Electron trapping in evolving coronal structures during a large gradual hard X-ray/radio burst

Gradual hard X-ray/radio bursts are characterized by their long duration, smooth time profile, time delays between peaks at different hard X-ray energies and microwaves, and radiation from extended sources in the low and middle corona. Their characteristic properties have been ascribed to the dynamic evolution of the accelerated electrons in coronal magnetic traps or to the separate acceleration of high-energy electrons in a second step process. The information available so far was drawn from qualitative considerations of time profiles or even only from the common occurrence of emissions in different spectral ranges. This paper presents model computations of the temporal evolution of hard X-ray and microwave spectra, together with a qualitative discussion of radio lightcurves over a wide spectral range, and metric imaging observations. The basic hypothesis investigated is that the peculiar gradual features can be related to the dynamical evolution of electrons injected over an extended time interval in a coronal trap, with electrons up to relativistic energies being injected simultaneously. The analyzed event (26 April, 1981) is particularly challenging to this hypothesis because of the long time delays between peaks at different X-ray energies and microwave frequencies. The observations are shown to be consistent with the hypothesis, provided that the electrons lose their energy by Coulomb collisions and possibly betatron deceleration. The access of the electrons to different coronal structures varies in the course of the event. The evolution and likely destabilisation of part of the coronal plasma-magnetic field configuration is of crucial influence in determining the access to these structures and possibly the dynamical evolution of the trapped electrons through betatron deceleration in the late phase of the event.     

On the formation of coronal cavities

We present a theoretical study of the formation of a coronal cavity and its relation to a quiescent prominence. We argue that the formation of a coronal cavity is initiated by the condensation of plasma which is trapped by the coronal magnetic field in a closed streamer and which then flows down to the chromosphere along the field lines due to lack of stable magnetic support against gravity. The existence of a coronal cavity depends on the coronal magnetic field strength; with low strength, the plasma density is not high enough for condensation to occur. Furthermore, we suggest that prominence and cavity material is supplied from the chromospheric level. Whether a coronal cavity and a prominence coexist depends on the magnetic field configuration; a prominence requires stable magnetic support.We initiate the study by considering the stability of condensation modes of a plasma in the coronal streamer model obtained by Steinolfson et al. (1982) using a 2-D, time dependent, ideal MHD computer simulation; they calculated the dynamic interaction between outward flowing solar wind plasma and a global coronal magnetic field. In the final steady state, they found a density enhancement in the closed field region with the enhancement increasing with increasing strength of the magnetic field. Our stability calculation shows that if the density enhancement is higher than a critical value, the plasma is unstable to condensation modes. We describe how, depending on the magnetic field configuration, the condensation may produce a coronal cavity and/or initiate the formation of a prominence.NRC Research Associate.     

Evolution of X-ray cavities

A wide range of recent observations have shown that active galactic nuclei (AGN) driven cavities may provide the energy source that balances the cooling observed in the centres of 'cool-core' galaxy clusters. One tool for better understanding the physics of these cavities is their observed morphological evolution, which is dependent on such poorly understood properties as the turbulent density field and the impact of magnetic fields. Here, we combine numerical simulations that include subgrid turbulence and software that produces synthetic X-ray observations to examine the evolution of X-ray cavities in the absence of magnetic fields. Our results reveal an anisotropic size evolution of the cavities that is dramatically different from simplified, analytical predictions. These differences highlight some of the key issues that must be accurately quantified when studying AGN-driven cavities, and help to explain why the inferred pV energy in these regions appears to be correlated with their distance from the cluster centre. Interpreting that X-ray observations will require detailed modelling of effects, including mass entrainment, distortion by drag forces and projection. Current limitations do not allow a discrimination between purely hydrodynamic and magnetically dominated models for X-ray cavities.     

Flow-tube dynamics and coronal holes

We reexamine the well-known polytropic flow-tube model of the expanding solar corona, and find that as the divergence of the flow tube increases the expansion speed increases throughout the flow, over a stated parameter range. Corresponding to a specified flow-tube geometry the terminal speed of the fluid may be far in excess of the value corresponding to purely spherically symmetric flow. The implications of the results for the modelling of high-speed streams emanating from coronal holes are discussed.     

Some aspects of coronal streamer dynamics

Observations of coronal streamers suggest that these configurations are stationary in terms of the convective time scale, but not with respect to the diffusive processes. In this paper, the diffusive time scale is estimated and the thickness of a stationary streamer where convective and diffusive processes are balanced turns out to be very small.Instead of balancing the terms in the magnetic equation to obtain a zero time derivative, we suggest that the diffusive part is negligible and therefore, the convective term must be zero in the largest part of the streamer. This can be obtained by introducing flux transport through the streamer.We show that the arising convective model is to be understood from evolutionary arguments; the upward flux transport leads to a time scale for the life of a streamer, related to the evolution of the underlying magnetic field. Slight differences occur between stationary models and the present model; these differences are shown not to be in conflict with radio observations.     

The X-ray signature of solar coronal mass ejections

The coronal response to six solar X-ray flares has been investigated. At a time coincident with the projected onset of the white-light coronal mass ejection associated with each flare, there is a small, discrete soft X-ray enhancement. These enhancements (precursors) precede by typically 20 m the impulsive phase of the solar flare which is dominant by the time the coronal mass ejection has reached an altitude above 0.5 R . We identify motions of hot X-ray emitting plasma, during the precursors, which may well be a signature of the mass ejection onsets. Further investigations have also revealed a second class of X-ray coronal transient, during the main phase of the flare. These appear to be associated with magnetic reconnection above post-flare loop systems.NCAR is sponsored by the National Science Foundation.     

X-ray observations of limb flare loops and post-flare coronal arch

We present observations of another post-flare arch following an eruptive flare, detected in X-ray lines above the western solar limb on 2 May 1985.     

Radio and X-ray imaging observations of solar flares and coronal transients

We present a review of selected studies based upon simultaneous radio and X-ray observations of solar flares and coronal transients. We use primarily the observations made with large radio imaging instruments (VLA, BIMA, Nobeyama, and Nançay) along with Yohkoh/SXT and HXT and CGRO experiments. We review the recent work on millimeter imaging of solar flares, microwave and hard X-ray observations of footpoint emission from flaring loops, metric type IV continuum bursts, and coronal X-ray structures. We discuss the recent studies on thermal and nonthermal processes in coronal transients such as XBP flares, coronal X-ray jets, and active region transient brightenings.Dedicated to Cornelis de Jager     

Distribution of lifetimes for coronal soft X-ray bright points

We have measured the lifetimes of all compact emission features visible on three sets of high time resolution soft X-ray images. The spectrum of lifetimes is found to be heavily weighted toward short lifetimes. The number of features present on the disk which live 2–48 hours is at least ten times as great as the number living more than 48 hours. The distribution of lifetimes can be fit in all three cases by a four-parameter function N(t) = N _s exp(-t/ _s ) + N _L exp(-t/ _L ), with _s = 8.7±0.2, _L = 35±4 and N _s 10N _L . Features living two days or less have a very broad latitude distribution (Golub et al., 1974, 1975) whereas nearly all longer-lived features are found within 30° of the equator. The growth rates of long-lived vs short-lived points are the same to within 20%, the major difference being that long-lived points continue to grow and generally reach larger sizes.Harvard College Observatory/Smithsonian Astrophysical Observatory.     

Soft X-ray observations of large-scale coronal active region brightenings

One-hundred fifty-six large-scale enhancements of X-ray emission from solar active regions were studied on full-disk filterheliograms to determine characteristic morphology and expansion rates for heated coronal plasma. The X-ray photographs were compared with H observations of flares, sudden filament disappearances, sprays and loop prominence systems (LPS). Eighty-one percent of the X-ray events were correlated with H filament activity, but only forty-four percent were correlated with reported H flares. The X-ray enhancements took the form of loops or arcades of loops ranging in length from 60 000 km to 520 000 km and averaging 15 000 km in width. Lifetimes ranged from 3 hr to >24 hr. Event frequency was 1.4 per day. X-ray loop arcades evolved from sharp-edged clouds in cavities vacated by rising H filaments. Expansion velocities of the loops were 50 km s^-1 immediately after excitation and 1–10 km s^-1 several hours later. These long-lived loop arcades are identified with LPS, and it is suggested that the loops outlined magnetic fields which were reconnecting after filament eruptions. Another class of X-ray enhanced loops stretched outside active regions and accompanied sprays or lateral filament ejections. H brightenings occurred where these loops intersected the chromosphere. Inferred excitation velocities along the loops ranged between 300 and 1200 km s^-1. It is suggested that these loops outlined closed magnetic fields guiding slow mode shocks from flares and filament eruptions.     

Strong coronal shocks and ‘thermal’ solar X-ray bursts

I find that a one-dimensional strong coronal shock (M _s 3) will grow outward until the Mach number (M _s ) ceases to increase with height (dM _s /dh = 0). The shock is driven by the pressure gradient and it is damped by gravity and by energy losses (radiative and conductive). The driving and damping terms reach equilibrium for M _s - 4.Standard shock jump conditions for M _s - 4 lead to post-shock temperatures in the corona in the range 10⁷ to 1.8 × 10⁷K and emission measures from 3.8 × 10⁴⁷ to 3.8 × 10⁴⁸ cm^-3. For isolated simple events, I predict an exponential decay of the emission measure with decay times in the range 1 6.5 min.In a detailed study of over 4000 X-ray bursts, Drake (1970) compares 1 to 6 keV X-ray data with 7.7 to 12.5 keV X-ray data (the thermal component) and finds ranges for the temperatures of 1.2 × 10⁷ to 1.8 × 10⁷K, for the emission measures of 5.1 × 10⁴⁷ to 3.8 x 10⁴⁸ cm^-3 and for the decay times 0.5 20 min. He also finds that the emission measure varies ... both from event to event and within the event, by more than a factor of two.The agreement between the predictions and the observations makes it appear that a strong shock in the corona will produce a post-shock state that yields the observed characteristics of the soft component of X-ray bursts (the thermal X-rays).I give several examples where sprays and fast eruptive prominences 1} \right)$$ " align="middle" border="0"> , that are not associated with solar flares, are associated with thermal X-ray bursts. There were two slow eruptive prominences (M 1) in the sample, and neither of them yielded a detectable X-ray burst.Now at the Dept. of Physics and Astronomy (NASA), Univ. of New Mexico; Albuquerque, N.M. 87106.     

X-ray bright points,coronal heating and the solar cycle

Recent Skylab observations about the bright points in the solar X-ray images seem to confirm an essential prediction of a model proposed by this author for the appearance and the disappearance of the photospheric fields during a solar cycle.The segments of the individually rising strands of the fundamental flux-loops proposed in the model may lead to the X-ray bright points with the observed properties.The emergence of such strands may substantially contribute to the coronal heating at different heights.     

Hard X-ray imaging of a solar gradual hard X-ray burst on April 1, 1981

An intense solar X-ray burst occurred on April 1, 1981. X-ray images of this gradual hard X-ray burst were observed with the hard X-ray telescope aboard the Hinotori satellite for the initial ten minutes of rise and maximum phases of the burst. The hard X-ray images (13–29 keV) look like a large loop without considerable time variation of an elongated main source during the whole observation period. The main X-ray source seems to lie along a ridge of a long coronal arcade 2 × 10⁴ km above a neutral line, while a tangue-like sub-source may be another large coronal loop although the whole structure of the X-ray source looks like a large semi-circular loop. Both nonthermal and hot thermal (3–4 × 10⁷ K) electrons are contributing to the source image. The ratio of these components changed in a wide range from 2.3 to 0.4 during the observation, while the image was rather steady. It suggests that both heating and accelerations of electrons are occurring simultaneously in a common source. Energetic electrons of 15–30 keV would be collisionally trapped in the coronal magnetic loops with density of the order of 10¹¹ cm^–3.     

Multifrequency radio observations of coronal holes,filaments and cavities on RATAN–600

On the basis of multifrequency solar radio observations made on RATAN–600 radiotelescope with high spatial resolution at nine wavelengths in the 2–32–wavelength range is shown that filaments and cavities are well detected on the solar scans at short centimeter wavelengths as the regions of low radio brightness with angular dimensions of 25′–80′ in E—W direction. The tendency of decreasing radio sizes for cavities and filaments from 2.0 to 8.0 cm is observed. The coronal hole (CH) is more contrast in the range of 8–32 cm. The radio size of CH in E—N direction increases from 2′ (at 8.2) to 5′.0 (at 31.6 cm). The spectra of the brightness temperature of CH and the quiet Sun are obtained. The brightness temperature of CH is twice lower than that of the quiet Sun at wavelength of 31.6 cm.     

Soft X-ray observation of a large-scale coronal wave and its exciter

Recent extreme ultraviolet (EUV) observations from SOHO have shown the common occurrence of flare-associated global coronal waves strongly correlated with metric type II bursts, and in some cases with chromospheric Moreton waves. Until now, however, few direct soft X-ray detections of related global coronal waves have been reported. We have studied Yohkoh Soft X-ray Telescope (SXT) imaging observations to understand this apparent discrepancy, and describe the problems in this paper. We have found good X-ray evidence for a large-scale coronal wave associated with a major flare on 6 May 1998. The earliest direct trace of the wave motion on 6 May consisted of an expanding volume within 20 Mm (projected) of the flare-core loops, as established by loop motions and a dimming signature. Wavefront analyses of the soft X-ray observations point to this region as the source of the wave, which began at the time of an early hard X-ray spike in the impulsive phase of the flare. The emission can be seen out to a large radial distance (some 220 Mm from the flare core) by SXT, and a similar structure at a still greater distance by EIT (the Extreme Ultraviolet Imaging Telescope) on SOHO. The radio dynamic spectra confirm that an associated disturbance started at a relatively high density, consistent with the X-ray observations, prior to the metric type II burst emission onset. The wavefront tilted away from the vertical as expected from refraction if the Alfvén speed increases with height in the corona. From the X-ray observations we estimate that the electron temperature in the wave, at a distance of 120 Mm from the flare core, was on the order of 2–4 MK, consistent with a Mach number in the range 1.1–1.3. Supplementary material to this paper is available in electronic form at http://dx.doi.org/10.1023/A:1022904125479 deceased     

X-ray resonance scattering in a spherically symmetric coronal model

In the solar corona the opacities of some of the prominent X-ray emission lines are on the order of 1 over typical coronal path lengths. We present and discuss a particular solution of the radiative transfer problem involving an extended, spherically symmetric coronal shell radiating isotropic, homogeneous emission in which single-scattering also takes place. Within the context of this simplified model we find that scattered radiation is an important contribution to the total emergent resonance line flux and that for the He-like family of resonance (r), intercombination (i), and forbidden (f) lines, the ratio G=(f + i)/r would decrease as a function of optical depth for disk-center emission in an extended spherically symmetric corona.     

Observational evidence of continual heating in X-ray emitting coronal loops

A 90 s time resolution study of the soft X-ray emission from three active region loops shows the emission to be constant to about two percent over the half hour period of observation. Soft X-ray observations in two wavebands are used to deduce the temperature and density of these loops. The data unambiguously demonstrate that energy is supplied to each loop during the observations. If heating is due to discrete events, the time interval between events is shown to be less than 10 min, which is short relative to the radiative cooling time of the loops.Skylab Solar Workshop Post-Doctoral Appointee, 1975–1977. The Skylab Solar Workshops are sponsored by NASA and NSF and managed by the High Altitude Observatory, National Center for Atmospheric Research.     

Oscillations of coronal electron traps inferred from hard X-ray data

The work of Brown and Hoyng (1975) on the betatron acceleration of hard X-ray source electrons trapped in a vibrating flux tube is generalised to include Fermi acceleration by the varying transverse field. This development can explain the trajectory of bursts in a plot of equivalent thick-target electron flux versus spectral index as inferred from observations obtained by ESRO TD1A. Specifically the loops observed in this , diagram, unexplained in Brown and Hoyng's anslysis are accounted for by a changing phase relationship between the varying field strength f(t) and magnetic scale length g(t)- Application of the formalism to the detailed TD1A observations of the large events of 1972, August 4 and 7, allows inference of the evolution of f(t), g(t) in these events.Now at: Space Sciences Laboratory, University of California, Berkeley, U.S.A.