Critical progress in our understanding of high energy emission from AGN has been determined in the last 10 years by X-ray monitoring campaigns with many space missions, notably ROSAT, ASCA, RXTE, BeppoSAX, and XMM, often in conjunction with observations at other frequencies. The emphasis of the present review is on recent findings about X-ray variability of blazars. Among AGN, these exhibit the largest amplitude variations of the X-ray emission, often well correlated with variations at higher energies (GeV and TeV radiation). The accurate sampling of the X-ray spectra over more than three decades in energy, made possible by the wide energy range of BeppoSAX, has also shown strong spectral variability in blazar active states, suggesting extreme electron energies and leading to the identification of a class of ‘extreme synchrotron’ sources.

This paper presents the results of a multiwavelength observational study of the active young F-type star HR 1817. The star was monitored at 4.80 and 8.64 GHz over 2 × 12 h allocations with the Australia Telescope Compact Array on 8 and 9 December, 2000. The Anglo-Australian Telescope was used for simultaneous optical spectropolarimetry during a 2 h period on 9 December.

The low levels of observed radio emission have characteristics that are similar to those seen in other active stars, and a gyrosynchrotron mechanism is proposed to explain them; this is supported by the relatively low fractions of circular polarisation measured in HR 1817.

Comparison of the emissions from 4.80 and 8.64 GHz shows a very strong cross-correlation peak, indicative of a common origin, although the shift of this peak indicates that 8.64 GHz variations tend to precede those at 4.80 GHz by, typically, ˜20 min.

The optical spectropolarimetry reveals polarisation signals characteristic of surface magnetic fields, with profile changes indicating a complex dynamo-type magnetic topology is present on the star. This result makes HR 1817 the star with the earliest spectral type on which dynamo magnetic fields have been detected directly up to now.

Thermal instabilities can cause a radiative shock to oscillate, thereby modulating the emission from the post-shock region. The mode frequencies are approximately quantised in analogy to those of a vibrating pipe. The stability properties depend on the cooling processes, the electron–ion energy exchange, and the boundary conditions. This paper considers the effects of the lower boundary condition on the post-shock flow, both ideally and for some specific physical models. Specific cases include constant perturbed velocity, pressure, density, flow rate, or temperature at the lower boundary, and the situation with nonzero stationary flow velocity at the lower boundary. It is found that for cases with zero terminal stationary velocity, the stability properties are insensitive to the perturbed hydrodynamic variables at the lower boundary. The luminosity responses are generally dependent on the lower boundary condition.

Radio variability on timescales from a few hours to several days in extragalactic flat spectrum radio sources is generally classified as intraday variability (IDV). The origin of this short term variability is still controversial and both extrinsic and intrinsic mechanisms must be considered and may both contribute to the observed variations. The measured linear and circular polarisation of IDV sources constrains the low energy end of the electron population. Any population of cold electrons within sources at or above the equipartition temperature of 1011 K depolarises the emission and can be ruled out. Intrinsic shock models are shown to either violate the large fraction of sources displaying IDV or they do not relax the light travel time argument for intrinsic variations. From structure function analysis, we further conclude that interstellar scintillation also leads to tight size estimates unless a very local cloud in the ISM is responsible for IDV.

Synchrotron self-Compton (SSC), external Compton, and hadronic models of blazar emission all invoke particle acceleration at relativistic shocks as the dissipation mechanism seeding their non-thermal X-ray and gamma-ray emission. Studies of diffusive acceleration at such relativistic shocks are more sparse than those pertaining to their non-relativistic counterparts. This paper presents acceleration time results from the theory of relativistic shock acceleration that are pertinent to AGN observations. This temporal information interfaces critically with the observed rapid variability of blazars. Very recent theoretical results are presented, where it is determined that acceleration times can never become arbitrarily short in relativistic shocks, but are dominated by diffusion in the downstream region and couple to the particle's gyroperiod. This fundamental bound links to the variability timescale to generate a firm lower bound to the environmental magnetic field of blazars such as Mrk 421. Consistency of such a bound with SSC spectral models and flare decay times is discussed.

There is new evidence which identifies seasonal changes of the variability time scale in intraday variable (IDV) sources with refractive interstellar scintillation effects. Such a RISS model takes the annual change of the Earth's velocity relative to the scattering medium into account. In September 1998 we found a remarkable prolongation of the variability time scale in the IDV source 0917+624 with only small variations in flux density during a period of 5 days. This was explained as a seasonal effect, in which the velocity vector of the Earth and the interstellar medium nearly cancelled. In order to further investigate the applicability of the model for 0917+624, we performed an Effelsberg 6 cm-flux monitoring program over the course of one year. Since September 2000, the source appears to be remarkably inactive and yet (May 2001) no return to its normal, faster and stronger variability pattern is observed. Here, our observational results and a possible explanation for the current quiescence are presented.

We present the kpc-scale behaviour of the powerful extragalactic radio source Hercules A and the behaviour of the intracluster gas in which the radio source is situated. We have found that Hercules A exhibits a strong Laing-Garrington effect. The X-ray observations have revealed an extended X-ray emission elongated along the radio galaxy axis. The estimated temperature of the cluster is kT=2.45 keV and the central electron density is n○≃7.8×10−3 cm−3 which reveals a hot, dense environment in which Hercules A is situated. From the combined study of the radio and X-ray data we have estimated a central value of 3 ≲ B○(μG)≲9.

We also present the most recent results from the analysis of the radio data on the pc-scale structure of the radio galaxy, observed at 18 cm by the EVN–MERLIN array. A faint but compact radio source, coincident with the optical centre of Hercules A, was detected by the EVN at 18 mas resolution. The total flux density of the EVN core is 14.6 mJy. Its angular size is 18×7 mas with a position angle of ≃139°. There is also evidence for extended emission in the NW–SE direction, most probably from the eastern pc-scale jet. If this is true then there is a misalignment between the direction of the pc-eastern and the aligned kpc-scale jets of ≃35°.

III Zw2 was observed with XMM-Newton in July 2000. Its X-ray spectrum can be described by a power law of photon index Γ≈1.7 with a Gaussian line at 6.7 KeV. There is no significant evidence of intrinsic absorption within the source or of a soft X-ray excess. Multi-wavelength light curves over a period of 25 years show related variations from the radio to X-rays. We interpret the radio to optical emission as synchrotron radiation, self-absorbed in the radio/millimetre region, and the X-rays as mainly due to Compton up-scattering of low energy photons by the population of high energy electrons that give rise to the synchrotron radiation.

We discuss the observational and theoretical constraints on the brightness temperature of compact opaque synchrotron sources. We consider the implication of observed apparent velocities on the amount of Doppler boosting and compare this with values deduced from directly measured brightness temperatures from ground and space based VLBI observations, as well as the implications of intraday variability. We also discuss the maximum rest frame brightness temperature expected under conditions of both inverse Compton cooling and equipartition conditions.

Results are presented of long-term monitoring at the radio telescope RATAN–600 and a model interpretation of instantaneous 1–22 GHz spectra at six frequencies for a sample of 550 compact extragalactic radio sources. The sources are selected from the Preston et al. (1985) VLBI survey and are located in the declination range from −30° to +43°. Dependence of the index of variability on frequency is analysed for different subsamples. Factors of flux density variations range from 1 to 15. A typical behaviour of spectra evolution during strong radio outbursts in various objects is recognised suggesting a common physical nature of the variability. Invoking VLBI observations in spectra–structure analysis, we find that usually more than 70% of the total emission is coming from the mas scale at frequencies higher than 5 GHz. The observed variability can be explained in the model with a relativistic jet of parsec scale in the longitudinal magnetic field or in the shock-in-jet model.

Powerful water masers have been detected in 23 active galactic nuclei (AGN) to date. Most of the sources that have been monitored show evidence for variability, which may be either intrinsic (response of the masers to a time-varying pump or background source) or extrinsic (the effects of scintillation in the interstellar medium of the Galaxy). I briefly review the examples in which these mechanisms may be operating (interstellar scintillation in the Circinus galaxy, a time-varying background source in Mrk 348, and a time-varying AGN luminosity in NGC 1068).

Microquasars are Galactic X-ray binaries which show radio emission from relativistic outflows. Here we argue that the origins of radio emission from microquasars and compact extragalactic radio sources are similar. In particular, both can be explained qualitatively by a model in which shocks are propagating in the relativistic jets. The microquasar GRO J1655–40 is used to illustrate that the spectral evolution of its 1994 radio outburst can be explained by the growth phase of the generalised-shock model, similar to the case of the quasar 3C 273.

The evidence in support of interstellar scintillation as the predominant cause of intraday variations at centimetre wavelengths is reviewed. A new evaluation of the highest brightness temperature implied for AGN is found to be consistent with the Doppler factors in typical jet models, but requires the scattering to occur within about 30 pc of the Earth.

During its burst of intraday variation (IDV) in June 1996 quasar PKS 0405–385 exhibited rapid changes in the Stokes parameters Q and U that were even faster than the 1 hr timescale changes in I. Auto- and cross-correlation analyses of I,Q, and U at 8.6 GHz show systematic relations between the variations in I,Q, and U. These are well modelled by interstellar scintillation (ISS) of a 14 × 20 μ as source, with about 180 degree rotation of the polarisation angle along its long dimension. This success in explaining the remarkable IDV in polarisation confirms that ISS gives rise to the IDV in this quasar.

We present the relation between optical luminosity and spectral slope for eight BL Lac objects (3C66A, PKS 0422+00, S5 0716+71, OJ 287, ON 231, OQ 530, S5 1803+78 and BL Lacertae), derived from B, V, R, and I observations spanning a time interval of about 5 years. Four objects show a marked correlation between spectral slope and luminosity, being bluer when brighter, while for the other four the correlation is weaker or absent. Possible explanations are briefly discussed in the framework of current models of the BL Lac phenomenon.

We present results of VLBI monitoring observations of the blazars detected by ground-based TeV gamma-ray telescopes. Preliminary results have shown that the apparent speeds of the parsec-scale jet features in these sources are predominantly subluminal, and significantly slower than those in the GeV-peaked (EGRET) gamma-ray blazars. Here we add results from additional sources and additional epochs that increase the statistical significance of this comparison. These slower speeds may indicate rapid deceleration of the jets in TeV blazars between the TeV gamma-ray emitting scales and VLBI scales.

We compare the total flux density variations and the VLBI structural variations in a sample of 27 gamma-ray blazars. We find that all the radio variations are due to shocks; the flux of the underlying jet remains constant. A large fraction of the shocks grow and fade within the innermost 0.1 mas, appearing only as ‘core flares’. Comparisons with the EGRET data show that gamma-ray flares must come from the shocks, not from the jet. At the time of an EGRET flare, the shock is typically already over a parsec downstream from the radio core, beyond the accretion disk and/or the broad line region (BLR) photon fields. Thus, present models for gamma-ray production are inadequate, since they typically model the gamma-ray inverse Compton flux as coming from the jet, with significant disk or BLR external Compton components.

The effect of radiation drag on relativistic bulk flows is re-examined. Highly relativistic bulk flows in the nuclear region are subject to Compton drag, i.e. radiation deceleration as a result of inverse Compton scattering of ambient soft photon fields from emission from the accretion disk, broad line region, or dusty torus. Possible observational consequences of X-/γ-ray emission produced from Compton drag are specifically discussed.

This paper discusses the boundary layer and the emission spectrum from an accretion disk having a jet anchored at its inner radius, close to the black hole. We summarise our earlier work and apply it to the accretion disks of some blazars. We suggest that the ‘accretion disk with jet’ (ADJ) model could make the bridge between standard accretion disk models (suitable for quasars and FRii sources) and low-power advection dominated accretion disk models (suitable for some of the low-power BL Lacs and FRi sources).

The jet is collimated within a very narrow region close to the black hole (nozzle). In our model it is assumed that the boundary layer of the disk is the region between radius Rms — the last marginally stable circular orbit calculated for a Kerr geometry — and the radius Rjet, which gives the thickness of the ‘footring’, i.e. the base of the jet. We analyse the size of the boundary layer of the disk where the jet is fed with energy, mass, and angular momentum. As a consequence of the angular momentum extraction, the accretion disk beyond Rjet no longer has a Keplerian flow. A hot corona usually surrounds the disk, and entrainment of the corona along the flow could also be important for the energy and mass budget of the jet.

We assume that the gravitational energy available at the footring of the jet goes into the jet, and so the spectrum from the accretion disk gives a total luminosity smaller than that of a ‘standard’ accretion disk, and our ADJ model should apply for blazars with low central luminosities. Variations of the boundary layer and nozzle may account for some of the variability observed in active galactic nuclei.

We consider the shock structures that can arise in blazar jets as a consequence of variations in the jet flow velocity. There are two possible cases: (1) a double shock system consisting of both a forward and reverse shock, and (2) a single shock (either forward or reverse) together with a rarefaction wave. These possibilities depend upon the relative velocity of the two different sections of jet. Using previously calculated spherical models for estimates of the magnetic field and electron number density of the emission region in the TeV blazar Mkn 501, we show that this region is in the form of a thin disk in the plasma rest frame. It is possible to reconcile spectral and pair opacity constraints for Mkn 501 for Doppler factors in the range of 10–20. This is easiest if the corrections for TeV absorption by the infrared background are not as large as implied by recent models.