The origin of extended emission line regions

The origin of extended emission line regions

New Astronomy Reviews 51 (2007) 47–51 www.elsevier.com/locate/newastrev The origin of extended emission line regions a,* , M. Villar-Martı´n b, C. T...

680KB Sizes 2 Downloads 67 Views

New Astronomy Reviews 51 (2007) 47–51 www.elsevier.com/locate/newastrev

The origin of extended emission line regions a,*

, M. Villar-Martı´n b, C. Tadhunter a, J. Holt a, R. Morganti

K.J. Inskip

c

a

c

Department of Physics and Astronomy, University of Sheffield, Sheffield, S3 7RH, UK b Instituto de Astrofı´sica de Andalucı´a (CSIC), Aptdo. 3004, 18080 Granada, Spain Netherlands Foundation for Research in Astronomy, Postbus 2, 7990 AA Dwingeloo, The Netherlands Available online 6 December 2006

Abstract We present the results of VIMOS-Integral Field Unit (IFU) and FORS1 spectroscopy of two radio galaxies: PKS1932-46 and PKS2250-41. The two sources display highly extended emission line regions, with varied ionization states and gas kinematics. Both sources appear to lie in rich environments, and may be undergoing interactions with companion bodies. The more distant emitting material generally displays simple, narrow emission line profiles, often at significant velocity offsets from the system rest-frame, and may be formed from the tidal debris of a previous interaction.  2006 Elsevier B.V. All rights reserved.

Contents 1. 2. 3. 4. 5.

Introduction . Observations . PKS1932-46 . PKS2250-41 . Conclusions . References . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

1. Introduction Extended emission line regions (EELRs) are frequently observed around powerful radio sources (McCarthy et al., 1987), and are often aligned with the radio source axis. It has long been known that their properties are closely linked with those of the radio source itself (Rawlings and Saunders, 1991): major trends have been identified between the EELR properties and those of the radio *

Corresponding author. E-mail address: [email protected] (K.J. Inskip).

1387-6473/$ - see front matter  2006 Elsevier B.V. All rights reserved. doi:10.1016/j.newar.2006.10.004

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

47 48 48 50 51 51

source. In addition to higher line luminosities, the EELRs surrounding more powerful radio sources typically display more extreme gas kinematics and exist in a higher ionization state (Inskip et al., 2002a,b). The EELR properties (kinematics, ionization state, EELR size and luminosity) also display a strong evolution with the size/age of the radio source (e.g. Best et al., 2000; Inskip et al., 2002a; Moy and Rocca-Volmerange, 2002). For the most compact (<2 kpc) radio sources, extended emission is typically observed out to 20 kpc from the AGN (Axon et al., 2000; Holt et al., 2003), although the line emission is at its most luminous over similar scales

48

K.J. Inskip et al. / New Astronomy Reviews 51 (2007) 47–51

to the radio emission. The gas in the outer regions is kinematically quiescent and the predominantly AGN photoionized, although recent research (Holt et al., 2003) has detected broader emission from the central regions (plausibly associated with activity-related outflows). As the radio source grows, the complexity of the gas kinematics and emission line luminosities increase, as does the physical extent of the EELR, often remaining comparable in size to the radio source out to distances of over 100 kpc. On these intermediate scales, the emission line ratios are frequently consistent with shocks being the dominant ionization mechanism. For the largest radio sources (>150 kpc), shocks become less important and AGN photoionization is once again the dominant ionization mechanism. The EELRs of these large radio sources usually display relatively quiescent kinematics, and are rarely as extensive as those of smaller (<150 kpc) radio sources. However, there are many outstanding questions in this field. The emitting material is not always limited to lying along the radio source axis or within AGN ionization cones. Recent narrow band imaging observations (VillarMartı´n et al., 2005; Solo´rzano-In˜arrea et al., 2002; Tadhunter et al., 2000) have identified extensive emission line regions lying almost perpendicular to the radio axis, which may perhaps be related to the extended Lyman-a halos observed at higher redshifts. The origin of the extended emission line gas – whether it exists in situ well before the onset of the radio source, is produced via galaxy interactions/mergers associated with the radio source triggering event, or is formed from material driven out of the host galaxy by AGN/radio source/starburst related outflows – is a particularly pertinent problem, which may be closely linked with both the radio source activity and the host galaxy evolution. Detailed integral field spectroscopy (IFS) of EELRs have the potential to greatly improve our understanding of these systems, particularly in terms of building

up a consistent explanation for the nature, distribution and origin of the emission line gas, and the links with the radio source triggering mechanism. 2. Observations IFS observations of PKS1932-46 (Fig. 1a) and PKS2250-41 (Fig. 1b) were carried out using VIMOS (Le Fe´vre et al., 2003; Scodeggio et al., 2005) on the VLT UT3 during 2004/2005. Our data were obtained using the medium resolution (MR-orange) grism; in this configuration, the IFU consists of 1600 microlenses coupled to 0.67-arcsec diameter fibres, covering a total sky area of 27 · 27 arcsec2, with a useful wavelength range of 4500– ˚ . The total integration times were 12450 s and 9000 A 12100 s for PKS1932-46 and PKS2250-41, respectively. Data reduction was carried out using the VIMOS Interactive Pipeline and Graphical Interface (VIPGI) package (Scodeggio et al., 2005; Zanichelli et al., 2005). IDL routines were used to locate and model the line emission in each fibre; see Inskip et al. (in preparation) for full details. Long-slit spectroscopic observations of PKS2250-41 at a PA of 69 were obtained using FORS1 at the VLT UT2, and reduced using standard IRAF routines. The total integration time was 6720 s, and the data have a spatial resolu˚/ tion of 0.2 arcsec/pixel, a spectral resolution of 1.07 A ˚ and a useful pixel, an instrumental profile of 5.9 ± 0.1 A ˚ . Cosmological parameters spectral range of 5520–7410 A of X0 = 0.3, XK = 0.7 and H0 = 70 km s1 Mpc1 are assumed throughout. 3. PKS1932-46 PKS1932-46 (Figs. 1a and 2) is an FRII BLRG at z = 0.231. Previous imaging and spectroscopic observations (Villar-Martı´n et al., 1998, 2005) have shown that it

Fig. 1. Left – 5.8 GHz radio contours on [OII] + continuum image for PKS1932-46, from Villar-Martı´n et al. (1998), illustrating the relative size of the radio source and optical emission. Right – 15 GHz radio contours on [OIII]5007 emission line image for PKS2250-41, from Tilak et al., 2005.

K.J. Inskip et al. / New Astronomy Reviews 51 (2007) 47–51

49

Fig. 2. (a) [OII] + continuum image of PKS1932-46 (from Villar-Martı´n et al., 1998) illustrating the field of view of the VIMOS IFU data and object ˚ flux in ergs s1 cm2. (c) [OIII] emission line FWHM (km s1). (d) Velocity shift of the [OIII] emission relative to a identifications. (b) Extracted [OIII]5007 A rest-frame redshift of z = 0.231 (km s1).

has an extensive, knotty emission line region, comparable in size to the radio source itself. Interestingly, several knots lie at large distances (up to 20 kpc in our IFS data) from the radio source axis, well outside the AGN ionization ˚ + continuum image, cone. Fig. 2a displays an [OII]3727 A illustrating the field of view of our IFS data. Extracted spectra confirm that the spiral galaxy to the north lies at a redshift of z = 0.229, similar to that of the radio source ˚ emission itself. Fig. 2b displays the extracted [OIII]5007 A line flux, with the emission line FWHM and velocity shifts displayed in Fig. 2c and Fig. 2d, respectively. The IFS data reveal previously undetected line emitting material throughout the western radio lobe, with a very knotty morphology. The knots seem to lie preferentially around the edges of the cocoon rather than directly on the jet axis. The more distant knots display fairly quiescent kinematics (150 kms1), with line widths comparable to the instrumental resolution, although the knots are redshifted by roughly 300 kms1 relative to the host galaxy rest-frame. The largest line widths are observed towards the centre of the host galaxy, although this is complicated by the presence of several different velocity structures in this region. The ‘‘arm’’ structure (noted by Villar-Martı´n et al. (1998)) extending eastwards from the galaxy is slightly redshifted, while blueshifted material with fairly narrow line widths (lying at the same redshift as the companion galaxy to the north) extends in a north-south direction from the host galaxy, perpendicular to the radio source axis.

We have also examined the ionization state of the EELR gas. Emission line diagnostics imply little variation in the gas properties in different regions; we typically derive electron densities of 100 cm3 and temperatures of 1–2 · 104 K. [OIII]/Hb and [NII]/Ha ratios are generally consistent with AGN photoionization for the western knots, but the most central regions tend to exist in a lower ionization state, which, together with the varied kinematics, is suggestive of shocks. The [NII]/Ha ratios of the blueshifted gas components are also relatively low, placing them closer to the maximum starburst track of Kewley et al. (2001). In their spectroscopic study of the PA 9 material, Villar-Martı´n et al. (2005) suggested that the best explanation for the properties of this EELR was a series of compact, star-forming objects associated with the tidal debris of a merging system. We concur with this explanation for the various knots extending north/south from the host galaxy. Star formation is not limited to the EELR knots, but is also clearly ongoing in the spiral companion galaxy to the north. In addition to very low [NII]/Ha line ratios in the companion galaxy spectra, Spitzer MIPS photometry (Tadhunter and Dicken, priv. comm.) of PKS1932-46 and its surrounds reveal that the companion galaxy is more luminous than the radio source at 24 lm, and substantially more so at 70 lm, suggestive of vigorous star formation in the companion. Our IFU and Spitzer data confirm that PKS1932-46 lies in a rich environment, and that mergers/interactions may still be ongoing for this source.

50

K.J. Inskip et al. / New Astronomy Reviews 51 (2007) 47–51

4. PKS2250-41 PKS2250-41 (Figs. 1b and 3) is an FRII radio galaxy at z = 0.308, and, like PKS1932-46, it also displays an EELR comparable in size to the radio source. The main emission line features are knotty emission regions extending to the east, and a luminous arc structure to the west, coincident with the western radio lobe. The ionization state of the gas clearly varies with position, as can be seen in our ˚ and [OII]3727 A ˚ fluxes and the images of the [OIII]5007 A [OII]/[OIII] emission line ratio (Fig. 3b–d). In Fig. 3e and

˚ emission line widths f, we display the observed [OIII]5007 A and velocity shifts. The large emission line arc to the west of the host galaxy (Tadhunter et al., 1994; Clark et al., 1997) is redshifted relative to the rest-frame of the system, and the largest line widths in this region lie close to the western radio hotspot, also coincident with the lowest ionization state gas observed in this source. These results are consistent with the findings of previous studies of this source (Villar-Martı´n et al., 1999;Clark et al., 1997), which suggest that radio source shocks are responsible for ionizing the gas around

˚ + continuum image of PKS2250-41 Clark et al., 1997 illustrating the location of the 100 FORS1 slit. (b) Extracted [OII]3727 A ˚ flux Fig. 3. ((a) [OIII]5007 A ˚ flux for PKS2250-41, in ergs s1 cm2. (d) for PKS2250-41, in ergs s1 cm2 (scaled and aligned with Fig. 3a). (c) Extracted [OIII]5007 A ˚ /[OIII]5007 A ˚ ) emission line ratio. (e) [OIII]5007 A ˚ emission line FWHM (km s1). (f) Velocity shift of the [OIII]5007 A ˚ emission relative Log10([OII]3727 A to a rest-frame redshift of z = 0.308 (km s1).

˚ emission line (the continuum Fig. 4. Left: Inner 400 (18 kpc) region of the FORS1 spectrum (PA 69), illustrating the continuum-subtracted [OIII]5007 A 1 ˚ centroid location is indicated by the dashed line). Centre: [OIII]5007 A emission line velocity shift (km s ) as a function of offset from the continuum ˚ emission line FWHM (km s1). centroid (arcsec). Right: [OIII]5007 A

K.J. Inskip et al. / New Astronomy Reviews 51 (2007) 47–51

the hotspot. However, even larger line widths are observed between the arc and the host galaxy, aligned with the radio jet axis. The line emission in these regions is best modelled by a combination of gaussian components, of which the broader emission component is blueshifted. Similar features are observed in our FORS1 spectrum of PKS225041 (Fig. 4), where the innermost regions of the EELR display highly asymmetric line profiles, requiring multiple gaussian components. The observed velocity curve is very smooth, suggestive of a 20 kpc rotating structure, and seems unaffected by the more extreme kinematics in the central regions of the galaxy. The gas kinematics have measured FWHM varying from 150–410 km s1, with a maximum value near the continuum centroid; these velocities are in agreement with those observed in our IFS data, which are also consistent with a rotation model in this region (Fig. 3e and f). A fainter, more diffuse band of emission (resolved out in the higher resolution emission line images of Tilak et al. (2005)) with unresolved line widths extends north-eastwards from the host galaxy (Fig. 3a) towards another nearby object. We also observe the companion galaxy in our FORS1 spectrum, which displays low S/N Balmer absorption features at a similar redshift to PKS2250-41. 5. Conclusions Both PKS1932-46 and PKS2250-41 display a wide variety of EELR features. We observe evidence for jet-cloud interactions, shocks (particularly clear-cut for PKS225041) and other interactions between the EELR gas and the radio source: for PKS1932-46, the EELR morphology appears to trace the edges of the western radio lobe. The broadest line emission is observed close to the host galaxy, but not always perfectly aligned along the radio source axis. At larger distances, the gas often traces a fairly smooth velocity profile with narrow line widths, although some knots can lie at significant velocity offsets from the galaxy rest-frame. This is unusual for sources displaying such extensive line emission; the lack of disturbed kinematics in some of the more distant parts of

51

the EELR is one of the more interesting features of these sources. The other notable feature of these sources is the presence of companion galaxies at the same redshift. In the case of PKS1932-46, the companion galaxy is actively forming stars, as are many of the knots in the EELR. Taken together with the gas kinematics, we suggest that both sources lie in rich, merger environments, and that the much of the material in their EELRs may originate from the tidal debris from previous interactions – which may, plausibly, also have been responsible for triggering the radio source activity. References Axon, D.J., Capetti, A., Fanti, R., Morganti, R., Robinson, A., Spencer, R., 2000. AJ 120, 2284. Best, P.N., Ro¨ttgering, H.J.A., Longair, M.S., 2000. MNRAS 311, 23. Clark, N.E., Tadhunter, C.N., Morganti, R., Killeen, N.E.B., Fosbury, R.A.E., Hook, R.N., Siebert, J., Shaw, M.A., 1997. MNRAS 286, 558. Holt, J., Tadhunter, C.N., Morganti, R., 2003. MNRAS 342, 227. Inskip, K.J., Best, P.N., Rawlings, S., Longair, M.S., Cotter, G., Ro¨ttgering, H.J.A., Eales, S., 2002a. MNRAS 337, 1381. Inskip, K.J., Best, P.N., Ro¨ttgering, H.J.A., Rawlings, S., Cotter, G., Longair, M.S., 2002b. MNRAS 337, 1407. Inskip, K.J. et al., 2006. MNRAS, in preparation. Kewley, L.J., Dopita, M.A., Sutherland, R.S., Heisler, C.A., Trevena, J., 2001. ApJ 556, 121. Le Fe´vre, O. et al., 2003. SPIE 4841, 1670. McCarthy, P.J., van Breugel, W., Spinrad, H., Djorgovski, S., 1987. ApJ 321, L29. Moy, E., Rocca-Volmerange, B., 2002. A&A 383, 46. Rawlings, S., Saunders, R., 1991. Nature 349, 138. Scodeggio, M. et al., 2005. PASP 117, 1284. Solo´rzano-In˜arrea, C. et al., 2002. MNRAS 331, 673. Tadhunter, C., Shaw, M., Clark, N., Morganti, R., 1994. A&A 288, 21. Tadhunter, C.N., Villar-Martı´n, M., Morganti, R., Bland-Hawthorn, J., Axon, D., 2000. MNRAS 314, 849. Tilak, A., O’Dea, C.P., Tadhunter, C., Wills, K., Morganti, R., Baum, S.A., Koekemoer, A.M., Dallacasa, D., 2005. AJ 130, 2513. Villar-Martı´n, M., Tadhunter, C., Morganti, R., Clark, N., Killeen, N., Axon, D., 1998. A&A 332, 479. Villar-Martı´n, M., Tadhunter, C., Morganti, R., Axon, D., Koekemoer, A., 1999. MNRAS 307, 24. Villar-Martı´n, M., Tadhunter, C., Morganti, R., Holt, J., 2005. MNRAS 359, 5. Zanichelli, A. et al., 2005. PASP 117, 1271.