The X-ray observations of the ROSAT satellite (see Trümper 1983) provide the largest database to date for an X-ray study of stellar coronae. Most recent investigations have dealt mainly with the amplitude of the X-ray emission and its dependence on basic stellar parameters (see e.g. Fleming et al. 1995; Schmitt et al. 1995). In this study we will focus our interest on the analysis of the X-ray spectra and the determination of the coronal temperature structure.
A similar study has been performed by Schmitt et al. (1990), using X-ray spectra obtained with the IPC detector on board of the EINSTEIN observatory. The ROSAT data, however, offer considerable advantages over the EINSTEIN data: First, the spectral resolution of the ROSAT Positional Sensitive Proportional Counter PSPC ( at 1 keV; see Pfeffermann et al. 1987), is significantly higher as compared to that of the EINSTEIN IPC (). Second, much more late type stars have been observed in deep ROSAT pointings than in EINSTEIN observations.
We are aware that the amount of information that can be inferred from the moderate resolution PSPC spectra is quite limited and observations with higher spectral resolution would allow a much more detailed study of the coronal temperature structure. However, the handful of stars for which high resolution X-ray spectra were taken e.g. with the EINSTEIN Focal Plane Crystal Spectrometer or the EXOSAT transmission grating are much too few to derive general stellar spectral properties. In the last years, the EUVE satellite (see Bowyer & Malina 1991) has performed high resolution extreme-ultraviolet observations of several late type stars. The EUVE data are well suited for a determination of the coronal temperature distribution (for a review see Drake 1996), but unfortunately the number of observed stars again is very small. Thus, only the ROSAT data allow one to study a reasonably large sample of late type stars. However, we will keep in mind the EUVE results and refer to them whenever needed.
We will restrict our study to late type stars (spectral type between F and M), since for this class of stars a coronal origin of the X-ray emission is well established (see Pallavicini 1989). The ages of the stars in our sample range from years for the T Tauri stars (TTS) up to the age of the Sun ( years). To explain the motivation of this study, one just has to compare the X-ray properties of TTS (for a recent review see Montmerle et al. 1993) to those of the Sun: Most TTS are vigorous X-ray emitters with X-ray luminosities of up to erg/sec, which exceeds the solar level by about three orders of magnitude; furthermore, for many TTS the coronal temperatures inferred from the X-ray spectra are very high, K, as compared to typical solar coronal temperatures of just a few K. It has to be noted that the high levels of X-ray emission can not at all be simply explained by the higher coronal temperatures: since the radiative cooling function of the coronal plasma decreases for temperatures between K and K (cf. Fig. 1 in Bruner & McWhirter 1988), a higher coronal temperature takes even more emission measure to produce the required level of emission.
One of the questions we want to address is, whether the coronal properties are a a function of age. Another point we will investigate is, what change in condition from the Sun can produce the high X-ray luminosities and coronal temperatures. We are especially interested in what determines the coronal temperatures.
Young stars are also known to show very powerful X-ray flares with a total energy output exceeding that of large solar flares by up to 5 orders of magnitude (e.g. Montmerle et al. 1983; Preibisch et al. 1993, 1995). In this study we explicitly exclude X-ray flares, since we want to study the quiescent corona (for an overview on X-ray flares on young stars see e.g. Preibisch & Neuhäuser 1995).
© European Southern Observatory (ESO) 1997
Online publication: June 30, 1998