VERITAS observations of M87 from 2007 to present
aa r X i v : . [ a s t r o - ph . C O ] J u l PROCEEDINGS OF THE 31 st ICRC, Ł ´OD ´Z 2009 1
VERITAS observations of M87 from 2007 to present
Chiumun Michelle Hui ∗ for the VERITAS collaboration †∗ University of Utah, Department of Physics and Astronomy, Salt Lake City, Utah 84112, USA ([email protected]) † see R. A. Ong et al (these proceedings) or http://veritas.sao.arizona.edu/conferences/authors?icrc2009 Abstract . M87 is a nearby radio galaxy and becauseof its misaligned jet, it is possible to correlatedetailed spatially-resolved emission regions in theradio, optical to X-ray waveband with unresolvedbut contemporaneous flux measurements in the TeVregime. Hence, M87 provides a unique opportunityto reveal the emission mechanisms responsible forhigh energy gamma-ray emission from active galacticnuclei. Observations with VERITAS since 2007 haveresulted in 90 hours of data while 2008 observa-tions were part of a concerted effort involving thethree major atmospheric Cherenkov observatories:H.E.S.S., MAGIC and VERITAS. As a result of theTeV campaign, a high flux state of M87 was detectedin February 2008 showing multiple flares with rapidvariability. We will present the comprehensive resultsfrom VERITAS observations since 2007 and alsoshow preliminary results from the 2009 campaign.
Keywords : gamma rays: observations - galaxies:individual (M87, VER J1230+123)
I. I
NTRODUCTION
M87 is a giant elliptical galaxy located 16 Mpc away(redshift z = 0 . ) near the center of the Virgocluster. It has been observed at all wavelengths rangingfrom radio to TeV gamma rays. Its core is an activegalactic nucleus (AGN) powered by a supermassiveblack hole of ∼ . × M ⊙ [20]. The jet of M87does not point along our line of sight; however apparentsuperluminal motion has been observed in both radio[12] and optical [11] for different features along the jet,constraining the jet orientation to < ◦ at the locationof the knot HST-1. M87 is described as a misalignedBL Lac [27]. The proximity of M87 and its misalignedjet have enabled the study of its jet morphologies,which are similar in radio, optical, and X-rays [22].Flaring activities have been observed at these energiessimultaneously and in different jet features [12], whichrevealed many characteristics of relativistic jets in AGN.TeV emission from M87 was discovered by theHEGRA collaboration from their 1998-1999 observa-tions [5] and was confirmed by the H.E.S.S. collabo-ration [6], which additionally reported year-scale anddays-scale flux variability during a high state of gamma-ray activity in 2005. The observed variability timescalesdisfavor large scale gamma ray production models such as the dark matter annihilation model [8] and the in-teracting cosmic ray proton scenario [23], and favorthe immediate vicinity of the M87 black hole as theTeV production site. However, the angular resolution ofimaging atmospheric Cherenkov telescopes (IACTs) isinsufficient to resolve any structure in M87. During thesame period of the flare observed by H.E.S.S. in 2005,the Chandra X-ray observatory detected the knot HST-1( ∼ . ” away from the nucleus) at an intensity morethan 50 times that observed in 2000. HST-1 was thensuggested as a more likely source of TeV emission thanthe core [16].The knot HST-1 has been demonstrated as a possiblelocation for jet reconfinement where photons can beupscattered to TeV energies via the inverse-Comptonprocess [25]. Several models with emission originatingat the inner jet region have also been proposed, whereTeV emission can be produced via inverse-Comptonscattering [14] or via synchrotron self-Compton pro-cesses involving more complex jet structures [18] [26].Leptonic models involving the electromagnetic field ofthe black hole [21] [24] with TeV emission coming fromthe vicinity of the black hole and not the inner jet havealso been suggested.VERITAS confirmed TeV emission above 250 GeVfrom M87 in the 2007 dataset, but at a lower flux thanwhat was reported by H.E.S.S. in 2005, and no vari-ability was detected for 2007 [3]. Day-scale variabilitywas reported in 2008 by the MAGIC collaboration [7]with a 13-day flare during the 2008 TeV joint moni-toring campaign by H.E.S.S., MAGIC, and VERITAS[9]. VERITAS observation was intensified following atrigger alert issued by the MAGIC collaboration andanother flare was detected [2]. Chandra X-ray data takenin the same month (6-week sampling frequency) showedhistorical maximum activity coming from the core whilethe nearby knot HST-1 remained quiescent [15]. Thecontemporaneous gamma-ray and X-ray flares suggestthe core as the more probable TeV emission region, incontrast to the 2005 flaring activity in TeV observed byH.E.S.S. and in the knot HST-1 in X-rays by Chandra.Publication for the results from this joint campaign isforthcoming [1].In 2009 the TeV monitoring campaign is contin-ued with MAGIC and VERITAS. In this paper, wewill present preliminary results of the VERITAS 2009dataset, along with the results from observations begin-ning in 2007. C. M. HUI et al.
VERITAS OBSERVATIONS OF M87
II. VERITAS
OBSERVATIONS
VERITAS, the Very Energetic Radiation ImagingTelescope Array System, is an array of four 12 m diam-eter imaging atmospheric Cherenkov telescopes locatedat the Fred Lawrence Whipple Observatory at MountHopkins in southern Arizona. Each telescope is equippedwith a camera comprising 499 photomultiplier tubesarranged in a hexagonal lattice covering a field of viewof . ◦ . The array is sensitive from 100 GeV to morethan 30 TeV. It has an effective area of ∼ m andan angular resolution of ∼ . ◦ ( containment). Formore details of VERITAS, see [4].M87 was observed with VERITAS for over 115 hoursbetween February 2007 and April 2009 at a range ofzenith angles from ◦ to ◦ . Observations in spring2007 were carried out during the construction phaseand only 3-telescope data (94% of spring 2007 data)are used in the spectral analysis. Later observations(fall 2007 onward) were achieved with 4 telescopes.All observations were performed in wobble mode whereM87 is tracked with a . ◦ offset to the camera center.After eliminating bad weather observations and unstabletrigger rate data, over 90 hours of quality live datawere then processed with several independent analysispackages [13] with slightly different algorithms. Allanalysis packages yield consistent results.Shower images are first corrected in gain and tim-ing using parameters obtained from the nightly lasercalibration data. Then the images are passed througha two-threshold cleaning. Each shower image is thenparametrized [17], and the shower direction is recon-structed using the stereoscopic technique. Events areselected as gamma-ray like if at least two images passedcuts optimized for a Crab Nebula flux source. Thesource region is defined by a . ◦ radius disk centeredon the source coordinates, and all the gamma-ray likeevents within this region are summed to the ON count;the background is estimated from seven identically sizedregions reflected from the source region around thecamera center, and is summed to the OFF count [10].The ON and OFF counts are then used in the Li &Ma formula 17 [19] to calculate the significance of theexcess. III. R ESULTS
In 2007, M87 was detected at a statistical significanceof . σ after 44 hours of observations between Februaryand April with a 3-telescope array. An average fluxof (3 . ± . × − cm − s − for energies above250 GeV was measured from this dataset, correspondingto ∼ of the Crab Nebula flux. No significant short-term flux variability was detected [3].In 2008, M87 was detected at . σ after 41 hoursof observations between December 2007 and May 2008.An average flux of (2 . ± . × − cm − s − above250 GeV was recorded, corresponding to ∼ of theCrab Nebula flux. During 4 days in February 2008 (MJD54505 - 54509) flaring activity was observed in 6 hours of data, resulting in a . σ detection and an average fluxthat corresponded to 5.3% of the Crab Nebula flux. 15hours of observations were performed before the flareperiod and resulted in a marginal detection ( < σ ) ofM87 at 2.0% of the Crab Nebula flux; after the flareperiod, 19 hours of observations yielded no detection ofM87 and an upper limit of < . of the Crab Nebulaflux. The spectral index ( Γ ) of the differential spectrumpower-law fit with the form d Φ /dE = Φ ( E/T eV ) − Γ showed no significant variation between pre-flare andflare period (see Table I).In 2009, 18 hours of observations between Januaryand April yielded a marginal detection of M87 and anupper limit of < . Crab Nebula flux. No flaringactivity was observed (see figure 2).
TABLE ID
IFFERENTIAL SPECTRUM POWER - LAW FIT OF THE FORM d Φ /dE = Φ ( E/T eV ) − Γ data Φ Γ ( − cm − s − T eV − )2007 . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . Energy (TeV)1 ) T e V - s - c m g d N / d E ( E -13 -12 VERITAS 2008VERITAS 2007MAGIC 2008HESS 2005HESS 2004
Fig. 1. VERITAS M87 energy spectrum of the entire dataset, incomparison to TeV energy spectra reported in the past. The spectralindex of all datasets ranged from 2.22 to 2.62, and are compatiblewithin statistical errors.
IV. D
ISCUSSION
VERITAS observations of M87 spanning three ob-serving seasons (from 2007 to present) have shown M87in steady emission state and flaring state. The spectraobtained from these observations show no significantchanges in the spectral index. During the 2008 jointmonitoring campaign of M87, VERITAS observed agamma-ray flare in February 2008 which spanned 4days, constraining the emission region size to R ≤ R var = δ c ∆ t/ (1 + z ) = δ cm ∼ . δ R s where δ is the relativistic Doppler factor and R s theSchwarzschild radius of the M87 black hole. Rapidvariability reported previously [6] [7] constrained thesize of the TeV emission region to < . δR s . ROCEEDINGS OF THE 31 st ICRC, Ł ´OD ´Z 2009 3
Fig. 2.
Upper panel:
Chandra X-ray lightcurves measured from the core and the knot HST-1.
Lower panel:
Nightly fluxes for energies above250 GeV from 2007 to present. Grey area highlights the 2008 flare period observed by VERITAS and the corresponding X-ray fluxes observedby Chandra. For details on the 2008 flare period, see [1]
Even though the gamma-ray observing technique can-not resolve individual features of M87, the TeV emissionsize constraint has narrowed down the most probableTeV emission location to the unresolved core region andthe knot HST-1. The 2008 gamma-ray flare coincidedwith the Chandra observation of historically high fluxcoming from the core while the nearby knot HST-1appeared to be inactive (figure 2) [1]. The contempo-raneous gamma-ray and X-ray flares suggest the core ismore likely the TeV emission region, in contrast to the2005 flaring activity in TeV observed by H.E.S.S. andin the knot HST-1 in X-rays by Chandra [16].The 2008 multi-wavelength observations of M87 in-cluded concurrent radio, X-ray, and TeV gamma-raycoverage of M87 flaring activity from the core regionin early 2008 [28] [1]. From the 2008 multiwavelengthdata, the TeV emission region is likely the unresolvedcore. However, the knot HST-1 is still a possible can-didate for the 2005 flare. Current models do not favorone over the other, and both the core and HST-1 remainas candidates for TeV emission. As of the end of April2009, the 2009 monitoring work has shown no flaring activity from M87. Further multi-wavelength monitoringcan potentially provide additional constraints on theenvironment of M87 and more insights into the emissionmechanism of AGN.A
CKNOWLEDGMENTS
This research is supported by grants from the USDepartment of Energy, the US National Science Foun-dation, and the Smithsonian Institution, by NSERC inCanada, by Science Foundation Ireland, and by STFCin the UK. We acknowledge the excellent work of thetechnical support staff at the FLWO and the collaborat-ing institutions in the construction and operation of theinstrument. R
EFERENCES[1] VERITAS, VLBA 43 GHz M87 monitoring team, the HESS, andMAGIC 2009, accepted[2] Acciari, V. A. et al. (VERITAS collaboration) in prep[3] Acciari, V. A. et al. (VERITAS collaboration) 2008, ApJ, 679,397[4] Acciari, V. A. et al. (VERITAS collaboration) 2008, ApJ, 679,1427[5] Aharonian, F. et al. (HESS collaboration) 2003, A&A, 403, L1
C. M. HUI et al.