Observations of Normal Modes from 84 Recordings of the Alaskan Earthquake of 1964 March 28‐II. Further Remarks based on new Spheroidal Overtone Data

Summary Using the same records as described in Paper I of this study, we have found 86 spheroidal overtones with periods between 285 and 100 s, radial order numbers from 3 to 25 and phase velocities from approximately 10 km s −1 to over 200 km s −1 . We discuss the implications of the normal mode‐body wave analogy and show how to translate the properties of a mode determined by the energy balance analysis (Dziewonski & Gilbert) and phase and group velocities into an equivalent body wave phase. Corresponding calculations for the observed modes indicate that over one‐half of the modes reported here correspond to the rays that bottom in the inner or outer core. Since the normal mode‐body wave analogy must also extend to the resolving power, it is clear that the amount of information on the structure of this region of the Earth is significantly increased. We report observations of an inner core mode ( 11 S 2 ) coupled with a compressional mode of nearly identical phase velocity ( 10 S 2 ; the observed periods are 246.89 and 247.74 s, respectively). This provides direct proof of solidity of the inner core, because inner core modes could not exist if the inner core were liquid. An argument, based on the coincidence of 10 S 2 and 11 S 2 is presented that the average shear speed in the inner core must be close to 3.6 km s −1 , and that dissipation in the inner core is low. Nearly all of the published average overtone periods have been derived from recordings of the Alaskan earthquake; if there existed a strong source bias, the Earth models derived from these data could be erroneous. We investigate this possibility by analysing the consistency of eight sets of average periods of the o S 1 series for 4 ≤ℓ≤ 50. It appears that for this series the source bias may be of the order of 0.03–0.05 per cent. We present an argument that the bias of the overtone periods should not be greater, and probably is less, than this figure. Appendix I contains a table with the Earth models C198 and B497 (Gilbert, Dziewonski & Brune) which were derived using the normal mode data of Dziewonski & Gilbert and those listed in this report; travel‐time data and toroidal overtone data (Brune & Gilbert 1973) were also used in derivation of the model B497.