Practical Suggestions for Assessing Rates of Seismic-Moment Release

Abstract

This article explores how to interpret observations of rates of seismic- moment release by evaluating simulated earthquake sequences generated assuming Gutenberg–Richter (gr), truncated gr, or exponentially tapered gr distributions. For earthquakes generated assuming a gr distribution, expected moment rates depend strongly on the largest event observed and increase indefinitely over time, never approaching a stable value. For events generated with truncated or tapered distributions expected moment rates increase with time but approach a stable value if moments near the corner moment M C are thoroughly sampled. Thus, interpreting reported moment rates requires knowledge of the corner moment M C and the number N of contributing earthquake observations. This article discusses how to estimate a critical number of events N large where the approach to stability begins; N large depends more strongly M C than on the β -value, and only weakly on whether one simulates the observations with a truncated or tapered gr distribution. For situations where N is less than N large , the article discusses how to adjust moment rates to account for the expected time dependence and also explores ways for estimating rates that reduce the dependence on the largest event.