The large flare/CME event that occurred close to the west solar limb on 3 November 2003 launched a large-amplitude large-scale coronal wave that was observed in Hα and Fe xii 195 Å spectral lines, as well as in the soft X-ray and radio wavelength ranges. The wave also excited a complex decimeter-to-hectometer type II radio burst, revealing the formation of coronal shock(s). The back-extrapolation of the motion of coronal wave signatures and the type II burst sources distinctly marks the impulsive phase of the flare (the hard X-ray peak, drifting microwave burst, and the highest type III burst activity), favoring a flare-ignited wave scenario. On the other hand, comparison of the kinematics of the CME expansion with the propagation of the optical wave signatures and type II burst sources shows a severe discrepancy in the CME-driven scenario. However, the CME is quite likely associated with the formation of an upper-coronal shock revealed by the decameter-hectometer type II burst. Finally, some six minutes after the launch of the first coronal wave, another coronal disturbance was launched, exciting an independent (weak) decimeter-meter range type II burst. The back-extrapolation of this radio emission marks the revival of the hard X-ray burst, and since there was no CME counterpart, it was clearly ignited by the new energy release in the flare. Moreton waves and type II bursts are very closely related phenomena (e.g., Harvey et al. 1974;Klassen et al. 2000;Khan & Aurass 2002;Warmuth et al. 2004b), indicating the common nature of the underlying disturbance. The physical background of the relationship was explained by Uchida (1974): the coronal fast-mode MHD shock that propagates along "valleys of low Alfven velocity" excites type II bursts in the corona, whereas the Moreton wave is a "surface track" of the shock front propagation (cf., Uchida 1974, and references therein).Generally, the coronal large-amplitude MHD disturbance could be generated by flares as well as by CMEs, and quite likely, both types of processes happen . The excellent timing association of Moreton waves and type II bursts with the impulsive phase of associated flares (e.g., Harvey 1965;Švestka & Fritzova-Švestkova 1974;Vršnak et al. 1995;Vršnak 2001;Klassen et al. 1999Klassen et al. , 2003) strongly suggests that the waves are ignited by flares. This is also supported by a number of relatively well-defined correlations between various wave characteristics and the flare energy release characteristics (e.g.,Article published by EDP Sciences and available at http://www.edpsciences.org/aa or http://dx

Aims. We study selected properties of solar energetic particle (SEP) events as inferred from their associated radio emissions. Methods. We used a catalogue of 115 SEP events, which consists of entries of proton intensity enhancements at one AU, with complete coverage over solar cycle 23 based on high-energy (∼68 MeV) protons from SOHO/ERNE. We also calculated the proton release time at the Sun using velocity dispersion analysis (VDA). After an initial rejection of cases with unrealistic VDA path lengths, we assembled composite radio spectra for the remaining events using data from ground-based and space-borne radio spectrographs. We registered the associated radio emissions for every event, and we divided the events in groups according to their associated radio emissions. In cases of type III-associated events, we extended our study to the timings between the type III radio emission, the proton release, and the electron release as inferred from VDA based on Wind/3DP 20-646 keV data. Results. The proton release was found to be most often accompanied by both type III and II radio bursts, but a good association percentage was also registered in cases accompanied by type IIIs only. The worst association was found for the cases only associated with type II. In the type III-associated cases, we usually found systematic delays of both the proton and electron release times as inferred by the particles' VDAs, with respect to the start of the associated type III burst. The comparison of the proton and electron release times revealed that, in more than half of the cases, the protons and electrons were simultaneously released within the statistical uncertainty of our analysis. For the cases with type II radio association, we found that the distribution of the proton release heights had a maximum at ∼2.5 R . Most (69%) of the flares associated with our SEP events were located in the western hemisphere, with a peak within the well-connected region of 50• -60• western longitude. Conclusions. Both flare-and shock-related particle release processes are observed in major proton events at >50 MeV. Typically, the protons are released after the start of the associated type III bursts and simultaneously or before the release of energetic electrons. Our study indicates that a clear-cut distinction between flare-related and CME-related SEP events is difficult to establish.

No abstract

Context. Narrow-band bursts appear on dynamic spectra from microwave to decametric frequencies as fine structures with very small duration and bandwidth. They are believed to be manifestations of small scale energy release through magnetic reconnection. Aims. We analyzed 27 metric type IV events with embedded narrow-band bursts, which were observed by the ARTEMIS-IV radio spectrograph from 30 June 1999 to 1 August 2010. We examined the morphological characteristics of isolated narrow-band structures (mostly spikes) and groups or chains of structures. Methods. The events were recorded with the SAO high resolution (10 ms cadence) receiver of ARTEMIS-IV in the 270-450 MHz range. We measured the duration, spectral width, and frequency drift of ∼12 000 individual narrow-band bursts, groups, and chains. Spike sources were imaged with the Nançay radioheliograph (NRH) for the event of 21 April 2003. Results. The mean duration of individual bursts at fixed frequency was ∼100 ms, while the instantaneous relative bandwidth was ∼2%. Some bursts had measurable frequency drift, either positive or negative. Quite often spikes appeared in chains, which were closely spaced in time (column chains) or in frequency (row chains). Column chains had frequency drifts similar to type-IIId bursts, while most of the row chains exhibited negative frequently drifts with a rate close to that of fiber bursts. From the analysis of NRH data, we found that spikes were superimposed on a larger, slowly varying, background component. They were polarized in the same sense as the background source, with a slightly higher degree of polarization of ∼65%, and their size was about 60% of their size in total intensity. Conclusions. The duration and bandwidth distributions did not show any clear separation in groups. Some chains tended to assume the form of zebra, lace stripes, fiber bursts, or bursts of the type-III family, suggesting that such bursts might be resolved in spikes when viewed with high resolution. The NRH data indicate that the spikes are not fluctuations of the background, but represent additional emission such as what would be expected from small-scale reconnection.

Context. Metric type II bursts are the most direct diagnostic of shock waves in the solar corona. Aims. There are two main competing views about the origin of coronal shocks: that they originate in either blast waves ignited by the pressure pulse of a flare or piston-driven shocks due to coronal mass ejections (CMEs). We studied three well-observed type II bursts in an attempt to place tighter constraints on their origins. Methods. The type II bursts were observed by the ARTEMIS radio spectrograph and imaged by the Nançay Radioheliograph (NRH) at least at two frequencies. To take advantage of projection effects, we selected events that occurred away from disk center. Results. In all events, both flares and CMEs were observed. In the first event, the speed of the shock was about 4200 km s −1 , while the speed of the CME was about 850 km s −1 . This discrepancy ruled out the CME as the primary shock driver. The CME may have played a role in the ignition of another shock that occurred just after the high speed one. A CME driver was excluded from the second event as well because the CMEs that appeared in the coronagraph data were not synchronized with the type II burst. In the third event, the kinematics of the CME which was determined by combining EUV and white light data was broadly consistent with the kinematics of the type II burst, and, therefore, the shock was probably CME-driven. Conclusions. Our study demonstrates the diversity of conditions that may lead to the generation of coronal shocks.