Abstract

The Hubble Space Telescope (HST) has been producing data from regularly scheduled observations for some time now. At this juncture, it is an appropriate time to review some of the areas where instruments such as the Faint Object Spectrograph (FOS), when coupled with ground-based data, may improve studies of Active Galactic Nuclei (AGN) and, hopefully, stir the imaginations of the theorists! We show examples of objects where the data indicate that the emission line properties of high- and low-luminosity objects are similar, lending support to relatively simple models in which the emission line regions of AGN scale homologously. We note, however, that the definition of what the ‘normal’ properties of AGN are rather than individual peculiarities will have to await complete samples of objects with data covering as large a wavelength range as possible.

Line Ratios.

It has become feasible in recent years to study the Balmer lines of high redshift QSOs in some detail and to determine velocity-resolved ratios such as those of Lyα/Hα – see, for example,. Even more recently, it has become possible to regularly obtain UV data with the HST and to perform more detailed analysis than was possible using IUE spectra. As an example, in Figure 1 we show a comparison between the Lyα/Hα line ratios in two radio-loud objects of similar linewidth and velocity shifts.

The 3C273 data are from [4] and the PKS 1448 – 232 are taken from [2].

Abstract

We analyse the properties of a sample of polarized radio galaxies with a wide range of redshifts. Our results suggest that a large fraction of UV light in distant radio galaxies is scattered anisotropic nuclear radiation.

Introduction

High redshift radio galaxies (HZRGs) have elongated morphologies (emission lines and continuum) aligned with the radio axis. The alignment of the continuum is strong in the ultraviolet and weak in the infrared around 1µ (rest frame) [2]. While the IR round component can be explained with an evolved stellar population, the nature of the UV aligned component is still a matter of debate. Two origins have been proposed for the UV light : (a) star formation induced in the ISM by the radio jet (b) dust/electron scattering of anisotropic nuclear light. We present new data which show that at least a considerable fraction of the UV light is not stellar.

The sample and the correlations

We have collected a sample of 41 radio galaxies with 0.1≤z≤1.2 taken from our survey ([1]) and from the literature. We find that the degree of linear polarization P increases with redshift and with the total radio power at 178 MHz (Pr), while it decreases with the rest frame wavelength of the observation (λrest). The main problem in the interpretation of these correlations is that the quantities z−Pr−λrest are interdependent and it is not possible to clarify which correlations are intrinsic and which derived.

Abstract

The possible relation between observed variability behaviour and slim disk stability properties is examined. It is argued that processes which give rise to QPOs in galactic sources are operative in AGN as well. Thus, it may be that unstable acoustic modes in the inner part of the accretion disk give rise to both the quasi-periodic short-term X-ray variability in NGC 6814 and the horizontal branch oscillations (HBOs) in X-ray binaries. The estimated central mass in NGC 6814 is ∼ 106M⊗.

Introduction

The majority of compact galactic and extragalactic sources seem to accrete at a rate ṁ ∼ 1, where ṁ = L/LE, LE = 1038m erg s−1 being the Eddington accretion rate and where m = M/M⊗ is the central mass in solar units. Some examples corresponding to AGN are shown in Figure 1. It follows that the standard Shakura-Sunyaev model (Shakura & Sunyaev 1973, 1976) is simply inadequate when it comes to a relevant description of, especially, the inner accretion disk in these sources, where the bulk of the luminosity is generated. One may also note that Shakura-Sunyaev disks contain an artificial singularity at the inner edge, due to an improper neglect of some inertial terms in the radial structure equations. Thus, any model which attempts to combine the effects of magnetic fields, electron/positron pairs, winds or whatever, with a Shakura-Sunyaev Keplerian disk, is bound to yield questionable results.

Abstract

Luminous accretion discs around black holes are expected to be optically thick and radiate much of their emission in the EUV and soft X-ray bands. Quasiblackbody emission consistent with such discs is observed in many Seyfert 1 galaxies and from Galactic black hole candidates such as Cygnus X-1. The harder, rapidly variable, X-rays from such objects must originate above the disc, probably from non-thermal processes involving magnetic fields. The disc is therefore irradiated by a hard X-ray continuum. Backscattering and fluorescence from the disc produce a reflection spectrum, which is now observed in X-rays. Features in the reflection spectrum act as a diagnostic of the geometry and conditions of the inner disc, offering the strong possibility that it can be mapped in the near future.

Introduction

We begin by reviewing the case for the presence of accretion discs in many Active Galactic Nuclei (AGN), such as the Seyfert 1 galaxies. Here we are concentrating on the inner disc within radii R ≲ 100Rs, where Rs is the Schwarzschild radius of the central object (assumed here to be a black hole). Such discs were first detected from the UV excess and in particular by the variable soft X-ray emission that they produce. Further rapid progress has been hindered by the unfortunate coincidence that most of the direct thermal radiation produced by accretion discs around massive objects is emitted in the EUV, where photoelectric absorption by the interstellar medium of our Galaxy is strong.

Abstract

Recent detailed data analysis of the last two Ginga observations of the Seyfert nucleus NGC6814 (Leighly et al. 1992, hereafter LKT92) has not only reconfirmed the periodicity of the fastest X-ray variability reported earlier for this source (Done et al. 1992, hereafter DMM92), but also has shown several very unique, definite characteristics which severely constrain any acceptable models. Consequently, various existing models have been ruled out (LKT92). Here we present, as a natural and self-consistent physical model which satisfies these detailed observational constraints, the occultation of the central X-ray source by matter overflowing the Roche lobe of a low mass star orbiting around a supermassive black hole. The importance of careful, detailed comparison of this type of model with further observations is emphasized, because the result may lead to strong circumstantial evidence for the presence of a supermassive black hole in the central engine of active galactic nuclei (AGN).

NGC6814 is among the most interesting Seyfert galaxies in the sense that its nuclear X-ray emission was found to be most rapidly variable, with the timescale of ∼ 300 seconds, and moreover the fastest variability exhibits periodicity of ∼ 12000 seconds (DMM92). The recent detailed data analyses of the last two Ginga observations, in April and October 1990, respectively, have shown several new detailed characteristics, such as the spectral variability and lags in flux between different energy bands (LKT92).

This volume contains the proceedings of the 33rd Herstmonceux Conference, the latest in a venerable series initiated by the Royal Greenwich Observatory in its former home at Herstmonceux Castle. It is the second conference in the series to have been jointly organized by the RGO and the Institute of Astronomy at Cambridge. However, it also marks a beginning. Both the timing and the subject matter of the meeting in Cambridge were co-ordinated with a companion conference in Paris. Together, the two meetings marked the inauguration of the European Association for Research in Astronomy. This grouping links together the Institute of Astronomy, the Institut d'Astrophysique de Paris, and Leiden Observatory with the intention of encouraging scientific exchanges between the three laboratories and enhancing their collaborative research activities. The Paris conference, entitled First Light in the Universe: Stars or QSO's, took place during July 7–11, 1992 at the Institut d'Astrophysique and was concerned with the cosmological evolution of galaxies and quasars, with particular emphasis on the alternative rôles played by starbursts and active galactic nuclei. In Cambridge, our aim was to focus in detail on the sources which power active galactic nuclei themselves.

Active galactic nuclei (AGN) are undoubtedly the most spectacular objects known to astronomy yet the nature of the fundamental power source remains elusive, despite many years of intensive research. Indeed, the somewhat ambiguous conference title reflects the fact that the conventional black hole–accretion disk paradigm is now being strongly challenged by the starburst hypothesis.

A galaxy which radiates strongly at 25 µm is likely to have an active nucleus. For Seyfert galaxies R=F (25 µm)/F(60 µm) is typically 0.2 to 0.5 (Miley & Neugebauer, 1985; De Grijp et al., 1985) and for quasars first detected by IRAS R ranges from 0.2 to 1.1 (Clowes, Leggett & Savage, 1991); whereas for IRAS galaxies for which star formation is the principal source of emission R is usually in the range 0.05 to 0.2. It appears that a ‘high’ value of R (» 0.2) strongly suggests the presence of an active nucleus while a ‘low’ value (R ≃ 0.2) does not exclude this. Hill, Becklin & Williams (1988) note that galaxies with high R tend to be compact at 10 µm which supports the idea that high R may be associated with nuclear activity.

The origin of the strong 25 µm emission from galaxies with active nuclei is unclear. Thermal emission from hot dust surrounding and heated by a power law source is an obvious possibility: however spherically symmetric models do not always provide good fits to the spectral energy distribution of Seyfert galaxies (Rowan-Robinson & Crawford 1989) and it appears that a disc geometry, perhaps combined with a high optical depth, may be needed in some cases. Other factors that may influence the models include the possible destruction of the very small grain component close to the AGN and the clumpiness of the dust distribution.

Introduction

Philosophy

It seems that every time a new population of extragalactic object is discovered, the first reaction of astronomers is to construct a luminosity function. Beyond sheer botany, this serves the useful purpose of giving a check on the completeness of surveys. The long-term motivation is Physics: the hope that the luminosity function and its change with redshift (the nearest we can get to an evolutionary track for a single object) will tell us something about how these spectacular sources operate. However, it has to be said that this objective remains in the far distance, despite nearly three decades of effort.

Why the radio waveband? Apart from the weight of history (radio astronomers take the blame for starting AGN research), the lack of foreground extinction and the lack of catalogue contamination by galactic objects are still very powerful advantages.

Notation

There are a few (arbitrary) conventions commonly adopted in the literature on this subject. The comoving density of objects per unit log10 power is denoted by ρ. The Hubble constant, where quoted explicitly, is given in the form h = H0/100 kms−1Mpc−1. Unless otherwise specified, Ω = 1 and h = 0.5 are assumed.

Radio astronomers' who's who

Radio astronomy takes the brutalist line of ordering the Universe according to output.

Abstract

We have embarked on a search for temporal velocity shifts, on timescales of a year or so, in QSO broad lines. Eighteen quasars in a total sample of thirty have been analysed so far, and no evidence for velocity shifts has been found. If no shifts are found in the remaining twelve QSOs, then our results will call into question the identification of certain QSO pairs as lensed systems and the limits on the sizes and spatial distribution of Lyman α clouds derived from such systems.

Introduction

Steidel and Sargent (1991) recorded high signal–to–noise ratio spectra of the QSO pairs Q1634+267A,B and Q2345+007A,B in order to determine whether the systems were gravitationally lensed. They concluded that the systems were both lensed on the strength of detailed comparisons of the line profile and continuum shapes. However, they also discovered that the Ly α, N V λ1240, C IV λ1549, and Si IV λ1400 emission lines of the two images of Q1634+267A,B exhibit a relative velocity shift of as much as ∼ 1000 km s−1. The authors favored the explanation that the redshifts of individual lines in QSO spectra vary with time and that, due to the roughly one year time delay between the light paths of the two images, one is seeing the individual QSO at two different times. Here, we describe our efforts to verify this prediction. We have reobserved 30 objects from the Mg II λ2800 survey of Steidel and Sargent (1992), being careful to use an identical instrument configuration and to obtain similar quality spectra.

Abstract

We report new measurements of the BL Lac object ON 231 (W Com) from radio to optical wavelengths. This source was found to be at its highest brightness in the near IR and optical bands for many years.

Introduction

ON 231 is one of the targets in the observational programme in the near IR that we have been carrying out since 1986 at the 1.5 m Italian IR Telescope at Gornergrat (TIRGO, 3150 m a.s.l.). The results of the measurements performed up to the spring of 1988 are given in. At the beginning of February 1992 we found the source in the highest state for about 20 years and therefore, in addition to the IR and optical measurements, we observed ON 231 at radio frequencies to obtain a more complete picture of its spectral distribution. Radio observations were carried out with the 32 m dish of Istituto di Radioastronomia (CNR - Bologna, Medicina) at a frequency of 22.2 GHz.

Results

The results of the new measurements are presented in Fig. 1 (open circles) together with those of Landau et al. performed in the spring 1983 (crosses). Some remarkable facts are evident: i) the flux in the near IR and optical bands in 1992 is generally higher than in 1983; ii) at variance with the above, the flux at 22.2 GHz does not show a significant change with respect to the previous measurement; iii) the maximum power emitted by ON 231 in the high state lies between 250 and 500 THz, while in 1983 it was at a lower frequency, likely below 100 THz; iv) the spectral slopes in the optical and near IR bands are flatter when the luminosity is higher.

Abstract

The Starburst model for radio-quiet Active Galactic Nuclei (AGN) postulates that the activity seen in most AGN is powered solely by young stars and compact supernova remnants (cSNR) in a burst of star formation at the time when the metal rich core of the spheroid of normal early type galaxies was formed. In this model, the broad permitted lines characteristic of the Broad Line Region (BLR) and their variability are originated in these cSNR. Combined analytic and numerical hydrodynamic simulations, with static photoionization computations have shown that cSNR can reproduce most of the basic properties of the BLR in low luminosity AGN.

We have explored the hypothesis that QSOs are the young metal rich cores of massive elliptical galaxies forming at z ≳ 2.0. Only a small fraction (∼ 5%) of the total mass of a normal spheroid, the core mass, is needed to participate in a burst to explain the observed luminosities and luminosity function of Quasars at z ≳ 2.0. We predict that the progenitors of QSOs should look as dusty starbursts and about 4 times more luminous than QSOs themselves.

Introduction

The hypothesis that a Starburst can power the most extreme forms of nuclear activity has been proposed several times in the past (Shklovskii 1960, Field 1964, McCrea 1976), but was not favoured mainly because it failed to explain satisfactorily the observed large luminosity and variability of quasars, their radio emission, unresolved images, the presence of extremely broad permitted emission lines in the spectrum and their observed intensity ratios.

Introduction

We have heard many times during this conference that variability is a powerful tool to investigate the nature of AGN. The goal of this talk is to demonstrate that the clues derived from variability studies may be partial, disappointing and even misleading, if some characteristics of variability are overlooked and the proper statistical tools are not adopted. To convince you that such a caveat, although obvious, is important and productive, we should like to focus on a few issues, particularly important from our point of view.

The stochastic nature of the light curves

First of all, let's draw your attention to the fact that light curves of quasars have generally random behaviour, i.e., that the knowledge of the value of a time series at a given instant does not allow (in a broad sense) one to forecast the future evolution of the light curve. What does it mean from a physical point of view? It means that quasars are dynamical systems of high dimensionality, systems whose temporal evolution is described by an extremely large set of differential equations (or by partial differential equations). In other words, quasars are dynamical systems whose evolution is determined by a large number of external factors. To understand this point let us consider an example close to common sense: a pendulum. As is well known, the temporal evolution of such a system is described by a second order ordinary differential equation.

Abstract

The first ROSAT X-ray spectra of two high-redshift quasars reveal unexpectedly strong absorption when compared with similar luminosity objects at lowredshift. A third quasar shows none. A fourth, low-redshift, radio-loud quasar (3C351) with extended radio structure, shows absorption possibly due to a warm absorber with a strong OVII absorption edge.

Introduction

X-ray spectral observations of quasars have been confined to low redshift objects (z≤0.5) whose proximity makes them bright enough to study and also to those with relatively bright X-ray flux (αox≲1.5). ROSAT, with its high sensitivity, enables us to observe the spectra of high redshift (z>2) and large αox quasars for the first time. We have begun a ROSAT observing program to study the X-ray spectra of quasars selected to cover the full range of continuum properties. In particular this sample includes objects at high redshift, with relatively faint X-ray flux and with a full range of radio properties: strong, weak, extended and compact. We are also carrying out a follow-up observing program to obtain multi-wavelength (infrared – ultra-violet) data for all our ROSAT-observed quasars.

Sampling the full quasar population with ROSAT

To date we have received and analysed data for > 25 quasars. Their spectra are generally steeper than those seen at higher (e.g. Einstein IPC) energies, as observed in general with ROSAT. Our current sample includes 4 high-redshift (z>2.8) quasars with sufficient counts (> 350) to obtain spectral information (Table 1).

Abstract

The variability properties of a sample of over 300 optically–selected quasars near the South Galactic Pole (SGP) have been studied using a series of eleven UKST Bj plates at seven epochs, spanning 16 years. Quasars of high luminosity show significantly less variation than those with low luminosity. A similar, though much weaker, trend with redshift was found; lower redshift quasars varying proportionally more than high redshift quasars. The observed trends are a consequence of an intrinsic dependence of quasar variability on luminosity combined with the effects of time–dilation and have strong implications for quasar samples selected solely on variability.

Introduction

Variability provides a simple yet powerful means for investigating the physical processes at work in the inner regions of AGN. The primary diagnostics for optical variability are: the dependence on absolute magnitude and redshift, the timescale of variations in the quasar rest frame and the degree of coherence of individual quasar light curves — in our case taken as an ensemble. In addition to providing insight into quasar models an important feature of such a study is the ability to predict selection effects for quasar samples chosen purely on the basis of variability (e.g., Hawkins 1986). In this paper we summarise our method and results: a more detailed account is given in Hook et al. (1991) and Hook et al. (1992).

Data

The sample of quasars was taken principally from the catalogue of Hewitt & Burbidge (1989) with additional objects from other surveys.