Effects of a gap-filling method on p-mode parameters

The quality of helioseismological ground-based data strongly depends on the presence of a gap in the observational window. In order to address that problem in the case of full-disc low-degree p-mode velocity measurements, Fossat et al. proposed a gap-filling method called 'Repetitive music'. The autocorrelation function of the velocity signal shows a correlation of more than 70 per cent at about 4 h because of the quasi-periodicity of p-mode peaks in the Fourier spectrum. The method then consists of filling gaps of the velocity signal with data, when they exist, located 4 h before or after. By using Monte Carlo simulations, we assess the effects of the gap-filling method on p-mode parameters and their errors. A way to remove the modulation in the power spectrum resulting from the gap-filling method is proposed; its effects on p-mode frequencies, linewidths, amplitudes and asymmetries are discussed as a function of frequency and signal-to-noise ratio.