We analyze the world-wide consistency of teleseismic reporting, completeness of the seismicity record, and homogeneity of magnitude determination for strong shallow earthquakes (magnitude, M ≧ 6; depth, h ≦ 70 km) in the period 1904 to 1980. For earthquakes with magnitude M ≦ 7, we used the catalog given by Abe (1981), which lists the surface-wave magnitude (Ms) of every reported large event that occurred between 1904 and 1980, according to the original formulation of Gutenberg (1945). Under the postulate that the rate of earthquake occurrence for the entire world as a whole is consistent on a time scale of decades, and that since the installation of the World-Wide Standardized Seismograph Network in the early 1960s the earthquake catalog for strong (M ≧ 6) shocks is complete and the seismicity rates are typical of all periods in the century, we find that, for instrumental-related reasons, the Ms of large events prior to 1908 have been systemically overestimated by at least 0.5 magnitude unit, and the Ms of events in the period 1908 to 1948 have been consistently overestimated by 0.2 unit, relative to the Ms assigned to shocks occurring after 1948. When these corrections are taken into account, the catalog of events with Ms ≧ 7.0 is shown to be largely complete (i.e., nearly all the large shallow shocks which occurred in the earth are listed) for most of the world since early in the century.
For earthquakes as the magnitude 6 level, we used the International Seismological Center (ISC) tape and Regional Catalogues of Earthquakes. This file compiles the earthquake data given by Gutenberg and Richter (1954) for the first half of the century, and data from various international agencies after ≈ 1945. We find a highly incomplete seismicity catalog at this magnitude level for the first half of the century, mainly due to numerous increases and decreased in the earthquake detection and reporting capabilities; a more complete but extremely inhomogeneous catalog for the period 1950 to 1963, due to the use of different formulations and criteria to calculate and report the parameter “magnitude” by the various seismological agencies; and a largely complete and homogeneous (in Ms) record since 1964.