Next Contents Previous

4. ORIGIN OF THE EMITTING GAS, KINEMATICS AND GAS FLOWS

In the previous sections, we have seen that extended Lyα emission is a common phenomenon around bright high redshift quasars (with possible differences below z ∼ 3 for radio-quiet systems), radio-galaxies and in overdense regions of the Universe. The ubiquity of AGN and massively star forming galaxies associated with these nebulae and simple analytical considerations discussed in section 3 suggest that most of the emission may be due to recombination radiation from dense and cold clouds within the nebulae. Because the emission is sensitive to gas density squared, the emitting gas could be a small fraction of the total gas in and around the massive halos associated with these systems, both in terms of volume and mass. In this section, we will look at the kinematical signatures derived from Lyα spectra of these sources in order to address the questions about the possible origin and fate of this cold gas component.

The proposed origins for the emitting gas and its relation to gas flows in the literature include: i) cosmological accretion from the IGM, ii) outflowing material from galaxy and AGN feedback, iii) in-situ formation from hot gas condensation. The suggested options about the fate of the gas include: i) accretion or "recycling" into galaxies (inflow) or into gaseous disks (rotation), ii) expulsion from the galaxy halos towards the IGM, iii) disruption and thermalization into the hot halo gas. Each of these hypothesis could potentially leave an imprint into the observed gas kinematics in terms of Lyα spectral profile shapes, velocity shears and velocity dispersion. However, it is important to stress that Lyα line is a resonant line and therefore any spectral information may also reflect complex radiative transfer effects rather than kinematics, especially if the gas is not highly ionized as it may be the case for a Lyα cooling origin of the emission.

The first spectroscopic measurements on giant Lyα nebulae were obtained on radio-galaxies and radio-loud quasar nebulae by McCarthy et al. (1987), McCarthy et al. (1990), Heckman et al. (1991b), McCarthy et al. (1996). The properties of these systems from a kinematical point of view appeared remarkably similar: the Full Width Half Maximum (FWHM) of the Lyα emission presented large values typically around 1000-1500 km s−1 for all radio-loud nebulae. No hints of velocity shear in excess of 500 km s−1 were found, although radio-galaxy nebulae seemed to show a more complex kinematical structure. Given the alignment effect between the radio-jets and Lyα emission discussed in section 2.2, these large FWHM were interpreted preferentially as being associated with powerful outflows from these AGN as a result of the jet-gas interaction (e.g., van Ojik et al. 1997, Humprey et al. 2006). However, it was early recognized that such large velocity widths could have been also caused by gravitational motion in very massive halos (e.g., Heckman et al. 1991b). More recent spectroscopic observations of radio-loud systems confirmed these large FWHM (e.g., Villar-Martin et al. 2003, Humphrey et al. 2006, Villar-Martin et al. 2007, Roche et al. 2014) but found that some radio-galaxies show, in addition, extended and lower-SB halos that appear more kinematically quiet, e.g. with FWHM ∼ 500 km s−1. Moreover, some of these extended "quiescent" regions show indication for velocity shifts of few hundred km s−1 that in some cases have been interpreted as a possible sign of rotating disks (e.g., Villar-Martin et al. 2007) or cosmological gas infall (e.g., Humphrey et al. 2007). These models are based on the correlation seen in several radio-galaxy nebulae i.e. that the more redshifted side of the nebula is the brighter in both Lyα and radio flux, and on the interpretation that this side of the radio-lobes and associated nebula is closer to the observer. This is based on the fact that the radio jet directed towards the observer will be Doppler-boosted and on the assumption that the Lyα emitting gas closer to the observer will be less "absorbed", i.e. scattered to a larger projected area, by neutral gas in the halo. The relative redshift of this near-side relative to the observer would then imply that the emitting nebula has a significant component of infall towards the galaxy.

Kinematical signatures in radio-galaxy and radio-loud quasars are however difficult to interpret because of the complex interaction between the possibly accreting gas and the massive energy input of the radio jet. The analysis of extended, rest-frame optical emission lines from several HzRG, e.g. Nesvadba et al. (2006), suggests indeed that kinematics in the brightest parts of the radio-loud nebulae should be dominated by powerful outflows. These observations, together with the measured line ratios using, He II, C IV, O III and S II with respect to Hα, Hβ and Lyα, all suggest that this outflowing material is metal rich (solar or super-solar), dusty (A(Hβ) ∼ 1-4 mag) and very dense (ne ∼ 500 cm−3) (see e.g., Villar-Martin et al. 2003, Nesvabda et al. 2008).

Long-slit and integral field spectroscopy of radio-quiet nebulae show instead that the large majority of these systems have much smaller FWHM, i.e. FWHM ∼ 300−500 km s−1 than radio-loud systems (e.g., Weidinger et al. 2005, Christensen et al. 2006, Arrigoni-Battaia et al. 2015, Borisova et al. 2016), unless they are associated with very overdense environment like in the case of SSA22-LAB1 and SSA22-LAB2 (e.g., Matsuda et al. 2006) or in the case of the "Jackpot" Nebula (Hennawi et al. 2015). These smaller velocity widths are indicative of better environments where signs of infall or rotation could be studied more in detail without the need to disentangle them from the broad component associated with the radio-jets and powerful outflows.

In particular, some radio-quiet giant Lyα Nebulae, such as the Slug Nebula (Cantalupo et al. 2014) are extended by several hundred of kpc and therefore the kinematics in their external parts are less likely to be "contaminated" by AGN outflows, if present. Long-slit spectroscopy (Arrigoni-Battaia et al. 2015) and integral field observations (Martin et al. 2015) have revealed relatively narrow Lyα emission (FWHM < 350 km s−1) at distances larger than 100 kpc from the quasar and apparent velocity shears of about 800 km s−1 that seem coherent over the brightest part of the Nebula (i.e., over 200 kpc in projected length). These velocity shears have been interpreted as evidence for rotation and would imply therefore that a massive, gaseous disk-like structure with size of about 150 kpc could be present in the brightest part of the Nebula (Martin et al. 2015) while the more tenuous, extended part have been interpreted as signatures of gas infall from the Intergalactic Medium. In particular, the velocity profile seems consistent with the rotation curve of a disk within a NFW profile of a massive halo but only if the "center" of the disk-like structure is located 25" away from the main quasar, in a region of low SB. Considering only the high SB regions, the spectra could be also interpreted however as arising from projection effects of two different structures located at about 500-800 km s−1 in velocity space away from each other. The presence of a relatively bright continuum source in the "redshifted" part of the velocity profile could support this interpretation. Ongoing observations of other emission lines such as He II, C IV (e.g., with MUSE) and Hα (e.g., with Keck/MOSFIRE) will help disentangling these two possibilities.

Other spectroscopic observations of radio-quiet nebulae (e.g., Martin et al. 2014, Prescott et al. 2015) have found signatures of coherent velocity fields over scales of several tens of kpc that have been interpreted as evidence of infall (e.g., Martin et al. 2014) or large scale rotation in a disk (Prescott et al. 2015). In particular, the observations of Prescott et al. (2015) of coherent velocity shears of about ∼ 500 km s−1 in both Lyα and in non-resonant lines such as He II within the central 50 kpc of a 80 kpc-sized Lyα nebula at z ∼ 1.67 provide evidences that the extended gas in this system is produced in situ by recombination radiation and settled in a rotating disk that is at least partially stable against collapse. In a larger spectroscopic study of eight small radio-quiet nebulae at z ∼ 2.3 including rest-frame optical emission lines (mostly OIII and Hα), Yang et al. (2014) found instead no significant evidences for bulk motions such as inflow, rotation or outflows and suggested that the gas should be "stationary" or slowly outflowing at speed less than 250 km s−1 with respect to the central galaxies in these systems.

The MUSE observations of about 17 bright radio-quiet quasars and 2 radio-loud quasars at 3 < z < 4 presented by Borisova et al. (2016) provided the first large statistical sample of giant (> 100 kpc) Lyα nebulae with full kinematical information from integral field spectroscopy over their full detectable extent. Figures 3 and 4 show, respectively, the maps of the first and second moment of the flux distribution, i.e. the flux-weighted velocity centroid shift and the dispersion relative to the peak of the integrated Lyα emission for each of the MUSE Quasar Nebulae (MQN) (figures taken from Borisova et al. 2016). While some systems, e.g. MQN15, show possible evidences of rotation in a disk-like structure with a velocity shear of about 800 km s−1, the majority of the MQN do not show clear evidences of rotation or other ordered kinematic patterns. Several MQN, especially the largest ones, show instead coherent kinematical structures over scales as large as 100 kpc, e.g. MQN01 and MQN03. The velocity dispersions, expressed in terms of Gaussian-equivalent FWHM, clearly shows the main difference between radio-quiet and radio-loud systems: the large majority of radio-quiet nebulae are narrower in Lyα emission (FWHM ∼ 500 km s−1) than radio-loud systems (FWHM > 1000 km s−1) in agreement with previous results discussed in this section. The only exception is MQN06 but this nebula is peculiar and more similar to radio-loud nebulae in terms of all the properties studied in Borisova et al. (2016), including the SB profiles and higher He II/Lyα and C IV/Lyα ratios with respect to the non detection for radio-quiet systems.

Figure 3

Figure 3. "Velocity maps" of the MUSE Quasar Nebulae presented in Fig. 2 obtained from the first moment of the flux distribution (see Borisova et al. 2016 for details). This is the largest sample to date of kinematical maps of giant Lyα nebulae obtained with integral-field-spectroscopy. As discussed in section 4, while some systems (e.g. MQN15) show possible evidences of rotation in a disk-like structure the majority of the nebulae do not show clear evidences of rotation or other ordered kinematic patterns. Several nebulae show instead coherent kinematical structures over scales as large as 100 kpc (e.g. MQN01 and MQN03). (Figure reproduced with permission from Borisova et al. 2016).

Figure 4

Figure 4. "Velocity dispersion maps" of the MUSE Quasar Nebulae presented in Fig. 2 obtained from the second moment of the flux distribution and expressed in terms of Gaussian-equivalent FWHM (see Borisova et al. 2016 for details). This figure shows that the large majority of radio-quiet nebulae are narrower in Lyα emission (FWHM ∼ 500 km s−1) than radio-loud systems, i.e. MQN-R1 and MQN-R2 (FWHM > 1000 km s−1), in agreement with the overall kinematical results discussed in section 4. (Figure reproduced with permission from Borisova et al. 2016).

The emerging picture from these observations seems to suggest therefore that kinematics in radio-loud nebulae may be dominated by ionized outflows of relatively cold and metal-enriched material within at least the inner 30-50 kpc from the AGN, while, on average, the ionized and clumpy gas in radio-quiet nebulae may be in a more "stationary" situation and in some cases settled in a possibly rotating structure. Clear evidences from Lyα emission for gas accretion into galaxies from these cold gas reservoirs are not currently detected, either because "washed-out" by Lyα radiative transfer effects or because their magnitude and projection effects could make them to small to be detected with current facilities. Future deep surveys using other bright, non-resonant lines such as hydrogen Hα or He II 1640 would be extremely helpful to search for small velocity shears and therefore for signature of cosmological gas accretion onto galaxies and AGN.

Next Contents Previous