Forum Springer Astron. Astrophys.
Forum Whats New Search Orders

Astron. Astrophys. 360, 1148-1156 (2000)

Previous Section Next Section Title Page Table of Contents

1. Introduction

Some of the most widely used techniques for measuring the physical parameters of optically thin plasmas involve the use of extreme ultraviolet emission line intensities. These techniques consist in comparing the observed line intensities and intensity ratios with theoretical estimates calculated as a function of the relevant physical parameter. Using these techniques, electron temperature, electron density, plasma Emission Measure and Differential Emission Measure, and element abundances have been extensively investigated in the literature. These diagnostic techniques have permitted to address some of the most important unresolved issues in the physics of the solar and stellar coronae.

However, any plasma diagnostic technique involving EUV line intensities requires the knowledge of a large amount of atomic data and transition probabilities in order to be carried out; these are necessary to calculate the theoretical line intensities for a given ion to be compared with the observations. It is also usually assumed that the plasma is in ionization equilibrium; this assumption, which can be misleading in highly dynamic plasma, allows the use of the ion fraction datasets found in the literature.

Thus, together with the atomic data and transition probabilities necessary to calculate ions' emissivity, ion fractions are a possibile source of uncertainty in plasma diagnostic studies.

The ion fractions datasets available in the literature have been calculated by a number of authors using the state-of-the-art ionization and recombination rates available at the time of publication. However, progress in the theoretical models for ionization and recombination processes have led to significant changes in ionization and recombination rates. The differences in the rates used in the calculations may cause differences in the resulting ion fractions, and therefore may have important effects on the plasma parameters determined adopting these datasets.

Electron density and temperature of optically thin plasmas are usually determined by intensity ratios between lines emitted by the same ion (Mason & Monsignori Fossi 1994), so that any effect of the ion fraction is ruled out. However, many authors have used intensity ratios from lines of different ions to determine the electron temperature in in the solar corona under different physical conditions, so that uncertainties and changes in the ion fraction need to be taken into account.

Element abundance measurements and DEMdiagnostics are usually carried out using lines from several different ions, so that ion fractions might be an important source of uncertainty.

A few studies have been carried out in order to assess the uncertainties of ion fraction datasets and their effects on plasma diagnostics. Cheng et al. (1979a,b) compared ion abundance calculations from Jordan (1969, 1970), Summers (1974) and Jacobs et al. (1977) by means of Si viii, Fe xi, Fe xii and Fe xxi line width studies from solar spectra. They find that observations indicate that Summers (1974) ion abundances are probably less accurate than the other two. The same conclusion has been drawn by Feldman et al. (1981) using EUV line intensities from solar flares. All these studies indicate that differences in plasma diagnostics are expected when these ion fraction datasets are used.

In the recent past, Masai (1997) investigated the impact of uncertainties in the ionization and recombination rates on X-ray spectral analysis, finding that differences in the rates led to significant differences in iron abundance and plasma temperature measurements. Phillips & Feldman (1997) have used Yohkoh flare observations to check the ion fractions of He-like ions, concluding that the observed spectra were consistent with the adopted ion fractions at the 50% level of precision, and this led to changes in plasma diagnostic results. Allen 2000 investigated the effect of different ion fraction datasets on temperature diagnostics in an isothermal, quiet solar region off the solar disc using EUV line intensities, finding that the use of different ion fractions did not alter the measured temperature values. Finally, Landi & Landini (1999) investigated the effect of different ion fractions in the calculation of radiative losses of optically thin plasmas: they found differences in the results up to 40% at coronal temperatures.

The scope of the present work is to address the problem of the effect of uncertainties in the ion fraction datasets on plasma diagnostics, in order to assess the stability of the measured plasma parameters against ion fraction uncertainties. In the present work element abundances of elements with First Ionization Potential (FIP) greater than 10 eV (high-FIP elements) and smaller than 10 eV (low-FIP elements) are measured by means of EUV spectral line diagnostics using three different ion abundance datasets. We analyze SOHO-CDS observations of an active region on the solar disk which is relatively stable in time and shows a large number of structures in the field of view. Since the three ion fraction datasets used show remarkable differences, it is expected that results will change. As the diagnostic technique used in the present work involves DEManalysis, the present work also allows to check the effects of different ion fractions on DEMdiagnostics.

This check on element abundances and DEMdiagnostics is important. In fact the DEMis an important physical quantity for plasma modelling of solar and stellar coronae.

Element abundances measurements are necessary to address the FIP effect . This effect consists in the difference between the element abundances in the solar photosphere/chromosphere and in the overlying corona, which has been observed by means of spectroscopic analysis and solar energetic particles (SEP) data. This difference seems to be related to the FIP of the elements, in the sense that the abundance of elements with FIP [FORMULA] 10 eV is enhanced by a factor between 3 and 4 relatively to that of elements with FIP [FORMULA] 10 eV. Recent reviews of this effect may be found in Feldman (1992) and Feldman & Laming (2000). Similar effects have been found in stellar spectra (Laming et al. 1996; Drake et al. 1997; Laming & Drake 1999).

To date, no comprehensive theory has been developed which is able to account for such a behaviour of element abundances.

The present paper is structured as follows: the basic theory of emission line intensity is outlined in Sect. 2, and a quick comparison between different ion fractions datasets of a few ions relevant to the present study is carried out in Sect. 3. The observations are described in Sect. 4; Sect. 5 reports the results, which are discussed in Sect. 6.

Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 2000

Online publication: August 23, 2000