ABSTRACT

We examine the distribution and kinematics of atomic and molecular gas mapped in a number of galaxies suspected to be in the process of merging. In most cases, the nuclear region of the merger has a high concentration of molecular gas, and a deficiency of atomic gas as compared with larger radii. Thus the total surface mass density of gas often has a minimum at an intermediate radius. In cases where the gas rotation curve is measured, the transition from regions dominated by molecular gas to those of atomic gas corresponds to abrupt changes in rotation characteristics. We propose that the merger is efficiently converting ISM from atomic into molecular form in central region of these galaxies, and that the dense clouds are experiencing radial accretion at a higher rate than diffuse gas.

INTRODUCTION

Since the early work of Toomre and Toomre (1972), much progress has been made in understanding dynamical processes of merger galaxies (cf. Barnes and Hernquist 1993). One result that appears consistently in theoretical studies and numerical simulations alike is that the end product of the merger is an early type galaxy. Increasingly, this scenario has acquired observational support as well. For example, the K-band light profiles of many Arp galaxies, mostly advanced mergers, show r1/4 law typical of ellipticals (e.g., Wright et al. 1990; Stanfrd and Bushouse 1991). Yet, the process in which the merging disks shed their abundant gas mass remains unclear, and numerical simulations are far from adequately resolving this problem given the enormous dynamical range required to mimic the changes in the ISM.

ABSTRACT

I review recent work on galaxy formation and relate it to questions concerning the formation and fuelling of active galactic nuclei. The theory of galaxy formation has developed dramatically in recent years as a result of new analytic methods coupled with substantial programs of direct numerical simulation. Many aspects of how galaxies might form in a universe where structure grows by hierarchical clustering are understood quite well. Others, particularly those that are closely linked to the star formation process, remain highly uncertain. Nevertheless, it is now possible to calculate formation and interaction rates for galaxies with some confidence in a wide variety of cosmogonies. It seems likely that nuclear activity, either starburst or AGN, is an inevitable consequence of the violent, asymmetric, and time-dependent processes which occur during the assembly of galaxies.

INTRODUCTION

The idea that quasars might be related to galaxy formation followed quickly after the first measurements of QSO redshifts but was somewhat neglected after the near-universal acceptance of the argument that QSO luminosities are more easily explained by accretion onto a supermassive black hole than by starlight. In such a model black hole formation and fuelling are major issues which must be addressed before QSO's and galaxy formation can be linked. There has always been some dissent from this model, most notably in recent years from R. Terlevich and his collaborators (e.g. Terlevich and Boyle 1993), but recent discussions of quasar formation have tended to emphasize how late the onset of quasar activity may be in comparison with the initial collapse of a protogalaxy (see, for example, Turner 1991).

ABSTRACT

It is generally believed that galaxy interactions induce bursts of star formation. We observed a sample of galaxies undergoing different types of interactions in the expectation that the location and nature of the induced star formation could be related to the dynamics of the interaction. We found instead that in almost all galaxies the star formation is concentrated in the nucleus or nuclei, sometimes to a remarkable degree. It appears that extra-nuclear star formation is either difficult to trigger or so short-lived as to be rarely observed. We discuss in detail two galaxies: NGC 5253, site of the most concentrated star formation region yet known, and Arp 30, the exception where star formation is broadly distributed and is seen in both nuclei and in clumps in the bridge connecting them.

INTRODUCTION

Studies of the global properties of large samples of galaxies have shown that interactions between galaxies correlate with bursts of star formation and have lead to the generally accepted belief that interactions can trigger such bursts. Attempts to correlate the local star formation properties of individual galaxies and the interactions they have undergone have been much less successful. We therefore undertook a multiwavelength study of star formation in a sample of interacting or post-interaction galaxies, in the hope of being able to relate the location and type of star formation to the interaction history. The observations include infrared spectra, radio continuum maps, and images in continuum bands, the Wolf-Rayet feature and Hα.

Introductory remarks

In previous chapters, we have considered microscopic instabilities, such as the two-stream instability, and MHD instabilities. The third major category of instability is that of ‘resistive instabilities.’ They differ from two-stream instabilities in that they may be treated by fluid equations, and they differ from MHD instabilities in that they are due essentially to the fact that the resistivity is nonzero. A nonzero resistivity allows the magnetic-field lines to move independently of the plasma, so that the ‘frozen flux theorem’ is not applicable in this context. A crucial consequence of this effect is that ‘magnetic reconnection,’ shown schematically in Fig. 17.1, is allowed. In addition, field lines can ‘vanish.’ Consider, for instance, the azimuthal magnetic field set up in a linear pinch. If the resistivity is nonzero, the current will dissipate. As a result, the circular field lines will shrink in diameter and the innermost loops will shrink to a point and disappear (see Fig. 17.2). If there is no external energy supply, the decrease in magnetic energy is converted into joule heating.

Current sheet configuration

The resistive instability that is of special importance in astrophysics is the ‘tearing-mode’ instability that can develop in a current sheet configuration. We shall find that this instability leads to development, in a field configuration of the type shown in Fig. 17.3(a), of ‘magnetic islands,’ as shown in Fig. 17.3(b).

ABSTRACT

The Galactic Center shows evidence for the presence of three important AGN ingredients: a black hole (M* ∼ 106 M⊙), an accretion disk (10-8.5 – 10-7 M⊙ yr-1) and a powerful jet (jet power ≥ 10% disk luminosity). However, the degree of activity is very low and can barely account for the energetics of the central region.

INTRODUCTION

The dynamical center of the Galaxy is the radio point source Sgr A*, which is also the center of the central star cluster (Eckart et al. 1993). Investigations of the enclosed mass in the central region show that there is evidence for a mass concentration of the order of 106 M⊙ within the central arcsecond (Genzel and Townes 1987). There is good reason to assume that this “dark mass” indeed is the mass of a massive black hole (BH) powering Sgr A*. The total spectrum of this source from radio to NIR was compiled by Zylka et al. (1992). There is a flat radio spectrum up to 7 mm, and a steeply rising submm spectrum, which Zylka et al. interpret as thermal emission from a dust torus surrounding the BH. In the FIR one finds a spectral break at 30 μm indicated by upper limits and a third spectral component rising in the NIR, which has been interpreted as emission from an accretion disk around the BH.

HERTZSPRUNG-RUSSELL DIAGRAM FOR THE SGR A* DISK

Because of strong obscuration in the galactic plane we probably will never be able to measure exactly the optical and UV part of Sgr A*, which is needed to discriminate between different disk models.

ABSTRACT

The effect of an active nucleus on its host galaxy (Shanbhag 1991; Shanbhag and Kembhavi 1988; Begelman 1985) can be large. In particular, heating by the radiation from the nucleus strongly affects the hydrodynamic evolution of gas in the interstellar medium (ISM) of the host galaxy. Enhanced star formation activity on a galaxy-wide scale can be induced. The proximity of NGC 1068 makes it an interesting candidate to study such interaction. The model explains extended X-ray emission and some properties of the diffuse ionized medium (DIM) observed in the galaxy.

INTRODUCTION

NGC 1068 is a nearby Sy 1 galaxy disguised as Sy 2 with its “buried” nucleus obscured from direct view (Antonucci and Miller 1985). The unobscured solid angle as seen from the nucleus, is inferred to be ∼ π (Krolik and Begelman 1986). The emission from the nucleus is modeled by bipolar conical outflows of opening angle ≃ 82° with the symmetry axis inclined to the plane of the galaxy by ≃ 35° (Cecil et al. 1990). Its inferred intrinsic luminosity (Sokolowski et al. 1991) is ≃ 7 x 1043 erg s-1. Many observations (Bergeron et al. 1989; Cecil et al. 1990) show existence of activity in the direction of the cone open angle; implying a connection with the active nucleus. The motivation behind this paper is to understand the nature of such interaction.

ABSTRACT

The infrared luminous galaxy MK 231 appears to exhibit characteristics similar to those of active galactic nuclei — it has been classified as a Seyfert 1 system. However, it has been shown to contain 3 ×x 1010 M⊙ of gas via CO observations. If the CO is confined to the region shielded by dust as in galactic molecular clouds, the molecular gas occupies a much smaller volume than previously thought. The ultimate questions are which characteristic is primarily responsible for most of the luminosity and whether AGN and starbursts are interdependent or coincidental.

THE SUPERGIANT MOLECULAR CLOUD

Scoville et al. (1989) and Radford et al. (1991) have discussed the physical conditions implied by the molecular observations of MK 231. It is inferred from CO observations, that 3 × 1010 M⊙ of molecular gas (H2) resides within a volume of radius RCo < 3 kpc. However, recent infrared observations reveal that the emission from dust at λ ≃ 10 μm arises within a volume of size RIR < 400 pc (Keto et al. 1992). This is an important complementary result to the CO observations because studies within our galaxy show clearly that the molecular gas is largely confined to the region shielded by dust extinction (Young et al. 1982).

ABSTRACT

It has been claimed that spherically symmetric, isothermal (ρ ∼ r-2) halos of invisible material surround certain galaxies and extend to between ≈ 40 kpc (Rubin et al. 1985) and ≈ 1 Mpc (Charlton and Salpeter 1991) from their centers. In this work in progress we consider the tidal effects due to the presence of such halos in binary galaxies. Having investigated the predicted frequency of tidal distortions we then compare with binary galaxy surveys to search for evidence of these effects.

TIDAL RADIUS

A tidal radius was calculated based upon a simplified model of a binary galaxy system. Each member of the system is identical and in a circular orbit about the system's center of mass. The dark matter halos extend to the point where they just begin to overlap and the visible disk of each galaxy, despite rotating on its axis, is taken to be approximately spherically symmetric. Furthermore, the rotational angular momentum vector of the material contained in each visible disk is taken to be parallel to the orbital angular momentum vector for the entire system. This information was used to determine the radial equation of motion for a test particle located at the outer edge of either visible disk and from this equation the tidal radius is obtained (≈ 40 kpc). The tidal radius is the separation between members of the binary system at which the test particle just begins to leave the visible disk.

In previous chapters, we have studied the behavior of charged particlesi using various approximations, in given electric and magnetic field configurations. For instance, we have studied waves in a medium that is assumed to be homogeneous and static. In reality, an astrophysical plasma (such as the solar wind) is neither homogeneous nor static. Furthermore, we have only incomplete information about the structure and time-evolution of the medium. Another important class of problems involves systems that are mildly unstable, in which perturbations grow from some initial small level to a finite but significant level. An example of such a situation is provided by the study of the growth of plasma oscillations in the quasilinear approximation (see, for instance, Drummond and Pines, 1962, or Nicholson, 1983).

In problems such as these, we must develop procedures for representing and perhaps calculating the fluctuations that arise in a plasma. In many cases, we can conveniently represent these fluctuations as a superposition of waves (normal modes) in the plasma. However, the individual particles in such a plasma may behave in a manner that differs in important respects from the behavior of particles in the simpler situations that we have studied so far.

In discussing collision theory, we took account of the fact that a plasma is highly random at the microscopic level that is relevant for the discussion of binary collisions, etc. We now introduce the concept that the plasma may also be random on a macroscopic level, and we study the implications of such randomness.

ABSTRACT

I discuss the various processes which may affect the transfer of angular momentum in gas within the inner parsec surrounding a supermassive black hole. Even after gas has been brought into the gravitational sphere of influence of the black hole, it still has 100–1000 times too much angular momentum to reach the event horizon. Angular momentum loss cannot be accommodated in a scaled-up standard thin accretion disk model, because local self-gravitational instabilities will lead to fragmentation of the disk, which will decrease the efficiency of angular momentum transport. Instead, some nonlocal mechanism or “external” trigger for angular momentum transfer is needed, such as a large-scale magnetized wind, stirring by winds from massive stars and supernovae, or global gravitational instabilities.

INTRODUCTION

Most participants in this conference seem to agree that large-scale, non-axisym-metric gravitational disturbances can be very effective in transferring angular momentum on scales of hundreds of parsecs or larger. These disturbances might be driven by tidal encounters or may be “self-starting”, as in the “bars-in-bars” scenario. What is less certain is whether similar mechanisms can be effective all the way into the nucleus, say, into the inner parsec surrounding a supermassive black hole. Gravitational triggers for inflow usually involve at least a mild form of self-gravitational instability, which requires that the stars plus gas constitute a significant fraction of the total mass enclosed within the region under consideration (see, e.g., Friedli, these proceedings).

ABSTRACT

The apparent low covering factor of the narrow line region (NLR), the nearly “empty” intermediate region between the broad line region (BLR) and the NLR, and the size of the BLR, are all naturally explained if dust is embedded in the narrow line emitting gas. Such dust, together with the observed blue-excess asymmetry of the narrow line profiles, imply that the NLR gas has a net inflow motion.

THE BASIC QUESTIONS

In this paper we address the following three questions: 1) Why does RBLR ∼ 0.1 L½46 pc? This scaling is indicated by the remarkably similar emission spectra of AGNs over a very large luminosity range. 2) What is the actual covering factor (C) of the cold gas in the NLR? Photon counting using the Hβ recombination line indicates C ∼ 2%, assuming optically thick clouds. The 3 – 30 μm IR continuum, if due to dust reprocessing of the UV continuum, indicates C ∼ 20 – 40%. High resolution imagings of the NLR by the HST in nearby AGNs indicate, if the clouds are resolved, C ≥ 20%. 3). Why is there an apparent gap in the gas distribution between the BLR and the NLR? A gap is indicated most clearly in Seyfert 1.5 galaxies and some quasars where the permitted line profiles show distinct narrow and broad components.

ABSTRACT

We used optical spectra to investigate the nuclear regions of the starburst galaxies NGC 2782, NGC 4102 and NGC 6764. In addition to the central starburst, we find evidence for extranuclear shock-ionised gas, with kinematical properties consistent with outflow along the minor axis. The observations are consistent with the presence of dense shockionised shells, formed by the starburst-driven winds. The shells, observed in NGC 2782, NGC 4102 and NGC 6764 respectively, appear to be in a different evolutionary phase, which is explained by differences in age and strength of the central starbursts.

INTRODUCTION

Evolutionary models of starburst nuclei predict the formation of a thin, dense, shock-ionised shell, surrounding a hot cavity (e.g. Tomisaka and Ikeuchi 1988). The shell expands predominantly perpendicular to the galactic plane, gradually elongating until it finally breaks open at the top. The nearby starburst galaxies NGC 253 and M82 are probably examples of the broken-shell phase.

We investigated the starbust galaxies NGC 2782, NGC 4102 and NGC 6764 with the aid of long-slit spectra with high spatial resolution. All three galaxies have been classified previously as starburst galaxies, but the optical line ratios derived from 1-dimensional spectra put the galaxies close to the borderline between starbursts and LINERs in diagnostic diagrams.

RESULTS

From the spatial behaviour of the line ratios in our spectra we find, that the high line ratios are due to the presence of an extranuclear high-ionisation component in addition to the central starburst. The line ratios of the high-ionisation component are in agreement with shock-ionisation; its kinematical properties are consistent with outflow along the minor axis.

ABSTRACT

Theoretical studies of gas dynamics in disk galaxies are reviewed in relation to the fueling of nuclear activity. Importance of self-gravitational effects in the interstellar gas component is emphasized.

FUELING PROBLEM

Recent observations have revealed that some galaxies show an unusual level of activity near the nuclear regions. One type of activity is the enhanced star formation in the central few kpc regions around galaxy nuclei. Another is the non-thermal activity which originates from sub-parsec region, which is called active galactic nuclei (AGN). This type of activity is believed to be powered by mass accretion onto supermassive black holes located in the galactic nuclei (e.g., Begelman et al. 1984). Both types of activity require adequate source of fuel. In typical AGNs, a gas supply rate of ∼ 1 M0 yr-1 is needed to generate the radiation of the observed amount. CO line observations (e.g., Kenney 1990) have detected large accumulation of a potential fuel in the form of molecular gas in the central kpc in many starburst galaxies.

Possible candidate sources of fuel are divided into local and global ones according to their spatial distribution. The difficulties for local sources such as compact star clusters have been pointed out by several works (e.g., Shlosman et al. 1990). In this article we concentrate on global sources, especially the interstellar gas in the galactic disks. Because interstellar gas is distributed on the 10 kpc scale, one problem is how the material can be channelled into the nuclear region with a sufficiently high efficiency to maintain the observed level of activity.

ABSTRACT

Galaxies with elevated metabolic rates get energy from their gaseous food by extracting its nuclear energy (in stars), and its gravitational energy (via accretion onto massive black holes). There is strong evidence that interactions with other galaxies trigger star formation activity, and weaker evidence that it triggers black hole accretion (nuclear activity). We review the processes by which interactions can remove angular momentum from gas, particularly gravitational torques and the m = 2, m = 1, Jeans and fission instabilities that give rise to them. There is ample evidence, both theoretical and observational, that these can remove enough angular momentum to move much of a galaxy's gas from ∼ 3 kpc to ∼ 300 pc. This is still many decades from the ∼ 10-5 pc scales of stars and black hole horizons. We discuss star formation, the interpretation of simulations, and cosmological implications. The evolution of binary supermassive black holes, and the problem of forming a dense (≲ 1 pc) nuclear star cluster are examined.

WHAT IS MASS-TRANSFER INDUCED ACTIVITY IN GALAXIES?

Before this conference, I wasn't sure. After this conference, I am sure I am not sure. Let me nevertheless attempt a definition, starting from the easy end, the back of the phrase. Galaxies are of course the island universes within which reside most of the stars, much of the gas, and a little of the mass in the cosmos. Activity in Galaxies, like that in animals, is defined by the metabolic rate. When this is well above the average ‘resting’ (aka. basal) level, a galaxy or animal is said to be active.

ABSTRACT

We discuss the results from a study in progress of radio–loud IRAS galaxies. We have discovered a class of gas-rich AGN which are characterized by large IR luminosities, and which are intermediate in radio luminosity, IR colour, and optical spectral class between the IRAS ultraluminous galaxies and powerful quasars. These objects form an important sample in studies of the AGN-starburst connection and may be the evolutionary link between the ultraluminous galaxies and quasars.

INTRODUCTION

The relationship between nuclear activity and galaxy evolution is poorly understood. Speculation on the subject ranges between ideas that Seyfert activity is due to nuclear starbursts, to the theory that the ultraluminous (LFIR ζ 1012 L⊙) far-infrared galaxies like Arp 220 are dust enshrouded, young quasars (Sanders et al. 1988). Unfortunately there exists no sample of active galaxies that can be used to test these various hypotheses since most samples used in these studies are selected either on the basis of optical emission-line equivalent width, or IRAS colour selection criteria, IRAS luminosity, etc. There is no ideal sample, but a good way of studying possible nuclear activity/evolution relationships is to select gas-rich galaxies with nuclear activity. We have selected sources from the IRAS survey (which is biased toward dusty, gas-rich objects) by using a completely independent indicator of galaxy activity: strong non-thermal radio emission. This sample of gas-rich radio galaxies also allows us to investigate the complex relationships between the ambient interstellar medium, the radio source, and the active nucleus.

ABSTRACT

We have studied the luminosity function (LF) of H II regions in the disk of the Seyfert 1 galaxy NGC 6814. We find that the LF is very similar to LFs of other late-type, not necessarily active, spiral galaxies. Although the Seyfert nucleus shows its character by emitting strongly in Hα, the disk H II regions seem not to be influenced by the active nucleus.

NGC 6814 is an Sbc galaxy with well-defined spiral arms, of type Seyfert 1. Because of the strong X-ray emission, it is considered a key object for understanding nuclear activity.

We have obtained new Hα observations with the 4.2m William Herschel Telescope (WHT) on La Palma, using the TAURUS instrument in imaging mode. The final continuum subtracted Hα image has high sensitivity (H II region detection limit is L = 1036.9 erg s-1) and resolution (0.8 arcsec, or about 100 pc at the distance of NGC 6814). From the image, we have measured positions, diameters and fluxes of a total number of 735 H II regions (Knapen et al. 1993). We found that the nucleus is a strong Hα emitter, of luminosity L = 1039.9 erg s-1.

From our catalog of H II regions we constructed a luminosity function (LF), for all the H II regions in the disk of the galaxy, and for arm and interarm H II regions separately. Figure 1 shows the total LF. The slope of the LF is a = –2.37 ± 0.09, well within the range of slopes measured in the literature for galaxies of similar morphological type. The arm and interarm LF slopes are equal within the fitting uncertainties.