Annotation of the results obtained in 2018
The scientific group continues experimental study of the theoretically predicted effect of the apparent radio core position shift in active galaxies as a function of observing wavelength. It is explained by frequency-dependent synchrotron self-absorbtion of radio emission in their jets. This effect is determined by density of emitting particles of non-thermal relativistic plasma, magnetic field strength, and geometry of a collimated outflow. Therefore, measuring the effect provides one of the most powerful tools to probe the extreme physical conditions in relativistic jets in the immediate vicinity of supermassive black holes. We solve the problem using a wide range of different approaches: - multi-frequency VLBI observations of selected sources during bright flares, - dual-frequency global VLBI observations of a large sample of sources to study the effect statistically, - investigation of variability of the core shift effect using multi-epoch long-term VLBI data for a number of sources, - measuring the effect by a method of relative astrometry using triplets of nearby sources on the sky, - measuring the effect by a method of time delays of emission at different frequencies from light curves, - a mass estimate of the effect from a comparison of the high-precision coordinates of the galactic nuclei measured in the radio and optical ranges of the electromagnetic spectrum, - taking into account the results of our analysis for the accuracy of measurements, given the inevitable random and possible system components of uncertainty. Using the methods of machine learning, the possibility of a more accurate estimation of the core shift effect was shown. Namely, the systematics arising from the use of simplified models in the analysis of observational data can be compensated if it is known how the parameters of these models depend on the parameters of the model of inhomogeneous jet, which describes the core shift effect. We simulated VLBI observations of a sample of artificial sources using a inhomogeneous emission model and evaluated this dependence using machine learning methods. For the first time, the evaluation of the physical parameters of emissions from AGN jets was done by fitting the physically based model of inhomogeneous emission directly to VLBI data. In fact, this marks a new stage in the development of methods for studying AGNs outflows. The development of VLBI methods and computational statistics now makes it possible to directly compare theoretical predictions and observational data obtained with ultra-high angular resolution. Measuring the frequency-dependent shift of the core allows us to investigate a number of important physical parameters of a relativistic jet, such as a magnetic field strength, density of emitting particles, jet shape, and even the bulk plasma speed if one can measure time delays of the emission at different frequencies or the evolution of the turnover frequency of the synchrotron spectrum of the VLBI core. Using these methods, we measured the jet speed in a number of sources. This is especially important for the objects having either very compact structure or showing a morphology that consists of several standing shocks, since in both these cases the standard kinematic analysis based on changes in the coordinates of such quasi-stationary components does not reflect the real velocity of the jet. In addition to a frequency-dependent shift of the core, we also found that there is an effect of the frequency-independent shift of the position of the brightest and most compact component of the jet — its apparent origin. The accretion onto the center of the source activity — a supermassive black hole — is non-constant. It results in a changing density of relativistic electrons being injected in the jet. This in turn affects the synchrotron self-absorbtion properties of plasma and leads to the variability of the position of the core as an emission region with an optical thickness of about 1, like a photosphere. In the framework of the project, we also used one of the most elegant methods for measuring the core shift (it was historically the first) - the phase-referencing method. It is based on alternate observations of a target and a bright nearby source acting as a phase calibrator and defining a reference point on the sky. We applied this unique technique to study the quasar 0850+581, which, as we discovered earlier, showed the largest corer shift among other AGNs. Since the position of the phase calibrator is also subject to the effect of a core shift, to improve the accuracy of measurements we have proposed and successfully implemented a modified phase-referencing method based on the observations of two nearby reference sources, making together with a target object a close triplet of sources, in which they are connected by one phase solution. We initiated observations on the European VLBI network including three Russian telescopes of the Kvazar-KVO system at four frequency bands for 8 such triplets of compact sources. We developed a method to measure the corresponding core shifts and examined their frequency dependencies. We found significant offsets between radio (VLBI) and optical (Gaia) position of active galaxies. It turned out that this shift is predominantly parallel to the directions of their jets. The shift from the radio to the optics in the direction of the central machine was found at about or less than 2 mas and is explained by synchrotron self-absorption in the core and an extended structure in the radio band. Moreover, color analysis showed that in this case the accretion disk dominates in optical radiation. In this case, the radio->optics shift has a value from 0 to more than 10 mas. We explain this by the presence of bright extended optical jets on parsec scales. Thus, these results can be considered the first massive observational indication of the existence of bright, extended optical jets in active galaxies. These results should be taken into account in order to increase the accuracy of alignment of inertial reference frames constructed in the radio and optical bands. The analysis of the shifts and colors for different types of active galaxies (quasars, BL Lacertae objects, Seyfert galaxies) agrees remarkably well with the predictions of the unified scheme. This effect opens up new unique opportunities for massive restoration and the study of the properties of the central regions of quasars (accretion disk and jet) on parsec scales.

 

Publications

---

Annotation of the results obtained in 2016
We have started experimental study of the theoretically predicted effect of the apparent radio core position shift in active galaxies as a function of observing wavelength. It is explained by frequency-dependent synchrotron self-absorbtion of radio emission in their jets. This effect is determined by density of emitting particles of non-thermal plasma, magnetic field strength, and geometry of a collimated outflow. Therefore, measuring the effect provides a powerful tool to probe the extreme physical conditions in relativistic jets in the immediate vicinity of supermassive black holes. We attack the problem using different approaches: - multi-frequency VLBI observations of selected sources during bright flares, - dual-frequency global VLBI observations of a large sample of sources to study the effect statistically, - investigation of variability of the core shift effect using multi-epoch VLBI data for a number of sources, - measuring the effect by a method of relative astrometry using triplets of nearby sources on the sky, - measuring the effect by a method of time delays of emission at different frequencies from light curves for ultra-compact sources, - analysis of the accuracy of the core shift measurements, taking into account the inevitable random and possible systematic components of the uncertainty. Changing the position and size of the apparent jet base (VLBI core) with observing frequency, as predicted by the Blandford-Koenigl jet model, has been brilliantly confirmed by experiment. The model was originally designed to explain the flat spectra and fast flux density variability observed in the compact extragalactic radio sources. The basic idea ot the Blanford-Koenigl model is that the observed bright region at the base of the jet (VLBI core) is the photosphere, i.e. the area where the jet becomes transparent to its own synchrotron radiation at a frequency of observations. An alternative explanation of the nature of the radio core as a standing shock wave does not predict its position depending on the frequency of observation, while the Blanford-Koenigl model naturally explains different positions of photosphere at different frequencies due to a gradient of the magnetic field and particle density along the gradually expanding jet. The shift position measurements and change the size of the VLBI core with frequency allow, within the model, to constrain the magnetic field configuration and density of the emitting particles in the jet. In addition, analyzing the VLBI core size and its separation from the true jet apex one may derive the form of the jet (conical/parabolic/hyperbolic) in the area of the observed jet regions. Measurements of the frequency-dependent core shifts allow to make assessments of the fundamental parameters of a jet, such as the magnetic field strength and distance from the apparent jet base to its true origin. According to the analysis of the multi-frequency VLBA observations of the quasar 3C273, its jet can be described by the Blandford-Koenigl model, with the separation of VLBI core from the true jet base being inversely proportional to frequency. The magnetic field strength measured at the VLBI core does not change from observation to observation and is about 0.3 G. The VLBI core undergoes a powerful flare. Assuming that the magnetic field is constant the flarecan be explained by an increase in particle density of 1000 times. This is consistent with the detected movement of the VLBI core during the flare. Increased density of the particles causes not only the flare, but also to an increase in optical thickness. The emission can become transparent if particle density decreases due to jet expansion. Therefore, it occurs downstream the jet in accordance with the detected core shift. The magnitude of this effect in 3C273 is about 4 pc. We have analyzed the multifrequency VLBA observations of the quasar S4 1030+61 at frequencies from 5 to 43 GHz in several epochs. We estimated the magnetic field (2 G) and plasma density (100 cm^-3) in the jet at a distance of one parsec from the supermassive black hole. These quantities decrease with distance from the jet vertex in accordance with expansion of a conical shaped jet. The frequency-dependent shift of VLBI core position can be measured by combining achromatic regions of a source at different frequencies and taking into account the difference between the core position in relation to them. The image registration has been performed using a two-dimensional cross-correlation method, while the core position has been derived by model fitting the source brightness distribution. We have developed a fully automatic method for core shift measurements, which can be applied to large data sets. The method displays results at the level of the best methods that require expert involvement. We have studied variability of the core shift effect and for the first time established its connection with bright nuclear flares. The core shift effect should be taken into account in high-precision astrometric studies. However, the method of measuring the frequency-dependent shift of the core relative to the optically thin source structure cannot be applied for high-compact objects, e.g. ICRF sources. In this case, a method based on relative astrometry makes it possible to detect and measure the effect. In 2008, VLBI observations of 8 compact radio sources from the ICRF2 catalog were observed with the European VLBI network with participation of the Russian telescope system Kvazar-KVO at four wavelengths. Each object was observed together with two nearby sources on the sky, which were used as the phase calibrators. Based on these observations, we measured the frequency-dependent core shifts for the source-calibrator pairs. For those pairs of sources, for which we were able to define jet position angle, the relative core shift was decomposed in the jet directions. This way we obtained an assessment of the core shift effect for each source separately. Bright flares of radio emission in blazars at low frequencies appear later than on high frequencies due to the effect of synchrotron self-absorption in relativistic jets. The time delays of the flares thus reflect actual physical conditions in jets: magnetic field strength and particle density. We have analyzed light curves (flux density variations) of five blazars: 0235+164, 0716+714, 0851+202, 1633+382, and 1730-130 obtained at five radio frequencies (from 5 to 37 GHz) in the University of Michigan, Metsähovi, and Owens Valley observatories. We derived time delays of flares at different frequencies. This will allow us to estimate apparent jet speeds in the VLBI core region (a few parsecs from the supermassive black hole) by a new proposed method. Uncertainty ("error") of the core shift effect consists of several components. To determine the accuracy of our core shift measurements and, if possible, to increase this accuracy we have done the following. First, it was implemented a tool to assess the accuracy of the random component of the overall uncertainty associated with the thermal noise in the observed VLBI data. This is not a trivial task due to specifics of VLBI observations and the complexity and/or non-linear data reduction algorithms. We used the statistical bootstrap and have shown that the resulting uncertainty estimates at least for VLBI images are statistically more optimal than traditional that produced on the basis of statistics of restored VLBI images. Using this tool, we found that some systematic components dominate in the total uncertainty of the effect. To reduce them, we have proposed methods that allow, as shown by the analysis of the characteristic of real data and simulations, to virtually eliminate the contribution of these effects in the resulting core shift error. Finally, images simulations of physically-plausible models and processing of the corresponding artificially created VLBI data indicate a non-significant taxonomy associated with the model approximations used to analyze the structure of the central regions of quasars. We got a bit lucky. An international team of the space telescope Gaia has released the first version of the catalog of highly accurate optical positions of quasars already in 2016. We have quickly performed an analysis and published two Letter-type papers of the early results of radio-to-optical band comparison of astrometric positions. We have found a sample of about 400 active galaxies with significant non-zero radio-optical offsets. We have found that there is a statistically significant excess of sources with radio to optical position offset in directions along and opposite to the jet. Offsets along the jet vary from zero to tens of mas. Offsets in the opposite direction do not exceed 3 mas. The presence of strong extended parsec-scale optical jet structure in many AGNs is required to explain all observed VLBI-Gaia offsets along the jet direction. The offsets in the opposite direction shorter 1 mas can be explained either by a non-point-like VLBI jet structure or the "core-shift" effect due to synchrotron opacity. This result makes our RSF-supported project even more timely and important not only for astrophysical but also for astrometric applications.

 

Publications

---

Annotation of the results obtained in 2017
The scientific group continues observational study of the theoretically predicted effect of the apparent radio core positional shift in active galaxies as a function of observing wavelength. It is explained by frequency-dependent synchrotron self-absorption of radio emission in their jets. This effect depends on the density of emitting particles of non-thermal plasma, magnetic field strength, and geometry of a collimated outflow. Therefore, measuring the shift provides a powerful tool to probe the extreme physical conditions in relativistic jets in the immediate vicinity of supermassive black holes. We attack the problem using different approaches: - multi-frequency VLBI observations of selected sources during bright flares, - dual-frequency global VLBI observations of a large sample of sources to study the effect statistically, - investigation of variability of the core shift effect using multi-epoch VLBI data for a number of sources, - measuring the effect by a method of relative astrometry using triplets of nearby sources on the sky, - measuring the effect by a method of time delays of emission at different frequencies from light curves, - an estimate of the effect from a comparison of the high-precision coordinates for a huge sample of galactic nuclei measured in the radio and optical bands of the electromagnetic spectrum, - taking into account the results of our analysis for the accuracy of measurements, given the inevitable random and possible system components of uncertainty. The question of high-precision and correct measurements has always been and remains a non-trivial task and measuring the core shift is no exception. This is a rather subtle effect, but the level of technology of the VLBI experiment nowadays is so high that it allows detecting such a shift. The measurement always has an uncertainty, which can be divided into random (calculated from the measurements) and systematic (hidden). For this reason, additional analysis is required to investigate the presence of a systematic error, to determine its magnitude and to account for it. Making use of modern statistical methods and computer simulations, we performed such an analysis and found systematics in the core shift assessments if simple models of a source morphology are used, and it can be significant and thus must be taken into account to obtain correct estimates of the key physical parameters of a jet, such as the magnitude of the magnetic field and the density of the emitting particles. One of the most elegant methods applied by us for measuring the core shift effect is based on long-term observations in a single-dish mode and does not require the participation of a system of aperture synthesis to achieve ultra-high angular resolution. Moreover, the method works equally well both for sources with rich morphology, and for ultra-compact objects, where most other methods are powerless. We conducted a study of the well-known "point-like" blazar 0235+164 on the basis of light curves obtained at five different observation frequencies and determined the time lags of bright flares as well as the corresponding values of the position shift of the radio core. For two dozens of quasars, the observational data was so rich that allowed us to study for the first time the variability of the core shift effect. Most of these objects, which had up to 70 epochs of observations on a time scale of up to 20 years, showed a significant variability in the frequency-dependent shift of the position of the core with time, with an amplitude comparable to the average value of the effect itself - up to about 1 milliarcsecond between 2 and 8 GHz. This allows us to investigate the nature of the source activity, as well as to trace flares and their propagation downstream the outflow. This has also important practical implications in the construction of a high-precision rest frame and other astrometric problems. As a result of comparison of the high-precision positions of the nuclei of active galaxies, according to measurements of very long baseline interferometer (VLBI) and the Gaia optical space telescope, it was found that 7% of objects have a significant difference between their radio and optical positions. It turned out that this shift is predominantly co-aligned with jets. The shift in the direction of the central engine is detected at a level about 1 milliarcsecond and results from synchrotron self-absorption of the core and non-point radio structure. The shift in the direction of the jet ranges from 0 to more than 10 milliarcseconds and is explained by the presence of bright extended optical jets on parsec scales. In fact, this is the first massive detection of optical jets in quasars which was possible due to the difference in observational techniques of VLBI and Gaia. The extended structure of jets affects the coordinates in the optical band more efficiently than in radio (VLBI method). Gaia detects total radiation power using CCD cameras while VLBI measures the correlated quantity and, therefore, is most sensitive to compact structures of an object. This effect must be taken into account for the high-precision alignment of celectial rest frames constructed in the radio and optical bands. In addition, it opens a new exciting window of opportunity for massive studies of central regions in quasars at parsec scales - accretion disk and jet. This can be achieved by modeling data on variability of the object positions in radio (VLBI) and optical (Gaia) domains as well as the variable optical luminosity and color.

 

Publications

---