aa r X i v : . [ a s t r o - ph . C O ] J un Fermi meets Jansky – AGN in Radio and Gamma-RaysSavolainen, T., Ros, E., Porcas, R.W., & Zensus, J.A. (eds.)June 21–23, 2010, Bonn, Germany
The TANAMI Program
Roopesh Ojha ⋆ , Matthias Kadler , , , Moritz B¨ock , Faith Hungwe , , Cornelia M¨uller , Joern Wilms ,Eduardo Ros , , and the TANAMI Team NVI/United States Naval Observatory, 3450 Massachusetts Ave, NW, Washington, DC 20392-5420 Dr. Karl Remeis-Sternwarte & ECAP, Sternwartstrasse 7, 96049 Bamberg, Germany CRESST/NASA Goddard Space Flight Center, Greenbelt, MD 20771, USA USRA, 10211 Wincopin Circle, Suite 500 Columbia, MD 21044, USA Department of Physics & Electronics, Rhodes University, PO Box 94, Grahamstown 6140, South Africa Hartebeesthoek Radio Astronomy Observatory, PO Box 443, Krugersdorp 1740, South Africa Department d’Astronomia i Astrofisica, Universitat de Val´encia, E-46100 Burjassot, Spain MPIfR, Auf dem H¨ugel 69, 53121 Bonn, Germany
Abstract.
The TANAMI (Tracking AGN with Austral Milliarcsecond Interferometry) program provides compre-hensive VLBI monitoring of extragalactic gamma-ray sources south of declination -30 degrees. Operating at tworadio frequencies (8 and 22 GHz), this program is a critical component of the joint quasi-simultaneous observationswith the
Fermi
Gamma-ray Space Telescope and ground based observatories to discriminate between competingtheoretical blazar emission models. We describe the TANAMI program and present early results on the 75 sourcescurrently being monitored.
1. Introduction
Very Long Baseline Interferometry (VLBI) observationsplay a unique role in unraveling the physics of active galac-tic nuclei (AGN). They provide the only direct measure-ments of relativistic motion in AGN, thus measuring jetspeeds, Doppler factors, opening and inclination anglesof jets. With their unmatched resolution, VLBI observa-tions can allow us to associate γ -ray flaring activity withstructural changes on millarcsecond scales (such as jet-component ejections) helping to identify the location andextent of emission regions.VLBI observations have acquired particular salience inthe age of Fermi . Data from
Fermi /LAT in combinationwith other space and ground-based telescopes have madepossible the quasi-simultaneous observations across theelectromagnetic spectrum that have long been consideredessential to distinguish between different models of AGNemission. The close connection between VLBI and
Fermi observations is impressively demonstrated by the largenumber of VLBI-
Fermi papers published and submittedin the past year, including many from the
Fermi /LAT col-laboration, to which VLBI observations have contributedcrucially needed data for the proper interpretation of γ -rayresults. With VLBI data we have started to address someof the most crucial questions raised by the association of γ -ray emission with blazars.
2. The TANAMI Program
The indispensable role of parsec-scale monitoring of radio-and γ -ray bright AGN has lead to the establishment of a ⋆ [email protected] number of highly successful VLBI monitoring programs(see Lister et al. 2010 for a review) but all of these pro-grams use northern hemisphere arrays that cannot ob-serve much of the southern hemisphere. The TANAMIprogram is the only parsec scale monitoring program tar-geting AGN south of declination − ◦ . Further, uniquelyamong comparable VLBI programs, TANAMI observa-tions are made at two frequencies (8.4 and 22 GHz). Thislets us monitor the parsec-scale spectra of the cores andthe brightest jet features, allowing us to contribute ra-dio spectral indices of jet features to Fermi multiwave-length studies (e.g., Abdo et al. 2010a, Chang et al. 2010)besides probing emission processes along AGN jets (e.g.,M¨uller et al. 2010, Hungwe et al. 2010).Since it covers that third of the sky not observed byother VLBI monitoring programs, TANAMI significantlyimproves the statistics for jet kinematics and flare-ejectionstudies. This region of the sky includes many interest-ing AGN (see below) and newly discovered γ -ray AGNcan be followed up, often for the first time, with VLBI(e.g., Abdo et al. 2009). The TANAMI collaboration hasalso begun work with the ANTARES (Coyle 2010) andKM3NeT (Piattelli 2010) consortia, two neutrino tele-scopes that target the southern sky. Fermi γ -ray variabil-ity data and TANAMI-determined jet-ejection epochs willhelp develop data-filtering techniques to search for extra-galactic neutrino point sources. This could usher us intoan era of multimessenger astronomy.TANAMI observations are made using the telescopesof the Australian Long Baseline Array (LBA ; e.g., The Long Baseline Array is part of the Australia Telescopewhich is funded by the Commonwealth of Australia for opera-tion as a National Facility managed by CSIRO. Roopesh Ojha et al.: The TANAMI Program
10 5 0 −5 −10 −15 −20−10−505101520 10 pc
10 5 0 −5 −10 −15 −20−10−505101520 10 pc -10
15 10 5 0 −5 −10−10−5051015 10 pc
15 10 5 0 -5151050-5
15 10 5 0 −5 −10−10−5051015 10 pc
Fig. 1.
TANAMI images of four
Fermi sources. Starting from the top, the four rows show images of 0047 − − −
477 respectively. In each row the left image shows the image at 8 GHz and the image on the right at 22 GHz at the sameepoch. In the center of each row is the spectral index image made from the simultaneous images at these two frequencies. Boththe axes in all plots are labeled in milliarcseconds from the center of the image. The hatched ellipse at the bottom left of eachcontour image represents the synthesized beam of the observing array. The color coding in the spectral index image representsthe spectral index defined as F ν ∼ ν α oopesh Ojha et al.: The TANAMI Program 3 Table 1.
VLBI array for TANAMI observations.Telescope Diameter(meters)Parkes, NSW, Australia 64Narrabri, NSW, Australia 5 × a b b c c a Not available since Sept 2008. Likely return Sept 2010. b Operated by the Deep Space Network of the NationalAeronautics and Space Administration, USA c Operated by the German Bundesamt f¨ur Kartographie undGeod¨asie (BKG)
Ojha et al. 2005) and affiliated telescopes. TANAMI wasable to significantly improve the ( u, v )-coverage of theLBA by obtaining access to International VLBI Service(IVS) telescopes in Antarctica and Chile as well asDeep Space Network telescopes in Tidbinbilla, Australia.All telescopes that participate in TANAMI observationsare listed, along with their diameters, in Table 1. Ateach epoch and each frequency, every source is typi-cally observed for 6 scans of about 10 minutes each.Typical ( u, v )-coverage at both frequencies are shown inM¨uller et al. 2010. Our augmentation of the LBA has leadto the highest fidelity images for most of the sources ob-served by TANAMI.The initial sample of 44 TANAMI sources wereselected based on previous (EGRET) γ -ray detectionand/or radio flux density and luminosity. Under an MoU(Memorandum of Understanding) with the Fermi col-laboration TANAMI started monitoring observations ofnew
Fermi sources through 2009 adding the new sourcesto our observing schedule while decreasing the observingcadence of sources showing limited radio-structural vari-ability. The current TANAMI sample includes 75 sourcesof which 55 have been detected by
Fermi . 53 TANAMIsources have 1FGL (Abdo et al. 2010c) associations while2 are tentative new detections (B¨ock et al. 2010). To date,12 epochs (most at both frequencies) have been ob-served. Correlation, processing and imaging are progress-ing smoothly. Images and other results are available at ourwebsite as soon as they are finalized. For further detailsof the TANAMI program including details of calibrationand imaging see Ojha et al. 2010. http://pulsar.sternwarte.uni-erlangen.de/tanami/
3. Results
TANAMI is routinely producing VLBI images of highquality at 8 and 22 GHz (X and K-band respectively).We show examples for three sources in Fig. 1. For eachsource we show the 8.4 GHz image and the 22 GHz imagefrom the same epoch (on the left and right respectively).In the center of each row is shown the corresponding two-frequency spectral index image. It is important to notethat the resolution of the lower frequency image is often better than that of the higher frequency image becausethe trans-oceanic telescopes in Antarctica and Chile can-not observe at 22 GHz.These spectral index images were made by aligningthe brightest pixels in the X and K-band images from thesame epoch. The images at both frequencies have beenconvolved with the larger of the two beams of the individ-ual images. The larger beam has also been used to producethe overlaid contours on these images. The color codingdepicts the spectral index defined as F ν ∼ ν α i.e., a posi-tive spectral indicates an inverted spectrum. Thus we areable to measure spectral indices of the cores and individ-ual jet features and we are using these data to measurecore shift, localize the central engine, calculate the opac-ity towards the central engine and identify the emissionalong the jet. In combination with data at other wave-lengths we are modeling the SEDs of AGN. Note that thefigures shown here are not corrected for coreshift.For a growing number of sources in our sample wehave enough epochs of data to study their kinematics. Weare fitting Gaussian components to jet features to trackjet trajectories, measure their speeds, and derive their in-trinsic parameters. When combined with SED modeling,these kinematic data address the relationship between theDoppler-boosting parameters for the radio and γ -ray emit-ting regions of the jets.TANAMI data have been and are being used in a num-ber of studies that can broadly divided into two categories,individual source studies and statistical studies of the fullsample or some subset, which are briefly described below. Studies of individual TANAMI sources include: • One of the first
Fermi /LAT publications addresses abright γ -ray flare of the poorly studied source PKS 1454-354 (Abdo et al. 2009). TANAMI contributed the firstdeep 8.4 GHz VLBI image of this source revealing its core-jet structure. • TANAMI data on nine
Fermi /LAT sources wereused to generate SEDs of the γ -ray selected LBASblazars and investigate their broadband spectral propeties(Abdo et al. 2010a). • TANAMI data were used to construct the SED ofPKS 2052 −
47 during a LAT multiwavelength campaign(Chang et al. 2010). • TANAMI data are being used to study the highly vari-able BLLac 0537 −
441 which is one of the most lumi-
Roopesh Ojha et al.: The TANAMI Program nous γ -ray blazars detected in the southern sky so far(Hungwe et al. 2010) • TANAMI data were used to constrain the size of the γ -ray emitting region and for SED modeling of the nearestgalaxy Centaurus A (Abdo et al. 2010b). A multi-epoch,dual-frequency analysis of the innermost regions of thissource is in progress (M¨uller et al. 2010) First epoch 8.4 GHz results for the initial sampleof 43 sources have been analyzed and presented inOjha et al. 2010. Using the classification scheme ofKellermann et al. (1998), the initial sample has 33 single-side (SS) and 5 double-sided (DS) sources with just one ex-ample each of the compact (C) and irregular (Irr) morpho-logical types. Three sources do not have an optical identi-fication. All of the quasars and BL Lacertae objects in thesample have an SS morphology while all 5 DS sources aregalaxies. The lone C source is optically unidentified whilethe only Irr source is a GPS galaxy 1718 − γ -ray selectedsubsamples. The core brightness temperature ( T B ) limitof all initial TANAMI sources was calculated. The highend of the distribution of calculated brightness temper-atures is dominated by quasars and the low end by BLLacertae objects and galaxies. Of the 43 sources in thesample, 14 have a maximum T B below the equipartitionvalue of 10 K (Readhead 1994), 30 below the inverseCompton limit of 10 K (Kellerman 1969), putting abouta third of the values above this limit. There is no signifi-cant difference in the brightness temperature distributionof LBAS and non-LBAS sources.A link between γ -ray emission and the parsecscale morphology of AGN has been sought (e.g.,Taylor et al. 2007). We fit circular Gaussians to the visi-bility data and measured the angle at which the innermostjet component appears relative to the position of the corei.e. the opening angle. Of the LAT AGN Bright Sample(LBAS) sources 78% have an opening angle >
30 de-grees while only 27% of non-LBAS sources do. This resultshould be treated with great caution as the sample size forthis analysis is currently small but Pushkarev et al. (2009)report similar results. If confirmed, the above result presents two possibil-ities: either the LBAS jets have smaller Lorentz fac-tors (since the width of the relativistic beaming cone ∼ / Γ) or LBAS jets are pointed closer to the lineof sight than γ -ray faint jets. The former scenario ap-pears unlikely, indeed the opposite effect is reported byLister et al. (2009), Kovalev et al. (2009).
4. Conclusions
Fermi sources in the southern third of the sky are beingmonitored by the TANAMI program at about every twomonths. These high quality, dual frequency observationsare producing spectral index images at milliarcsecond res-olutions which are a crucial element in the multiwave-length study of AGN physics. For a subset of the TANAMIsample, the number of observed epochs is now sufficientfor kinematic modeling to begin. When combined withjet-speed measurements, SED modeling across the electro-magnetic spectrum will let us probe the relation betweenthe Doppler-boosting parameters for the radio and γ -rayemitting regions of the jet.Studies of several individual AGN detected by Fermi have been enriched by data from the TANAMI programand multiwavelength analysis of a number of interestingsources are in progress. Statistical analysis of the growingTANAMI sample is providing broader insight into the tiebetween the low- and high-energy radiation from AGN.
Acknowledgements.
We thank the
Fermi /LAT AGN groupfor the good collaboration. This research has been partiallyfunded by the Fermi Guest Investigator Program. This re-search has been partially funded by the Bundesministeriumf¨ur Wirtschaft und Technologie under Deutsches Zentrum f¨urLuft- und Raumfahrt grant number 50OR0808.