Astron. Astrophys. 323, 235-242 (1997)

5. Conclusions

We are confident that the fitting strategy described in Sect 3.1 gives an unbiased determination of the frequencies below . From the frequency tables presented here we determine values of and that are in good agreement with our previous work and with other sets of observational frequencies. By direct subtraction of mode frequencies found in 1989 and 1992, we find a decrease in mode frequencies of . The error bars, computed using a Monte Carlo method, are somewhat higher than previous published values (Elsworth et al., 1994). This is not only due to IRIS network's lower 50% duty cycle, but also because the Monte Carlo method more completely accounts for additional uncertainties in the identification of the p-mode frequencies. Such errors are underrated by the classical formulas. For data sets with higher duty cycles (above 85 %) the Monte-Carlo estimations of the error bars converge toward the values given by the theoritical formulas.

We have analyzed four summer seasons of IRIS network data. Eight years of data already exist, so we shall soon be able to obtain a much more complete list of full disk frequencies that sample the solar cycle from maximum to minimum. Although the now fully-deployed IRIS network has not been able to get as close to the ideal 100% duty cycle as originally intended, many collaborations have been developed, and it is presently possible to merge full disk data not only from the IRIS instruments, but also from Cacciani's MOF, from the BiSON Mark-1 at Tenerife, from GONG integrated images, and possibly from the LOWL integrated images, so that we expect to be close to 100% starting in 1995. This will improve the statistics, reduce the uncertainty levels, and allow us to extend the current limits of p-mode detectability at both low and high frequencies. Of course the extremes of the spectrum will be better investigated using the data from the helioseismology experiments on the SOHO spacecraft, which began operation in 1996. The GOLF experiment, in particular, is most like the IRIS instruments and has proved to be much less noisy at low frequency and has the advantages of total continuity and homogeneous data quality at all frequencies.

© European Southern Observatory (ESO) 1997

Online publication: June 5, 1998

helpdesk.link@springer.de