Introduction

It is unlikely that many seismologists given a choice between seismic magnitude and some physical measure of seismic-source size would choose magnitude. As Thirlaway (1973) notes, some of the founders of seismology regarded seismic amplitudes as not worth serious study and since then others have argued that magnitudes should no longer be used (Jones, 2000). The main disadvantage of the magnitude scale is that it is not based on the physics of the source. It is an arbitrary scale; there is no wrong answer. Seismologists are therefore free to propose modifications to the scales. Scales defined on one type of recording system or component of ground motion, for example, are simply applied without change to new systems or components, which often then leads to attempts to relate the different scales.

Further, how A and T should be measured is often only loosely defined. Consequently, what is measured depends on analyst interpretation of the rules, and rarely are corrections applied to allow for: departures in the propagation path from some average structure; station effects; or the effects of earthquake radiation patterns. So it is easy to make the argument for abandoning magnitude as a measure of the size of a seismic disturbance and turning to physical measures such as seismic moment, which is based on the source physics and in theory at least, is independent of radiation patterns and Earth structure.