Detection of 160 min solar intensity variations: Sampling effect

The hypothesis that the 2h40m period in the Crimean and Stanford global solar velocity oscillation measurements, which is close to a 1/9 of a day, might be produced partly by a 1-day sampling regularity in observations, is checked using the IR center-limb intensity variations measurements. We find: a) The power spectrum of the series of these data does show a peak near the 2 h 40 m period. b) Power spectra of the series with observing windows 'filled' with a constant do not exhibit a significant peak at the 9th harmonic of a day, nor do the power spectra of the set with observing windows 'filled' with randomly generated numbers. Accordingly, we conclude that the analysis performed on the really observed data is not biased by the effect of data windows and a 1-day sampling regularity of observations. (orig.)