
whenthesourceisdetectedinallthethreefilterswhileforWISEJ141046.00+740511.2wecomputedthe[4.6]-[12]µmcolor-lightcurvealso
using the upper limits (red triangles down). Right panel) Same of Figure 5 but for the two RWBL selected: WISE J064459.38+603131
(purplesquares)andWISEJ141046.00+740511.2(magentadiamond). Thevariationsfoundinthesingle-epochWISEobservationswhen
bothsourcesaredetected inthefirstthreemid-IRfiltersisclearlyconsistentwiththeboundariesoftheportionofthe“WISEGamma-ray
strip”forWISEJ141046.00+740511.2andpartiallyinagreementforWISEJ064459.38+603131.
Radio Weak BL Lac objects in the Fermi era 9
(v.3.0.0) developed at the ASI Science Q9 Data Center
(ASDC) and distributed within the HEASOFT package
(Rwvee.rs6ee.1ac8rac)hrbrCiyeedtnhtoeeuNrt(AHinSEAtAhHSeiAgphRhCoEt)no.enArglclyotuAhnesttXirnoRgpTh(yPosCbics)serrAvearactdhiooinuvest 44:00.042:00.0 SWIFT - XRT 44:00.042:00.0 SWIFT - UVOT (M2)
mode. Event files were calibrated and cleaned apply- 40:00.0 40:00.0
ing standard filtering criteria with the XRTPIPELINE 38:00.0 38:00.0
taskandusingthelatestcalibrationfilesavailableinthe 36:00.0 36:00.0
SwwitihftgCraAdLeDs B0-.1E2vwenertes isneletchteedenienrgtyheraanngaely0s.i3s–a1c0cokredV- 34:00.0 34:00.0
ingtothestandardguidelines. Exposuremapswerealso 32:00.0 32:00.0
createdusingXRTPIPELINE.TheSWIFTobservations 60:30:00.0 60:30:00.0
available for both sources have variable exposures, we 28:00.0 28:00.0
mergedcleanedeventfilesobtainedwiththeaboveproce- 26:00.0 26:00.0
dureusingXSELECTtaskandconsideringonlySWIFT
30.0 45:00.0 30.0 6:44:00.0 30.0 43:0030.0 45:00.0 30.0 6:44:00.0 30.0 43:00
observationswithtelescopeaimpointfallinginacircular
region of 12 arcmin radius centered in the median of the
individual aim points, to get a uniform exposure. The
corresponding merged exposure maps were then gener-
ated by summing the exposure maps of the individual
observations with XIMAGE package. Source detection
ofin theXRTimages wasperformed usingthedetection 18:00.0 SWIFT - XRT 18:00.0SWIFT - UVOT (U)
algorithm DETECT. 16:00.0 16:00.0
We collected 13 counts for WISE J064459.38+603131 14:00.0 14:00.0
in the merged XRT event file while there are 36 counts 12:00.0 12:00.0
fwnouermllWbdeeIrtSeEoctfeJdc1o4au1bn0ot4sv6e.a0tv0ha+eil7a54bσ0le5t,h1r1ce.o2srh,roemlsdpa.oknHindogiwngebvoettorhtahseocuporuocnoesrt 10:00.008:00.0 10:00.008:00.0
rate of 0.004 cts/s and 0.003 cts/s, respectively, did 06:00.0 06:00.0
not permit us to carry out a detailed spectral anal- 04:00.0 04:00.0
ysis. Thus assuming an absorbed power-law model 02:00.0 02:00.0
wdtirmiothgaetenadpchotholuetmonunniandbdesenoxsribtoyefd2(K.X2a-larbaneydrlaflwueitxthaolt.fhe2i0nG05at)hlaecatn0idc.5h-e1ys0-- 74:00:00.03:58:00.0 13:00.0 30.0 12:00.0 30.0 11:00.0 30.0 14:10:00.0 30.0 09:00.0 74:00:00.03:58:00.0 13:00.0 30.0 12:00.0 30.0 11:00.0 30.0 14:10:00.0 30.0 09:00.0
keV for WISE J064459.38+603131 and of for WISE
J141046.00+740511.2 using WebPIMMS8.
Details on the UVOT reduction procedure used here Fig. 8.—Toppanels)TheXRT(left)andtheUVOT(M2filter)
are described in Massaro et al. (2008b) and Maselli et imagesofthefieldwhereWISEJ064459.38+603131lies. Theyel-
low dashed ellipse marks the positional uncertainty region of the
al. (2008). Weperformedthephotometricanalysisusing
closest 2FGL unassociated source while the white one the error
a standard task UVOTSOURCE available in the HEA- ellipse from the 3FGL catalog, both at 95% level of confidence;
SOFT package (v.6.18). Counts were extracted from a green circles mark the location of the X-ray counterpart. Bottom
panel) Same of top panels for WISE J141046.00+740511.2. The
5 arcseconds radius circular region in the V, B, and U
UVOT image in this case is in the optical energy range (i.e., U
filters as well as for the UV filters (UVW1,UVM2, and filter)insteadoftheUV.
UVW2). The count rate was corrected for coincidence
loss and the background subtraction was performed by fidence, from the 2FGL and the 3FGL catalogs for the
estimating its level in an offset annular region centered nearby UGSs. WISE J064459.38+603131 lies within the
on the source position with inner radius of 7.5 arcsec- error ellipse of the 2FGL, as occurred when it was se-
onds and outer radius of 15 arcseconds. Since estimate lected (Massaro et al. 2012a) but not within the one of
of magnitude errors is complex due to possible instru- the 3FGL, while WISE J141046.00+740511.2 that is as-
mentalsystematicsandcalibration,hereweonlyshowed sociabletothe3FGLsource,itwasnotfoundapotential
a typical uncertainty of 5% for all filters. counterpart of the 2FGL UGS.
The values of the observed AB magnitude for WISE SWIFTobservations,bothintheUVandintheX-ray
J064459.38+603131 in the UV band at 1.36×1015Hz energyrangesareusedinthefollowingtobuildtheSEDs
is 19.69 mag (M2 UVOT filter) while for WISE for both the RWBLs selected.
J141046.00+740511.2 the AB magnitudes measured in
the UVOT filters are: 19.23 mag (U), 19.85 mag (W1), 7. SPECTRALENERGYDISTRIBUTIONS
20.24 mag (M2) and 20.31 mag (W2), respectively. For the two RWBL selected we also built the SEDs
In Fig. 8 we show the XRT event file in the 0.3-10 to check if they are similar or not to those of WELQs
keV energy range and one of the UVOT images for both (Lane et al. 2011). AnalysisoftheSEDsisequivalentto
the RWBL selected. We overlaid to the SWIFT images comparing broad band spectral indices that could give
the positional uncertainty region, at 95% level of con- an additional indication of the differences between BL
Lacs and WELQs (see e.g Wu et al. 2012).
8 https://heasarc.gsfc.nasa.gov/cgi- In the case of WISE J064459.38+603131, the source
bin/Tools/w3pimms/w3pimms.pl has an upper limit in the NVSS catalog of 0.13
10 F. Massaro, E. Marchesini, R. D’Abrusco, N. Masetti, I. Andruchow & Howard A. Smith 2016
mJy/beam while WISE J141046.00+740511.2 1.69 • In addition to them 2QZ J215454.3-305654 could
mJy/beam. The former is detected in all the WISE alsopotentiallybeaRWBL,butlackingofthemid-
bandsandinthe2MASS.FromtheSWIFTobservations IR detection at 12µm it is not possible to confirm
we also got an estimate of the UV flux density in the its nature using the mid-IR color-color plot. In
M2 filter and the X-ray flux. The latter source still has additionthetwocontaminants: WISEJ001514.87-
a WISE counterpart detected at all mid-IR frequencies 103043.4 and WISE J150427.69+543902.2, found
but it does not have a 2MASS correspondence. SWIFT in the WELQ sample used here, deserve a deeper
observationsprovidedusfluxdensityestimatesintheop- investigation to clarify their nature.
tical(U)andinalltheUVfilters(W1,M2,W2)together
with the X-ray one.
Wehighlightthatthestrategyusedtosearchforpoten-
Wecorrectedthealltheinfrared, opticalandultravio-
tial counterparts of UGSs via X-ray follow up observa-
letfluxdensitiesfortheGalacticabsorptionfollowingthe
tions has been crucial in both cases to discover WISE
Draine(2003)andthereddeningcorrectionsofSchlafly&
J064459.38+603131 and WISE J141046.00+740511.2,
Finkbeiner(2011),withtheonlyexceptionfortheWISE
thus emphasizing that combining X-ray and mid-IR ob-
magnitudes at 12µm and 22µm.
servations could be crucial for future searches of low en-
In Fig. 9 we show the SEDs for both WISE
ergy counterparts of gamma-ray sources. However, we
J064459.38+603131 and WISE J141046.00+740511.2.
also stress that such discovery has been done thanks
For the former SED the infrared emission ap-
to optical spectroscopic follow up campaigns that fi-
pear to be compatible with a log-parabolic shape
nally revealed the source nature. Since both these two
(Massaro et al. 2004;Massaro et al. 2006)withtheonly
sources are not only associable with 3FGL objects but
exception of the UVOT flux density in the M2 filter.
alsowithFermisourcesdetectedabove10GeVtheirexis-
This could be due to the presence of the host galaxy
tencechallengescurrentgamma-rayassociationmethods
and/or to variability since UV and IR observations were
mostlybasedonradiosurveysthatdonotallowtocatch
not taken simultaneously and both of them are inte-
these potential counterparts. The existence of RWBLs
grated over different epochs. The latter source: WISE
will also strong impacts on jet physics since they could
J141046.00+740511.2 has also an SED well described by
allowustostudysourceswhoseemissionisdominatedby
a log-parabolic shape but in this case the SWIFT opti-
relativistic jets that do not tend to form extended radio
cal and UV data seems to mark the presence of an host
structures as their parent population of radio galaxies.
galaxy. In both panels of Fig. 9 we also report the un-
Finally,weremarkthatdeeperradiofollowupobserva-
absorbed X-ray flux computed from the source counts of
tionsthanthoseavailabletodatecouldpotentiallyreveal
the XRT event file. Since this X-ray flux is an estimate
their low energy emission at GHz frequencies. However
we neglect second order corrections to convert the X-ray
the ratio between the infrared and the radio emission of
flux into a νF SED value.
ν
socalledRWBLs,beingbelowthefluxlimitofcurrentra-
The agreement between the multifrequency data and
dio surveys is unexpectedly unusual. In Fig. 10 we show
the broad band log-parabolic SED support the in-
the distribution of such ratio to highlight how the two
terpretation of WISE J064459.38+603131 and WISE
selected RWBLs lie in the tail of the BL Lac population
J141046.00+740511.2 as RWBLs.
being below the 0.5 value.
8. SUMMARYANDCONCLUSIONS
• The mid-IR colors of BL Lacs appear to be clearly
different from those of WDs and WELQs. This We thank the anonymous referee for useful comments
strongly indicate that the WISE color-color dia- that led to improvements in the paper. F.M. grate-
gram built with the mid-IR magnitudes at 3.4µm, fully acknowledges the financial support of the Pro-
4.6µm and 12µm together with optical spectro- gramma Giovani Ricercatori – Rita Levi Montalcini –
scopic observations are a powerful diagnostic tool Rientro dei Cervelli (2012) awarded by the Italian Min-
to distinguish these three source classes. We only istry of Education, Universities and Research (MIUR).
found few contaminants within the WD and the H.A.S.acknowledgespartialsupportfromNASAGrants
WELQ catalogs used in our analysis. However, for NNX15AE56G and NNX14AJ61G. This research has
all these cases, the combination of mid-IR and op- made use of data obtained from the high-energy Astro-
ticalspectroscopicdataallowedustoexcludethem physics Science Archive Research Center (HEASARC)
as BL Lacs. provided by NASA’s Goddard Space Flight Center. The
NASA/IPACExtragalacticDatabase(NED)operatedby
• Applying this mid-IR color-based analysis we con- the Jet Propulsion Laboratory, California Institute of
clude that WISE J064459.38+603131 and WISE Technology, under contract with the National Aeronau-
J141046.00+740511.2, lying within the positional ticsandSpaceAdministration. Partofthisworkisbased
uncertainty regions of two distinct UGSs, are the on the NVSS (NRAO VLA Sky Survey): The National
firsttwoRWBLsdetectedsofar. Bothofthemare RadioAstronomyObservatoryisoperatedbyAssociated
also associable with 1FHL sources thus strength- Universities, Inc., under contract with the National Sci-
ening our interpretation, since the high energy sky ence Foundation and on the VLA Low-frequency Sky
above10GeVismostlydominatedbyBZBs. Both Survey(VLSS).Thispublicationmakesuseofdataprod-
sources appear to be variable in the mid-IR as ex- uctsfromtheWide-fieldInfraredSurveyExplorer,which
pect for BZBs and they also show a broad-band is a joint project of the University of California, Los An-
SEDs, built thanks to the WISE and the SWIFT geles, and the Jet Propulsion Laboratory/California In-
observations,consistentwithalog-parabolicshape. stitute of Technology, funded by the National Aeronau-