JET-SHOCKED H₂ AND CO IN THE ANOMALOUS ARMS OF MOLECULAR HYDROGEN EMISSION GALAXY NGC 4258

Radio jet–interstellar medium (ISM) interactions may cause either positive or negative feedback on star formation in galaxies (Wagner & Bicknell 2011). Jet feedback may explain the truncation of the galaxy mass function and quenching of star formation via expulsion or heating of the ISM (Croton et al. 2006). At larger scales, radio jets may prevent the cooling and accretion of intergalactic medium (Fabian et al. 2000). Hydrodynamical simulations of radio jets demonstrate that the jet deposits a large amount of energy, inflating a cocoon of hot gas that may encompass and shock the entire host galaxy ISM (Sutherland & Bicknell 2007). Molecular gas may be dissociated by shocks and driven out of the galaxy by outflows, preventing star formation altogether. Alternatively, jet-driven turbulence may support the gas against collapse, preventing star formation.

Ogle et al. (2010, 2007) find that 30% of 3C radio galaxies have extremely luminous H₂ emission from large masses (up to 10¹⁰ M_☉) of warm molecular gas. The H₂ is heated to temperatures of 100–1500 K. UV excitation in photodissociation regions (PDRs) is ruled out by a very high ratio of L(H₂ 0–0 S(0)–S(3))/L(PAH 7.7 μm)>0.04. Heating by X-rays from the active galactic nucleus (AGN) is ruled out by insufficient X-ray luminosity. The remaining viable sources of heating are shocks driven into the ISM by the radio jet, or cosmic rays (Ogle et al. 2010).

Similar conditions and IR spectra are seen in the Stephan's Quintet intergalactic shock, where there is a close correspondence between X-ray, Hα, H₂, [C ii], and CO emission (Appleton et al. 2006, 2013; Cluver et al. 2010; Guillard et al. 2012a). The large range of densities, temperatures, and cooling timescales are characteristic of a shocked multi-phase medium, where X-ray-emitting gas occupies the lower-density spaces in the post-shocked gas (Guillard et al. 2009).

X-ray emission from jet-shocked gas is seen in several radio molecular hydrogen emission galaxies (MOHEGs), including Per A (Fabian et al. 2006), Cen A (Kraft et al. 2007), and 3C 305 (Massaro et al. 2009; Hardcastle et al. 2012). Some radio MOHEGs display jet-driven neutral and ionized outflows with velocities up to 1200 km s⁻¹ (Morganti et al. 2005; Emonts et al. 2005; Guillard et al. 2012b). The limited spatial resolution of mid-IR H₂ observations of distant radio galaxies has made it difficult to determine the relationship between the radio jet, X-ray-emitting gas, outflows, and the H₂ emission region. It is unknown whether the H₂ emission comes from a distinct shock front or is distributed more widely in the galaxy. It is also unclear whether the H₂ emission arises from preexisting molecular gas that has been shocked in-situ, entrained in an outflow, or created in the post-shock gas.

NGC 4258 (Messier 106, z = 0.001494, d = 7.2 Mpc) is a nearby galaxy hosting a Seyfert 1.9 AGN (Ho et al. 1997). It is known for its anomalous spiral arms, which emerge from the galaxy nucleus and appear to intersect the regular spiral arms of the galaxy (Courtes & Cruvellier 1961). It is famous for water maser emission from a nuclear disk that enables accurate measurement of the central black hole mass (Miyoshi et al. 1995; Greenhill et al. 1995).

The anomalous spiral arms of NGC 4258 emit strongly in radio continuum (van der Kruit et al. 1972), optical nebular lines, and X-rays. Optical emission line ratios (Cecil et al. 1995a) and X-ray spectroscopy (Cecil et al. 1995b; Wilson et al. 2001; Yang et al. 2007) indicate 10⁵–10⁷ K shocked gas in the anomalous arms. Large lobes of X-ray-emitting plasma extend above and below the galactic disk, from a jet-driven outflow (Cecil et al. 1995b; Wilson et al. 2001). A series of Hα-emitting streamers of ionized gas branch off of the anomalous arms, marking cooler gas entrained by the hot outflow.

The three-dimensional orientation of the inner radio jet relative to the maser disk is well-constrained by very long baseline interferometry observations. Jet hot spots indicate a consistent jet orientation at larger scales, at an angle of 30° to the plane of the galaxy disk (Cecil et al. 2000). Wilson et al. (2001) propose that the anomalous arms are the projection of the jet along the disk, along a "line of destruction." Gas is blown out the opposite side of the disk to the jet.

We created an excitation diagram for the H₂ pure rotational lines, and used this to estimate the mass of warm H₂ as a function of temperature, assuming thermal equilibrium. Three temperature components are required to fit the excitation diagram, with M(H₂) =8.9 ± 0.3 × 10⁶ M_☉, 5.3 ± 2.4 × 10⁵ M_☉, and 2.5 ± 0.8 × 10⁴ M_☉ at T = 240 K, 450 K, and 1040 K, respectively. The total warm H₂ mass at T > 200 K is therefore M_tot(H₂) =9.4 ± 0.4 × 10⁶ M_☉.

The mid-IR (MIR) line luminosities inside our IRS extraction region are given in Table 1. The total ionized gas MIR line luminosity is L(MIR, ionized) =2.27 × 10⁴⁰ erg s⁻¹, compared to the total H₂ 0–0 S(0)–S(7) rotational line luminosity of L(H₂ 0–0 S(0)–S(7)) =5.27 × 10⁴⁰ erg s⁻¹. H₂ emission is one of the main ISM cooling channels. The luminosity of H₂, summed over the first four transitions is L(H₂) =3.92 × 10⁴⁰ erg s⁻¹, while the PAH 7.7 μm luminosity is L(PAH 7.7) =1.07 × 10⁴¹ erg s⁻¹. The ratio is L(H₂)/L(PAH 7.7) =0.37, which qualifies NGC 4258 as a MOHEG (Ogle et al. 2010). A large ratio of H₂ to PAH 7.7 μm (>0.04) indicates that the H₂ is not heated by UV photons from hot stars in a PDR (Guillard et al. 2012b).

We measure an unabsorbed 2–10 keV luminosity of 7.3 × 10⁴⁰ erg s⁻¹ for the AGN from the Chandra image. The ratio of L(H₂)/L_X(AGN, 2–10 keV) = 0.54 rules out AGN heating of the H₂ by X-rays from the nucleus, which has a maximum ratio of 0.01 (Ogle et al. 2010). The extended thermal X-ray emission inside the IRS extraction region has a total 0.3–2 keV luminosity of 7.3 × 10³⁹ erg s⁻¹, which is a factor of ∼7 less than the total H₂ line luminosity, and therefore contributes a negligible amount of X-ray heating to the H₂.

Heating by mechanical energy from the jet via shocks or cosmic rays is the prime candidate for powering the H₂ emission. The radio jet power dissipated at its terminal hotspots is estimated to be L_mech ≃ 1.9 × 10⁴⁰ erg s⁻¹, based on the Hα luminosities of the N and S bow shocks, assuming an 0.82% conversion efficiency (Cecil et al. 2000). This is only 36% of the total H₂ power measured within the IRS spectral extraction region. In order to power the observed H₂ emission, the jet must be at least three times more powerful than the above estimate. Additional power may be dissipated along the entire length of the jet, particularly where the primary H₂ emission ridge is brightest, near the AGN (Figures 2 and 3).

We measure the BIMA CO 1–0 flux (Helfer et al. 2003) in our IRS extraction region to be 1783 Jy km s⁻¹. Using the standard Galactic conversion factor of X_CO = 2 × 10²⁰ cm⁻²/(K km s⁻¹) (Lebrun et al. 1983; Bolatto et al. 2013), we estimate a cold H₂ mass of 9.7 × 10⁸ M_☉. This is in agreement with previous CO measurements and H₂ mass estimates of ∼1 × 10⁹ M_☉(Cox & Downes 1996; Krause et al. 2007). However, we find evidence for a nonstandard X_CO value below, so this is likely an overestimate of the molecular gas mass.

The integrated 42–122 μm luminosity from the IRS extraction region is L(FIR) =2.2 × 10⁴² erg s⁻¹. We estimate a dust mass of M_dust = (0.8–1.7) × 10⁶ M_☉ by fitting the FIR SED with a modified blackbody with dust emissivity powerlaw index β = 1.5–2.0; in agreement with a dust mass of $M_\mathrm{dust}=1.0^{+0.2}_{-0.1} \times 10^6\, M_\odot$ derived by fitting the UV–FIR SED (Figure 1(b)) using MAGPHYS (da Cunha et al. 2008). We use this to estimate a total gas mass of 1.0 × 10⁸ M_☉, assuming a gas/dust mass ratio of 100, as found in the Milky Way (Savage & Mathis 1979). This is only 10% of the molecular gas mass estimated from CO using the standard X_CO value. The corresponding gas surface density is Σ_gas = 30 M_☉ pc⁻².

To reconcile the large molecular gas mass estimated from CO with the smaller total gas mass estimated from FIR dust emission, either the dust/gas ratio must be a factor of 10 smaller, or the X_CO value must be a factor of 10 lower than in the Milky Way. While we cannot rule out a low dust/gas ratio in the H₂ emission region, dust absorption is seen in optical images of this region with a morphology similar to the H₂ emission. In support of a low X_CO factor is the close spatial correspondence between CO 1–0 emission and H₂ emission. If most CO emission comes from the same 240–1040 K gas as the H₂ emission, we expect the CO emission to be enhanced by the high temperature and any additional turbulence. Some other nearby galaxies show similarly depressed X_CO values in their central regions (Sandstrom et al. 2013). The warm H₂ in NGC 4258 constitutes roughly 1% of the CO-derived molecular mass, or 10% of the dust-derived gas mass. This warm H₂ percentage is similar to that seen for LINER and Seyfert nuclei in the SINGS sample (Roussel et al. 2007).

The radio jet has a clear impact on the temperature and spatial distribution of molecular and atomic gas in the anomalous arms of NGC 4258. The question remains whether it has a significant impact on the gas content, star formation rate (SFR), and evolution of the central regions of NGC 4258 and the galaxy as a whole. The clearest indication that the jet is ejecting large quantities gas from the galaxy is seen in X-ray images, where bubbles of hot gas extend far above and below the plane. Integrating over the entire X-ray lobes and assuming a filling factor of f ≲ 1, we find 1.8 × 10⁸ M_☉f^0.5 of 0.2–0.9 keV gas cooling at a rate of 0.31 M_☉ yr⁻¹. An outflow of hot gas at this rate is necessary to sustain the X-ray emission from the lobes, assuming no additional source of heating. If the gas in the center of the galaxy is the source of this outflow, then most of it has already been ejected into the galaxy halo and the rest would be depleted in 3 × 10⁸ yr, if not replenished.

We estimate the SFR in the IRS extraction region using two methods. The MAGPHYS model (Figure 1(b)) gives a best-fit SFR of $0.084^{+0.015}_{-0.004}$ M_☉ yr⁻¹ and a stellar mass of $7.6^{+0.3}_{-1.9}\times 10^9\, M_\odot$. In agreement with this, we estimate SFR(PAH 7.7) =2.4 × 10⁻⁹L(PAH 7.7) =0.069 M_☉ yr⁻¹ from the luminosity of the 7.7 μm PAH feature (Roussel et al. 2001). In comparison, the Kennicutt–Schmidt (KS) law (Kennicutt 1998) predicts a SFR surface density of ∼0.03 M_☉ yr⁻¹ kpc⁻² for the dust-derived gas surface density of Σ(H₂) =30 M_☉ pc⁻², or a total SFR of ∼0.1 M_☉ yr⁻¹, in agreement with the observed SFR. However, if were to use the CO-based molecular gas mass density of Σ(H₂) =300 M_☉ pc⁻², the KS law would predict a 25 times greater SFR of 2.5 M_☉ yr⁻¹, in disagreement with the observed value. This demonstrates the crucial role the X_CO factor plays in estimates of SFR suppression, which may be greatly overestimated in MOHEGs where CO emissivity is enhanced and the X_CO value is low.