We use a long (300 ks), continuous Suzaku X-ray observation of the active nucleus in NGC 1365 to investigate the structure of the circumnuclear broad line region (BLR) clouds through their occultation of the X-ray source. The variations of the absorbing column density and of the covering factor indicate that the clouds surrounding the black hole are far from having a spherical geometry (as sometimes assumed), instead they have a strongly elongated and cometary shape, with a dense head (n ∼ 10 11 cm −3 ) and an expanding, dissolving tail. We infer that the cometary tails must be longer than a few times 10 13 cm and their opening angle must be smaller than a few degrees. We suggest that the cometary shape may be a common feature of BLR clouds in general, but which has been difficult to recognize observationally so far. The cometary shape may originate from shocks and hydrodynamical instabilities generated by the supersonic motion of the BLR clouds into the intracloud medium. As a consequence of the mass loss into their tail, we infer that the BLR clouds probably have a lifetime of only a few months, implying that they must be continuously replenished. We also find a large, puzzling discrepancy (two orders of magnitude) between the mass of the BLR inferred from the properties of the absorbing clouds and the mass of the BLR inferred from photoionization models; we discuss the possible solutions to this discrepancy.

We present a new analysis of a 9-d long XMM-Newton monitoring of the narrow-line Seyfert 1 galaxy Mrk 766. We show that the strong changes in the spectral shape, which occurred during this observation, can be interpreted as due to broad-line region clouds crossing the line of sight to the X-ray source. Within the occultation scenario, the spectral and temporal analyses of the eclipses provide precise estimates of the geometrical structure, location and physical properties of the absorbing clouds. In particular, we show that these clouds have cores with column densities of at least a few 10 23 cm −2 and velocities in the plane of the sky of the order of thousands of km s −1 . The three different eclipses monitored by XMM-Newton suggest a broad range in cloud velocities (by a factor of ∼4-5). Moreover, two iron absorption lines clearly associated with each eclipse suggest the presence of highly ionized gas around the obscuring clouds and an outflow component of the velocity spanning from 3000 to 15 000 km s −1 .

The pre-main sequence binary system V773 Tau A shows remarkable flaring activity around periastron passage. Here, we present the observation of such a flare at a wavelength of 3 mm (90 GHz) performed with the Plateau de Bure Interferometer . We examine different possible causes for the energy losses responsible for the e-folding time of 2.31 ± 0.19 h of that flare. We exclude synchrotron, collisional, and inverse Compton losses because they are not consistent with observational constraints, and we propose that the fading of the emission is due to the leakage of electrons themselves at each reflection between the two mirror points of the magnetic structure partially trapping them. The magnetic structure compatible with both our leakage model and previous observations is that of a helmet streamer that, as in the solar case, can occur at the top of the X-ray-emitting, stellar-sized coronal loops of one of the stars. The streamer may extend up to ∼20 R * and interact with the corona of the other star at periastron passage, causing recurring flares. The inferred magnetic field strength at the two mirror points of the helmet streamer is in the range 0.12−125 G, and the corresponding Lorentz factor, γ, of the partially trapped electrons is in the range 20 < γ < 632. We therefore rule out that the emission could be of gyro-synchrotron nature: the derived high Lorentz factor proves that the nature of the emission at 90 GHz from this pre-main binary system is synchrotron radiation.

Aims. Our aim is to show how variable Doppler boosting of an intrinsically variable jet can explain the long-term modulation of 1667 ± 8 days observed in the radio emission of LS I +61• 303. Methods. The physical scenario is that of a conical, magnetized plasma jet having a periodical (P 1 ) increase of relativistic particles, N rel , at a specific orbital phase, as predicted by accretion in the eccentric orbit of LS I +61• 303. Jet precession (P 2 ) changes the angle, η, between jet axis and line of sight, thereby inducing variable Doppler boosting. The problem is defined in spherical geometry, and the optical depth through the precessing jet is calculated by taking into account that the plasma is stratified along the jet axis. The synchrotron emission of such a jet was calculated and we fitted the resulting flux density S model (t) to the observed flux density obtained during a 6.5-year monitoring of LS I +61• 303 by the Green Bank radio interferometer. Results. Our physical model for the system LS I +61• 303 is not only able to reproduce the long-term modulation in the radio emission, but it also reproduces all the other observed characteristics of the radio source, the orbital modulation of the outbursts, their orbital phase shift, and their spectral index properties. Moreover, a correspondence seems to exist between variations in the ejection angle induced by precession and the rapid rotation in position angle observed in VLBA images. Conclusions. The peak of the long-term modulation occurs when the jet electron density is around its maximum and the approaching jet is forming the smallest possible angle with the line of sight. This coincidence of maximum number of emitting particles and maximum Doppler boosting of their emission occurs every ∼1667 days and creates the long-term modulation observed in LS I +61• 303.

Context. The young binary system V773 Tau A exhibits a persistent radio flaring activity that gradually increases from a level of a few mJy at apoastron to more than 100 mJy at periastron. Interbinary collisions between very large (>15 R * ) magnetic structures anchored on the two rotating stars of the system have been proposed to be the origin of these periodic radio flares. Magnetic structures extended over tens of stellar radii, that can also account for the observed fast decay of the radio flares, seem to correspond to the typical solar semi-open quite extended magnetic configurations called helmet streamers. Aims. We aim to find direct observational evidence for the postulated, solar-like, coronal topologies. Methods. We performed seven-consecutive-day VLBI observations at 8.4 GHz using an array consisting of the VLBA and the 100-m Effelsberg telescope. V773 Tau A was phase-referenced to QSO B0400+258. Results. Two distinctive structures appear in the radio images here presented. They happen to be associated with the primary and secondary stars of the V773 Tau A system. In one image (Fig. 2B) the two features are extended up to 18 R * each and are nearly parallel revealing the presence of two interacting helmet streamers. One image (Fig. 2E) taken a few hours after a flare monitored by the 100-m Effelsberg telescope shows one elongated fading structure substantially rotated with respect to those seen in the B run. The same decay scenario is seen in Fig. 2G for the helmet streamer associated with the other star. Conclusions. This is the very first direct evidence revealing that even if the flare origin is magnetic reconnection due to interbinary collision, both stars independently emit in the radio range with structures of their own. These structures are helmet streamers, observed for the first time in stars other than the Sun. The complete extent of each helmet streamer above the stellar surface is about 24 R * which implies that they can practically interact throughout the whole orbit, even rather close to apoastron where the stellar separation is 52 R * . However, the radio flares become stronger when the stars approach. Around periastron the stellar separation is only 30 R * , nearly covered by a single streamer: the two streamers overlap producing the observed giant flares.