For many years now, the analysis of increasingly precise p-mode data has been used to understand the internal structure of the Sun. Only the low- modes (those with ) are effective in providing information about the Sun's central regions and these can be observed using the integrated light from the entire solar image. Since these modes contain integral information from the entire Sun, inversion to give the core structure requires also a good precision for the higher- modes localised in the outer layers. It is the precise frequencies of the mode resonances which are important for determining the speed of sound in the interior of the Sun. On the other hand, the multiplet splittings can offer knowledge of the internal rotation, and the resonance widths carry information on the excitation and damping of the modes.
All of these parameters are determined through a fit of theoretically expected profiles to the Fourier spectrum of the observed oscillations. However, this procedure is complicated by a number of difficulties. These include the stochastic nature of the excitation process, the existence of a background solar noise or "continuum", the close multiplet structure of many modes due to rotational splitting and the effects of overlapping wings from the many neighbouring resonances. To these must be added "noise" introduced by the instrumentation and by any lack of continuity in the observing sequence.
A commonly accepted representation of the resonance profiles has been as pure Lorentzians in Fourier space as would result from the transform of a damped sine-wave (Anderson et al. 1990). However, in the last few years, attention has been drawn to the possible asymmetry of these resonances (Duvall et al. 1993, Gabriel 1993, Abrams & Kumar 1996, Toutain et al. 1997). As we discuss later, the fitting of asymmetric profiles results in a small systematic shift in the measured frequencies. The application of a uniform shift appears to have very little influence on the inverted solar core parameters, which are related more to differences between frequencies. To improve the interpretation of the core, we are therefore looking for the slope, or higher order effects, in the resultant corrections.
It is clear from the above that we are interested in the highest possible precision in measuring these frequencies. It is with the much improved observing conditions from space using the SOHO instruments that we can hope to advance these studies. A recent study by Toutain et al. (1998) has interpreted some first results of the low- asymmetries from VIRGO and MDI. Here we analyse the results from GOLF. The GOLF instrument has produced, up to the temporary loss of SOHO in June 1998, an 805 day time series of data with exceptional stability and continuity (greater than 99 %). In this paper we measure the corrections to frequencies, splittings and line-widths obtained from asymmetric fitting.
© European Southern Observatory (ESO) 2000
Online publication: March 9, 2000