Unveiling the submerged part of the iceberg: 6 radio-loud narrow-line Seyfert 1s with SKA 1 0 2 b e F M. Berton∗1, L.Foschini2, A.Caccianiga2, J.L.Richards3,S. Ciroi1, E.Congiu1, V. 4 Cracco1, G.LaMura1,L.Marafatto4,P. Rafanelli1 2 1 DipartimentodiFisicaeAstronomia"G.Galilei",UniversitàdiPadova,Vicolo ] dell’osservatorio3,35122,Padova,Italy A 2 INAF-OsservatorioAstronomicodiBrera,ViaE.Bianchi46,23807,Merate(LC),Italy; G 3 DepartmentofPhysicsandAStronomy,PurdueUniversity,525NorthwesternAvenue,West . Lafayette,IN47907,USA; h p 4 INAF-OsservatorioAstronomicodiPadova,Vicolodell’osservatorio5,35122,Padova,Italy. - E-mail: [email protected] o r t s Narrow-line Seyfert 1 galaxies (NLS1) are active galactic nuclei (AGN) known to have small a [ massesofthecentralblackholeandhighaccretionrates. NLS1saregenerallyradio-quiet,buta smallpartofthem(about7%)areradio-loud.Therecentdiscoveryofpowerfulrelativisticjetsin 2 v radio-loudNLS1s(RLNLS1s),emittingathigh-energyg -rays,openedintriguingquestions. The 1 observedluminosityofthejetisgenerallyweak,smallerthanblazars,althoughwhenrescaledfor 9 7 themassofthecentralblackhole,itbecomesofthesameorderofmagnitudeofthelatter. The 5 weakluminosity,andhenceobservedflux,resultedinasmallnumberofknownRLNLS1. From 0 . a recentsurveyof RLNLS1s, it was foundthat only 8 out of 42 sourceshad radio flux density 1 0 at 1.4 GHz greater than 100 mJy, while 21 out of 42 had flux density smaller than 10 mJy. In 6 addition,giventhestrongvariabilityatallwavelengths,withpresent-dayfacilitiesRLNLS1scan 1 : oftenonlybedetectedduringhighactivityperiods. TheSquareKilometerArray(SKA),withits v i superiorsensitivity,willbreakthislimit,allowingustounveilarelativelyunknownpopulationof X jettedAGN.Wepresenttheresultsofastudyaimedatevaluatingthescenariothatcouldemerge r a aftertheadventofSKA. EXTRA-RADSUR2015(*) 20–23October2015 Bologna,Italy (*)ThisconferencehasbeenorganizedwiththesupportoftheMinistryofForeignAffairsand InternationalCooperation,DirectorateGeneralfortheCountryPromotion(BilateralGrantAgreement ZA14GR02-MappingtheUniverseonthePathwaytoSKA) ∗Speaker. (cid:13)c Copyrightownedbytheauthor(s)underthetermsoftheCreativeCommons Attribution-NonCommercial-NoDerivatives4.0InternationalLicense(CCBY-NC-ND4.0). http://pos.sissa.it/ RLNLS1swithSKA M.Berton 1. Introduction Narrow-line Seyfert 1galaxies (NLS1s)are aclass ofactive galactic nuclei (AGN)that since its classification [18] has always presented new, intriguing challenges for AGN physics and mod- elling. Theyareclassifiedonthebasisofthefullwidthathalfmaximum(FWHM)oftheHb line, FWHM(Hb ) < 2000 km s−1, and of the ratio between [O III]/Hb < 3. The width of permitted lines is comparable to that of type 2AGN, but the presence of strong FeII multiplets in the spec- trum reveals that the BLRis actually visible, as in type 1AGN. Thelines narrowness is therefore attributed toalowrotational velocity around arelatively lowmassblack hole(106-108 M ,[17]). ⊙ Their bolometric luminosity is similar to that of broad-line Seyfert 1, and this translates in a very highEddingtonratio,possiblyasignofaveryhighaccretionrate[5]. Forallthesereasons,NLS1s aresometimesconsidered ayoungandfastlygrowingphaseintype1AGNevolution [13,17]. In radio NLS1s do not usually exhibit any intense emission. Indeed 93% of them are radio- quiet, while the remaining 7% is radio-loud [15]. Some of these radio-loud NLS1s (RLNLS1s) showseveralblazar-likeproperties,suchasahighbrightnesstemperatureandaflat-radiospectrum [22], and in recent years the Fermi Satellite discovered g -ray emission coming from 10 of them, indicating the presence of a relativistic beamed jet [1, 11, 8, 21, 16]. Misaligned jets were also found in a few steep radio-spectrum NLS1s (S-NLS1s, e.g. [6, 19]), and these steep-spectrum sources werepointed outaspartofthebeamedsourcesparentpopulation [4]. Oneofthebiggestissuesinthestudyofradio-emitting NLS1sisthesmallnumberofsources available. As shown in the following section, we found only 149 radio-emitting NLS1s up to z = 0.3. The aim of this work is to investigate how next generation instruments, and in particular the SquareKilometerArray(SKA),willaffectthesearchforthiselusiveclassofAGN. 2. Samples selection We decided to use two samples of relatively low z NLS1s. We extracted from SDSS DR7 all the NLS1s at z<0.3, using as classification criteria the FWHM(Hb )< 2000 km s−1 and the ratio [O III]/Hb < 3. Then we searched for a FIRST radio-source within a radius of 5 arcsec [3]. Finally we calculated their radio-loudness, defined as RL = F /F [14]. To derive B−band 5GHz the B-band magnitude, we convolved the optical spectra with a B-band filter template, and then estimatedtheintegratedflux. The5GHzfluxinsteadwasobtainedfromthe1.4GHzfluxofFIRST assuming aspectral indexa =0.5(Fn (cid:181) n −a [22]). SourceswithRL>10weredefinedasradio- loud, otherwise they wereconsidered radio-quiet. Inthis wayweobtained 117RQNLS1s,and32 RLNLS1s. We tested the samples by means of V/V test [2] to investigate their completeness. max Theradio-quietsamplehashV/V i=0.54±0.03,whiletheradio-loudhashV/V i=0.45±0.05. max max Both the samples are then in agreement with the uniform distribution of sources, and they can be considered statistically complete. 3. Results A useful tool to investigate the space density of sources is the logN-logS test, which does not require the knowledge of the distance of each source. The test assumes that all the sources 2 RLNLS1swithSKA M.Berton 2.0 2.0 1.5 1.5 1.0 1.0 ) ) 1− 1− sr sr N ( 0.5 N ( 0.5 g g o o l 0.0 l 0.0 −0.5 −0.5 −1.0 −1.0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 logS (mJy) logS (mJy) Figure 1: logN-logS test for radio-loud (left panel) and radio-quiet (right panel) NLS1s. The errors are poissonian.Theredsolidlineindicatesthelinearbest-fit. are above a minimum luminosity which always corresponds to an observable flux in the explored volume, an assumption that appears reasonable for samples limited to relatively low redshifts. The test evaluates the number of sources above each flux density S, so it studies the cumulative distribution of sources as a function of the flux. The number of sources brighter than some flux S should then be proportional to S−3/2, for constant density in a Euclidean space. This relationship isknownasthelogN-logStest,whichwasdeveloped totesttheevolution ofradiosources, soitis particularly wellsuitedforouraim. The sources counts are evaluated by considering the theoretical observing limit of SKA in declination (∼60◦ N), and the SDSS DR7 sky coverage (∼1/6 of the whole sky). The results are shown in Fig. 1. The errors are Poissonian. The distribution of RQNLS1s is the closer one to the Euclidean distribution (slope b =1.64±0.01), while RLNLS1s have a flatter distribution (b =0.68±0.01). TheSKA-1sensitivity1 at1.4GHzwillbe0.082mJyhr−1/2. Inourcalculations we assume that the resulting slope is remaining the same at lower fluxes. This can be done in absence of evolutionary effects, that we do not expect to be particularly strong in our relatively low redshift domain. This simple extrapolation yields very large numbers: SKA might be able to identify ∼1300RLNLS1and∼50000RQNLS1suptoz=0.3. This important result, nevertheless, is only an upper limit, since there are a few caveats that must be kept in mind. In particular, regarding RLNLS1s,a decrease ofone order of magnitude in radio flux must correspond to a similar decrease also in optical flux, otherwise the source would not be radio-loud anymore. It is then possible that SKA will not detect many more RLNLS1s, at least up to z = 0.3, although it will clearly allow a much deeper study of all currently known RLNLS1s. Conversely, at higher z, the number of new RLNLS1s can be largely increased, since therearemanyradio-loudsourceswithafluxdensity∼1mJythatatlargerdistancesarenotvisible forpresentdayobservatories. ForRQNLS1stherearealsoacoupleofcaveats. First,itisunlikelythatsomanyNLS1sexist uptoz=0.3,butthisonlymeansthatSKAwillbeabletodetectalltheexistingNLS1s. Butwhilst SKA willdetect many more sources, it will be difficult to identify them as NLS1swithout a deep 1SKA-TEL-SKO-0000229-ReportandOptionsforRe-BaseliningofSKA-1 3 RLNLS1swithSKA M.Berton optical observation. Theradioluminosity isproportional tothe[OIII]lineluminosity [9],hencea low radio flux corresponds to an equally low [O III] flux. This makes the classification more and moredifficultforweaksources. Anyway, evenwiththis optical limit, the knowledge increase onthe nature ofRQNLS1swill be very large. In particular, a deep radio investigation will provide helpful information on the dichotomybetweenradio-quietandradio-loudsources. Inturnitwillalsohelpustounveilwhether the origin of radio emission in RQNLS1s is due to the high star formation rate typical of NLS1s [20],ortosomesortofweakactivitythatmightbeongoing,suchasanabortedorfaintjet[12,10]. Nonetheless, it will also allow to study in detail the radio emission from RLNLS1s, investigating at very high resolution the morphology of the jets, and finally clarifying how strong the incidence of the starburst component is in these sources [7]. Last but not least, SKA will provide enough statistictoperformadetailedinvestigationonRLNLS1sbymeansoftheradioluminosityfunction, allowingustostudytheirevolution through cosmictime. References [1] Abdo,A.A.,etal.2009,ApJ,699,976 [2] Avni,Y.,&Bahcall,J.N.1980,ApJ,235,694 [3] Becker,R.H.,etal.1995,ApJ,450,559 [4] Berton,M.,etal.,2015A&A,578,A28 [5] Boroson,T.A.,&Green,R.F.1992,ApJS,80,109 [6] Caccianiga,A.,etal. 2014,MNRAS,441,172 [7] Caccianiga,A.,etal. 2015,MNRAS,451,1795 [8] D’Ammando,F.,etal.2015,MNRAS,452,520 [9] deBruyn,A.G.,&Wilson,A.S.1978,A&A,64,433 [10] Doi,A.,etal.2013,ApJ,765,69 [11] Foschini,L.,etal.2015,A&A,575,A13 [12] Ghisellini,G.,etal.2004,A&A,413,535 [13] Grupe,D.2000,New.Astron.Rev.,44,455 [14] Kellermann,K.I.,etal.1989,AJ,98,1195 [15] Komossa,S.,etal.2006,AJ,132,531 [16] Liao,N.-H.,etal.2015,arXiv:1510.05584 [17] Mathur,S.2000,MNRAS,314,L17 [18] Osterbrock,D.E.,&Pogge,R.W.1985,ApJ,297,166 [19] Richards,J.L.,&Lister,M.L.2015,ApJ,800,L8 [20] Sani,E.,etal.2010,MNRAS,403,1246 [21] Yao,S.,etal.2015,MNRAS,454,16 [22] Yuan,W.,etal.2008,ApJ,658,801 4