A neural network forecast of substorms caused by the impact of solar wind plasma flows on the Earth’s magnetosphere has been performed. For this, recurrent neural network models were created based on physical cause-and-effect relationships of the dynamics of high-latitude geomagnetic activity (according to the AL index) with the parameters of the interplanetary magnetic field (IMF) and solar wind plasma (SWP). Two parameters are used as input sequences: the bz-component of the IMF and the integral parameter Σ[NV 2], taking into account the prehistory of the process of pumping the kinetic energy of the solar wind into the magnetosphere, where N and V are the plasma density and solar wind velocity, respectively. The forecast of the AL index according to SWP and IMF for 10 min, etc. with 10 min discreteness individually by an individual artificial neural network (ANN) for each point corresponding to the dynamics of the AL index was completed. This means that the prediction of a continuous series of values AL index is achieved by a parallel running of the ANN package. The number of ANNs in the package is determined by the duty cycle of the required predictive series of the AL index, while taking 90 min of the history of input parameters in each of the networks into account provides a prediction of the values AL index with an accuracy of 80

Results are presented from comparing an analytical study of the radial distribution of magnetohydrodynamic disturbances of magnetic field components in the body of a model solar wind magnetic cloud to data on a real cloud recorded on a spacecraft. An analytical–numerical solution for disturbances was found using a cylindrical coordinate system for a cloud represented by a force-free cylindrical tube with a helical magnetic field. The general idea of the radial distribution of the magnetic field components in the body of the magnetic cloud is consistent with the registered parameters of magnetic field disturbances on patrolling spacecraft, as is confirmed by an objective correlation analysis. Joint post-processing of the count results and real data enhances the consistency, testifying to the adequacy of the magnetohydrodynamic approach to analyzing oscillations in the model body of the magnetic cloud.

The research of indices connection describing symmetrical (SYM) and asymmetric (ASYM) parts of a magnetic disturbance, created by a ring current, with the indices of auroral electrojet activity (AE, AU, AL) is carried out during the main phases and recovery phases for separate magnetic storms in period with 2000 on 2001. At the transition from the main phase of a magnetic storm to the recovery phase the signs of correlation of indices AU, AL with an index SYM changes. On the level of correlation it is possible to analyze the role of influence of cur-rent systems, arising during physical processes, causing asymmetry of a ring current. At the absence of temporary delay the highest anticorrelation between indices ASYM and AL exists, as the analysis displays, for temporary sectors «evening», «evening–night», «night», «night–morning» and «morning». So we can assume that the current system of the asymmetric part of ring current presents in these sectors and its direction is opposite to the direction of the westward electrojets current system. Correlation between ASYM and AU indices is positive and the strongest for “morning-day” and “day” sectors. So the asymmetric part of ring current is present in these sectors and its direction is the same as the eastward auroral electrojet direction (AU). For sectors “night” and “night-morning”, the correlation of these data is negative, so the corresponding current systems are counter directed.

Images are created of the causal relationship between substorm activity and the characteristics of such large-scale solar fluxes as magnetic clouds interacting with the Earth’s magnetosphere. An artificial neural network of the Kohonen layer type is used to classify these images. The results from classification identify selected classes of substorms with perturbations in parameters of the solar wind and interplanetary magnetic field that are typical of structural elements of magnetic clouds.

В исследовании рассмотрены особенности генерации высокоширотных геомагнитных возмущений, вызываемых оболочками магнитных облаков (МО) солнечного ветра представляющих собой горячую и плотную турбулентную плазму с сильными флуктуациями компонент межпланетного магнитного поля. Для этих целей выполнены нейросетевые классификационные эксперименты по сопоставлению динамики параметров оболочек магнитных облаков с динамикой аврорального AL-индекса еще до развития глобального геомагнитного возмущения. Результаты, полученные искусственным интеллектом, согласуются с физическими представлениями о процессах развития высокоширотной геомагнитной активности под действием турбулентной среды оболочек магнитных облаков на земную магнитосферу.

Simultaneous disturbances of the total electron content and the horizontal component of geomagnetic field in the range of Pc6 are registered during periods of magnetospheric substorms. Such disturbances arise in intervals of interaction between Earth’s magnetosphere and the turbulent sheath of magnetic clouds. Wavelet analysis of disturbances in the interplanetary magnetic field and the velocity and density of solar wind plasma testify to the existence of oscillations with similar periods in the interplanetary medium.

An analysis is performed of the statistical relationship between coronal mass ejections and solar flares to establish the sequence of these events. Periods of delay in the occurrence of coronal mass ejections relative to the associated flare-like manifestations of solar activity are determined. It is found that situations in which flares occur several hours ahead of coronal ejections predominate.

The auroral activity indices AU, AL, AE, introduced into geophysics at the beginning of the space era, although they have certain drawbacks, are still widely used to monitor geomagnetic activity at high latitudes. The AU index reflects the intensity of the eastern electric jet, while the AL index is determined by the intensity of the western electric jet. There are many regression relationships linking the indices of magnetic activity with a wide range of phenomena observed in the Earth's magnetosphere and atmosphere. These relationships determine the importance of monitoring and predicting geomagnetic activity for research in various areas of solar-terrestrial physics. The most dramatic phenomena in the magnetosphere and high-latitude ionosphere occur during periods of magnetospheric substorms, a sensitive indicator of which is the time variation and value of the AL index. Currently, AL index forecasting is carried out by various methods using both dynamic systems and artificial intelligence. Forecasting is based on the close relationship between the state of the magnetosphere and the parameters of the solar wind and the interplanetary magnetic field (IMF). This application proposes an algorithm for describing the process of substorm formation using an instrument in the form of an Elman-type ANN by reconstructing the AL index using the dynamics of the new integral parameter we introduced. The use of an integral parameter at the input of the ANN makes it possible to simulate the structure and intellectual properties of the biological nervous system, since in this way an additional realization of the memory of the prehistory of the modeled process is provided.

Comparison of wavelet spectrum (skeletons) local maxima for disturbed components of solar plasma flow parameters and geomagnetic field disturbances recorded along the meridional station chain during geomagnetic storm intervals is performed in the range of magnetohydrodynamic (MHD) waves. An algorithm for quantitative evaluation of analyzed skeletons consistency has been developed. It has been used to demonstrate the possibility of the type of solar wind plasma flow elaboration on unique spectral signs of Solar wind speed disturbances, density and interplanetary magnetic field. It is shown that the energy spectrum of oscillations for these parameters reflects the internal structure of the corresponding plasma formation. The skeletons application to the analysis of the interplanetary space main parameters made it possible to estimate the magnetosphere reaction time in geomagnetic field horizontal component oscillations at different latitudes on the disturbance. As a result, the distributed magnetosphere reaction over latitude was determined in the form of geomagnetic field oscillations on the disturbed solar flow parameters. It is shown that the dynamics of the components of the solar wind parameters disturbance spectra corresponding to plasma flows manifest themselves in the MHD spectra of high-latitude stations magnetograms and can be used as a diagnostic tool.

A Kohonen artificial neural network (ANN) was used to classify patterns of causal relationships between the level of geomagnetic activity in the auroral zone and plasma and magnetic field parameters in the body of an interplanetary magnetic cloud (IMO). Terrestrial and satellite observations during 33rd interplanetary magnetic clouds recorded from 1998 to 2012 are examined in detail. Experiments with the ANN during its fast training show that substorm discrimination by their intensity by three classes plus a "collector" for collecting atypical events is optimal for the study. An analysis of the classification result studies showed that each selected class of substorms corresponds to a specific set of perturbations of the plasma parameters and the magnetic field of the IMO. Using the integral characteristics of the plasma and the IMF components as input parameters of the ANN allowed us to detect the levels of the expected intensity of the AL index with an accuracy of up to 70%. The created ANNs can be used to restore the AL index both during periods of isolated magnetospheric substorms and during periods of a series of continuous successive substorms, one after another.

Впериоды магнитосферных суббурь зарегистрированы одновременные возмущения полного электронного содержания и горизонтальной компоненты геомагнитного поля в диапазоне периодов Рс6, возникающие в интервалы взаимодействия магнитосферы Земли с турбулентной оболочкой магнитных облаков. Вейвлет анализ возмущений межпланетного магнитного поля, скорости и плотности плазмы солнечного ветра свидетельствует о существовании в межпланетной среде колебаний с такими же периодами.

The study demonstrated the technology for searching nonlinear correlations between the intensity indices of auroral electric jets (AU, AL) and ring current intensity indices (SYM, ASY). The well-known connection of the western electrojet with the asymmetric part of the ring current and the eastern electrodett with the symmetric is confirmed. The technology of artificial neural networks was chosen as the main method. Thus, the relationship of magnetic disturbances in the auroral region and magnetic disturbances at middle and low latitudes in the main phase of a geomagnetic storm is investigated. The characteristic lead times in the development of mid-latitude magnetospheric processes relative to polar are determined. An acceptable neural network recovery of auroral electrojet intensity indices from ring current index data is demonstrated.

The efficiency of the previously developed algorithm for predicting extreme solar flares from their geomagnetic precursors has been tested on the example of a series of class X flares recorded on September 6, 2017. Over the past few years, these events of the solar flare activity are seen as the most intense. The algorithm consists in comparing the total histograms of the normalized number of synchronization moments for oscillations of long-period pulsations in the horizontal component of the geomagnetic field with extreme solar flare events in the intervals preceding them. The result of the algorithm is that extrema of histograms were detected a few days before the development of intense solar flares and can be considered as markers of the expected extreme flare activity. The study confirms the possibility of using the previously proposed algorithm for a short-term forecast of solar flare activity.

We propose a method for determining location and orientation of extended solar sources of magnetic clouds, using coronagraph data and SOHO EIT/MDI images of the photosphere. To estimate the probability of formation of magnetic clouds, we use a simple cylindrical force-free model. We have established that more extended sources and those having a slight inclination to the solar equator and located on the solar limb as compared to those that are nonextended and strongly inclined can generate expanding clouds, which with high probability can reach the magnetosphere like clouds from a source near the zero meridian and low latitudes. We determine the relationship between extreme values of substorm activity and parameters of solar sources under study during the impact of magnetic clouds on Earth’s magnetosphere from the AL index. We note that there are no substorms associated with extended sources outside the heliolatitude range ~5–20°. The established relationship between solar source coordinates and geomagnetic activity of the magnetic cloud sheath and body are consistent with the most probable distribution of magnetoactive regions over the solar disk.

The paper draws attention to the complex structure of the fast magnetic clouds of the solar wind, which, in addition to the actual cloud body, contain a turbulent transition layer (cloud sheath) with a large and irregular magnetic field following the shock wave. The orientation of the plane of the magnetic-cloud shock wave with respect to the interplanetary magnetic field modified by the shock wave propagating in the solar wind for 33 cases of the registration of fast magnetic clouds has been calculated. The dependence of the substorm activity in the auroral zone on the level of turbulent processes occurring in the sheath of magnetic clouds has been studied. It was taken into account that the turbulent phenomena in the sheath are largely determined by the orientation of the shock-wave plane with respect to the interplanetary magnetic field. It is shown that the level of magnetic activity in the auroral zone, which is characterized by the integral AL index, increases with a decrease in the angle between the direction normal to the shock-wave front and the vector of the interplanetary magnetic field. Thus, the most geoeffective are magnetic clouds with a quasi-parallel shock wave, and the least geoeffective are those with a quasi-perpendicular shock wave. It was concluded that the intensity of turbulent processes in the cloud sheath increases with a decrease in the magnetic field penetrating the sheath, which plays a stabilizing role for turbulent magnetohydrodynamic perturbations.