More than 140 isolated non-recurrent geomagnetic storms (GMSs) of various intensities from extreme to weak are considered, which are reliably identified with solar eruptive sources (coronal mass ejections, CMEs). The analysis aims to obtain a possibly complete picture of the relationship between the transit time of propagation of CMEs and interplanetary coronal mass ejections (ICMEs) from the Sun to the Earth (more precisely, the time interval dtp from the moment of an eruption until the peak of the corresponding GMS) and the maximum intensity of this GMS, as measured by the disturbance storm time geomagnetic index Dst. Two groups of events are singled out: one includes GMSs, the source of which was an eruption from an active region (AR events), the other GMSs caused by filament eruptions from quiescent areas of the Sun located outside ARs (QS events). The distribution of the large number of the analyzed events on a dtp - Dst plane confirms and substantially clarifies the known regularities. The AR events are characterized by a shorter transit time (dtp ~ 1-4 days) and much stronger GMSs (Dst up to -600 nT mainly) in comparison with the QS events (dtp ~ 3-5 days, Dst > -200 nT). For events of both groups, the shorter transit time of CMEs/ICMEs, the more intense GMSs; in particular, for AR events when dtp declines from 4 to 1 day, Dst decreases on average from -100 to -470 nT and can reach -900 nT. From the point of view of the nature of GMSs and their sources on the Sun, the obtained results mean that both the speed of CMEs/ICMEs and the strength of the magnetic field transferred by them are largely determined by the parameters of the corresponding eruptions, in particular, by the eruptive magnetic flux and the released energy.

On the approach to minimum of Solar Cycle 24, on 26 August 2018, an unexpectedly strong geomagnetic storm (GMS) suddenly occurred. Its Dst index reached -174 nT, that is the third of the most intense storms during the cycle. The analysis showed that it was initiated by a two-step long filament eruption, which occurred on 20 August in the central sector of the solar disk. The eruptions were accompanied by two large-scale divergent ribbons and dimmings of a considerable size and were followed by relatively weak but evident Earth-directed coronal mass ejections. In the inner corona, their estimated speed was very low of about 200-360 km/s. The respective interplanetary transients apparently propagated between two high-speed solar wind streams originated from a two-component coronal hole and therefore their expansion was limited. The resulting ejecta arrived at the Earth only on 25 August and brought an unexpectedly strong field of Bt ~ 18.2 nT with a predominantly negative Bz component of almost the same strength. The geospace storm also manifested itself in the form of a peculiar Forbush decrease (FD). Its magnitude was about 1.5%, which is rather small for the observed G3-class GMS. The main unusual feature of the event is that large positive bursts with an enhancement up to 3% above the pre-event level were recorded on the FD background. We argue that these bursts were mainly caused by an unusually large and changeable cosmic ray anisotropy combined with lowering of the geomagnetic cutoff rigidity in the perturbed Earth's magnetosphere under conditions of the cycle minimum.

Based on our tool for the early diagnostics of solar eruption geoeffectiveness (EDSEG tool; Chertok et al., 2013, 2015, 2017), we have analyzed space weather disturbances that occurred in early September 2017. Two flares, SOL2017-09-04T20:33 (M5.5) and SOL2017-09-06T12:02 (X9.3), accompanied by Earth-directed halo coronal mass ejections (CMEs) were found to be geoeffective. We extracted the associated EUV dimmings and arcades and calculated their total unsigned magnetic flux. This calculation allowed us to estimate the possible scales of the Forbush decreases (FDs) and geomagnetic storms (GMSs) in the range from moderate to strong, and they are close to the observed scales. More precisely, after the first eruption, an FD approximately equal to 2% and almost no GMS occurred because the Bz magnetic field component in front of the corresponding interplanetary CME (ICME) was northern. The stronger second eruption produced somewhat larger composite disturbances (FD ~ 9.3% and GMS with indexes Dst ~ -144 nT, Ap ~ 235) than expected (FD ~ 4.4%, Dst ~ -135 nT, Ap ~ 125) because the second ICME overtook the trailing part of the first ICME near Earth, and the resulting Bz component was more intense and southern. Both ICMEs arrived at Earth earlier than expected because they propagated in the high-speed solar wind emanated from an extended coronal hole adjacent to AR12673 along their entire path. Overall, the presented results provide further evidence that the EDSEG tool can be used for the earliest diagnostics of actual solar eruptions to forecast the scale of the corresponding geospace disturbances.

The powerful flares that occurred in the Sun on September 4-10, 2017 are analyzed on the basis of the technique for the quantitative diagnostics of proton flares developed in IZMIRAN in 1970-1980s. It is shown that the fluxes and energy spectra of protons with energy of tens of MeV coming to Earth qualitatively and quantitatively correspond to the intensity and frequency spectra of microwave radio bursts in the range 2.7-15.4 GHz. In particular, the flare of 4 Sept. with a peak radio flux S ~ 2000 sfu at the frequency f ~ 3 GHz (i.e., with a soft radio spectrum) was accompanied by a significant proton flux J (> 10 MeV) ~ 100 pfu and soft energy spectrum with the exponent gamma ~ 3.0, and the powerful flare of 10 Sept. with S ~ 21000 sfu at f ~ 15 GHz (i.e., with a hard radio spectrum) led to a very intense proton event with J (> 10 MeV) ~ 1000 pfu with a hard energy spectrum (gamma ~ 1.4), including a ground level enhancement (GLE72). This is further evidence that data on microwave radio bursts can be successfully used to diagnose proton flares, regardless of the particular source of acceleration of particles on the Sun.

In this note, we consider radio characteristics of three proton flares that caused discrete enhancements of solar energetic particles (SEPs) near Earth. The analysis confirmed that the flux density and frequency spectrum of microwave bursts, although the latter are generated by electrons propagating to the photosphere, reflect the number and energy spectrum of accelerated particles, including the 10-100 MeV protons coming to Earth.

Using data from the Geostationary Operational Environmental Satellites (GOES) spacecraft in the 1-8 \AA\ wavelength range for Solar Cycles 23, 24, and part of Cycles 21 and 22, we compare mean temporal parameters (rising, decay times, duration) and the proportion of impulsive short-duration events (SDE) and gradual long-duration events (LDE) among C- and $\geq$M1.0-class flares. It is found that the fraction of the SDE $\geq$M1.0-class flares (including spikes) in Cycle 24 exceeds that in Cycle 23 in all three temporal parameters at the maximum phase and in the decay time during the ascending cycle phase. However, Cycles 23 and 24 barely differ in the fraction of the SDE C-class flares. The temporal parameters of SDEs, their fraction, and consequently the relationship between the SDE and LDE flares do not remain constant, but they reveal regular changes within individual cycles and during the transition from one cycle to another. In all phases of all four cycles, these changes have the character of pronounced, large-amplitude "quasi-biennial" oscillations (QBOs). In different cycles and at the separate phases of individual cycles, such QBOs are superimposed on various systematic trends displayed by the analyzed temporal flare parameters. In Cycle 24, the fraction of the SDE $\geq$M1.0-class flares from the N- and S-hemispheres displays the most pronounced synchronous QBOs. The QBO amplitude and general variability of the intense $\geq$M1.0-class flares almost always markedly exceeds those of the moderate C-class flares. The ordered quantitative and qualitative variations of the flare type revealed in the course of the solar cycles are discussed within the framework of the concept that the SDE flares are associated mainly with small sunspots (including those in developed active regions) and that small and large sunspots behave differently during cycles and form two distinct populations.

In our previous articles (Chertok et al.: 2013, Solar Phys. 282, 175, and 2015, Solar Phys. 290, 627), we presented a preliminary tool for the early diagnostics of the geoeffectiveness of solar eruptions based on the estimate of the total unsigned line-of-sight photospheric magnetic flux in accompanying extreme-ultraviolet arcades and dimmings. This tool was based on the analysis of eruptions observed in 1996-2005 with SOHO/EIT and MDI. Empirical relationships were obtained to estimate the probable importance of upcoming space weather disturbances caused by an eruption, which just occurred, without data on the associated coronal mass ejections. It was possible to estimate the intensity of a non-recurrent geomagnetic storm (GMS) and Forbush decrease (FD), as well as their onset and peak times. After 2010-2011, data on solar eruptions are obtained with SDO/AIA and HMI. We use relatively short intervals of overlapping EIT-AIA and MDI-HMI detailed observations and a number of large eruptions over the next five years with the 12-hour cadence EIT images to adapt the SOHO diagnostic tool to SDO data. The adopted brightness thresholds select from the EIT 195 \AA\ and AIA 193 \AA\ image practically the same areas of arcades and dimmings with a cross-calibration factor of 3.6-5.8 (5.0-8.2) for the AIA exposure time of 2.0 s (2.9 s). For the same photospheric areas, the MDI magnetic flux systematically exceeds the HMI flux by a factor of 1.4. Based on these results, the empirical diagnostic relationships obtained from SOHO data are adjusted to SDO instruments. Examples of a post-diagnostics based on SDO data are presented. As before, the tool is applicable to non-recurrent GMSs and FDs caused by nearly central eruptions from active regions, provided that the southern component of the interplanetary magnetic field near the Earth is predominantly negative, which is not predicted by this tool.

We further study the relations between parameters of bursts at 35 GHz recorded with the Nobeyama Radio Polarimeters during 25 years, on the one hand, and solar proton events, on the other hand (Grechnev et al. in Publ. Astron. Soc. Japan 65, S4, 2013a). Here we address the relations between the microwave fluences at 35 GHz and near-Earth proton fluences above 100 MeV in order to find information on their sources and evaluate their diagnostic potential. A correlation was found to be pronouncedly higher between the microwave and proton fluences than between their peak fluxes. This fact probably reflects a dependence of the total number of protons on the duration of the acceleration process. In events with strong flares, the correlation coefficients of high-energy proton fluences with microwave and soft X-ray fluences are higher than those with the speeds of coronal mass ejections. The results indicate a statistically larger contribution of flare processes to high-energy proton fluxes. Acceleration by shock waves seems to be less important at high energies in events associated with strong flares, although its contribution is probable and possibly prevails in weaker events. The probability of a detectable proton enhancement was found to directly depend on the peak flux, duration, and fluence of the 35 GHz burst, while the role of the Big Flare Syndrome might be overestimated previously. Empirical diagnostic relations are proposed.

Around the peaks of substantial flares, bright artifact nearly horizontal saturation streaks (B-streaks) corresponding to the brightest parts of the flare sources appear in the STEREO/EUVI 195 A images. We show that the length of such B-streaks can be used for the solution of an actual problem of evaluating the soft X-ray flux and class of far-side flares registered with double STEREO spacecraft but invisible from Earth. For this purpose from data on about 350 flares observed from January 2007 to July 2014 (mainly exceeding the GOES M1.0 level) both with GOES and STEREO, an empirical relation is established correlating the GOES 1-8 A peak flux and the B-streak length. This allowed us for the same years to estimate the soft X-ray classes for approximately 65 strong far-side flares observed by STEREO. The results of this simple and prompt method are consistent with the estimations of Nitta et al. (Solar Phys., 288, 241, 2013) based on the calculations of the EUVI full-disk digital number output. In addition, we studied some features of the B-streaks in impulsive and long-duration flares and demonstrated that B-streaks in several consecutive EUVI images can be used to reconstruct a probable time history of strong far-side flares.

Multi-instrument observations of two filament eruptions on 24 February and 11 May 2011 suggest the following updated scenario for eruptive flare, CME and shock wave evolution. An initial destabilization of a filament results in stretching out of magnetic threads belonging to its body and rooted in the photosphere along the inversion line. Their reconnection leads to i) heating of parts of the filament or its environment, ii) initial development of the flare arcade cusp and ribbons, and iii) increasing similarity of the filament to a curved flux rope and its acceleration. Then the pre-eruption arcade enveloping the filament gets involved in reconnection according to the standard model and continues to form the flare arcade and ribbons. The poloidal magnetic flux in the curved rope developing from the filament progressively increases and forces its toroidal expansion. This flux rope impulsively expands and produces an MHD disturbance, which rapidly steepens into a shock. The shock passes through the arcade expanding above the filament and then freely propagates ahead of the CME like a decelerating blast wave for some time. If the CME is slow, then the shock eventually decays. Otherwise, the frontal part of the shock changes into the bow-shock regime. This was observed for the first time in the 24 February 2011 event. When reconnection ceases, the flux rope relaxes and constitutes the CME core-cavity system. The expanding arcade develops into the CME frontal structure. We also found that reconnection in the current sheet of a remote streamer forced by the shock's passage results in a running flare-like process within the streamer responsible for a type II burst. The development of dimming and various associated phenomena are discussed.

This is the first of four companion papers, which analyze a complex eruptive event of 18 November 2003 in AR 10501 and the causes of the largest Solar Cycle 23 geomagnetic storm on 20 November 2003. Analysis of a complete data set, not considered before, reveals a chain of eruptions to which hard X-ray and microwave bursts responded. A filament in AR 10501 was not a passive part of a larger flux rope, as usually considered. The filament erupted and gave origin to a CME. The chain of events was as follows: i) an eruption at 07:29 accompanied by a not reported M1.2 class flare associated with the onset of a first southeastern CME1, which is not responsible for the superstorm; ii) a confined eruption at 07:41 (M3.2 flare) that destabilized the filament; iii) the filament acceleration (07:56); iv) the bifurcation of the eruptive filament that transformed into a large cloud; v) an M3.9 flare in AR 10501 associated to this transformation. The transformation of the filament could be due to its interaction with the magnetic field in the neighborhood of a null point, located at a height of about 100 Mm above the complex formed by ARs 10501, 10503, and their environment. The CORONAS-F/SPIRIT telescope observed the cloud in 304 A as a large Y-shaped darkening, which moved from the bifurcation region to the limb. The masses and kinematics of the cloud and the filament were similar. Remnants of the filament were not observed in the second southwestern CME2, previously regarded as a source of the 20 November superstorm. These facts do not support a simple scenario, in which the interplanetary magnetic cloud is considered as a flux rope formed from a structure initially associated with the pre-eruption filament in AR 10501. Observations suggest a possible additional eruption above the bifurcation region close to solar disk center between 08:07 and 08:17 that could be the source of the superstorm.

We continue our study (Grechnev et al. (2013), doi:10.1007/s11207-013-0316-6; Paper I) on the 18 November 2003 geoffective event. To understand possible impact on geospace of coronal transients observed on that day, we investigated their properties from solar near-surface manifestations in extreme ultraviolet, LASCO white-light images, and dynamic radio spectra. We reconcile near-surface activity with the expansion of coronal mass ejections (CMEs) and determine their orientation relative to the earthward direction. The kinematic measurements, dynamic radio spectra, and microwave and X-ray light curves all contribute to the overall picture of the complex event and confirm an additional eruption at 08:07 - 08:20 UT close to the solar disk center presumed in Paper I. Unusual characteristics of the ejection appear to match those expected for a source of the 20 November superstorm but make its detection in LASCO images hopeless. On the other hand, none of the CMEs observed by LASCO seem to be a promising candidate for a source of the superstorm being able to produce, at most, a glancing blow on the Earth's magnetosphere. Our analysis confirms free propagation of shock waves revealed in the event and reconciles their kinematics with "EUV waves" and dynamic radio spectra up to decameters.

Solar coronal mass ejections (CMEs) are main drivers of the most powerful non-recurrent geomagnetic storms. In the extreme-ultraviolet range, CMEs are accompanied by bright post-eruption arcades and dark dimmings. The analysis of events of the Solar Cycle 23 (Chertok et al., 2013, Solar Phys. 282, 175) revealed that the summarized unsigned magnetic flux in the arcades and dimming regions at the photospheric level, $\Phi$, is significantly related to the intensity (Dst index) of geomagnetic storms. This provides the basis for the earliest diagnosis of geoefficiency of solar eruptions. In the present article, using the same data set, we find that a noticeable correlation exists also between the eruptive magnetic flux, $\Phi$, and another geomagnetic index, Ap. As the magnetic flux increases from tens to $\approx 500$ (in units of $10^{20}$ Mx), the geomagnetic storm intensity measured by the 3-hour Ap index, enhances in average from Ap $\approx 50$ to a formally maximum value of 400 (in units of 2 nT). The established relationship shows that in fact the real value of the Ap index is not limited and during the most severe magnetic storms may significantly exceed 400.

The geomagnetic superstorm of 20 November 2003 with Dst = -422 nT, one of the most intense in history, is not well understood. The superstorm was caused by a moderate solar eruptive event on 18 November, comprehensively studied in our preceding Papers I-III. The analysis has shown a number of unusual and extremely complex features, which presumably led to the formation of an isolated right-handed magnetic-field configuration. Here we analyze the interplanetary disturbance responsible for the 20 November superstorm, compare some of its properties with the extreme 28-29 October event, and reveal a compact size of the magnetic cloud (MC) and its disconnection from the Sun. Most likely, the MC had a spheromak configuration and expanded in a narrow angle of < 14 degree. A very strong magnetic field in the MC up to 56 nT was due to the unusually weak expansion of the disconnected spheromak in an enhanced-density environment constituted by the tails of the preceding ICMEs. Additional circumstances favoring the superstorm were (i) the exact impact of the spheromak on the Earth's magnetosphere and (ii) the almost exact southward orientation of the magnetic field, corresponding to the original orientation in its probable source region near the solar disk center.

Our analysis in Papers I and II (Grechnev et al., 2014, Solar Phys. 289, 289 and 1279) of the 18 November 2003 solar event responsible for the 20 November geomagnetic superstorm has revealed a complex chain of eruptions. In particular, the eruptive filament encountered a topological discontinuity located near the solar disk center at a height of about 100 Mm, bifurcated, and transformed into a large cloud, which did not leave the Sun. Concurrently, an additional CME presumably erupted close to the bifurcation region. The conjectures about the responsibility of this compact CME for the superstorm and its disconnection from the Sun are confirmed in Paper IV (Grechnev et al., Solar Phys., submitted), which concludes about its probable spheromak-like structure. The present paper confirms the presence of a magnetic null point near the bifurcation region and addresses the origin of the magnetic helicity of the interplanetary magnetic clouds and their connection to the Sun. We find that the orientation of a magnetic dipole constituted by dimmed regions with the opposite magnetic polarities away from the parent active region corresponded to the direction of the axial field in the magnetic cloud, while the pre-eruptive filament mismatched it. To combine all of the listed findings, we come to an intrinsically three-dimensional scheme, in which a spheromak-like eruption originates via the interaction of the initially unconnected magnetic fluxes of the eruptive filament and pre-existing ones in the corona. Through a chain of magnetic reconnections their positive mutual helicity was transformed into the self-helicity of the spheromak-like magnetic cloud.

Low-temperature plasma ejected in solar eruptions can screen active regions as well as quiet solar areas. Absorption phenomena can be observed in microwaves as 'negative bursts' and in different spectral domains. We analyze two very different recent events with such phenomena and present an updated systematic view of solar events associated with negative bursts. Related filament eruptions can be normal, without essential changes of shape and magnetic configuration, and 'anomalous'. The latter are characterized by disintegration of an eruptive filament and dispersal of its remnants as a cloud over a large part of solar disk. Such phenomena can be observed as giant depressions in the He II 304 A line. One of possible scenarios for an anomalous eruption is proposed in terms of reconnection of filament's internal magnetic fields with external large-scale coronal surrounding.

Proceeding from close association between solar eruptions, flares, shock waves, and CMEs, we analyze relations between bursts at 35 GHz recorded with the Nobeyama Radio Polarimeters during 1990-2012, on the one hand, and solar energetic particle (SEP) events, on the other hand. Most west to moderately east solar events with strong bursts at 35 GHz produced near-Earth proton enhancements of J(E > 100 MeV) > 1 pfu. The strongest and hardest those caused ground level enhancements. There is a general, although scattered, correspondence between proton enhancements and peak fluxes at 35 GHz, especially pronounced if the 35 GHz flux exceeds 10^4 sfu and the microwave peak frequency is high. These properties indicate emission from numerous high-energy electrons in very strong magnetic fields suggesting a high rate of energy release in the flare-CME formation process. Flaring above the sunspot umbrae appears to be typical of such events. Irrespective of the origin of SEPs, these circumstances demonstrate significant diagnostic potential of high-frequency microwave bursts and sunspot-associated flares for space weather forecasting. Strong prolonged bursts at 35 GHz promptly alert to hazardous SEP events with hard spectra. A few exceptional events with moderate bursts at 35 GHz and strong proton fluxes look challenging and should be investigated.

This study aims at the early diagnostics of geoeffectiveness of coronal mass ejections (CMEs) from quantitative parameters of the accompanying EUV dimming and arcade events. We study events of the 23th solar cycle, in which major non-recurrent geomagnetic storms (GMS) with Dst <-100 nT are sufficiently reliably identified with their solar sources in the central part of the disk. Using the SOHO/EIT 195 A images and MDI magnetograms, we select significant dimming and arcade areas and calculate summarized unsigned magnetic fluxes in these regions at the photospheric level. The high relevance of this eruption parameter is displayed by its pronounced correlation with the Forbush decrease (FD) magnitude, which, unlike GMSs, does not depend on the sign of the Bz component but is determined by global characteristics of ICMEs. Correlations with the same magnetic flux in the solar source region are found for the GMS intensity (at the first step, without taking into account factors determining the Bz component near the Earth), as well as for the temporal intervals between the solar eruptions and the GMS onset and peak times. The larger the magnetic flux, the stronger the FD and GMS intensities are and the shorter the ICME transit time is. The revealed correlations indicate that the main quantitative characteristics of major non-recurrent space weather disturbances are largely determined by measurable parameters of solar eruptions, in particular, by the magnetic flux in dimming areas and arcades, and can be tentatively estimated in advance with a lead time from 1 to 4 days. For GMS intensity, the revealed dependencies allow one to estimate a possible value, which can be expected if the Bz component is negative.

On 17 January 2010, STEREO-B observed in extreme ultraviolet (EUV) and white light a large-scale dome-shaped expanding coronal transient with perfectly connected off-limb and on-disk signatures. Veronig et al. (2010, ApJL 716, 57) concluded that the dome was formed by a weak shock wave. We have revealed two EUV components, one of which corresponded to this transient. All of its properties found from EUV, white light, and a metric type II burst match expectations for a freely expanding coronal shock wave including correspondence to the fast-mode speed distribution, while the transient sweeping over the solar surface had a speed typical of EUV waves. The shock wave was presumably excited by an abrupt filament eruption. Both a weak shock approximation and a power-law fit match kinematics of the transient near the Sun. Moreover, the power-law fit matches expansion of the CME leading edge up to 24 solar radii. The second, quasi-stationary EUV component near the dimming was presumably associated with a stretched CME structure; no indications of opening magnetic fields have been detected far from the eruption region.

We show examples of excitation of coronal waves by flare-related abrupt eruptions of magnetic rope structures. The waves presumably rapidly steepened into shocks and freely propagated afterwards like decelerating blast waves that showed up as Moreton waves and EUV waves. We propose a simple quantitative description for such shock waves to reconcile their observed propagation with drift rates of metric type II bursts and kinematics of leading edges of coronal mass ejections (CMEs). Taking account of different plasma density falloffs for propagation of a wave up and along the solar surface, we demonstrate a close correspondence between drift rates of type II bursts and speeds of EUV waves, Moreton waves, and CMEs observed in a few known events.

Plasma with a temperature close to the chromospheric one is ejected in solar eruptions. Such plasma can occult some part of emission of compact sources in active regions as well as quiet solar areas. Absorption phenomena can be observed in the microwave range as the so-called 'negative bursts' and also in the He II 304 A line. The paper considers three eruptive events associated with rather powerful flares. Parameters of absorbing material of an eruption are estimated from multi-frequency records of a 'negative burst' in one event. 'Destruction' of an eruptive filament and its dispersion like a cloud over a huge area observed as a giant depression of the 304 A line emission has been revealed in a few events. One such event out of three ones known to us is considered in this paper. Another event is a possibility.

We present a case study of the 13 July 2004 solar event, in which disturbances caused by eruption of a filament from an active region embraced a quarter of the visible solar surface. Remarkable are absorption phenomena observed in the SOHO/EIT 304 A channel; they were also visible in the EIT 195 A channel, in the H-alpha line, and even in total radio flux records. Coronal and Moreton waves were also observed. Multi-spectral data allowed reconstructing an overall picture of the event. An explosive filament eruption and related impulsive flare produced a CME and blast shock, both of which decelerated and propagated independently. Coronal and Moreton waves were kinematically close and both decelerated in accordance with an expected motion of the coronal blast shock. The CME did not resemble a classical three-component structure, probably, because some part of the ejected mass fell back onto the Sun. Quantitative evaluations from different observations provide close estimates of the falling mass, ~3 10^15 g, which is close to the estimated mass of the CME. The falling material was responsible for the observed large-scale absorption phenomena, in particular, shallow widespread moving dimmings observed at 195 A. By contrast, deep quasi-stationary dimmings observed in this band near the eruption center were due to plasma density decrease in coronal structures.

The extreme solar and SEP event of 20 January 2005 is analyzed from two perspectives. Firstly, we study features of the main phase of the flare, when the strongest emissions from microwaves up to 200 MeV gamma-rays were observed. Secondly, we relate our results to a long-standing controversy on the origin of SEPs arriving at Earth, i.e., acceleration in flares, or shocks ahead of CMEs. All emissions from microwaves up to 2.22 MeV line gamma-rays during the main flare phase originated within a compact structure located just above sunspot umbrae. A huge radio burst with a frequency maximum at 30 GHz was observed, indicating the presence of a large number of energetic electrons in strong magnetic fields. Thus, protons and electrons responsible for flare emissions during its main phase were accelerated within the magnetic field of the active region. The leading, impulsive parts of the GLE, and highest-energy gamma-rays identified with pi^0-decay emission, are similar and correspond in time. The origin of the pi^0-decay gamma-rays is argued to be the same as that of lower energy emissions. We estimate the sky-plane speed of the CME to be 2000-2600 km/s, i.e., high, but of the same order as preceding non-GLE-related CMEs from the same active region. Hence, the flare itself rather than the CME appears to determine the extreme nature of this event. We conclude that the acceleration, at least, to sub-relativistic energies, of electrons and protons, responsible for both the flare emissions and the leading spike of SEP/GLE by 07 UT, are likely to have occurred simultaneously within the flare region. We do not rule out a probable contribution from particles accelerated in the CME-driven shock for the leading GLE spike, which seemed to dominate later on.