Parsec-scale Jets in Lobe-dominated Quasars

David HoughTrinity University

Abstract: In order to test relativistic jet and AGN models over a wide range of orientations, we are conducting a VLBI survey of a complete sample of 3CR lobe-dominated quasars. With the benefit of intensive VLBA/HSA observations during the period 1995-2005, the nuclei in all 25 objects have been imaged at one or more frequencies between 5 and 22 GHz. In 22 objects, we find one-sided parsec-scale jets (three sources have barely resolved cores). The jets typically show bends of a few degrees in their inner few parsecs. Multiple-epoch observations have yielded at least rough speed estimates in 15 sources. Component speeds range from 0 to 10c, and there is a strong tendency for outer components to move faster than inner ones (3C245 is an exception, with alternating fast and slow components). Both the average and maximum speeds tend to correlate with R, the ratio of nuclear to extended emission at 5 GHz (emitted). Polarized emission has been detected in five objects: four show significant rotation measures in their cores/inner jets, two have long jets with longitudinal magnetic fields, and one has a distant jet feature with a transverse field. Implications for jet and AGN models will be discussed.

“Extragalactic Jets: Theory and Observation from Radio to Gamma Ray”May 21-24, 2007 – Girdwood, AK

Jets and AGN Unification

• In unification scenarios for active galactic nuclei (AGN), many lines of evidence suggest the “sub-unification” of powerful radio-loud sources, in which these objects are intrinsically similar but appear different due to the angle, , of their radio jet axes relative to our line-of-sight. In order of increasing , the sequence of object classes is blazars, core-dominated quasars (CDQs), lobe-dominated quasars (LDQs), and powerful Fanaroff-Riley Type II (FR-II) radio galaxies. While this sub-unification is generally accepted, the detailed physics of the parsec-scale jets in these sources is still not well understood.

• Great effort has gone into VLBI surveys of sources viewed at small – e.g., Pearson & Readhead 1988, ApJ, 328, 114; Kellermann et al. 2004, ApJ, 609, 539; Lister & Homan 2005, AJ, 130, 1389 – and these have been critical to the development of detailed parsec-scale jet models. To complement this work, studies of lobe-dominated quasars (LDQs) offer an important means of testing jet models over a wide range in orientation (e.g., 0o-60o). The nuclei in these objects are much brighter, on average, than those in FR-II radio galaxies, and can thus be imaged with standard VLBA observing modes.

The Lobe-dominated Quasar Sample

• Hough & Readhead (1989, AJ, 98, 1208) defined a complete, flux-density-limited sample of 25 LDQs from the 3CR catalogue. The objects were chosen on the basis of their low-frequency, extended lobe emission only, which should avoid orientation bias due to relativistic beaming in their compact cores. The ratio of core to extended emission at 5 GHz (emitted) is denoted by R; LDQs have R < 1. The resulting range of core flux density spans a factor of ~1000, from ~1 mJy to ~1 Jy!

• Hough et al. (2002, AJ, 123, 1258) report total-intensity VLBI observations of 24 of these objects. Al1 19 that were imaged show one-sided jets, often significantly curved. Jet speed estimates for 10 sources are in the range ~0-10c, but only 4 of these were well-determined, and they fall in the range ~0-5c. No more than a single VLBA “pilot” or imaging epoch was included for any source.

• Aars et al. (2005, AJ, 130, 23) report optical spectrophotometry for the complete sample that yields evidence for a disk-like component perpendicular to the radio jet. This particular study finds only a mildly restricted orientation range (0o-80o), broader than many other studies, e.g., maximum Barthel 1989, ApJ, 336, 606), but similar to the maximum found by Wardle & Aaron (1997, MNRAS, 286, 425).

• To advance our knowledge of LDQ jet morphology and speeds, and to make the first polarization-sensitive VLBI images of several of these sources, an intensive program of VLBA observations was conducted from 1995 to 2005.

VLBA Observations & Imaging• To complete imaging for all 25 LDQs in the 3CR sample, 8.4 GHz phase-referenced observations were

made for an additional six sources in the ~1-5 mJy range between 1997 and 2004: 3C9, 3C14, 3C181, 4C16.49, 3C432 with the VLBA, and 3C68.1 with the High Sensitivity Array (HSA).

• To measure jet speeds, most observations were done at 8.4 GHz; in some instances, 15/22 GHz were used for higher resolution on compact cores.

• Eight sources in a group covering a wide range of R were observed at numerous epochs (from 12 to 21) during 1995-2005: 3C204, 3C205, 3C207, 3C208, 3C212, 3C245, 3C249.1, 3C263. During ~2001, six epochs were observed with a shortened monitoring interval of 2-3 months to minimize ambiguities in component identifications and reduce uncertainties in component speeds; this was repeated in ~2003, and again in 2005 for 3C207 and 3C245 only.

• Five sources were observed well enough for useful imaging at two epochs between 1996 and 2001: 3C14, 3C190, 3C191, 3C275.1, 3C336.

• Polarization-sensitive imaging was done with simultaneous 5/8.4/15 GHz observations for seven sources between 1996 and 1999: 3C207, 3C212, 3C245, 3C263, 3C270.1, 3C275.1, 3C334.

• Data reduction and imaging were done with the NRAO AIPS and Caltech DIFMAP packages, with Gaussian modelfits to quantify image features.

• At 8.4 GHz, the typical resolution is ~1 mas and the typical image noise ~0.1 mJy/beam.

• Below are sample images of (1) 3C14, showing the jet in this faint ~6-mJy source for the first time; (2) 3C208, 20% super-resolved, showing the jet knots to the west of the core in this ~50-mJy source that were tracked from 1995 to 2003; and (3) 3C245, showing the C-X band rotation measure (color) and X-band polarization vectors on the X-band total intensity contours for the ~50-mas jet in this ~800-mJy source.

Jet Morphology and Speeds• All 25 nuclei in the 3CR LDQ complete sample have now been imaged. In 22 sources, we find a

one-sided parsec-scale jet fairly well aligned with the kiloparsec-scale jet (within a few degrees). Three sources – 3C68.1, 3C181, 3C432 – show only barely resolved VLBI cores.

• Useful multiple-epoch images have been obtained for 15 objects. For 10 sources, apparent transverse jet speeds have been well determined, and they fall in the range app= vapp/c ~0-10c. The number of jet knots per source ranges from 1 to 6. These 10 sources are used in the analyses below. (For each of the other 5 sources, app was measured for only a single component, and with a large uncertainty: 3C14 @ 5+/-5, 3C191 @ 4+/-4, 3C205 @ –4+/-7, 3C275.1 @ 2+/-2, 3C336 @ 2+/-2.)

• “Jet Knot Speed app vs. Projected Distance rproj Along Jet” is plotted below. (Proper motions and angular distances were converted into corresponding app and rproj using redshifts and a standard cosmology with qo=0.5, Ho=70 km/s/Mpc, m=0.3, and =0.7.) With the exception of 3C245, it is clear that component speeds tend to be faster at larger projected distances.

• Hough et al. (2002) defined a “pseudo-angle” based on core strength R and projected linear size Lproj, which they used to assign an orientation angle to each of the 3CR LDQs assuming a range from 10o to 45o. In a few cases, the pseudo-angle is inconsistent with app reported here, even as the true jet speed approaches c. Therefore, we have reassigned orientation angles on the basis of R alone. These angles have been used to find the “Jet Lorentz Factor vs. Deprojected Distance r Along Jet” shown below. There is a positive correlation between and r at the 99.9% confidence level.

• “Maximum Jet Knot Speed max vs. log R” is plotted below, and shows a positive correlation at

the 98% confidence level (98.5% for maximum and average speed, without the three single-speed sources).

Polarization Structure

• We attempted polarization imaging for seven sources; five objects show polarized emission:

3C207 – longitudinal magnetic field B in inner few mas of jet; rest frame rotation measure RM~600 rad/m2 in/near core.

3C245 – longitudinal B in long, ~50-mas jet; RM~1600 rad/m2 in/near core.3C263 – RM~3000 rad/m2 in/near core.3C275.1 – RM~2000 rad/m2 in/near core.3C334 – transverse B in isolated, distant knot ~100 mas from core.

• The fractional polarization m is typically ~0.01-0.05 in the cores, and ~0.1-0.5 in the jets.

Implications for Jet Models• The tendency of jet components to travel faster at larger distances from the core does not find

a ready explanation in a constant model.

• Alternative #1: Jet acceleration has usually been thought to occur on scales well below those probed by VLBI (<~0.1 pc) – e.g., Begelman (1995, Proc.Natl.Acad.Sci., 92, 11442). Such models (e.g., McKinney 2006, MNRAS, 368, 1561) would need to account somehow for any patterns in the terminal Lorentz factors of ejected components.

• Alternative #2: Acceleration/deceleration could be occurring on scales of tens of parsecs. There is

some prior observational evidence for this from proper motion studies (Piner et al. 2003, ApJ, 588, 716; Zensus, Cohen, & Unwin 1995, ApJ, 443, 35; Hough, Zensus, & Porcas 1996, ApJ, 464, 715; Homan et al. 2001, ApJ, 549, 840; Lobanov & Roland 2005, A&A, 431, 831). Vlahakis & Konigl (2004, ApJ, 605, 656) and Komissarov et al. (2007, MNRAS, submitted; astro-ph/0703146v1) have used semi-analytic and numerical relativistic MHD approaches to demonstrate “magnetic driving” of jet components on these scales. The essential feature of their models and simulations is acceleration by the gradient of the circumferential magnetic-field pressure. They can easily obtain accelerations resulting in ~ 5-10, and report terminal Lorentz factors as high as ~35. However, it is not clear if this model can explain the alternating “fast/slow” superluminal speeds in 3C245.

Implications for Jet Models (continued)• The correlation between max and log R is expected in simple constant beaming models, but it may

also be attributed to our ability to detect the fastest components far from the core in highly-beamed sources (specifically, 3C207 and 3C245).

• The longitudinal magnetic fields in the continuous jets of 3C207 and 3C245 suggest shear effects, while the isolated transverse-B knot in 3C334 suggests a shock.

• In the jets, the high fractional polarization implies a high degree of order in the B field.

• In the cores, the low fractional polarization and significant rotation measures may be due to depolarization by the Faraday screen.

• The two sources with the strongest cores, 3C207 and 3C245, are likely to have the smallest angles to our line-of-sight, perhaps ~10o; they have the lower RM values. The two sources with more modest cores, 3C263 and 3C275.1, are probably at larger perhaps ~20o; they have the higher RM values. The statistics, of course, are weak, but these results are consistent with jets at larger being viewed along a greater path length through the Faraday screen (Taylor 1998, ApJ, 506, 637; Taylor 2000, ApJ, 533, 95).

Acknowledgments

• This work was supported in part by a 2006-2007 AAS Small Research Grant and NSF grant AST-0098253. Collaborators include Christian Aars, Dan Homan, Tony Readhead, and Rene Vermeulen, as well as numerous Trinity students, including Erika Barth, Phil Beyer, Lara Cross, Eric Danielson, Stephen Escobedo, Hakima Ibaroudene, Justin Linick, Ellen Phifer, Ben Sadler, Alex Webb, Nickolaus Wing, and Lam Yu. Neal Pape of Trinity’s ITS Center maintains the astronomy workstations there.