Highlights of Recent Multiwavelength Observations of VHE Blazars with VERITAS
aa r X i v : . [ a s t r o - ph . H E ] J u l PROCEEDINGS OF THE 31 st ICRC, Ł ´OD ´Z 2009 1
Highlights of Recent Multiwavelength Observationsof VHE Blazars with VERITAS
J. Grube ∗ for the VERITAS collaboration †∗ School of Physics, University College Dublin, Belfield, Dublin 4, Ireland † see R.A. Ong et al. (these proceedings) or http://veritas.sao.arizona.edu/conferences/authors?icrc2009 Abstract . We present long-term observations ofseveral VHE (E >
100 GeV) blazars with VERITAS,together with contemporaneous
Swift and
RXTE
X-ray data. The observed targets include Mrk 421,Mrk 501, 1ES 2344+514. Strong flux and spectralvariability is seen in Mrk 421 on nightly time-scalesbetween January and June 2008, revealing a highlycorrelated X-ray to GeV/TeV connection. Modest X-ray variability is evident in Mrk 501. Observationsof 1ES 2344+514 in December 2007 show VHE γ -rayand X-ray flux doubling on nightly time-scales. Keywords : Gamma-ray Astronomy, Active GalacticNuclei, VERITAS
I. I
NTRODUCTION
Blazars (BL Lac objects and Flat Spectrum RadioQuasars) are active galactic nuclei (AGN) with a rela-tivistic plasma jet oriented close to the line of sight [1].These objects exhibit rapid variability and have broad-band spectral energy distributions (SEDs) characterized,in a ν F ν representation, by a synchrotron componentextending from radio to X-ray frequencies, and a secondcomponent peaking at γ -ray frequencies due to eitherinverse-Compton radiation or from hadronic processes.BL Lac type blazars are further sub-divided based onthe peak frequency of the synchrotron emission aslow-, intermediate-, or high-frequency-peaked BL Lacs(LBLs, IBLs, and HBLs). Currently, 20 HBLs from atotal of ∼
25 VHE blazars are detected at very highenergy (VHE, E >
100 GeV) γ -rays. Elsewhere in theseproceedings, the VERITAS blazar observing programis described [2], and the discovery of VHE emis-sion with VERITAS from two HBLs (RGB J0710+591and 1ES 0806+524), and from two IBLs (3C 66A andW Comae) is highlighted [3].Multiwavelength campaigns on BL Lacs are importantto sample the broadband flux and spectral variabilityon time-scales ranging from minutes to months. Recentjoint observing campaigns with VERITAS on two IBLsare presented elsewhere in these proceedings. Lightcurves and the broadband SED of 3C 66A are presentedfrom contemporaneous VERITAS,
Fermi , X-ray, andoptical data between September and November 2008 [4].Observations of W Comae in June 2008 reveal X-rayand VHE γ -ray flaring, with a marginal detection above100 MeV from AGILE observations [5]. In this paper TeVCat catalog of VHE γ -ray sources: http://tevcat.uchicago.edu ] - T e V - s - d N / d E [ c m -14 -13 -12 -11 -10 -9 Energy [TeV]
Very low stateLow stateMid stateHigh A stateHigh B stateHigh C stateVery high state
VERITAS (2008)
Fig. 1. Time-averaged VERITAS photon spectra of Mrk 421 fordiscrete flux levels. A power law with exponential cutoff model dN / dE = I o · (E / − Γ · exp( − E / E cut ) is fit to each spectrum. detailed multiwavelength results on the bright and well-studied HBLs: Mrk 421, Mrk 501, and 1ES 2344+514are presented. Multiwavelength results on the recentlyobserved HBL RGB J0710+591 will also be presentedat the conference.II. VERITAS AND X- RAY D ATA A NALYSIS
VERITAS is an array of four imaging atmospheric-Cherenkov telescopes located in southern Arizona. Thearray is sensitive over the energy range of 100 GeVto >
30 TeV, and can detect (5 σ level) a source fluxof 5% of the Crab Nebula flux in ∼ wobble mode, where the source is positioned at a fixedoffset of 0.5 ◦ from the camera center, was used forall observations presented here. Standard data reductionand γ -ray selection cuts are applied [6]. The resultspresented here agree well with those performed usingan independent VERITAS analysis package.Long-term X-ray observations with RXTE
PCA [7]and
Swift
XRT [8] were taken contemporaneously withthe VERITAS data. All
RXTE
PCA data were takenwith only PCU2 operational. For all
Swift
XRT data onMrk 421 the observations were taken in window timing(WT) mode, while for 1ES 2344+514 the XRT data wereall taken in photon counting (PC) mode. Data reductionis performed with the
HEAsoft
J. GRUBE et al.
MULTIWAVELENGTH OBSERVATIONS OF VHE BLAZARS WITH VERITAS
470 480 490 500 510 520 530 540 550 560 570 580 590 600 610 620 630 ] − s − c m − F ( > G e V ) [ Jan. 2008 Feb. 2008 Mar. 2008 Apr. 2008 May 2008 Jun. 2008VERITAS
470 480 490 500 510 520 530 540 550 560 570 580 590 600 610 620 630 ) − s − er g c m − · F [ − k e V ] ( RXTE PCASwift XRT
Date (MJD − 54000)470 480 490 500 510 520 530 540 550 560 570 580 590 600 610 620 630 I nd e x a RXTE PCASwift XRT
Fig. 2. VHE γ -ray and X-ray light curve of Mrk 421. Shown in the top panel are VERITAS F( >
300 GeV) fluxes from observations of 20minutes each. The middle panel shows the 2–10 keV fluxes from observations with
RXTE
PCA (circles) and
Swift
XRT (open squares). Thebottom panel shows the X-ray spectral indices ( a parameter, see text) from log-parabolic fits to the 0.4–10 keV Swift
XRT spectra and 3–20 keV
RXTE
PCA spectra. ] -1 s -2 erg cm -10 F(2-10 keV) [ 105 10 15 20 25 ] - s - c m - F ( > G e V ) [ RXTE PCASwift XRT
Fig. 3. VERITAS flux F( >
300 GeV) from 20 min exposures versusthe X-ray flux for observations of Mrk 421 taken within ± source regions are used to extract the Swift
XRT datain PC mode when photon pileup is evident [9].III. M RK = γ -ray energies. VERITAS ob-served Mrk 421 between January and June 2008 fora total live-time of 43.6 hours. An average analysisenergy threshold of 260 GeV is determined for theobservations that span a range of zenith angles from 6 ◦ –56 ◦ . In the whole data set an excess of ∼ γ -ray events is detected, corresponding to a significanceof > σ . Previously observed VHE γ -ray spectralvariability correlated with flux level [11] is investigatedby dividing the VERITAS data into subsets accordingto flux level. Figure 1 shows the VERITAS spectrain six subsets with best-fit curves for a power lawwith exponential cutoff model of the form dN/dE = I o · (E / − Γ · exp( − E / E cut ) . The nightly X-rayspectra from RXTE
PCA and
Swift
XRT data are bestfit with an absorbed log-parabolic model, which usesa fixed column density [10] and an energy dependentphoton index
Γ = a + b · Log(E / E o ). The mean reduced χ values from log-parabolic fits to the RXTE
PCAand
Swift
XRT spectra are 0.87 and 1.20 compared toreduced χ values of 1.35 and 1.80 for an absorbedpower law model.Figure 2 shows on the top panel the nightly VHE γ -ray fluxes F( >
300 GeV) from nightly VERITAS ob-servations. Shown on the middle and bottom panels areX-ray fluxes F(2–10 keV) and spectral indices from
RXTE
PCA and
Swift
XRT observations. Strong fluxvariability on nightly time-scales is seen over the entiresix month period in 2008, with exceptionally bright X-ray flaring in March to April, and VHE γ -ray flaringin early May. A measure of the integrated level offlux variability is the fractional root-mean-square (rms) ROCEEDINGS OF THE 31 st ICRC, Ł ´OD ´Z 2009 3
555 560 565 570 575 580 585 590 595 600 ] - s - c m - F ( > G e V ) [ Apr. 2008 May 2008 VERITAS
555 560 565 570 575 580 585 590 595 600 ] − s − er g c m − F ( − k e V ) [ RXTE PCA
Date [ MJD − 54000 ]555 560 565 570 575 580 585 590 595 600 G X − r ay I nd e x RXTE PCA
Fig. 4. VHE γ -ray and X-ray light curve of Mrk 501. Shown in the top panel are the nightly VERITAS F( >
300 GeV) fluxes. The middleand bottom panels shows the
RXTE
PCA 2–10 keV fluxes and power law indices, respectively. variability amplitude F var [12]. Significant VHE γ -rayflux variability of F var = (66 . ± . )% and X-rayflux variability F var = (57 . ± . )% was observed.Figure 3 shows the VHE γ -ray flux versus X-ray fluxfor observations taken within ± γ -ray fluxes on May 2 andMay 3 there were no simultaneous RXTE
PCA or
Swift data. Combining all VHE γ -ray flux versus X-ray fluxpoints, the correlation coefficient is r = 0 . ± . .Clear X-ray spectral hardening with increasing 2–10 keVflux is shown in figure 2, with the log-parabolic indexparameter a ranging from ∼ RK = γ -ray energies by the Whipple 10 m telescopein 1996 [13]. VERITAS observed Mrk 501 betweenApril and June 2008 for a total live-time of 6.2 hours.An excess with the statistical significance of 22 σ ismeasured for the total data set. Figure 4 shows thenightly VERITAS VHE fluxes F( >
300 GeV) and
RXTE
PCA 2–10 keV fluxes and power law indices. Marginalvariability is seen in the VERITAS F( >
300 GeV) fluxeswith F var = (14 ± )%, while moderate 2–10 keVflux variability is evident with F var = (12 . ± . )%.More detailed results from this campaign are presentedelsewhere in these proceedings [14]. V. 1ES 2344+5141ES 2344+514 is another close HBL (z = γ -ray energies by the Whipple10 m telescope in 1995 [15]. Figure 5 shows the nightlyVHE γ -ray and X-ray light curve of 1ES 2344+514 fromVERITAS, RXTE
PCA, and
Swift
XRT observations. Astrong VHE γ -ray flare is seen on December 7, 2007(54441.12 MJD) at a flux F( >
300 GeV) correspond-ing to 48% of the Crab Nebula flux. The measuredincrease in flux of a factor of 1.92 ± ∼ day time-scale VHE γ -ray variability from1ES 2344+514 since the initial Whipple 10 m detectionin 1995. Excluding the December 7 flaring event, theaverage F(E >
300 GeV) is 7.6% of the Crab Nebula flux.For the full VERITAS data set a high level of variabilityF var = (75 ± )% is implied. Excluding the flare night,a F var = (34 ± )% is determined.Figure 5 (lower panels) shows the 2–10 keV flux andphoton index Γ measured over 3–20 keV from RXTE
PCA and 0.4–10 keV from
Swift
XRT data. The X-ray flux is shown to be highly variable throughout thecampaign, with F var = (51 ± )%. In December 2007,large amplitude flaring is evident with flux doublingtime-scales of ∼ (6 . ± . × − erg cm − s − is seen from the SwiftXRT data on December 8, 2007, representing the highestX-ray flux ever measured for 1ES 2344+514. Figure 6shows the VERITAS flux F( >
300 GeV) versus
RXTE
PCA and
Swift
XRT 2–10 keV fluxes for nights with
J. GRUBE et al.
MULTIWAVELENGTH OBSERVATIONS OF VHE BLAZARS WITH VERITAS
370 375 380 385 390 395 400 405 410 415 420 425 430 435 440 445 450 455 460 465 470 475 480 ] − s − c m − F ( > G e V ) [ Oct. 2007 Nov. 2007 Dec. 2007 Jan. 2008VERITAS
370 375 380 385 390 395 400 405 410 415 420 425 430 435 440 445 450 455 460 465 470 475 480 ] − s − er g c m − F ( − k e V ) [ RXTE PCASwift XRT
Date [ MJD − 54000 ]370 375 380 385 390 395 400 405 410 415 420 425 430 435 440 445 450 455 460 465 470 475 480 G X − r ay I nd e x RXTE PCASwift XRT
Fig. 5. VHE γ -ray and X-ray light curve of 1ES 2344+514. Shown in the top panel are the nightly VERITAS F( >
300 GeV) fluxes. Themiddle panel shows the 2–10 keV fluxes from observations with
RXTE
PCA (circles) and
Swift
XRT (open squares). The bottom panel showsthe power law indices Γ from the 0.4–10 keV Swift
XRT spectra and 3–20 keV
RXTE
PCA spectra. ] −1 s −2 erg cm −11 F(2−10 keV) [ 101 2 3 4 5 6 ] − s − c m − F ( > G e V ) [ RXTE PCASwift XRT
Fig. 6. VERITAS γ -ray flux F( >
300 GeV) versus X-ray 2–10 keVflux for 1ES 2344+514 observations from nights with
RXTE
PCA(circles) and
Swift
XRT (open squares) data. observations in both energy bands. A Pearson coefficientof r = 0 . ± . is calculated for the VHE γ -ray toX-ray flux points, suggestive of correlated variability.Further results and discussion on this campaign are inpress [16]. VI. C ONCLUSIONS
Joint VHE γ -ray and X-ray observing campaigns onthe bright HBLs Mrk 421, Mrk 501, and 1ES 2344+514with VERITAS reveal significant flux variability on nightly time-scales. X-ray spectral hardening is shownat increasing flux levels. Results from synchrotron self-Compton (SSC) modeling of the broadband SED will bepresented at the conference.A CKNOWLEDGMENTSThis research is supported by grants from the US Department ofEnergy, the US National Science Foundation, and the SmithsonianInstitution, by NSERC in Canada, by Science Foundation Ireland,and by STFC in the UK. We acknowledge the excellent work of thetechnical support staff at the FLWO and the collaborating institutionsin the construction and operation of the instrument. R EFERENCES[1] Blandford, R. D., & K¨onigl, A. 1979, ApJ, 232, 34[2] Benbow, W. et al. 2009, these proceedings[3] Perkins, J. et al. 2009, these proceedings[4] Reyes, L. et al. 2009, these proceedings[5] Maier, G. et al. 2009, these proceedings[6] Acciari, V. A., et al. 2008, ApJ, 679, 1427[7] Johoda, K. et al. 1996, SPIE, 2808, 59[8] Burrows, D., et al. 2005, SSRv., 120, 165[9] Grube, J., et al. 2008, Proceedings of the 4 thth