We are leading a comprehensive multi-waveband monitoring program of 34 gamma-ray bright blazars designed to locate the emission regions of blazars from radio to gamma-ray frequencies. The "maps" are anchored by sequences of images in both total and polarized intensity obtained with the VLBA at an angular resolution of ~ 0.1 milliarcseconds. The time-variable linear polarization at radio to optical wavelengths and radio to gamma-ray light curves allow us to specify the locations of flares relative to bright stationary features seen in the images and to infer the geometry of the magnetic field in different regions of the jet. Our data reveal that some flares occur simultaneously at different wavebands and others are only seen at some of the frequencies. The flares are often triggered by a superluminal knot passing through the stationary "core" on the VLBA images. Other flares occur upstream or even parsecs downstream of the core.
💡 Deep Analysis
Deep Dive into Multi-waveband Emission Maps of Blazars.
We are leading a comprehensive multi-waveband monitoring program of 34 gamma-ray bright blazars designed to locate the emission regions of blazars from radio to gamma-ray frequencies. The “maps” are anchored by sequences of images in both total and polarized intensity obtained with the VLBA at an angular resolution of ~ 0.1 milliarcseconds. The time-variable linear polarization at radio to optical wavelengths and radio to gamma-ray light curves allow us to specify the locations of flares relative to bright stationary features seen in the images and to infer the geometry of the magnetic field in different regions of the jet. Our data reveal that some flares occur simultaneously at different wavebands and others are only seen at some of the frequencies. The flares are often triggered by a superluminal knot passing through the stationary “core” on the VLBA images. Other flares occur upstream or even parsecs downstream of the core.
📄 Full Content
One of the most fascinating properties of blazars is their ability to channel most of their apparent luminosity into high-energy photons. The γ-ray luminosity can be as much as three orders of magnitude higher than that at other wavebands. Furthermore, the emission regions must be very small, since the time-scales of variability can be as short as hours. While relativistic beaming with Doppler factors of 10-50 or even higher can bring down the luminosities and increase the upper limits to the sizes of the emission regions, we are still faced with two problems: (1) getting TeV photons out without rampant attenuation by pair production if they originate within ∼ 10 17 cm of the central engine, as many theorists favor, and (2) explaining the short time-scales of variability if the γ-rays are emitted parsecs from the black hole. In order to determine which of these difficulties we are facing, we need to specify where the γ-ray flares occur in blazars. We are leading a collaboration that has mounted a comprehensive program aimed to do just that.
The heart of our program (see www.bu.edu/blazars
) is monitoring 34 γ-ray bright blazars ∼ monthly with the VLBA at 43 GHz, which produces images of the compact jets with an angular resolution of ∼ 0.1 milliarcseconds. We also observe a subset of this sample, as well as some TeV blazars not in the sample, for 10-14 days during intense campaigns two times per year in order to examine the shortterm variability properties. Every blazar contains a bright feature at the upstream end that is referred to as the “core.” At lower frequencies, the images contain a “pseudo-core” that is usually the location where the jet becomes optically thick to synchrotron self-absorption. At millimeter wavelengths, however, the core in most blazars has the properties of a physical feature with a quasi-stationary position. Since the core is so prominent and is at least approximately stationary, it is very convenient to use as a reference for locating the sites of flares at higher frequencies, where images are far too coarse to resolve the nuclei.
Our technique for locating the flares relative to the core is most accurate when we can measure the optical linear polarization of a flare and find that the electric vector position angle χ is essentially the same as that of a feature seen (perhaps with a time delay) in the VLBA images. If that value of χ is unique to the feature, then we can identify the flare as occurring within that feature with a high level of confidence. If, as is very often the case, the optical flare is coincident with a γ-ray and/or X-ray flare, then we can conclude that the high-energy emission arises in the same feature. We can associate the flares based solely on near-sumultaneity of peaks in the light curves as well if such flares occur infrequently enough to render a chance coincidence highly unlikely.
Here we report the results of our monitoring program for some blazars that have exhibited flaring events at γ-ray energies as observed by the Fermi Gammaray Space Telescope. While some of the flares seem to occur between the 43 GHz core and the central engine, we find that many events must occur at or beyond the core, which is located parsecs downstream of the black hole.
Figures 1 and 2 present multi-waveband light curves of BL Lac and 3C 279, respectively. In the case of BL Lac, the X-ray, γ-ray, and radio fluxes all follow the same long-term rising trend starting in early 2009, while the optical light curve does not do so until mid-2009. At the onset of this rise, the X-ray flux increases Figure 1. Multi-waveband light curves of BL Lac. In the top panel, the data correspond to 5-day integrations (black data points, blue 2-σ upper limit arrows) or, during flares, 1-day integrations (red data points, green upper limits); flares discussed in the text are marked by vertical blue dashed lines. Times when superluminal knots passed the core are marked with upward arrows, with the horizontal line segments indicating the uncertainties in these times. The images in summer 2010 contain a knot whose motion has not yet been determined; if its apparent speed is the same as for previous knots, it passes through the core at the time indicated by the dotted arrow. From Marscher et al. (in prep.) by nearly a factor of 3 (Flare 1) over 40 days while the γ-ray emission fluctuates, with a 2-day peak that is simultaneous with the X-ray maximum. There is no sign, however, of the event at the optical or radio bands. Flare 2 is a major, sharp γ/X-ray flare and a major radio outburst that is delayed by one week, with only a hint of an optical counterpart. A superluminal knot passes through the 43 GHz core on the VLBA images as the radio flux rises. Another superluminal knot passes through the core near JD 2455090 during a more pronounced optical and modest γ-ray flare (3), with no obvious sign of the event at the X-ray energies. In early 2010, the very sharp Flare 4 occurs simultaneously at γ-ra
