Table Of Content

2011 Fermi Symposium, Roma., May. 9-12 1 Gamma-ray Clues to the Relativistic Jet Dichotomy EileenT.Meyer∗ andGiovanniFossati DepartmentofPhysicsandAstronomy,RiceUniversity,Houston,TX77005 MarkosGeorganopoulos DepartmentofPhysics,UniversityofMarylandBaltimoreCounty,Baltimore,MD21250and NASAGoddardSpaceFlightCenter,MailCode663,Greenbelt,MD20771,USA MatthewL.Lister DepartmentofPhysics,PurdueUniversity,WestLafayette,IN47907 2 Inexaminingaselectsampleofover200blazarsofknownjetkineticpower(Lkin)andwell-characterizedSEDs, 1 wefound[1]thatIntermediatesynchrotron-peaking(ISP)blazarsmayhavelowergamma-rayoutputthanhigh 0 synchrotron-peaking (HSP) blazars of similar Lkin, consistent with our hypothesis that ISP blazars are less- 2 beamedversionsofHSPblazars,ratherthanadistinctpopulation. Further,byusingtheradiocoredominance asameasureofrelativebeaming,wefindthatgamma-rayluminositydepends onbeaminginaconsistentway n forblazarsrangingoveralljetkineticpowers(1042 −1046 ergss−1). Were-examinethegamma-rayproperties a ofthiscoresampleofblazarsusingthe1-yearLATcatalog [2]. Wefindthatforweakjets,theratioofinverse J Comptontosynchrotronemissionremainsconstantwithincreasedbeaming,consistentwithanSSCmodelfor 6 thejetemission,whilethemostpowerfuljets showastrongincreaseinCompton dominancewithorientation, 2 consistentwithanexternalCompton(EC)emissionmodel. ] E 1. Introduction or molecular torus [7, 8, 9]). Generally low-power H Radio-loudactivegalacticnuclei(RLAGN)arebe- and lineless objects are believed to radiate by SSC at . h lieved to contain an actively accreting super-massive high energies, while individually, many high-powered -p black hole (107 - 1010 M⊙) which generates bi-polar FSRQhavebeenmoresatisfactorilyfitwithECmod- o jets of relativistic material reaching up to Mpc in els [10, 11]. However a consistent basis for which r scale and with luminosities up to 1049 ergs s−1, pow- sourcesrequireEChasnotbeendemonstratedforany st erful enough to heat the intra-cluster medium and particular class of blazars. a potentially produce a feedback mechanism in galaxy [ ThelocationoftheGeVemissionisalsoamatterof formation [3]. When these relativistic jets are seen active debate, as it comes from locations close to the 2 with the axis near to the line of sight, they are centralenginethatremainunresolved,evenwithvery v called blazars (see [4] for a review). The jet radia- 0 long baseline interferometry (VLBI). If the emission tion is strongly Doppler-boosted and dominates over 9 region is located within the sub-pc scale BLR, the the emission from the host galaxy, the dusty torus 4 GeV emission of blazars with strong lines is due to 6 (thought to surround the nuclear black hole), the ac- ECscatteringofBLR photons[12], whichcanleadto . cretion disk, and, except at very low frequencies, the 1 moreefficientcoolingandthuslowerpeakfrequencies, isotropic emission from the giant lobes of gas fed by 1 as seen in the blazar sequence [13]. However, others the jet overtime. Thus blazarsarenaturallyinterest- 1 agree that the high-energy emission may be coming 1 ing objects as the magnified view of the jet emission several pc downstream as suggested by some multi- : canallowustoprobethestructureandnatureofthese v wavelength observations of variability [14, 15]. The jets and ultimately to understand the physical mech- i EC versus SSC origin of the high-energy emission is X anisms behind their origin. animportant question as it can be used as a diagnos- r The broad-band, highly luminous and variable tic for the location of the gamma-ray emission, and a spectrum from blazars is usually characterized as a the inferred structure of the jet. ‘double-peaked’ one, with one broad emission com- ponent believed to be synchrotron radiation peaking TheFermi gamma-raytelescopeisgivingusanun- from sub-infrared energies to X-rays, and a second precedented look at the high-energy emission from peak from inverse Compton emissionat > MeV ener- blazars, with nearly 700 blazar associations in the 1- gies. This second high-energy peak in blazars can be yearcatalog(1-LAT)[2]. Withsuchalargesample,we explained with both synchrotron-self Compton (SSC, canbegin to put some orderin the phenomenologyof in which the particles in the relativistic jet up-scatter blazarsathighenergies. InSection2weintroducethe synchrotron photons, [5, 6]) and external Compton ‘BlazarEnvelope’asseenbyFermi anddiscusstheim- (EC) models, in the latter case with photons from portanceofthejetkineticpowerandorientationofthe either the accretion disk, broad line region (BLR), jet in determining the gamma-rayluminosity. In Sec- tion 3 we discuss a recent finding suggesting that the high-energy emission of high-power jets is dominated by EC rather than SSC. In Section 4, we summarize ∗Electronicaddress: [email protected] our findings. eConf C110509 2 2011 Fermi Symposium, Roma., May. 9-12 Figure 1: The synchrotron Lpeak - νpeak plane, in which Figure 2: The InverseCompton Envelope. Figure is theblazar sequencefirst found an anti-correlation adapted from [16]. The Fermi-detected AGN (large, dark between bolometric luminosity (with Lpeak) and the bluecircles and magenta squares) show that alignment synchrotron peak frequency. Using only well-sampled playsa strong effect in thegamma-ray output. SEDsreveals a possible alternative to thecontinuous Interestingly,while theFR II sources appear to drop sequence, in which two populations show dramatically directly below thepowerful FSRQsources, FR I radio different behaviors in this plane. Track (A) is the galaxies appear to follow a more horizontal track,similar de-beaming trail in synchrotron peak for a standard toour findingsin thesynchrotron envelope (Figure 1, single Lorentz-factor jet (‘strong’ type),track (B) shows left). The presenceof a forbidden zone (emptyregion at themarkedly horizontal movement typical of a upperright) suggests that there is a sequencein theIC decelerating jet model characterized byvelocity gradients plane, similar to that found in the synchrotron (Figure (‘weak’ type). Adapted from [1]. 1). 2. The Inverse Compton Envelope as This was the first work to consider both the jet ki- seen by Fermi netic power (L , as measured from the isotropic ra- kin The relativistic jets in RL AGN are generally con- dioemission)andradiocoredominance(R,ameasure fined to a small opening angle and thus most blazars ofjetorientation),asimportantfactors. Strikingly,we are seen within a small range of orientation angles foundthatthe blazarsequenceisactuallybrokeninto from the jet axis. It is often assumed that bright two populations in the L − ν plane (Figure peak peak blazarsin flux-limited samples are “well-aligned”and 1). The ‘weak’ jets consist entirely of sources with therefore the exact viewing angle is of little impor- L <1044.5 ergs s−1, and extend from mis-aligned kin tance (e.g., for modeling, often the assumption θ = FR I radio galaxies at low ν up to well-aligned peak 1/Γ is used). However, as surveys go deeper, it is (as measured by R) HSP BL Lacs. The ‘strong’ jets expected that more and more relatively misaligned comprised a population of low ν jets which drop peak sources will be observed. in peak luminosity rapidly with decreasing R. The Fossati et al. (1998) demonstrated an anti- critical transition between weak and strong jets ap- correlation between the synchrotron peak luminosity pears to occur at an (dimensionless) accretionrate of and the peak frequency, forming the now canonical m˙ ∼10−2 (see also [21, 22]), as estimated from the cr ‘blazarsequence’of(presumably)well-alignedsources ratio L /L 1, matching the divide suggested by kin Edd [17]. Lower-luminositysourcesappearinginthe space [23]. below the blazar sequence are expected due to some We note that the large number of blazars seen by sources being less aligned to the line of sight. Indeed, Fermi, as well as the detection of severalradio galax- [18]foundthatnewsourcestheyidentifiedmodifythe ies [16], implies that we should have a large range in blazar sequence to an envelope, with the area below orientationsintheFermi samples. AsseeinFigure2, the blazar sequence populated with sources. Similar an‘envelope’isindeedseenintheplaneofgamma-ray envelopes were found by Anton & Browne and Niep- pola et al. [19, 20]. In recent work, we filled the synchrotron L −ν plane using a much larger peak peak sample of jets over a wide range of orientations (from 1TheEddingtonluminositywascalculatedusingtheaverage blazarstosourcesmisalignedenoughtobeseenasra- blackholemassestimatefromtheliterature,aswillbediscussed dio galaxies), with much-improved SED sampling [1]. inaforthcomingpublication. eConf C110509 2011 Fermi Symposium, Roma., May. 9-12 3 3. Evidence for External Compton in High-Power Jets With a large sample of blazars spanning a range of orientations as shown in the previous section, it is possibletolookforsignaturesofdifferenthigh-energy emission mechanisms in a population of blazars. The Dopplerbeamingfactor,δ,isafunctionoftheLorentz factor Γ and the orientationangle, and the effect of δ on the observed (monochromatic) luminosity is: L=L δ3+α (1) 0 where L is the rest-frame luminosity (at δ = 1), 0 andαis the energyspectralindex atthe frequency of interest. The exponent 3+α is the value assumed for a‘movingblob’inthe jet. Ifthe emissioncomesfrom a standing shock, the exponent would be 2+α [24]. Figure 3: Total gamma ray output versusradio core In the case of emission by SSC, the IC peak has a dominance for the 130+ sources with good SED coverage detected by Fermi in the1-LAT.Jet power has a very beamingpatternwhichisidenticaltothesynchrotron strong effect on thegamma-ray luminosity, but when (i.e., it follows Equation 1). However,for EC models, sources are sorted bytheir jet powers, thedependenceon thebeamingathighenergiesgoesasL =L δ4+2α IC 0,IC radio core dominance also becomes apparent. The slopes [25, 26]. The larger exponent indicates that as a shown are statistically significant at the95% level. source with significant EC emission is aligned, the IC peakshouldbemoreandmoredominantoverthesyn- chrotron peak (i.e., the ratio L /L will increase IC sync luminosity (> 100 MeV) versus the LAT photon in- with R. We can measure the Compton Dominance dex(whichapproximatelytracksthepeakfrequency). R = log(L /L ) ∝log (δ(3+α)/(4+2α)), and see CD IC sync There is clearly a ‘forbidden zone’ at the upper right, that the dependence on δ goes with an exponent of 2 with the edge of the envelope presumably comprised whencomparingtheICandsynchrotronpeaks(where of the most aligned objects. The sources are divided α = 1). At radio frequencies, where the spectral in- into the typical higher-power and broad-lined flat- dex is confined to a fairly narrow range in values of spectrumradioquasars(FSRQ)andthe lower-power, α =0−0.5,the beaming exponent will be ∼ 3 - 3.5. r lineless BL Lac objects (BLL).While this figurelacks In Figure 4 we examine the relationship between the precise information needed to find the two popu- R and the radio core dominance, R, for two sub- CD lations suggested by [1], it is clear the the FR I radio samples of the 1-LAT blazars. The pink triangles are galaxies(whichareunifiedwithmostBLLacs)appear high-jet-power sources (log L > 45.5), while the kin at much lower IC peak frequencies, forcing a similar blue are low-power (log L < 43.5). The flat dis- kin horizontal track from the low-power, aligned BLLs, tribution in the latter group is consistent with the just as was found in the synchrotron plane. It is in- SSC models generally used to model these types of teresting to note that the FR II radio galaxies,which sources. For the high-power group, however, we find wouldbegenerallyunifiedwiththeFSRQpopulation, a significant trend of increasing Compton dominance arebothdirectlybeneaththeminpeakfrequency,but withincreasingradiocoredominance(i.e.,alignment). are also not particularly mis-aligned, which suggests ThiscanbeinterpretedasasignatureofanECprocess that these sources might drop out of Fermi detection for a population of jets characterized by high kinetic more quickly than the weak sources. powers. It is interesting to consider the effect of jet kinetic However, from the above relations, the expected power as well as orientation on the total LAT-band slope of the correlation should be close to 2/3, up luminosity. As shown in Figure 3, as a total popula- to a max of 1 under reasonable assumptions. The tion the 1FGL blazars appear as a scatter-plot in a slope in Figure 4 is clearly much higher, in fact con- plot of L versus R. However, when the jet kinetic sistentwith a value of ∼ 2. While a more detailed in- γ power is shown (color scale), it becomes clear the L vestigationisinprogress,wenotethatthishighslope γ increases with both radio core dominance (i.e., angle mightmeanthattheassumptionthattheLorentzfac- totheline-of-sight)and,evenmoredramatically,with tor of the plasma emitting in the gamma-rays is the the jet kinetic power. We note that L is calculated same as that emitting at GHz frequencies is incor- kin fromthelow-frequencyisotropicemissionandisthere- rect. If a roughly 2:1 ratio is assumed (i.e., Γ = 15 IC fore a completely independent measurementfromL , while Γ = 8), the higher slope can be explained. γ radio making it unlikely for this trend to be the result of Thisisconsistenttosomeextentwiththedifferencein the selection effects of the 1-LAT sample. variability timescales for very high energies (minutes eConf C110509 4 2011 Fermi Symposium, Roma., May. 9-12 France(http://simbad.u-strasbg.fr/simbad/),andthe NASA/IPAC Extragalactic Database (NED) which is operated by the Jet Propulsion Laboratory, Cal- ifornia Institute of Technology, under contract with the National Aeronautics and Space Administra- tion (http://nedwww.ipac.caltech.edu/). We utilized packagenpintheRlanguageandenvironmentforsta- tistical computing. GF and EM acknowledgesupport from NASA grants NNG05GJ10G, NNX06AE92G, and NNX09AR04G, as well as SAO grants GO3- 4147X and G05-6115X. MG acknowledges support from the NASA ATFP grant NNX08AG77G and NASA FERMI grant NNH08ZDA001N. The MO- JAVE project is supported under National Science Foundation grant 0807860-AST. References Figure 4: We compare strong-jet and weak-jet sources in terms of their inverse Compton dominance over [1] Meyer,E.T.,Fossati,G.,Georganopoulos,M.,&Lis- synchrotron emission. Wefind that for thestrong jets, ter, M. L.2011, ApJ, 740, 98 theinverse Compton dominance increases dramatically [2] Abdo, A.A., Ackermann,M., Ajello, M., et al. 2010, (up to a factor of 100) with increasing alignment, as VizieR OnlineData Catalog, 2188, 80405 measured by radio core dominance. Conversely, theweak [3] McNamara, B.R.andNulsen,P.E.J., 2007Ann.Rev. jets show nosuch increase. The IC peak is measured by Astron. & Astrophys.45, 117 fittingthe X-ray,gamma-ray, and TeV (where available) [4] Urry,C. M., & Padovani, P. 1995, PASP,107, 803 points with a parabola and taking thepeak luminosity. [5] Maraschi, L., Ghisellini, G., & Celotti, A. 1992, ApJ (Dashed line shows slope=2 for reference). Lett., 397, L5 [6] Marscher,A.P.,&Travis,J.P.1996,A&As,120,537 [7] Ghisellini, G., & Madau, P. 1996, MNRAS,280, 67 to hours) versus radio (much longer), which indicates [8] B laz˙ejowski, M., Sikora, M., Moderski, R., & Made- jski, G. M. 2000, ApJ,545, 107 thattheradioemissionis comingfromaregionwhich [9] Sikora,M.,Stawarz,L .,Moderski,R.,Nalewajko,K., is slower and/or at larger scales. & Madejski, G. M. 2009, ApJ,704, 38 [10] Ghisellini, G.,Tavecchio,F.,Foschini,L.,et al.2010, 4. Conclusions MNRAS,402, 497 WehaveshownthattheFermi gamma-raysatellite [11] Vercellone, S., Striani, E., Vittorini, V., et al. 2011, sees relativistic jets in RL AGN over a wide range ApJ Lett., 736, L38 of orientation angles, including sources seen as ra- [12] Sikora, M., Begelman, M. C., & Rees, M. J. 1994, dio galaxies. There is a ‘forbidden zone’ of high- ApJ, 421, 153 [13] Ghisellini, G., Celotti, A., Fossati, G., Maraschi, L., power, high-peak sources which remains empty, con- & Comastri, A. 1998, MNRAS,301, 451 sistent with a spectral sequence similar to that seen [14] Böttcher, M., Reimer, A., & Marscher, A. P. 2009, in the synchrotron peak luminosity - peak frequency ApJ, 703, 1168 plane. Further, the much lower IC peak frequencies [15] Agudo,I.,Marscher,A.P.,Jorstad,S.G.,etal.2011, found for the few detected FR I radio galaxies sug- ApJ Lett., 735, L10 gest that the low-power class of FR I/BL Lacs de- [16] Abdo, A.A., Ackermann,M., Ajello, M., et al. 2010, beamalongmorehorizontaltracksinthesynchrotron ApJ, 720, 912 and IC planes (Figures 1 and 2). We show that the [17] Fossati, G., Maraschi, L., Celotti, A., Comastri, A., total gamma-ray band luminosity depends on both & Ghisellini, G. 1998, MNRAS,299, 433 orientation(measuredthroughradiocoredominance) [18] Padovani, P., Perlman, E.S., Landt,H.,Giommi, P., and more heavily on the jet kinetic power (as mea- & Perri, M. 2003, ApJ, 588, 128 [19] Antón, S., & Browne, I. W. A. 2005, MNRAS, 356, sured from low-frequency, isotropic radio emission). 225 Finally,wepresentthe firstcollectiveevidenceforEC [20] Nieppola, E., Tornikoski, M., & Valtaoja, E. 2006, process in a sample of high-power jets (log L > kin A&A,445, 441 45.5), showing that as alignment (radio core domi- [21] Georganopoulos,M.,Meyer,E.T.,Fossati,G.,&Lis- nance) increases, the Compton dominance increases ter, M. L.2011, arXiv:1111.4711 dramatically. [22] Ghisellini, G., & Celotti, A. 2001, A&A,379, L1 [23] Narayan,R.,Garcia,M.R.,&McClintock,J.E.1997, ApJ Lett., 478, L79 Acknowledgments [24] Lind, K.R., & Blandford, R.D. 1985, ApJ, 295, 358 [25] Dermer, C. D.1995, ApJLett., 446, L63 This research has made use of the SIM- [26] Georganopoulos, M., Kirk, J. G., & Mastichiadis, A. BAD database, operated at CDS, Strasbourg, 2001, ApJ,561, 111 eConf C110509