Solar Chromospheric Flares: Observations in Ly-alpha and H-alpha and Radiative Hydrodynamic Simulations

Abstract

This thesis is divided into two main parts: a multi-wavelength observational study of solar flares, focusing mainly in the chromosphere in Ly-alpha and H-alpha, and an application of a radiative transfer code and a radiative hydrodynamic code, to compare the results obtained by observations with the simulated ones.

The Ly-alpha emission is a very interesting line because it is a natural tracer of the solar activity in the chromosphere. The Transition Region And Coronal Explorer satellite observed a small number of flares in the Ly-alpha passband, but apart from this, these events have not often been observed in this strong chromospheric line. Because TRACE has a broad Ly-alpha channel, in order to estimate the ``pure'' Ly-alpha emission, we had to apply an empirical correction.

We found that there is a reasonable coverage in TRACE 1216 A and the TRACE 1600 A for two different flares: on 8 September 1999 and on 28 February 1999. Studying them we estimated, for the first time, the pure Ly-alpha flare signature, being on the order of 10^{25} erg/s at the flare peak.

The study of the first flare gave us the possibility to calculate the electron energy budget using the X-ray data from Yohkoh/HXT in the context of the collisional thick target model, finding that the Ly-$\alpha$ power is less than 10% of the power inferred by the electrons.

The morphology and evolution of the second flare were described in different wavelengths by using imaging data acquired by TRACE and by BBSO in white light and in H-alpha. We studied the magnetic topology using the magnetic field provided by SOHO/MDI, extrapolating the photospheric magnetic field lines, assuming a potential field. We found different morphologies in the magnetic configuration before and after the flare, confirming the occurrence of a reconnection process.

The H-alpha line is the most important line in the chromosphere. We studied the H-alpha emission of a flare which occurred on 3 July 2002 using some spectroscopical observations from the Ondrejov Observatory. Analyzing the available data in other wavelengths, we made a morphological study of the active region from three hours before the flare to seven hours after it.

The results obtained by observations, both in the form of integrated intensity as a function of time, and detailed line profiles, motivated the second part of the thesis. In this, we used a radiative transfer code applying different atmospheric models as input parameters in order to compute the hydrogen spectral lines and study how they are affected by the temperature and microturbulent stratification. In particular, the intensity of the Ly-alpha and H-alpha lines is related to the temperature stratification of the atmospheric model, the position of the transition region being a key factor. The variation of the microturbulent velocity does not significantly affect the resulting intensities, but we observed that an increase of the microturbulent velocity broadens the line profiles.

The RADYN Radiative HydroDynamic code was applied to solar flares, modelling a flare loop from its footpoints in the photosphere to its apex in the corona by adding non-thermal heating at the lower atmosphere and soft X-ray irradiation. The majority of this work was to deal with investigating the dynamical response of the solar chromosphere to energy injected in the form of non-thermal electrons during solar flares. We studied the flare energy transport and radiation production in the chromosphere as well as the H-alpha and Ly-alpha emission. The Ly-alpha intensity is affected by the flux of the initial beam of electrons injected at the top of the loop, while the H-alpha intensity appears to be less affected by the flare model.

Comparing the observational results in Ly-alpha and H-alpha with the computed ones from the radiative code and the RADYN code, we found that the RADYN code fits better the H-alpha intensities to the observations than the Ly-alpha intensities, concluding that the code gives a better description of processes in the lower chromosphere than those in the upper layers.