The FTS has a dual role. By integrating the spectrum it provides a UV broad band photometric signal for the seismology, helping mode identification by providing information about propagation through the stellar atmosphere. Spectral information from the MgII lines, though not suitable for Doppler imaging as they are optically thick, provides important diagnostics on radiative losses from chromospheres, and are the only chromospheric lines that can be observed with sufficient signal-to-noise ratio for cool giants. In solar-type stars the optically thin SiIII 189.2nm line, formed in the upper chromosphere can be used for Doppler imaging.
The FTS produces spectra of all the stars in the field of view with a resolution of 30000. The design of the FTS is based on the Imperial College - Chelsea Instruments FT500 layout (Thorne et al. 1987) which was also adopted in the ESA SIMURIS study (Coradini et al. 1991), though at the moment we do not propose working below the spectrosil cut off at 175nm. In this layout both of the interferometer outputs are available, and camera mirrors image the field of view onto anti-reflection coated, thinned CCDs which provide sensitivity in this wavelength range ( Schaefer et al. 1990, Delamere et al. 1990). However, we recommend studying the possibility of extending the wavelength range, by using for example a grating beamsplitter, or a MgF2 substrate beamsplitter. Recent progress in this area is very promising, and if successful the UV line monitor would no longer be necessary.
The UV photometric signal is less precise than that of the white light photometer, but can still be used for mode identification. This has been demonstrated by ground based solar measurements, where high quality velocity measurements were compared with very low quality spectra from ground based intensity measurements (Jimenéz et al. 1990). The phase relationship found, though more noisy, agrees very well with more recent measurements made by comparing velocity data with very high precision intensity measurements from the IPHIR experiment (Schrijver et al. 1991).
By allowing the possibility of motion in both of the interferometer arms the instrument has redundancy, and by using both outputs the system is both more efficient, and more reliable. Should the translation mechanisms fail totally however, the UV photometric signal can still be recovered and be used for asteroseismology.