Appendix A is only available in electronic form at http://www.edpsciences.org

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

The 10.7-cm solar radio flux used in this analysis was downloaded through the website http://spidr.ngdc.noaa.gov/spidr/

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... data

The most prominent, outer peaks of each $\ell=3$ multiplet (those with |m|=3) have peak heights (maximum power densities) in the frequency domain that are $\approx$ 20 per cent of their nearest, $\ell =1$ , |m|=1 neighbours.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... lengths

This assumes that the background-to-signal ratio $\beta _{n\ell }$ remains unaltered for the various T. As has been shown by Chaplin et al. 2003a this is indeed the case provided the mode is well resolved (i.e., $T \gg \tau$ ). This is satisfied for the range of modes and T considered in our analyses.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.