DISCOVERY OF GIANT RELIC RADIO LOBES STRADDLING THE CLASSICAL DOUBLE RADIO GALAXY 3C452

A striking manifestation of episodic jet activity in massive elliptical galaxies is the so-called "double–double radio galaxies" (DDRGs), in which two pairs of synchrotron emitting lobes are seen to straddle the parent galaxy fairly symmetrically and nearly colinearly (e.g., Lara et al. 1999; Schoenmakers et al. 2000b; Saripalli et al. 2003). About 15 DDRGs are currently known (Saikia & Jamrozy 2009) and in essentially all cases the outer radio lobes lack bright hot spots, indicating that they are no longer being fed by the central engine via relativistic jets. Although DDRGs form only a tiny fraction of the more powerful and edge-brightened Fanaroff–Riley type II (FR II; Fanaroff & Riley 1974) radio source population, they are particularly important for testing models for the evolution of active galactic nuclei, as they provide vital evidence for recurring nuclear activity (e.g., Lara et al. 1999; Schoenmakers et al. 2000a, 2000b; Konar et al. 2006; for a review, see Saikia & Jamrozy 2009). The best evidence for the recurrence hypothesis so far has come from the RG B 0925 + 420 which is actually a "triple–double" source with three aligned lobes seen on each side of the nucleus (Brocksopp et al. 2007). Detailed modeling of DDRGs could therefore provide important clues to the duration and duty cycle of the jet activity. Additionally, they can provide vital information about the ambient medium through which successive generations of radio jets propagate. Most DDRGs are giant radio galaxies with projected linear extents exceeding 1 Mpc, but the inner lobes are seen on a wide range of scales (Saikia & Jamrozy 2009).

There has been renewed discussion about the nature of the younger, inner pair of lobes in DDRGs, motivated by the unusually narrow shape observed in a few cases. These narrow shapes have been linked to the circumstance that the jets powering them could well be advancing almost ballistically through the light relativistic plasma filling the outer radio lobes that are relics of the previous active phase (Brocksopp et al. 2007, 2011; Safouris et al. 2008). This possibility has previously been considered in different contexts (e.g., Clarke & Burns 1991; Stawarz 2004). These authors have proposed that each ballistically advancing inner jet does not create a significant hot spot by terminating at a Mach disk as envisaged in the standard model for FR II sources (e.g., Longair et al. 1973; Blandford & Rees 1974). This can occur if the external density in the first (outer) lobe is extremely low; then the recurrent jet could drive a narrow bow-shock at which the ambient plasma of the outer relic lobe produces brightened synchrotron emission because of the shock-induced compression. Thus, in this radical scenario for the formation of the inner lobe pair of DDRGs, these lobes are not completely filled with the material transported by the restarted inner jets, unlike the outer lobes which were inflated by the first-generation jets that propagated non-relativistically but supersonically through the (much denser) galactic halo and intergalactic medium (IGM). One possibility to avert the putative, nearly ballistic, motion of the restarted jets is that a substantial density enhancement occurs within the outer radio lobes, conceivably due to a gradual mixing of thermal gas, either through lobes' surface instabilities or the ablation of the narrow-line emitting clouds within the lobes (e.g., Kaiser et al. 2000; Safouris et al. 2008). In that case, the inner lobes would resemble classical FR II RGs. Here we present new evidence favoring this last scenario by showing that the well-known "canonical" FR II double radio source 3C452 is actually an inner double of a DDRG whose megaparsec-scale outer radio lobes with extremely steep spectra have now been seen in our sensitive meter-wavelength map presented here.

We have assumed a flat ΛCDM cosmology with H₀ = 71 km s⁻¹ Mpc⁻¹, Ω_m = 0.27, and Ω_Λ = 0.73. Thus, at the redshift of 3C452, with z = 0.0811, 1'' translates to a projected distance of 1.51 kpc.

The radio data were taken from the archive of the observations made with the Giant Metrewave Radio Telescope (GMRT; Swarup et al. 1991) and were originally mapped using AIPS and discussed by Nandi et al. (2010). The observations at 325 MHz were made on 2008 January 3, with an on-source time of 4.1 hr and using a total bandwidth of 32 MHz. The flux-density scale was set using 3C48, which is 43.43 Jy at 325 MHz on the Perley–Taylor 99 scale and the phase calibration was done using J2350+646. The data were reduced employing the imaging software AIPS++, following the procedure described in Sirothia et al. (2009) and the resulting map is shown in Figure 1(a). We also used the same procedure to make a GMRT map at 1314 MHz using the archival data of 2005 December 11 with an on-source time of 5.5 hr and a bandwidth of 4 MHz. The flux calibration of these visibilities was also done using 3C48 (16.88 Jy; taking the Perley–Taylor 99 scale), but J2202+422 was used for phase calibration. The resulting 1314 MHz map is shown in Figure 1(b), smoothed to the resolution of the 325 MHz map, and the corresponding full resolution false-color image at 1314 MHz is shown in Figure 2. This image clearly shows a resolved hot spot in each lobe as well as a pair of jets feeding these hot spots. These hot spots are recessed from the outermost (leading) edges of the lobes and are therefore not compatible with emission arising primarily from a bow-shock. In Figure 3 we give a one-dimensional cut through the 325 MHz map taken along the source axis. Note the steep fall in brightness just beyond the hot spots; however, the faint fossil radio emission is clearly visible, extending another ∼400 kpc on each side.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Thus, the striking new feature revealed by the present analysis of the 325 MHz GMRT observations of this well-known classical double radio source is that it is straddled by a pair of faint "outer" lobes which are joined by a bridge of radio emission (Figures 1(a) and 3). The outer lobes span a total extent of 1305 (1.1 Mpc). This projected size is unlikely to be much smaller than the true size, since the Very Long Baseline Interferometry map of the central core shows a fairly symmetric pair of closely aligned jets on a scale of tens of parsecs (Giovannini et al. 2001), indicating that this DDRG lies within 30° to the plane of the sky and hence its projected shape and size are close to the actual ones.

The inner lobes account for 27.7 Jy out of the integrated flux density of 34.4 Jy at 325 MHz, with an uncertainty of 15%. At 1314 MHz the integrated flux density, apparently completely emitted by the inner lobes, is found to be 11.37 Jy, with an uncertainty of 5%. Comparison of the surface brightness of the NRAO VLA Sky Survey (NVSS) image of 3C452 at 1.4 GHz (Condon et al. 1998), taking 3σ upper limits for the undetected outer lobes, with our GMRT 325 MHz image smoothed to the NVSS resolution (45'' × 45'') gives a spectral index upper limit of α < −2.3 for the eastern outer lobe and α < −2.4 for the western outer lobe, where S_ν∝ν^+α.

3.1. The Relic Outer Radio Lobes

3.2. The Inner Double Radio Source

Turning to the origin of the inner radio lobes in DDRGs, we note that the bow-shock hypothesis has so far been invoked for the inner lobes of four DDRGs based on models from Kaiser (2000): B0925+420 (Brocksopp et al. 2007); PKS1545−321 (Safouris et al. 2008); B1450+333; and B1834+620 (Brocksopp et al. 2011). In B0925+420 the existing relatively low-resolution images of the inner lobes appear consistent even with a regular hot spot description, as was also noted in Brocksopp et al. (2007); however, these authors argue that the bow-shock interpretation is preferable. For 3C452, the average hot spot advance speeds are 0.024c, based on the spectral ageing estimate of the source lifetime of 2.7 × 10⁷ yr (Burch 1977; Nandi et al. 2010) and a total extent of 400 kpc, and thus much slower than the modestly relativistic motions required in the bow-shock model, even more so because "dynamical ages" are typically 2–4 times larger than such "spectral ages" (e.g., Scheuer 1995).

Alternatives must be considered while searching for the best models for the morphologies of DDRGs. Many of these were investigated in the papers discussing B0925+420 (Brocksopp et al. 2007), B1545−321 (Safouris et al. 2008), B1450+333, and B1834+620 (Brocksopp et al. 2011), and we summarize them here. (1) Their inner lobes are indeed bow-shock-dominated, though the observed morphologies do not force this conclusion. (2) Thermal gas entrainment into the outer cocoon has occurred and the inner jets are propagating non-ballistically through a medium denser than a pure synchrotron plasma, thereby allowing a standard FR II source to develop. Provided that their Mach number is high enough, a narrow cocoon is expected to form behind a jet termination hot spot (e.g., Carvalho & O'Dea 2005), without invoking a bow-shock. (3) Even if the jet entrainment is minimal, a significant quantity of thermal gas may be mixed throughout the outer lobes via the ablation or shredding of warm gas clouds engulfed by them (Kaiser et al. 2000; Brocksopp et al. 2007). (4) The restarted jets are significantly weaker than the original ones, though this option does not seem to be possible for the case of B1834+630 (Brocksopp et al. 2011). We suggest another related possibility. (5) The restarted jets are comparably powerful but significantly lighter than the original jets, hence they still propagate through a relatively denser medium, producing inner lobes with normal hot spots and cocoons. If thermal gas infusion into the outer cocoon is negligible, this last alternative still could be realized if the restarted jets are launched as purely electromagnetic flows, i.e., Poynting-flux-dominated (e.g., Clarke et al. 1986; Lovelace & Romanova 2003). These jets would only require a small mass loading to emit synchrotron radiation.

To gain some insight into these possibilities, we now consider the dynamics of the restarted jets which are responsible for the inner jets in 3C452. Considering that these current jets are moving through the relic lobes from the earlier active phase, it is very likely that they are propagating in quite uniform media. Therefore, assuming that the jets on each side are intrinsically very similar, any asymmetry in the distances of the hot spots from the galactic center should be produced by light-travel time effects (e.g., Scheuer 1995). Then, we have

$$\begin{equation} D \equiv \frac{d_1}{d_2} = \frac{1 + \beta _{{\rm hs}} \cos \theta }{1 - \beta _{{\rm hs}} \cos \theta }, \end{equation} \tag{ 1 }$$

or

$$\begin{equation} \beta _{{\rm hs}} = \frac{1}{\cos \theta } \frac{D-1}{D+1}, \end{equation} \tag{ 2 }$$

where d₁ and d₂ are the observed (projected) separations of the more and less distant hot spots, respectively, from the core, β_hs is the (assumed common) speed of advance of the hot spots in units of the speed of light, c, and θ is the angle between the jets and the line of sight. For 3C452 we have for the western hot spot d₁ = 196 kpc, while for the eastern hot spot d₂ = 191 kpc (Isobe et al. 2002), so D = 1.026 and β_hscos θ = 0.0119. As discussed previously, we expect θ to be a moderately large angle, and for θ = 60° we find β_hs = 0.024, in excellent agreement with the estimates arising from spectral ageing (Burch 1977; Nandi et al. 2010). We can estimate the cocoon density, ρ_c, through which the jets propagate, via the balance between ram pressure and the hot spot pressure, p_hs, (Safouris et al. 2008)

$$\begin{equation} \rho _c = \frac{3}{4} \frac{p_{{\rm hs}}}{p_{\min }} \frac{p_{\min }}{c^2} \frac{(1 -\beta _{{\rm hs}} \cos \theta)^2}{\beta _{{\rm hs}}^2}, \end{equation} \tag{ 3 }$$

where the minimum pressure in the hot spots is p_min = u_e/3 + B²/8π, with u_e the energy density in relativistic electrons and B the magnetic field strength.

A unique advantage afforded by 3C452 is that not only are the speeds of its hot spots well determined but their internal pressures are also known without resorting to the assumption of minimum energy. By combining the radio and X-ray data, Isobe et al. (2002) have shown that its hot spots are significantly particle-dominated and have an energy density u_e ≈ 5 × 10⁻¹² erg cm⁻³ and B ≈ 3μG, so p ≈ 2 × 10⁻¹² dyne cm⁻². This yields ρ_c ≃ 2.9 × 10⁻³⁰ g cm⁻³, taking the nominal values of β_hs and θ used above, while for θ = 80°, hence β_hs = 0.068, we have ρ_c ≃ 3.7 × 10⁻³¹ g cm⁻³. Since the synchrotron plasma in those outer lobes is expected to be even less dense than this value (cf. Carvalho & O'Dea 2005), it does appear as if some significant entrainment of external gas (or ablation of IGM clouds) occurred during the first phase of activity in 3C452. While the corresponding number density of ∼1.8 × 10⁻⁶ cm⁻³ is ∼3 dex lower than the estimates available for the media around powerful FR II radio sources (e.g., Wellman et al. 1997), they are comparable to the density of the IGM in modestly overdense regions of the universe (e.g., Safouris et al. 2008; Stawarz 2004 and references therein). Interestingly, the above estimate of ambient density for the inner lobes of 3C452 is comparable to the value ρ_c ≲ 1.6 × 10⁻³⁰ g cm⁻³ estimated by Schoenmakers et al. (2000a) for the external medium of the inner lobes of the DDRG B 1834+620.

We recall the conclusion of Brocksopp et al. (2011) in the context of the inner lobes: "the presence of hot spots in a DDRG would rule out the bow-shock model, providing an observational test." This statement was, however, made in the context of the conditions under which the bow-shock model could be applied, namely ambient densities and pressures that are insufficient to significantly decelerate the restarted jets and create hot spots. We have shown that the well-known classical RG, 3C452, is in fact the inner pair of lobes of a DDRG, each of which possesses a distinct hot spot whose offset from the lobe's extremity is entirely consistent with the "dentist's drill" model for the hot spots (Scheuer 1982) but disagrees with the bow-shock hypothesis (Figure 2). In this case there is enough matter in the outer lobes to slow down the jets. Our high-resolution dual-frequency radio maps of 3C452 show that its lobes have shapes characteristic of classical double sources and that its hot spots are moving with small, non-relativistic speeds, and not ballistically with moderately relativistic speeds. These findings make clear that the bow-shock mechanism for the origin of the inner lobes of DDRGs is not generally applicable, though it may certainly be preferred for some cases, as argued by Brocksopp et al. (2007, 2011) and Safouris et al. (2008).

Finally, if future deep low-frequency radio observations, for example, with LOFAR or Murchison Widefield Array, reveal such giant radio fossils lurking around a significant fraction of edge-brightened double radio sources, this would have serious implications for the numerous studies based on radio galaxy sizes. This is because the actual sizes, energy contents, and active lifetimes of radio galaxies would then be much greater than inferred from existing maps. Edge-brightened radio galaxies have long been used as tools to probe the cosmological evolution of radio galaxies and the physical conditions in the universe (e.g., Kapahi 1975; Buchalter et al. 1998; Chen & Ratra 2003; Daly et al. 2009), including the impact of their mechanical energy output on the cosmic structure formation (e.g., Fabian 2012). Likewise, the use of the linear sizes of radio galaxies and quasars for assessing the orientation-based unification scheme of radio-loud active galaxies (e.g., Barthel 1989; Gopal-Krishna et al. 1996; Antonucci 2013) would have to be re-examined.

We thank the staff of the GMRT that made these observations possible. GMRT is run by the National Centre for Radio Astrophysics of the Tata Institute of Fundamental Research. This research used the NASA/IPAC Extragalactic Database (NED) which is operated by the Jet Propulsion Laboratory, California Institute of Technology, under contract with the National Aeronautics and Space Administration.