“…Structure in the Upper Mantle 1.21.6.1 Normal Modes and Long Period Surface Waves 1.21.6.1.1 Early studies In the 1970s, after the occurrence of the 1960 Great Chile Earthquake (Mw 9.6), and with the accumulation of data from the World-Wide Standardized Seismographic Network (WWSSN) network, measurements of long-period surface-wave attenuation, then based on relatively few records, showed large disagreements (e.g., Ben-Menahem, 1965;Smith, 1972;Mills and Hales, 1978;Jobert and Roult, 1976). At that time it was realized that: (1) elastic effects were important in limiting the accuracy of Q measurements; (2) anelastic attenuation caused frequency dependence of elastic velocities, which needed to be taken into account (Luh, 1974;Randall, 1976;Liu et al, 1976;Akopyan et al, 1976;Kanamori and Anderson, 1977); and (3) large systematic lateral variations existed, correlated with those of phase velocities and suggesting significant differences in attenuation under different tectonic regions, with low attenuation under shields and high attenuation under oceans and tectonically active provinces (e.g., Nakanishi, 1978Nakanishi, , 1979bNakanishi, , 1981Lee and Solomon, 1979;Roult, 1982;Dziewonski and Steim, 1982).…”