Session 87 - Star Parameters.
Oral session, Thursday, June 13

## [87.02] Stellar Transition Region Line Profiles, Redshifts, and Densities Based on GHRS Data

B. E. Wood (JILA, U. Colorado)

Our analysis of high-resolution GHRS spectra of active late-type stars (e.g. AU Mic and HR 1099) shows that transition region (TR) emission lines formed near 10^5 K possess broad, non-Gaussian wings. These profiles can be modeled as the sum of two Gaussians --- a narrow component and a very broad component. GHRS spectra of relatively inactive stars such as \beta Gem and \alpha Cen A+B show weak or no detectable broad components. For the main sequence stars, the widths of the narrow components suggest subsonic nonthermal velocities, and there appears to be a correlation between these nonthermal velocities and stellar surface gravity. For the evolved stars with lower surface gravities, the nonthermal velocities suggested by the narrow component are at or just above the sound speed. Nonthermal velocities computed from the widths of the broad components are always supersonic. We find that the fractional contribution of the broad component to the total line flux is well correlated with both the C IV surface flux, a good measure of the TR heating rate, and the X-ray surface flux, a good measure of the coronal heating rate. We propose that the narrow components are diagnostics for shock wave heating, most likely due to MHD waves, and that the broad components are diagnostics for microflare heating. A solar analog for the broad components might be the explosive events'' detected by the HRTS experiment.

We also discuss the velocities of the UV emission lines seen in our GHRS data, and the TR densities suggested by O IV] density diagnostics. For eight of the ten stars in our sample, the O IV] lines suggest electron densities between \log n_e=9.5 and \log n_e=10.0. The exceptions are the inactive K0 III star \beta Gem, which appears to have \log n_e<9.0, and the very active dMe star AU Mic, which may have densities of \log n_e>10.4. We find that the TR lines of all the stars in our sample are redshifted, similar to redshifts seen on the Sun. Surprisingly, the magnitude of these redshifts is not correlated with stellar activity.