H. Pietilä has concluded a study of the evolution of binary black hole systems in the nuclei of galaxies. The problem of the timescale of the evolution of a binary black hole in the center of the galaxy is not simple. The most troublesome points are the wide and skewed distributions of the changes of energy and especially angular momentum in a single encounter of a star and the binary. A definite solution requires a more detailed model of the galaxy core and its evolution.

The work suggests that the interactions of the binary with the stars in the nucleus do not make the binary coalesce, i.e. the binaries can survive over the Hubble time in the nucleus of an isolated galaxy. P. Heinämäki has studied what happens to the binary system when the galaxy is not isolated but merges with other major galaxies. Then three or four black hole systems may form in the centre of the merger. Evolutionary scenarios have been constructed using the cosmological halo merger models of Lacey & Cole. It was found that the frequency of two sided black hole ejections is rather close to the observed frequency of double radio sources, and that the symmetry properties of the ejected pairs can be made to match the symmetry properties of double radio sources if reasonable observational selection biases are assumed.

M.Valtonen, P.Heinämäki, H. Lehto, and A.Chernin studied the basic physical characteristics of chaos in three-body dynamics, in connection with the concept (put forward earlier by Valtonen and coworkers) of multiple supermassive black holes as the most convincing source of energy in quasars and active galactic nuclei. With the use of long time series generated by the model of three black holes confined in the potential well of a galaxy, the phenomenon of intermittence was found in three-body chaos. The shape of the series proved to be similar to one of type-III intermittence known in hydrodynamical flows with transition to turbulence. It was argued that the physical parameters of three-black-hole chaos could be revealed in the gravitational radiation of quasars and AGNs. The space projects of giant low-frequency gravitational antennas planned for the next century (LISA and others) could make possible the detection of gravitational radiation "light curves" from these objects which might be expected to have the same shape as the long time series (above a threshold level of the antenna sensitivity) studied in the models.

The violently variable quasar OJ287 has kept several staff members busy during the period. The expected fade in February 1998 was observed in optical by H. Pietilä et al. but not in radio. The significance of this observation has been studied from the point of view of the binary black hole model (which made the prediction) and by using the Marscher & Gear jet model. The conclusions of this work by M. Valtonen, H. Lehto and H. Pietilä is that the optical and radio emitting regions in the jet are closer than 0.1 pc from the central black hole. The radio emitting region seems to be directly connected with the accretion disk at about 12 Schwarzschild radii from the centre, i.e. the jet is probably magnetically collimated by a magnetic field which is connected to this region of the disk. The viewing angle relative to the jet line appears to be very small, smaller than the jet opening angle. H. Pietilä has calculated a new solution for the binary black hole model using the fade data. It specifies the parameters of the model very accurately, and also makes predictions of the future and past (both known and unknown) outburst events with the accuracy of a few days.

Radio variability of active galactic nuclei

The largest and longest-running astronomical research project in Finland, total flux density monitoring of active galaxies, has been going on since 1980. Using the Metsähovi 13.7 m radio telescope (Kirkkonummi, Finland) and the 15-m SEST telescope (LaSilla, Chile), over 40 000 individual millimeter flux measurements of over 100 active galaxies have been obtained since the beginning of the joint program between the Tuorla Observatory and the Metsähovi Radio Observatory (HUT). The only comparable effort is the University of Michigan monitoring program, which, however, monitors only the cm-flux variations and therefore cannot penetrate as deep into the radio core regions.

The program is led jointly by E. Valtaoja and S. Urpo (HUT), with M. Tornikoski (HUT) responsible for the SEST observing programs. Over twenty senior and junior researchers and students from Tuorla, HUT and Helsinki University have participated in the observations and related investigations. This unique database is also extensively used in international collaborative projects. Over 70 refereed publications, fifteen of them in 1997-1998, and five doctoral theses attest to the vigor of this venerable research program.

E. Valtaoja and M. Lainela, together with A. Lähteenmäki and H.Teräsranta from HUT, have developed a new technique for modelling the total flux density variations in AGN. They have found that the radio variations can be well modelled by a superposition of a small number of exponential flares plus a constant contribution from the underlying unperturbed jet. To a good approximation, each synchrotron flare produced by a new shock can be modelled with just two parameters: the variability timescale and the peak flux. Considering the wide variety of AGN luminosities, the diversity of observed VLBI structures and relativistic jet speeds, this is a rather surprising result, and its physical reasons are as yet unknown.

The flux decomposition technique opens up a wealth of possibilities. Valtaoja, Lähteenmäki and K. Wiik have estimated the Doppler boosting factors for a number of radio sources using their variability brightness temperatures obtained from the exponential flare fits. Comparing these with traditional (sychrotron-self-Compton) boosting factors obtained using VLBI and X-ray data, they have found that the highest intrinsic brightness temperatures in synchrotron sources seem to reach only the equipartition value of approx. 5 × 10¹⁰ K instead the usually assumed Compton catastrophe limit of approx. 10¹² K.This changes significantly all the previous estimates of relativistic effects, jet flow speeds, viewing angles, etc., in compact radio sources. Investigations of the improved Doppler boosting factors and their impact on, e.g., AGN unification-scenarios, are in progress.

Very long baseline interferometry

The Finnish VLBI program is a spin-off of the AGN monitoring program. High resolution interferometry is a natural companion to high frequency flux density observations, since both probe the parsec-scale structure around the supermassive black holes. Turku University has contributed over 1.5 million FIM towards the instrumentation required for VLBI observations with the Metsähovi telescope, and both Institutes continue to participate jointly in the Russian-led RADIOASTRON and the Japanese HALCA (formerly VSOP) space VLBI programs. The difficult financial situation in Russia has forced the postponement of the RADIOASTRON launch. HALCA, however, was launched successfully in February 1997 and continues to produce high-resolution images of radio sources. Several HALCA observing programs have co-investigators from Tuorla and Metsähovi.

E. Valtaoja and K. Wiik (HUT) have continued the analysis of the first Finnish ground based VLBI observations, a three-epoch survey of 15 AGN. As part of a Target-of-opportunity program, Wiik and Valtaoja also observed the source AO 0235+164 with the global CMVA (Coordinated Millimeter VLBI Array) network during a strong flare in 1998. Another mm-VLBI target was the quasar CTA 102 (in collaboration with F. Rantakyr", SEST). The millimeter VLBI-observations can provide angular resolutions exceeding even space VLBI, but the low sensitivity necessitates that the observations are done when the sources are flaring - another justification for the monitoring program.

Plans for the next generation space VLBI mission, ARISE, were finalized in 1998 and presented to NASA towards the end of the year. If the mission is approved, the orbiting radio telescope with a 25-m inflatable mirror will start observing around 2008. E. Valtaoja is one of the international members of the ARISE Science Advisory Group, which drafted the proposal to NASA.

Gamma rays and other high-frequency radiation from AGN

E. Valtaoja, M. Lainela and A. Lähteenmäki (HUT) have continued to investigate the connections between high radio frequency variability and gamma ray flares detected by the EGRET instrument abroad the COMPTON gamma ray satellite. An analysis of common sources in the full Metsähovi and EGRET databases, using the flux decomposition technique, indicates that there is a close connection between the radio and the gamma ray states, with significant amounts of gamma rays detected mainly when the AGN in question is in the initial stages of a strong millimeter flare.The most straightforward explanation is that the gamma rays are first order SSC (synchrotron self-Compton) radiation from shocks forming in the relativistic jets.

The participation in the large international EGRET blazar collaboration has continued to provide complementary radio data for the program sources and targets of opportunity. In general, these multifrequency studies have strengthened the case for SSC origin of gamma rays. Our monitoring has also enabled the identification of several previously unknown gamma ray sources.

Together with graduate students M. Hanski and P. Koivisto, E.Valtaoja is applying the flux decomposition technique to comparisons between radio and optical variations from various monitoring programs, and to comparisons between radio and X-ray data from the ROSAT satellite. Both the optical and the X-ray data are in general too sparse for the usual correlation analysis techniques, but each individual data point can be compared with the contemporary radio flare parameters obtained from the radio flux decompositions.

Optical AGN studies

E. Valtaoja and A. Lähteenmäki have continued their studies of optical hotspots in the radio lobes of active galaxies. Their earlier NOT observations revealed the first known optical double hotspot, in the quasar 3C 351. The single radio and optical hot spots are presumed to be the working surfaces of jets boring through the intergalactic medium, but the formation mechanism of double hot spots is still disputed. Optical synchrotron radiation must be generated by highly relativistic electrons accelerated in situ, and the double hot spot quasar 3C 351 thus offers a laboratory for studying the interactions between the jet and the ambient medium. New, very deep polarized images of the 3C 351 hot spots were obtained with NOT in June, 1997, and their analysis is ongoing.

K. Nilsson, M. Valtonen, H. Lehto together with L.R. Jones (Southampton) and W.C. Saslaw (Virginia and Cambridge) have studied the optical counterparts of hot spots in Cygnus A. They interpret the results in terms of a double-twin-jet model where there are two terminal surfaces of jets in each radio lobe. M. Valtonen has studied the model further and has shown that the properties of large scale jets, including the depolarization correlations, can be very nicely explained.

Valtaoja, Flynn and X. Wan have analyzed the NOT gravitational lens polarization monitoring data. With the good resolution provided by NOT, the optical polarizations of individual lensed components can be measured, and gravitational lenses can in principle be used as 'polarization magnifiers' for studying the magnetic fields of AGN.

Valtaoja and M. Tornikoski (HUT), together with collaborators from the U.K., have obtained NOT images of a complete sample of Compact Steep Spectrum (CSS) sources previously mapped with the MERLIN radio array. The CSS sources with small radio sizes may be either young progenitors of larger radio sources, or alternatively sources permanently confined to subgalactic dimensions by high gas pressures in the parent galaxies. Analysis of the complete optical sample should differentiate between the 'young' and the 'frustrated' source models.

Multifrequency variability campaigns

L.O. Takalo, A. Sillanpää, K. Nilsson together with the students at the observatory have taken part in a large number of multifrequency monitoring campaigns of blazars. Most of these campaigns have involved simultaneous observations by RXTE (Akn120, OJ 287), BeppoSax (Mkn 501) and EGRET (3C 279) satellites. They have used the one meter telescope at Tuorla and also the NOT on La Palma for the optical monitoring of selected blazars monitored also by the satellites. Several blazars (e.g ON 231,Mkn 421, 3C279 and BL Lac) have been observed in outburst during the last two years. On the other hand, both of the OJ-94 project objects,OJ 287 and 3C66A, have declined in brightness since their long outbursts. The OJ-94 project meetings were held at Perugia (1997) and Turku (1998). The monitoring campaigns have enhanced the collaboration with other blazar monitoring groups. Results from some of these monitoring campaigns have been published in a series of articles, mostly in A&A. At the Perugia meeting it was also decided to start a new electronic journal called Blazar Data for publishing observational results on blazars.

Host galaxies

In collaboration with R. Falomo (Padova) and R. Scarpa (STSCI), J. Kotilainen undertook a near-infrared imaging survey of a complete sample of flat spectrum radio quasars (FSRQ) at 0.5 < z < 1.0. The host galaxy was clearly detected for six FSRQs, marginally for six FSRQs, and remained unresolved for eight FSRQs. The galaxies hosting FSRQs are very luminous, much brighter than the typical galaxy luminosity L*, and at least as bright as the brightest cluster galaxies. The luminosities of the FSRQ hosts are intermediate between the hosts of low redshift radio-loud quasars and BL Lac objects, and the hosts of high redshift radio-loud quasars, in good agreement with current unified models for radio-loud AGN, taking into account stellar evolution in the elliptical host galaxies. There is a trend between the host and nuclear luminosity for the most luminous FSRQs, supporting the suggestion based on studies of lower redshift AGN, that there is a minimum host galaxy luminosity which increases linearly with the quasar luminosity.

Together with Falomo and Scarpa, Kotilainen also studied the near-infrared properties of the host galaxies of a sample of BL Lac objects. The host galaxy was clearly detected in all low redshift (z < 0.25) BL Lacs. The galaxies hosting the low redshift BL Lacs are large and luminous, slightly brighter than L*, but slightly fainter than the brightest cluster galaxies. The optical/near-infrared colour and colour gradient of the BL Lac hosts are similar to normal ellipticals, indicating that the nuclear activity has little effect on the star formation history of the hosts. The nucleus-to-galaxy luminosity ratio of the BL Lacs is similar to that of low redshift radio galaxies and consistent with what is found in previous optical studies of BL Lacs. However, it is smaller that that of the FSRQs, suggesting there is a difference in the intrinsic brightness of the nuclear source or in the Doppler beaming factor between the two types of blazars.

L. Takalo, A. Sillanpää, K. Nilsson, T. Pursimo and collaborators (J. Heidt, Heidelberg, Germany, S.A. Laurent- Muehleisen, LLNL, USA and W. Brinkmann, MPI, Germany) have been conducting a project to study the host galaxies and environments of BL Lac objects. During the last few years several studies have shown that at least some BL Lacs belong to small clusters and many objects seem to have close companion galaxies. This has raised the question about the role of the close environment to the BL Lac phenomenon. To answer this question a number BL Lac samples are being studied using the Nordic Optical Telescope (NOT). These include the 1 Jy radio-selected sample, the EMSSX- ray sample and the both radio and X-ray selected RGB (ROSAT - Green Bank) sample. Observing time to study the gamma ray selected EGRET BL Lacs has also been received.

So far more than 100 objects have been imaged at the NOT, mostly in good seeing conditions (FWHM ~1.0 arcsec or better). The main aim of these studies is to separate the host galaxy light from the nuclear light to study the host galaxy properties and the immediate environment of the BL Lac. This is achieved by fitting two-dimensional galaxy + nucleus models to the observed images. Methods to analyze the images are currently being developed and tested at Tuorla Observatory. These methods have already been successfully applied to a small sample of EINSTEIN Slew Survey BL Lacs (by Heidt et al.).

The objects imaged so far seem to lie in very interesting neighborhoods. Most of the objects have close companion galaxies and sometimes clear signs of gravitational interaction is seen. In one case we found a blue arc close to the BL Lac (Fig. 1). Further observations at the NOT showed that this arc is a gravitationally distorted image of a more distant background galaxy, the first of its kind found close to a BL Lac object. Closer to the nucleus many interesting features have been seen. For instance, an optical jet coincident with the radio jet was found in 3C 371 and in BL Lac itself a disk component was discovered.

A composite colour image of RGB1745+398 (the large reddish galaxy left of the center) and the gravitational arc close to it (the bluish arc southeast of the galaxy).

In collaboration with Yu. Baryshev (St.Petersburg), V. Sokolov (SAO), and S. Zharikov (SAO), M. Hanski, K. Nilsson, and P. Teerikorpi have investigated the nature of the host galaxy of the gamma ray burst GRB970508. First optically identified at the NOT, the optical transient of GRB970508 was photometrically measured by the 6-m telescope in BVR_cI_c bands several times during the year following the outburts in 1997-98. These observations allowed one to detect the contribution that the faint host galaxy gave to the photometry and derive its B- magnitude and broad-band colours, and also the K_B correction (at z = 0.835, the redshift measured by Keck). The K-corrected B-absolute magnitude of the host is about -18.5 mag (for H₀ = 60 km/s/Mpc), 2.5 mag fainter than the knee in the local general Schechter luminosity function.

The broad-band spectrum of the host is incompatible with the spectra of normal local galaxies earlier than Sbc. It fits best with a Scd starburst galaxy and a blue compact galaxy. The host's position in the absolute magnitude vs. size diagram indicates that its surface brightness is significantly higher than is typical for local late-type galaxies from the LEDA extragalactic database.

In another deep universe collaboration with Russian astronomers, going on for a few years, the RATAN radio galaxies have been studied by T. Pursimo, K. Nilsson, and P. Teerikorpi together with Yu. Baryshev (St.Petersburg) and the RATAN group of Yu. Parijskij (St.Petersburg). In 1998 the analysis of the NOT subarcsecond observations of about 20 radio galaxies was completed. These morphological and colour observations form a part of the effort to create a complete sample of faint, very distant radio galaxies around the narrow RATAN strip in the sky.

The linear Hubble redshift-distance law and the fractal density-distance law with D ~ 2 in same spatial scales. The displayed Hubble law is based (in the order of increasing distance) to galaxies with Cepheid distances, galaxies with Tully-Fisher distances, and to galaxies with distances from Supernovae of Ia type. The fractal law is based on the gamma function analysis of several redshift surveys (CfA1, Perseus- Pisces, LCRS, ESP, LEDA). This surprising co-existence of two cosmological laws was studied by Baryshev et al. (1998).

Seyfert galaxies

J. Kotilainen presented optical B-I continuum colour maps of a large sample of Seyfert 2 galaxies. Extended blue continuum components were detected in the circumnuclear region of several galaxies, either elongated or forming a double structure across the nucleus. They are closely aligned with the radio and emission line axes of these galaxies and probably arise from scattering of nuclear continuum light by extranuclear mirrors. The colours of the blue maxima are consistent with those expected from scattering off dust or electrons. These findings thus strongly support the current unified models of AGN.

BL Lac Phenomenon meeting

During June 22-26, 1998 L.O. Takalo and A. Sillanpää organized an international meeting in Turku on "BL Lac Phenomenon". This meeting was the result of some discussions about two years earlier, during which they were asked to organize such a meeting. The meeting was the third in a series of BL Lac meetings, previous ones being Pittsburgh 1978 and Como 1988. It was attended by 120 astronomers from over 20 countries, to discuss our current knowledge and the future of BL Lac research. Meeting presentations will be published in the PASP conference series.