A formula for magnitude distribution of earthquakeslog n(M)=a-bM+log(c-M)has an advantage over the well-known Gutenberg-Richter formula for it can represent a sharp decrease of frequency near the upper limit of magnitude c, which is recognized in some earthquake groups. In the range where M<<c, the above formula is practically the same as the G-R formula. A method for obtaining the maximum likelihood estimates of b and c in the above formula has been developed and the accuracy of the estimates has been investigated. The a-, b-, and c-values have been calculated for earthquakes in various regions of Japan and in the world. The c-value seems to exhibit regional variations. It is suggested that in some cases the spatial or temporal variations in b-value for the G-R formula or in the average magnitude M are interpretable as due to the variation in the upper limit of magnitude c.
No takes yet. Share an insight, caveat, or question.
Tokuji Utsu (1974) studied this question.
Synapse has enriched 5 related papers. Consider them for comparative context: