Which are typical limitations of using recurrence intervals for long-term forecasting in seismology?

• A. They predict exact dates of events.
• B. They account for all possible fault behaviors.
• C. They are limited by short data length, non-stationary fault behavior, clustering, and long-term forecast uncertainty.
• D. They provide precise magnitudes.

Answer: (C)

Recurrence intervals test your understanding that earthquakes on a fault might occur at roughly regular gaps, with a constant average rate. But real faults don’t behave in that simple way. Fault behavior is not stationary: stress levels, fault properties, and interactions with other faults can change over time, so the spacing between big earthquakes can speed up or slow down. The historical record we rely on is usually short, so our estimated average interval carries substantial uncertainty. Earthquakes also tend to cluster—periods of higher activity followed by quieter times—violating the idea of evenly spaced events. When predicting far into the future, that initial uncertainty compounds, making long-term forecasts increasingly unreliable. Also, recurrence intervals don’t provide exact dates for the next event, nor do they give precise magnitudes. They’re a rough, simplified way to think about long-term behavior, but their usefulness for long-term forecasting is limited by these issues: short data length, non-stationary fault behavior, clustering, and growing long-term uncertainty.