SpringerLink
Forum Springer Astron. Astrophys.
Forum Whats New Search Orders


Astron. Astrophys. 356, 1067-1075 (2000)

Previous Section Next Section Title Page Table of Contents

1. Introduction

It is widely accepted that a solar flare results from the rapid release of free magnetic energy near the top of active region loops. The dissipation of this energy in the coronal release sites leads to acceleration of electrons and ions to high energies and to direct heating of the plasma and plasma motions. There are basically two forms of energy transport from the primary energy release sites to the regions of energy dissipation in the low-corona and chromosphere, at the footpoints of the flaring loops. The study of the morphology and timing of the H[FORMULA] line emission with respect to that of the microwave and hard X-ray emission provides a useful test through which one can discriminate between these forms of energy tranport, provided by the different flare models.

The thermal model assumes that most of the energy released goes into the impulsive heating of the plasma near the release site to temperatures above 108 K (Brown et al. 1979; Smith & Lillequist 1979; Batchelor et al. 1985). Conduction fronts are formed in the loops and move at the ion sound speed (typically 100-1000 km s-1). The observed microwave and HXR emission arise then predominently from expanding hot sources confined behind the conduction front. H[FORMULA] starts to rise when the conduction front hits the chromosphere [FORMULA] 10 to 20 s after the onset of the microwave and HXR emission for typical loop parameters.

The second scenario, known as the non-thermal thick target model, assumes that most of the energy release goes into the acceleration of particles near the release site (e.g. Brown 1971). Non-thermal electrons which stream down along the magnetic field reach the low corona and upper chromosphere in less than 1 s where they produce the HXR bremsstrahlung emission by thick-target interaction with ambient ions. Below, in the chromosphere, the same electrons are dumped and heat the ambient medium which leads to an enhanced H[FORMULA] emission. The rises of the H[FORMULA], microwave and HXR emission are thus expected to be simultaneous within less than 1 s. It should be noted that [FORMULA] 0.5 MeV protons can similarily produce enhanced H[FORMULA] emission (Simnett 1986). Unfortunately, there are no unambiguous radiative signatures of such low energy protons in the solar atmosphere. Observations of linear polarization in the H[FORMULA] line may, in principle, help to discriminate between electrons and protons (e.g. Vogt & Hénoux 1999), but the time resolution (1 to several minutes) of the few observations made so far is much larger than the typical time scale of energy transport by particles. Canfield & Gayley (1987) have computed the temporal evolution of the H[FORMULA] line profile using models of Fisher et al. (1985a, 1985b) which simulate the dynamic response of a loop atmosphere to Coulomb heating by a beam of accelerated electrons. They show that the temporal H[FORMULA] response arises from three distinct physical processes whose relative importance varies over the line profile: a temperature response, an ionization response and a hydrodynamic response. The first two processes are responsible for the fastest time scales, typically [FORMULA] 0.1 s for the temperature response which has its largest amplitude at line center, and [FORMULA] 0.5 s for the ionization response which is most apparent in the blue wing. The hydrodynamic response, which is related to the formation of a chromospheric condensation, gives rise to longer time scales (several seconds) and is most apparent in the red wing. Heinzel (1991) who performed similar calculations by using atmosphere models of Karlický (1990) further predicts an anticorrelation between HXR and H[FORMULA] sub-second pulses.

Studies of the relative timing between H[FORMULA] and HXR (or microwave observations), all made with [FORMULA] 1 s time resolution, have shown that both forms of energy transport may exist at different sites during the same flare (e.g. Kämpfer & Schöchlin 1982; Kämpfer & Magun 1983; Wülser & Kämpfer 1986). Indeed, at some sites H[FORMULA] is delayed by 10 s or more with respect to the microwave or HXR while at other sites a synchronism (within 1 to 2 s) is observed between these different emissions. The latter supports the models describing the heating of the chromosphere by non-thermal electrons. A close similarity between the H[FORMULA] and HXR time profiles was also found in several flares where H[FORMULA] data with sub-second time resolution were available (Wülser & Marti 1989; de la Beaujardière et al. 1992; Neidig et al. 1993). However, the data were smoothed with integration times above 1 s for noise reduction, so that a study of subsecond H[FORMULA] structures was not possible.

In this paper we present the comparison of fast (0.2 s) H[FORMULA] line center imaging observations together with fast microwave (0.1 s) and HXR (0.1-1 s) spectral measurements during the 1991 March 13 flare at 08 UTC. Sects. 2 and 3 present the instruments and the analysis of the observations, respectively. It appears that the time evolution of each of the four H[FORMULA] kernels observed during this flare can be related in a simple way to the HXR time profile. This finding is discussed in Sect. 4 in the framework of an electron-beam-heated chromosphere. The conclusions are summarized in Sect. 5.

Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 2000

Online publication: April 17, 2000
helpdesk.link@springer.de