A joint analysis of high-energy neutrons and neutron-decay protons from a flare

A joint analysis of neutron monitor and GOES data is performed to study the production of high-energy neutrons at the Sun. The main objects of the research are the spectrum of >50 MeV neutrons and a possible spectrum of primary (interacting) protons which produced those neutrons during the major 1990 May 24 solar flare. Different possible scenarios of the neutron production are presented. The high magnitude of the 1990 May 24 neutron event provided an opportunity to detect neutron decay protons of higher energies than ever before. We compare predictions of the proposed models of neutron production with the observations of protons on board GOES 6 and 7. It is shown that the precursor in high-energy GOES channels observed during 20:55–21:09 UT can be naturally explained as originating from decay of neutrons in the interplanetary medium. The ratio of counting rates observed in different GOES channels can ensure the selection of the model parameters.The set of experimental data can be explained in the framework of a scenario which assumes the existence of two components of interacting protons in the flare. A hard spectrum component (the first component) generates neutrons during a short time while the interaction of the second (soft spectrum) component lasts longer. Alternative scenarios are found to be of lesser likelihood. The intensity-time profile of neutron - decay protons as predicted in the framework of the two-component exponential model of neutron production (Kocharov et al., 1994a) is in an agreement with the proton profiles observed on board GOES. We compare the deduced characteristics of interacting high-energy protons with the characteristics of protons escaping into the interplanetary medium. It is shown that, in the 100–1000 MeV range, the spectrum of the second component of interacting protons was close to the spectrum of the prompt component of interplanetary protons. However, it is most likely that, at 300 MeV, the interacting proton spectrum was slightly softer than the spectrum of interplanetary protons. An analysis of gamma-ray emission is required to deduce the spectrum of interacting protons below 100 MeV and above 1 GeV.     

Proton Flare Project, 1966

The paper summarizes observations of solar and space phenomena related to the McMath region Number 8461 which passed over the solar disk during the 1966 Proton Flare Project period, from August 21 to September 4, and produced two important solar particle events on August 28 and September 2. The most important results are reviewed and interpretation of some of them is suggested.Items of particular interest: Occurrence of proton-active regions when two or more rows of activity approach each other (Section 3). Possible stimulation of activity by magnetic fields of decaying regions that had been active before (4.2a, 5.1a). Significantly increased correlation of flares with X-ray bursts during the proton-active transit of the region (5.3b). Striking difference in the flare response in radio frequency range before and after August 26 (5.2b). Hardening of the X-rays (5.3a), increase in radio flux (5.2a), change in sunspot configuration (5.1c), and increased capability of the region for particle acceleration (5.1b, 5.2b), starting about three days prior to the proton flare. Clear evidence that some flares that occurred on or after August 26, but prior to the proton flare of August 28, already were sources of 1 MeV protons (5.2b, 8). Anomalous deficiency in metric component of radio bursts produced in the region (5.2c, 9.4d, 11.4b). Strong radio storm on meter waves immediately preceding the proton flare on August 28 (5.2a, 9.1b), coincident with preflare rising dark filament (9.1a) and slight preflare rise in flux of 1 MeV protons (10.2). Two phases of expansion (fast and slow) of the bright flare ribbons (9.2c). Coincidence of hard X-ray burst with the formation and fast separation of the bright flare ribbons. It is suggested that this is the time of particle acceleration in the flare (9.5b). Short-lived burst of UV radiation (9.6). Visible flare wave in the flare of August 28 (9.3b), and complexity of motions in this flare (9.4b). Suggested electron release by means of a blast wave (10.1a). Electron-proton splitting in the delayed shock-wave-associated maximum of the particle flux on August 29 (10.2c). First brightening of both proton flares in a similar position between the regions 8461 and 8459 (11.2c). Existence of a unique, low elevation coronal condensation three days after proton flare occurrences (7.2). Very strong flux of protons in energy range of the order of 100 MeV producing the largest PCA since July 1961, and unusually steep energy spectrum above 100 MeV in the flare of September 2 (12.2a, b, 12.4). Unusually long rise to the maximum flux, inconsistent with Burlaga's theory of anisotropic diffusion (12.2b). Interpretation of the undisturbed flux decay from September 2 to September 8 (12.2c). A corotating modulation phenomenon on September 8 (12.2d). Detection of medium nuclei, with He/M ratio 50 ± 11 (12.3a). Evidence against a purely velocity-dependent mode of particle propagation (12.3b). Electrons as the possible cause of the first PCA phase (12.4). Plasma disturbance due to permanent proton flux from the region (13.1). Electron injection into inner radiation belt during the geomagnetic storm associated with the September 2 flare (13.3).Section 14 brings a time scheme of the most important phenomena associated with the complex of activity and the active region in question, and some unsolved problems of particular interest are pointed out in Section 15.     

Solar Proton Rigidity Spectra From 1 to 10 gv of Selected Flare Events Since 1960

We have deduced the power-law rigidity spectra, J(P)=AP ^–, and the spectral evolution of the solar flare events that occurred in the present solar activity cycle on 6 November 1997, 14 July 2000, and 15 and 18 April 2001. The implications of these results for the acceleration of high-energy protons are discussed. The analysis is based on the ratios of the Mt. Washington to the Durham neutron monitor count-rate increases during the solar flare events. These two neutron monitors are located at different elevations (828 and 1030 g cm^–2, respectively) but at approximately the same geographical latitude and longitude. The proton spectra from 1 to 10 GV determined from the ratios of the count rate increases of the two neutron monitors are found to agree with those deduced from the global neutron monitor network or selected neutron monitors in 10 solar flare events from 1960 to 1990 for which comparative results are available. Thus the ratio method is quick, easy and reliable for deducing the spectral shape of solar flare protons at neutron monitor rigidities and for obtaining the spectral evolution as a function of time.     

The 1990 May 24 solar cosmic-ray event

This paper presents an integrated analysis of GOES 6, 7 and neutron monitor observations of solar cosmic-ray event following the 1990 May 24 solar flare. We have used a model which includes particle injection at the Sun and at the interplanetary shock front and particle propagation through the interplanetary medium. The model does not attempt to simulate the physical processes of coronal transport and shock acceleration, therefore the injections at the Sun and at the shock are represented by source functions in the particle transport equation. By fitting anisotropy and angle-average intensity profiles of high-energy (>30 MeV) protons as derived from the model to the ones observed by neutron monitors and at GOES 6 and 7, we have determined the parameters of particle transport, the injection rate and spectrum at the source. We have made a direct fit of uncorrected GOES data with both primary and secondary proton channels taken into account.The 1990 May 24–26 energetic proton event had a double-peaked temporal structure at energies 100 MeV. The Moreton (shock) wave nearby the flare core was seen clearly before the first injection of accelerated particles into the interplanetary medium. Some (correlated with this shock) acceleration mechanism which operates in the solar corona at a height up to one solar radius is regarded as a source of the first (prompt) increase in GOES and neutron monitor counting rates. The proton injection spectrum during this increase is found to be hard (spectral index 1.6) at lower energies ( 30 MeV) with a rapid steepening above 300 MeV. Large values of the mean free path ( 1.8 AU for 1 GV protons in the vicinity of the Earth) led to a high anisotropy of arriving protons. The second (delayed) proton increase was presumably produced by acceleration/injection of particles by an interplanetary shock wave at height of 10 solar radii. Our analysis of the 1990 May 24–26 event is in favour of the general idea that a number of components of energetic particles may be produced while the flare process develops towards larger spatial/temporal scales.Visiting Associate from St. Petersburg State Technical University, St. Petersburg 195251, Russia.     

On proton and electron acceleration by shock waves during large solar flares

The data on optical, X-ray and gamma emission from proton flares, as well as direct observations of flare-associated phenomena, show energetic proton acceleration in the corona rather than in the flare region. In the present paper, the acceleration of protons and accompanying relativistic electrons is accounted for by a shock wave arising during the development of a large flare. We deal with a regular acceleration mechanism due to multiple reflection of resonance protons and fast electrons from a collisionless shock wave front which serves as a moving mirror. The height of the most effective acceleration in the solar corona is determined. The accelerated particle energy and density are estimated. It is shown in particular that a transverse collisionless shock wave may produce the required flux of protons with energy of 10 MeV and of relativistic electrons of 1–10 MeV.The proposed scheme may also serve as an injection mechanism when the protons are accelerated up to relativistic energies by other methods.     

The 1991 March 22 flare: Possible anisotropy of high-energy neutral emission

We made a parameter fit to the Haleakala neutron monitor counting rate during the 1991 March 22 solar flare (Pyle and Simpson, 1991) using the time profiles of -rays at 0.42–80 MeV obtained with the GRANAT satellite (Vilmeret al., 1994) and the microwave data from Owens Valley Radio Observatory. We use a two-component neutron injection function to find that either an impulsive injection or the impulsive-plus-prolonged neutron injection is possible. In both cases, the number of > 300 MeV neutrons emitted towards the Earth is estimated as 2 × 10²⁷ sr^–1, which is less than that of the 1990 May 24 flare by an order of magnitude.We tested if such a big difference in neutron number detected on the Earth can be accounted for solely by their different positions on the solar disk. For the estimation of the degree of anisotropy of high-energy secondary emission, we made use of macroscopic parameters of the flare active region, in particular, the vector magnetogram data from the Big Bear Solar Observatory. In our result, the anisotropy factor for the neutral emissions of the 1991 March 22 flare is only 1 – 10, which is rather small compared with previous theoretical predictions for a disk flare. Such a moderate anisotropy is due to the relatively large inclination angles of the magnetic fields at the footpoints of the flaring loop where accelerated particles are trapped. We thus concluded that the smaller number of neutrons of the 1991 March 22 flare would be not only due to its location on the disk, but also due to fewer protons accelerated during this event as compared with the 1990 May 24 limb event. For a more precise determination of the anisotropy factor in a flare, we need a detailed spectrum of electron bremsstrahlung in 0.1 – 10 MeV and the fluence of -ray emission from the ⁰-decay.Visting Associate from St. Petersburg State Technical University, St. Petersburg, 195251, Russia.     

Energetic protons from the solar flare of March 24, 1966

Ten to 100 meV protons from the solar flare of March 24, 1966 were observed on the University of California scintillation counter on OGO-I. The short rise and decay times observed in the count rates of the 32 channels of pulse-height analysis show that scattering of the protons by the interplanetary field was much less important in this event than in previously observed proton flares. A diffusion theory in which D = M _r is found to be inadequate to account for the time behavior of the count rates of this event. Small fluctuations of the otherwise smooth decay phase may be due to flare protons reflected from the back of a shock front, which passed the earth on March 23.     

Neutron monitor and Pioneer 6 and 7 studies of the January 28, 1967 solar flare event

A discussion of the January 28, 1967 solar flare event is presented. High energy data from several neutron monitor stations are supplemented by low energy data from the interplanetary space probes Pioneers 6 and 7. A study of the data obtained from these three observation stations widely separated in solar azimuth has shown (1) the most probable location for the responsible flare was 60 ° beyond the western solar limb, (2) other than the large emitted particle flux, the phenomena associated with the January 28 activity are not atypical of other solar flare effects, (3) both the 0.5 GeV and 7.5 MeV fluxes observed at the earth were isotropic, indicative of particle diffusion across the interplanetary magnetic field lines, (4) the spectral exponent of the differential rigidity spectrum at high energies was - 4.8 ± 0.2, and (5) there was an indication of low energy solar injection prior to the high energy event of January 28.A technique is also described for obtaining the differential rigidity spectral index for an isotropic flux as a function of the relative enhancements of any pair of neutron monitors sufficiently separated in latitude.     

The solar-flare infrared continuum

We consider potential sources of infrared (1 to 1 mm) continuum in solar flares. Several mechanisms should produce detectable fluxes: in the 350 window for ground-based observations, impulsive emission will arise in synchrotron radiation from 1–10 MeV electrons, and possibly thermal (free-free) continuum from the source of the white-light flare; the hot flare plasma responsible for soft X-ray emission will also emit detectable fluxes of free-free continuum in the largest flares. At shorter wavelengths the dominant infrared emission will come from the H flare itself. Observations in the infrared wavelengths will help to complete our picture of flare structure in both the impulsive and gradual phases.     

Dynamics of the earth radiation belts during strong magnetic storms based on <Emphasis Type="Italic">CORONAS-F</Emphasis> data

The results of an experimental study of the variations in the intensity of the fluxes of the Earth radiation belt (ERB) particles in 0.3–6 and 1–50 MeV energy intervals for electrons and protons, respectively, are reported. ERBs were studied during strong magnetic storms from August 2001 through November 2003. The results of the CORONAS-F mission obtained during the magnetic storms of November 6 (D _st = ?257 nT) and November 24, 2001 (D _st = ?221 nT), October 29–30 (D _st = ?400 nT) and November 20, 2003 (D _st = ?465 nT) are analyzed. The electron flux is found to decrease abruptly in the outer radiation belt during the main phase of the magnetic storms under consideration. During the recovery phase, the outer radiation belt is found to recover much closer to Earth, near the boundary of the penetration of solar electrons during the main phase of the magnetic storm. We associate the decrease in the electron flux with the abrupt decrease of the size of the magnetosphere during the main phase of the storm. Note that, in all cases studied, the Earth radiation belts exhibited rather long (several days) variations. In those cases where solar cosmic-ray fluxes were observed during the storm, protons with energies 1–5 MeV could be trapped to form an additional maximum of protons with such energies at L >2.     

The decay phase of solar flare events

A study of the properties of the cosmic radiation of energy - 10 MeV generated by solar flares is reported. Data from four Pioneer spacecraft in interplanetary orbits, and separated by 180° in heliocentric longitude are employed. Attention is restricted to the properties evident at times in excess of 1 day after the occurrence of the parent flare. The anisotropic character of the radiation; the gradients in heliocentric longitude; the decay time constants; and the energy spectra of the radiation are all studied in detail.It is found that the equilibrium anisotropy assumes a direction - 45° E of the satellite-Sun line at very late times. It is suggested that the anisotropy at such times is parallel to E × B. This observation confirms that convection is the determining process in the escape of the solar cosmic rays from the solar system. It indicates that a positive radial gradient of solar cosmic radiation density has builtup at orbit of Earth some 4 days after a flare. This results in an effective convective velocity of approximately 1/2 the solar wind velocity. Direct measurements indicate the presence of strong gradients in heliocentric longitude even at very late times ( 4 days). These gradients are essentially invariant with respect to time, e-folding angles of n - 30° have been observed at - 10 MeV. The presence of these gradients has a major effect on the temporal variation of the cosmic ray flux during the decay phase of the flare effect. Thus, the observed decay time constant is either increased or decreased relative to the convective value depending on the position of the observer relative to the centroid of the cosmic ray population injected by the flare. The effect of the gradient becomes more pronounced at lower energies, and may even exceed the convective removal rate. The observed decay time constant, the characteristics of the anisotropy, and the gradient in longitude are shown to be inter-related as demanded by theory. It is shown that the exponent of the cosmic ray spectrum is dependent on the location of the observer relative to the centroid of the cosmic ray population injected by the parent flare. At a given point in the frame of reference of the cosmic ray population, the spectral exponent is invariant with time.Now at CSIRO, G.P.O. Box 124, Port Melbourne, Victoria 3207, Australia.On leave from Physical Research Laboratory, Ahmedabad, India.     

Cosmic ray intensity variations and two types of high speed solar streams

This study deals with the short-term variations of cosmic ray intensity during the interval 1973–78. Daily means of high latitude neutron and meson monitors from the same station and those of a low latitude neutron monitor have been analysed using the Chree method of superposed epochs. The zero epoch for the Chree analyses corresponds to the day of a substantial increase (V 200 km s^–1) in the solar wind speed to values of 550 km s^–1 and which persists at such high values for an interval of at least three days. The investigation reveals the existence of two types of cosmic ray intensity variations with distinctly different spectral characteristics. During the interval 1973–76, relative changes in the neutron and meson monitor rates are nearly equal indicating an almost flat rigidity spectrum of variation. During 1977–78, however, the spectrum acquires a negative spectral character similar to that observed for Forbush decreases. We suggest that events of the interval 1973–76 are essentially due to high speed streams associated with solar coronal holes and that events of the interval 1977–78 are due to fast streams from solar active regions with flare activity.     

Multispacecraft observations of the east-west asymmetry of solar Energetic Storm Particle events

Multispacecraft observations of energetic protons (E _p 500 keV) were obtained by the APL/JHU instruments on board the IMP-7 and 8 spacecraft and the Voyager-1 and 2 deep space probes, in order to study the generation of solar flare Energetic Storm Particle (ESP) events at widely separated locations on the same shock front. These locations are presumably characterized, on the average, by different interplanetary magnetic field-shock front configurations, i.e. quasi-perpendicular (quasi-parallel) shocks for eastern (western) solar flare sites. The multispacecraft energetic proton observations show that substantial differences in the ESP proton intensity enhancements (defined as the ratio of intensity increases near the shock over the ambient solar proton population) are detected at these energies for locations on the shock front with wide heliolongitude separations. In particular, large ESP proton intensity enhancements are detected at locations on the shock front for which the solar flare site generating the shock is to the east of the spacecraft meridian, whereas only weak ESP events are observed at locations on the same shock for which the flare site is to the west of the spacecraft meridian. The results indicate that acceleration of ESP protons to E _p 500 keV takes place exclusively at the quasi-perpendicular shock front domain, consistent with the shock drift acceleration mechanism (Armstrong et al., 1977).     

Observations of energetic X-rays and solar cosmic rays associated with the 23 May 1967 solar flare event

On 23 May 1967 energetic (10–50 keV) solar flare X-rays were observed by the OGO-III ion chamber during the period 1808–2100 UT. The time-intensity profile for the X-ray event showed three distinct peaks at 1810, 1841 and 1942 UT. The second peak, which is equivalent to 2.9 × 10^–3 ergs cm^–2sec^–1 above 20 keV, is the largest X-ray burst observed so far by the OGO-I and OGO-III ion chambers. The soft (2–12 Å) X-ray observations reported by Van Allen (1968) also show similar peaks, roughly proportional in magnitude to the energetic X-ray peaks. However, the intensity of energetic X-rays peaked in each case 5–10 min earlier than the soft X-ray intensity indicating a relatively hard photon energy spectrum near the peak of the energetic X-ray emission. The corresponding time-intensity profile for the solar radio emission also showed three peaks in the microwave region nearly coincident with the energetic X-ray peaks. The third radio peak was relatively rich in the metric emission. Beyond this peak both the energetic X-rays and the microwave emission decayed with a time constant of 8 min while the corresponding time constant for the soft X-rays was 43 min. In view of the earlier findings about the energetic X-rays it is indicated that the 23 May solar X-ray event was similar to those observed earlier. During the 23 May event the integral energy flux spectrum at the time of peak intensity is found to be consistent with the form e ^–E/E ₀, E ₀ being about 3.4 and 3.7 keV for the peaks at 1841 and 1942 UT, respectively. Assumption of a similar spectrum during the decay phase indicates that the spectral index E ₀ decreased nearly exponentially with time.The OGO-III ion chamber, which is also sensitive to protons 12 MeV, observed a solar particle event starting at 2100 UT on 23 May. It could not be determined uniquely which of the two principal X-ray peaks was associated with the particle event, and in fact both may have contributed. The particle intensity reached its maximum value at 1003 UT on 25 May 1967. The equivalent peak radiation dosage was 24 R/hour behind the 0.22 g cm^–2 thick aluminum wall of the chamber. This peak radiation dosage was considerably smaller than the maximum dosage (60 R/hour) during the 2 September 1966 solar particle event, the largest event observed so far by the OGO-I and OGO-III satellites. The temporal relationship between the solar X-ray and particle events on 23 May 1967 was similar to that observed in the solar flare events on 7 July 1966, 28 August 1966 and 27 February 1967.     

Theoretical model of flares and prominences

A theoretical model of flare which explains observed quantities in H, EUV, soft X-ray and flare-associated solar wind is presented. It is assumed that large mass observed in the soft X-ray flare and the solar wind comes from the chromosphere by the process like evaporation while flare is in progress. From mass and pressure balance in the chromosphere and the corona, the high temperature in the soft X-ray flare is shown to be attained by the larger mass loss to the solar wind compared with the mass remained in the corona, in accord with observations. The total energy of 10³² erg, the electron density of 10^13.5 cm^–3 in H flare, the temperature of the X-ray flare of 10^7.3K and the time to attain maximum H brightness (600 s) are derived consistent with observations. It is shown that the top height of the H flare is located about 1000 km lower than that of the active chromosphere because of evaporation. So-called limb flares are assigned to either post-flare loops, surges or rising prominences.The observed small thickness of the H flare is interpreted by free streaming and/or heat conduction. Applications are suggested to explain the maximum temperature of a coronal condensation and the formation of quiescent prominences.     

Non-thermal processes in large solar flares

We analyze particle acceleration processes in large solar flares, using observations of the August, 1972, series of large events. The energetic particle populations are estimated from the hard X-ray and γ-ray emission, and from direct interplanetary particle observations. The collisional energy losses of these particles are computed as a function of height, assuming that the particles are accelerated high in the solar atmosphere and then precipitate down into denser layers. We compare the computed energy input with the flare energy output in radiation, heating, and mass ejection, and find for large proton event flares that:

1. The ～10–10² keV electrons accelerated during the flash phase constitute the bulk of the total flare energy.
2. The flare can be divided into two regions depending on whether the electron energy input goes into radiation or explosive heating. The computed energy input to the radiative quasi-equilibrium region agrees with the observed flare energy output in optical, UV, and EUV radiation.
3. The electron energy input to the explosive heating region can produce evaporation of the upper chromosphere needed to form the soft X-ray flare plasma.
4. Very intense energetic electron fluxes can provide the energy and mass for interplanetary shock wave by heating the atmospheric gas to energies sufficient to escape the solar gravitational and magnetic fields. The threshold for shock formation appears to be ～10³¹ ergs total energy in >20 keV electrons, and all of the shock energy can be supplied by electrons if their spectrum extends down to 5–10 keV.
5. High energy protons are accelerated later than the 10–10² keV electrons and most of them escape to the interplanetary medium. The energetic protons are not a significant contributor to the energization of flare phenomena. The observations are consistent with shock-wave acceleration of the protons and other nuclei, and also of electrons to relativistic energies.
6. The flare white-light continuum emission is consistent with a model of free-bound transitions in a plasma with strong non-thermal ionization produced in the lower solar chromosphere by energetic electrons. The white-light continuum is inconsistent with models of photospheric heating by the energetic particles. A threshold energy of ～5×10³⁰ ergs in >20 keV electrons is required for detectable white-light emission.

The highly efficient electron energization required in these flares suggests that the flare mechanism consists of rapid dissipation of chromospheric and coronal field-aligned or sheet currents, due to the onset of current-driven Buneman anomalous resistivity. Large proton flares then result when the energy input from accelerated electrons is sufficient to form a shock wave.     

Evidence for a two-component injection of cosmic rays from the solar flare of 1969, March 30

The solar flare of 1969 March 30, occurring 20° behind the west limb, produced very extensive 80 MHz radio emission at the Sun, and gave rise to the deployment of cosmic radiation over 360°long, in interplanetary space. The wide spread of this event may reflect a similar spread of coronal magnetic fields from the flare site. We interpret the solar proton data recorded by spacecraft at two separate points both at 1 AU, in terms of a two-component injection of particles at the Sun consisting of: (i) a soft component which arrived promptly; (ii) a harder component which arrived later. The radio spectral and positional data provide evidence of shock waves which propagated far and wide from the flare; we attribute the precursor injection of the soft ( 10 MeV) proton component to one of these shock waves.Radiophysics Publication RPP 1590, May, 1972.Now at University of California, LASL, Los Alamos, N.M., U.S.A.     

Relativistic electrons from the Sun observed by IMP-4

Data are presented from the IMP-4 satellite of 0.3–12 MeV electrons from the Sun between May 24, 1967 and May 2, 1969. Correlations with contemporary proton intensity increases at energies above 1 MeV are studied. Classical solar flare events such as those frequently observed from 30°W–60°W in solar longitude are not discussed. Categories of unusual events are defined and examples of each type are given. Discussion of these events centers around the emission and propagation of energetic particles from the point of origin on the Sun to the Earth. The results of this study are the following: (1) The differential electron energy spectrum (0.3–12 keV) from solar flares appears to be a constant of the flare process, with the spectral index = (-)3.0 ± 0.2. (2) Particle emission from solar flares contains a prompt component, which is injected into the interplanetary medium beyond the Sun and which is responsible for the diffusion characteristics of solar particle events, and a delayed component which is effectively contained in the lower solar atmosphere where it diffuses typically ± 100° in longitude and gradually escapes into interplanetary space. The delayed component gives rise to the corotating features commonly observed after the impulsive and diffusive onset from the prompt component. This is not the same as the two component model discussed by Lin (1970a) in which 40 keV electrons are often observed as a separate phenomenon and frequently precede higher energy particles observed at 1 AU. (3) Storage of electrons > 300 keV and protons > 1 MeV is essential to explain emission and propagation characteristics of solar particle events. In some rare cases the storage mechanism appears to be very efficient, culminating in a catastrophic decay of the trapping region. (4) The events with low proton/electron ratios all occur at least three weeks after the previous relativistic electron producing flare.     

Does solar flare activity lag behind sunspot activity?

Recently, Wheatland and Litvinenko (2001) have suggested that over the solar cycle both the flaring rate and the magnetic free energy in the corona lag behind the energy supply to the system. To test this model result, we analyzed the evolution of solar flare occurrence with regard to sunspot numbers (as well as sunspot areas), using H flare data available for the period 1955–2002, and soft X-ray flare data (GOES 1–8 Å) for the period 1976–2002. For solar cycles 19, 21, and 23, we find a characteristic time lag between flare activity and sunspot activity in the range 1015 months, consistent with the model predictions by Wheatland and Litvinenko (2001). The phenomenon turns out to be more prominent for highly energetic flares. The investigation of solar activity separately for the northern and southern hemisphere allows us to exclude any bias due to overlapping effects from the activity of both hemispheres and confirms the dynamic relevance of the delay phenomenon. Yet, no characteristic time lag >0 is found for solar cycles 20 and 22. The finding that in odd-numbered cycles flare activity is statistically delayed with respect to sunspot activity, while in even-numbered cycles it is not, suggests a connection to the 22-year magnetic cycle of the Sun. Further insight into the connection to the 22-year magnetic cycle could possibly be gained when a 22-year variation in the energy supply rate is taken into account in the Wheatland and Litvinenko (2001) model. The existence of a 22-year modulation in the energy supply rate is suggested by the empirical Gnevyshev – Ohl rule, and might be caused by a relic solar field.     

Gamma-ray and microwave evidence for two phases of acceleration in solar flares

Relativistic electrons in large solar flares produce gamma-ray continuum by bremsstrahlung and microwave emission by gyrosynchrotron radiation. Using observations of the 1972, August 4 flare, we evaluate in detail the electron spectrum and the physical properties (density, magnetic field, size, and temperature) of the common emitting region of these radiations. We also obtain information on energetic protons in this flare by using gamma-ray lines. From the electron spectrum, the proton-to-electron ratio, and the time dependences of the microwave emission, the 2.2 MeV line and the gamma-ray continuum, we conclude that in large solar flares relativistic electrons and energetic nuclei are accelerated by a mechanism which is different from the mechanism which accelerates 100 keV electrons in flares.Research supported by NASA Grant 21-002-316 at the University of Maryland, College Park.