4. Calculation of the stellar velocity
4.1. Radial velocity
The radial velocity was measured in two different ways. First, by doing gaussian fits to a single, isolated photospheric absorption line (Fe I 6546.25Å) and, second, by applying a cross-correlation technique.
The main problems with determining accurate velocities from gaussian fits are the intrinsic line profile variability and the blending of photospheric lines due to a combination of rapid rotation and late-spectral type. The heliocentric velocity derived by using this method was km s-1, a value which is close to that obtained by Jeffries, James & Bromage (1994), km s-1. In both cases, the error is given as a standard deviation.
This result was confirmed by cross correlating individual spectra with the spectrum of HIC 67155, a non active star with a M3 spectral type, comparable to that of RE 1816+541. From the shift of the cross-correlation peak an average velocity of km s-1 was derived. This does not include results from the data obtained during the last night since they gave much larger uncertainties and the quality of the cross correlation was considerably worse. For this reason, the value finally used to refer the spectra to the stellar rest frame was km s-1, obtained as an average of the velocities from cross correlations for the first two nights and the velocity from gaussian fits to photospheric lines done on the last night.
4.1.1. Rotational velocity
As a measure of rotation, the sini has been calculated from cross correlation analyses, using again the spectrum of HIC 67155 as a template. In order to obtain a relationship between the width of the cross-correlation peak and sini, the template spectrum has been broadened to different velocities. By cross-correlating the resulting spectra with respect to the unbroadened template spectrum, a width value was obtained for each velocity. The calibration between width and rotational velocity was then established by doing a linear least-squares fit to those points. The sini value obtained for RE 1816+541 is km s-1, in agreement with results found previously by Jeffries, James & Bromage (1994).
© European Southern Observatory (ESO) 1998
Online publication: August 27, 1998