Giant nebulae around distant radio galaxies (z~2-3)
M.M.Villar-Martín (IAA-CSIC); S.F.Sánchez (CAHA); R.F. Peletier (Kapteyn Astronomical Institute, Groningen).


Our current belief is that the hosts of powerful radio galaxies in the distant Universe (z>2) are destined to become the giant ellipticals of today (e.g. McLure et al. 1999, Sánchez et al. 2003). While some may have commenced their formation at very high redshift, the process of assembly is likely to be very active at z~2-3. By looking at high redshift radio galaxies (HzRG) we are likely to be witnessing the formation process of giant elliptical galaxies (Stevens et al. 2003).

It is well known that HzRG are associated with regions of ionized gas that extend for tens of kpc. This gas shows very perturbed kinematics with line FWHM >~1000 km/s (e.g. Villar-Martin et al. 2003, McCarthy et al. 1996). In a study based on long-slit medium resolution (FWHM~10-14 Å) Keck II/VLT spectroscopy of a sample of 10 ultra steep spectrum HzRG (z~2.5) Villar-Martín et al. (2002, 2003) found that these objects are embedded in giant (often >100 kpc) low surface brightness halos (LSBHs) of metal rich ionized gas with much more quiescent kinematics , with typical line FWHM and velocity shifts of few hundred km/s. The halos often extend well beyond the brighter perturbed regions and sometimes beyond the radio structures. Compared with the extended emission line regions of low z (<0.6) powerful radio galaxies (median extent ~10 kpc), the quiescent LSBHs are giant structures with enormous luminosities (>100-1000 times more luminous) (Baum & Heckman 1989, Baum et al.1988).

The discovery of the giant quiescent halos in all the radio galaxies of our sample suggested that they are a common ingredient of powerful HzRG. Villar-Martin et al. proposed that they are the gas reservoir from which the galaxy started to form and may eventually settled into a rotating disk. At some point, the quasar switched on and the halo became observable thanks to strong line emission powered by the ionizing quasar continuum. Such structures could be the progenitors of the giant rotating disks found around some low redshift powerful radio galaxies (e.g. García-Lorenzo et al. 2005, Oosterloo et al. 2002).

Figure 1
Narrow-band image centred on the wavelength of Lya at the redshift of the radio-galaxy 4C40.36 (field-of-view of 40"X40"), reconstructed from PPAK data (2h integration using the V300 grating). The position of maximum Lya brightness is indicated with a black square. The approximate positions of the radio-lobes are indicated with two black crosses Lya emission is detected over a large area. It extends for at least ~9 arcsec (~100 kpc), not only along the radio axis, but also perpendicular to it. The two bright objects to the south-east and north-east from the radio-galaxy are probably field stars.


This work was seriously limited by the lack of spatial information in directions other than the radio axis. As an example, although we found evidence for rotation in some objects, the lack of spatial information in two dimensions (2D) made it difficult to obtain any information on the rotation axis, or even reject alternative scenarios such as radial motions. In addition, as most spectroscopic studies, the slit was always located along the radio axis and it is often the case that interactions between the radio and gas structures produce kinematic perturbance and morphological distortions of the gas (e.g. Clark et al. 1998; Villar-Martín 1999). It became clear that spectroscopy in two spatial dimensions is essential.

For this reason, we are carrying out a program of 3D integral field spectroscopy with PMAS/PPAK on the 3.5m telescope at Calar Alto Observatory, to characterize the general (morphological, kinematic, ionization, chemical composition) properties of the giant halos in two spatial dimensions using the main UV emission lines (from Lya to CIII]1909, which are observable in the optical window at z~2.5). We are especially interested on the study of the quiescent gas, whose kinematic pattern, chemical composition and spatial distribution will provide valuable information about the formation process and early evolution of the host galaxy. We will also gain a much better understanding of the nature and general properties of the giant nebulae around high redshift radio galaxies.

PMAS/PPAK on the 3.5m is an ideal instrument for this project. It comprises ~331 science fibers of 2.7''/fiber with an hexagonal field of view ~74 x 65 arcsec^2, which is well suited for the expected size of the giant halos (<~25 arcsec). Thanks to the large size and efficiency of the optical fibers of PPAK, we can detect structures as faint as ~1e-17 erg s-1 cm-2 arcsec-2 and reach S/N levels equivalent to those obtained with the traditional long slit spectroscopic technique using 10m class telescopes in similar exposure times.

From May 30th to June 5th we carried out our first observations with PMAS/PPAK. We obtained low (FWHM~10 Å, grating V300) and intermediate (FWHM~5 Å, grating V600) resolution spectra of the radio galaxies 4C40.36 (z=2.27) and 4C48.48 (z~2.34). Figure 1 shows an image of the Lya nebula surrounding 4C40.36, obtained by coadding the 4 exposures of 1800s using the V300 grating and interpolating to a 0.5'' resolution using E3D (Sánchez 2004). The data were reduced using R3D, a new IFS reduction package also presented in this newsletter. It can be clearly seen that the Lya emission is extended for ~9 arcsec (~100 kpc) not only along the radio axis, but also perpendicular to it. Lya emission seems to be detected also outside any plausible ionization cones. Further work is needed to investigate the mechanism responsible for the Lya emission in these regions.

We have compared the efficiency of PMAS/PPAK with that of the Low Resolution Imaging Spectrometer (LRIS) currently installed on the 10m Keck I telescope. Fig. 2, bottom-panel, shows our PMAS/PPAK spectrum of the fiber (2.7 arcsec diameter) centered on the highest surface brightness regions of the radio galaxy, for the V300 grating. The total exposure time was 2 hr. The top-panel shows the long slit spectrum obtained in 1998 with Keck (see Vernet et al. 2001 for a more detailed description). The exposure time was 3 hr. With a 1 arcsec wide slit the spectral resolution for this spectrum is ~10 Å , similar to that obtained with PMAS/PPAK.

Figure 2
Comparison between the spectrum of 4C40.36 taken with LRIS at Keck, with a 1" wide slit (3h integration, FWHM ~ 10 Å) (top panel) and the spectrum taken with PMAS/PPAK (2h integration, FWHM ~ 10 Å) at the CAHA 3.5m telescope (bottom panel). From blue to red, the Lya, NV1240, CIV1550 and HeII1640 lines are seen. The Keck spectrum was extracted from a 2.7 arc sec aperture along the 1arc sec slit located along the radio axis and centred on the Lya peak of emission. We show here a coadded spectrum extracted using a ~2.7 arcsec aperture, matching, as much as possible, a single PPAK fiber (2.7" diameter). The PPAK spectrum corresponds to a single fiber centred on the peak of the Lya emission, arbitrary scaled to match the Keck spectrum.



The PPAK spectrum shown in Fig.2 is the results of a preliminary reduction performed during the observations and we expect the final spectra to have better signal-to noise. Nevertheless, the main UV lines Lya, CIV1550 and HeII1640 are detected with similar signal to noise as in the Keck LRIS spectrum (with a shorter exposure time!). This is possible thanks to the large PPAK fibers (2.7'' diameter): we loose spatial resolution, but gain in depth and, in addition, we have 2D spatial information. We are in the process of developing new dithering techniques that would allow us to sample the object down to the original fiber sampling. These very promising preliminary results show that 3.5m telescopes can be competitive with 10m class telescopes using the appropriate instrumentation and technique.

References


Baum S., Heckman T., Bridle A. et al., Extended optical-line-emitting gas in radio galaxies - Broad-band optical, narrow-band optical, and radio imaging of a representative sample 1988, ApJSS, 68, 833
Baum S., Heckman T., 1989: Extended optical line emitting gas in powerful radio galaxies: what is the radio emission-line connection, ApJ, 336, 702
Clark N., Axon D., Tadhunter C. et al., 1998: Jet-induced Shocks in 3C 171: an Intermediate-Redshift Analog of High-Redshift Radio Galaxies, ApJ, 494, 546
García-Lorenzo B., Sánchez S.F., Mediavilla E. et al., 2005 : Integral field spectroscopy of the central regions of 3C 120: Evidence of a part merging event, ApJ, 621, 146
McCarthy P., Baum S., Spinrad H., 1996, Emission-Line Properties of 3CR Radio Galaxies. II. Velocity Fields in the Extended Emission Lines ApJS, 106, 281
McLure R., Kukul M., Dunlop J., et al. 1999 :A comparative HST imaging study of the host galaxies of radio-quiet quasars, radio-loud quasars and radio galaxies - I MNRAS, 208, 377
Ossterloo T., Morganti R., Sadler E. et al.: Extended H I Disks in Dust Lane Elliptical Galaxies 2002, AJ, 123, 729
S.F.Sánchez, J.I.González-Serrano, 2003: The Near-infrared properties of Host Galaxies of radio-loud QSOs, AA, 406, 435
Stevens J., Ivison R., Dunlon J. et al. 2003: The formation of cluster elliptical galaxies as revealed by extensive star formation, Nature, 425, 264
Tadhunter C., Fosbury R., Quinn P., 1989: The kinematics of ionized gas around powerful radio galaxies, MNRAS, 240, 225
Villar-Martín M., Tadhunter C., Morganti R. et al., 1999: PKS 2250-41 and the role of jet-cloud interactions in powerful radio galaxies, MNRAS, 307, 24
Villar-Martín M., Vernet J., di Serego Alighieri S., et al., 2002: Giant low surface brightness haloes in distant radio galaxies: USS0828+193 , MNRAS, 346 273
Villar-Martín M., Vernet J., di Serego Alighieri S., et al., 2003: Kinematically quiet haloes around z~2.5 radio galaxies. Keck spectroscopy,MNRAS, 346, 273