Discovery of Variability in the Very High Energy γ-Ray Emission of 1ES 1218+304 with VERITAS V. A. Acciari1, E. Aliu2, M. Beilicke3, W. Benbow1, D. Boltuch4, M. B¨ottcher5, 0 1 S. M. Bradbury6, V. Bugaev3, K. Byrum7, A. Cesarini8, L. Ciupik9, P. Cogan10, W. Cui11, 0 2 R. Dickherber3, C. Duke12, A. Falcone13, J. P. Finley11, G. Finnegan14, L. Fortson9, n a A. Furniss15, N. Galante1, D. Gall11, K. Gibbs1, R. Guenette10, G. H. Gillanders8, J 5 S. Godambe14, J. Grube16, D. Hanna10, C. M. Hui14, T. B. Humensky17, A. Imran18,∗, 1 P. Kaaret19, N. Karlsson9, M. Kertzman20, D. Kieda14, H. Krawczynski3, F. Krennrich18, ] E H M. J. Lang8, S. LeBohec14, G. Maier10, S. McArthur3, A. McCann10, P. Moriarty21, . h T. Nagai18, R. A. Ong22, A. N. Otte15, D. Pandel16, J. S. Perkins1, A. Pichel23, M. Pohl18, p - o J. Quinn16, K. Ragan10, L. C. Reyes24, P. T. Reynolds25, E. Roache1, H. J. Rose6, r t s M. Schroedter18, G. H. Sembroski11, A. W. Smith7, D. Steele9, S. P. Swordy17, a [ M. Theiling1, S. Thibadeau3, V. V. Vassiliev22, S. Vincent14, S. P. Wakely17, 1 v T. C. Weekes1, A. Weinstein22, T. Weisgarber17, D. A. Williams15 0 9 5 2 . 1 0 0 1 : v i X r a – 2 – 1Fred Lawrence Whipple Observatory, Harvard-Smithsonian Center for Astrophysics, Amado, AZ 85645, USA 2Department of Physics and Astronomy, Barnard College, Columbia University, NY 10027, USA 3Department of Physics, Washington University, St. Louis, MO 63130, USA 4Department of Physics and Astronomy and the Bartol Research Institute, University of Delaware, Newark, DE 19716, USA 5Astrophysical Institute, DepartmentofPhysics andAstronomy, OhioUniversity, Athens, OH 45701, USA 6School of Physics and Astronomy, University of Leeds, Leeds, LS2 9JT, UK 7Argonne National Laboratory, 9700 S. Cass Avenue, Argonne, IL 60439, USA 8School of Physics, National University of Ireland Galway, University Road, Galway, Ireland 9AstronomyDepartment, AdlerPlanetariumandAstronomyMuseum, Chicago, IL60605, USA 10Physics Department, McGill University, Montreal, QC H3A 2T8, Canada 11Department of Physics, Purdue University, West Lafayette, IN 47907, USA 12Department of Physics, Grinnell College, Grinnell, IA 50112-1690, USA 13Department of Astronomy and Astrophysics, 525 Davey Lab, Pennsylvania State Uni- versity, University Park, PA 16802, USA 14Department of Physics and Astronomy, University of Utah, Salt Lake City, UT 84112, USA 15Santa Cruz Institute for Particle Physics and Department of Physics, University of Cal- ifornia, Santa Cruz, CA 95064, USA 16School of Physics, University College Dublin, Belfield, Dublin 4, Ireland 17Enrico Fermi Institute, University of Chicago, Chicago, IL 60637, USA – 3 – Received ; accepted To appear in ApJLetter 18Department of Physics and Astronomy, Iowa State University, Ames, IA 50011, USA 19Department of Physics and Astronomy, University of Iowa, Van Allen Hall, Iowa City, IA 52242, USA 20Department of Physics and Astronomy, DePauw University, Greencastle, IN 46135-0037, USA 21Department of Life and Physical Sciences, Galway-Mayo Institute of Technology, Dublin Road, Galway, Ireland 22Department of Physics and Astronomy, University of California, Los Angeles, CA 90095, USA 23Instituto de Astronomia y Fisica del Espacio, Casilla de Correo 67 - Sucursal 28, (C1428ZAA) Ciudad Autnoma de Buenos Aires, Argentina 24Kavli Institute for Cosmological Physics, University of Chicago, Chicago, IL 60637, USA 25DepartmentofAppliedPhysics andInstrumentation, CorkInstituteofTechnology, Bish- opstown, Cork, Ireland *Corresponding author: [email protected] – 4 – ABSTRACT We present results from an intensive VERITAS monitoring campaign of the high-frequency peaked BL Lac object 1ES 1218+304 in 2008/2009. Although 1ES 1218+304 was detected previously by MAGIC and VERITAS at a persistent level of ∼6% of the Crab Nebula flux, the new VERITAS data reveal a prominent flare reaching ∼20% of the Crab. While very high energy (VHE) flares are quite commoninmanynearbyblazars, thecaseof1ES1218+304(redshift z = 0.182)is particularly interesting since it belongs to a group of blazars that exhibit unusu- allyhardVHE spectra considering their redshifts. When correcting the measured spectra for absorption by the extragalactic background light, 1ES 1218+304 and a number of other blazars are found to have differential photon indices Γ ≤ 1.5. The difficulty in modeling these hard spectral energy distributions in blazar jets has led to a range of theoretical γ-ray emission scenarios, one of which is strongly constrained by these new VERITAS observations. We consider the implications of the observed light curve of 1ES 1218+304, which shows day-scale flux varia- tions, for shock acceleration scenarios in relativistic jets, and in particular for the viability of kiloparsec-scale jet emission scenarios. Subject headings: BL Lacertae objects: individual(1ES 1218+304 = VER J1221+301) — galaxies: active — gamma rays: galaxies – 5 – 1. INTRODUCTION The very high energy (VHE) γ-ray sky has seen a dramatic increase in the number of sources, with 26 detected active galactic nuclei1 (AGNs) over redshifts of 0.0018 -0.536. All, except two, are blazars with their jets closely aligned to our line of sight. Blazars as a class are comprised of BL Lacertae objects, optically violent variables and other radio-loud quasars (Urry & Padovani 1995). Non-thermal emission dominates the broad-band continuum spectra of blazars. The characteristic double-peaked spectral energy distribution of blazars is usually attributed to synchrotron radiation in the radio to X-ray wavebands, and inverse- Compton up-scattering of soft target photons by relativistic electrons at GeV and TeV energies (Maraschi et al. 1992; Bloom & Marscher 1996). Alternative emission scenarios involve hadronic interactions producing neutral pions which decay into photons (Mannheim & Bierman 1992) and proton synchrotron radiation (Aharonian 2000). The latter models provide a link between AGNs and possible sources of ultra-high-energy cosmic rays. A better understanding of the underlying acceleration and emission processes requires a larger sample of sources, which is now being provided by the Fermi Gamma-ray Space Telescope (Fermi) at 0.1 GeV to hundreds of GeV and by the imaging atmospheric Cherenkov telescopes (IACT), namely the MAGIC, H.E.S.S. and VERITAS Observatories, that provide coverage primarily in the 100 GeV−10 TeV regime. To date, blazars detected at VHE energies are predominantly high-frequency peaked BL Lacertae objects (HBLs). Nevertheless, a few objects with lower peak energies, namely low-frequency-peaked BL Lacertae (LBLs; (Albert et al. 2007; Teshima et al. 2008)) and intermediate-frequency-peaked BL Lacertae objects (IBLs) (see Acciari et al. 2008a,b, 1For updates see http://tevcat.uchicago.edu – 6 – 2009b; Ong et al. 2009) were detected recently by IACTs at TeV energies. The γ-ray luminosity of LBLs and IBLs generally peaks at sub-GeV to tens of GeV in energy, and they are therefore easily detectable by Fermi while their detectability in the TeV regime by IACTs is more difficult due to their soft VHE spectra. Among the HBLs, which are typically the domain of the TeV telescopes, a number of objects exhibit unusually hard intrinsic power law energy spectra (dN/dE ∼ E−Γi) after correcting for the γ γ → e+ e− absorption by the cosmological diffuse EBL radiation field. While the measured spectral indices, Γ , of these blazars (1ES 1101–232, m 1ES 0347–121, 1ES 0229+200, 1ES 1218+304, RGB J0710+591) range from 2.5 to 3.1 (see Aharonian et al. 2006a, 2007a,b; Albert et al. 2006; Acciari et al. 2009a; Perkins et al. 2009), the absorption-corrected spectral indices suggest very hard intrinsic source spectra in the VHE regime with Γ ≤ 1.28±0.28 (Krennrich et al. 2008), based on recent lower i limits to the EBL from galaxy counts (Levenson & Wright 2008). Similar hard energy spectra are found by Fermi, where, for example, the measured spectral indices for RGB J0710+591 and 1ES 1440+122 are found to be Γ = 1.21±0.25 and 1.18±0.27 over energies from 0.2 GeV to tens of GeV (Abdo et al. 2009a). While the findings of very hard VHE spectra are based on EBL constraints from galaxy counts, the Fermi spectra directly resemble the intrinsic source spectra since absorption by the EBL is minimal for the covered energy range. The photon spectrum of 1ES 1218+304 measured by Fermi yields a spectral index of Γ = 1.63±0.12 between 0.2 GeV and ∼300 GeV, but absorption may already play a role at the high-energy end (Abdo et al. 2009a). Diffusive shock-acceleration theory (for a review see Malkov & Drury 2001) generally yields a limit to the spectral index of GeV – TeV photon spectra resulting from inverse Compton scattering of Γ ≥ 1.5. Only recently, numerical studies by Stecker et al. (2007) i indicate that sufficiently hard electron spectra could be generated by diffusive shock – 7 – acceleration in relativistic shocks, for the production of photon spectra 1.0 < Γ < 1.5. i However, B¨ottcher et al. (2008) suggest that even for a hard-spectrum electron population, in the framework of a synchrotron self-Compton (SSC) scenario, the resulting GeV–TeV γ-ray spectra should experience substantial softening from Klein–Nishina effects making γ-ray spectra with Γ < 1.5 less likely. i Other approaches to explain the hard γ-ray spectra are offered by ad hoc assumptions about the electron distribution, an additional absorption component in the source, or a postulate of new physics describing the propagation of γ-rays on extragalactic distance scales. Katarzyn´ski et al. (2006) invoke a high low-energy cutoff in the electron distribution that could give the appearance of a hard γ-ray spectrum for a given energy regime. Aharonian et al. (2008) show that γγ absorption in the source due to a narrow-band emission component from the AGN could lead to unusually hard VHE spectra. Finally, S´anchez-Conde et al. (2009) suggested an axion like particle (ALP) that would distort the γ-ray spectrum through ALP/photon mixing on cosmological distances in the presence of intergalactic magnetic fields. A more easily testable model that involves known physics was recently proposed by B¨ottcher et al. (2008). The authors attribute the hard photon spectra to inverse-Compton up-scattering of ambient photons from the cosmic microwave background (CMB), occurring in a kiloparsec-scale jet. In this case, a substantial fraction of the jet power is transported by hadrons to the outer regions of the jet, where it is dissipated into ultra-relativistic electrons. Inverse-Compton scattering in the Klein–Nishina regime is no longer the limitation at the highest energies since low-energy target photons from the CMB are abundant. The low magnetic field in the large-scale jet also avoids the overproduction of synchrotron radiation and allows Compton emission to dominate. Furthermore, the synchrotron emission in the radio to X-ray wavebands is assumed to originate on sub-parsec scales. A similar – 8 – mechanism was previously suggested to explain the hard X-ray spectra observed from the kiloparsec-scale jets of radio quasars (Tavecchio et al. 2007). To date, all of the blazars exhibiting very hard spectra appear to emit at a baseline level that is consistent with this picture. However, it is also possible that the sensitivity of the current IACTs is the limiting factor in detecting day scale variations and that the underlying emission may contain flares. The question as to whether or not the γ-ray emission from 1ES 1218+304 consists of flares or corresponds to a constant baseline emission level was one of the main motivations for VERITAS to monitor this object over a period of 5 months in the 2008/2009 season. The other motivation for deep exposures of relatively large redshift (z = 0.1−0.2) blazars at TeV energies is to provide better constraints on the EBL spectrum. The observed VHE photon spectrum can be used to constrain the EBL in the near to mid-IR. This is particularly promising for hard-spectrum blazars that provide the best sensitivity to any possible absorption feature. 2. OBSERVATIONS AND DATA ANALYSIS VERITAS is an array of four 12 m IACTs located at the base camp of the F.L. ◦ ◦ Whipple Observatory in southern Arizona (31.68N, 110.95W, 1.3 km above sea level). Each ◦ VERITAS focal plane instrument has a 3.5 field of view and consists of 499 photomultiplier tubes. The stereoscopic system has been fully operational since mid-2007 and is capable of detecting sources with a flux of 1% of the Crab Nebula in < 50 hr of observation. In Summer 2009, Telescope 1 was relocated and the improved array configuration allows us to detect a point source at 1% of the Crab Nebula flux under 30 hr of observations. The large effective area (∼ 105 m2) of the array enables VERITAS to be sensitive over a wide range of energies (from 100 GeV to 30 TeV) with an energy resolution of 15%−20% above 300 GeV. For further details about VERITAS, see e.g., Holder et al. (2006). – 9 – 1ES 1218+304 was observed by VERITAS from 2008 December to 2009 May. All data were taken in wobble mode where the source is positioned 0◦.5 from the center of the camera in order to allow simultaneous background estimation. A quality selection was applied to the data to reject observations affected by poor weather conditions and other anomalies, resulting in a total live time of 27.2 hr for analysis. The mean zenith angle of ◦ the observations was 12 . The data are analyzed using standard VERITAS calibration and analysis tools. Initially, the observed data set is calibrated by flat-fielding the camera gains with nightly laser runs. Following calibration and cleaning, the shower images are parameterized using a second-moment analysis (Hillas 1985). Images from individual telescopes are combined to reconstruct both arrival direction and impact position for each event. The γ-ray-like events are separated from the cosmic ray background events using cuts on the mean-reduced-scaled length and mean-reduced-scaled width parameters (see Aharonian et al. 2006b for a full description of the parameters). The cuts for γ-ray/hadron separation are optimized a priori for a source with a 10% Crab-like flux. A circular region of radius θ is applied around the target to estimate the ON-source events. The background or OFF-source counts are estimated with off-source reflected regions (Berge et al. 2007). The significance of any excess (ON–OFF) events is calculated following Equation (17) in Li & Ma (1983). The angular distribution of excess events from the direction of 1ES 1218+304 (Figure 1) is comparable with the expectation from the Monte Carlo (MC) simulation of a point source. MC simulations are used to generate a lookup table for estimating the energies of the primary γ-rays as a function of the image size and impact parameters. For spectral analysis, a looser cut (spectrum cuts) on the arrival direction of γ-rays (a region with θ2 = 0◦.03) is applied along with three off-source reflected-regions. This has the benefit of retaining a higher number of γ-ray candidates in addition to reducing the systematic uncertainties – 10 – associated with the MC detector model. Both the standard cuts (θ2 = 0◦.0225) and the spectrum cuts yield compatible results. Consistent results are attained with independent analysis packages. 3. RESULTS Table 1 summarizes the results of the VERITAS observations of 1ES 1218+304. For the spectral analysis, we report an excess of 1155 events with a statistical significance of 21.8 standard deviations, σ, from the direction of 1ES 1218+304 during the 2008-2009 campaign (2808 signal events, 4959 background events with a normalization of 0.33). Figure 2 shows the corresponding time-averaged differential energy spectrum. The spectrum extends from 200 GeV to 1.8 TeV and is well described (χ2/dof = 8.2/7) by a power law, dN E = (11.5±0.7)×10−12 ( )−3.07±0.09stat cm−2s−1TeV−1 (1) dE 500GeV The errors bars are statistical only. Figure 3 shows the light curve of 1ES 1218+304 for the 2008-2009 season (binned by night). The average integral photon flux above 200 GeV is Φ(E > 200 GeV) = (18.4±0.9 )×10−12 cm−2 s−1 corresponding to ∼7% of the Crab Nebula flux above the stat same threshold. The light curve from 1ES 1218+304 shows strong nightly flux variations between 2009 January 25 and 2009 February 5. The highest γ-ray flux is observed on MJD 54861 (2009 January 30) and is ∼20% of the Crab Nebula flux. The inset within Figure 3 shows the light curve above 200 GeV for nights with the highest flux. A fit of a constant to the emission during the entire season yields a χ2/dof of 106/21 (Figure 3, dashed dotted line), corresponding to a very small χ2 probability (∼ 10−13). The data taken during the active period in January/February of 2009 indicate variability on a time-scale of days.