Why do seismic hazard models worldwide appear to overpredict historical intensity observations?

Area of Science:

• Earthquake science
• Seismology
• Geophysics

Background:

• Probabilistic seismic hazard assessments (PSHA) are crucial for earthquake-resistant building codes.
• The accuracy of PSHA in predicting actual earthquake shaking intensity is not well understood.

Purpose of the Study:

• To evaluate the predictive accuracy of recent PSHA for California, Japan, Italy, Nepal, and France.
• To identify potential systematic biases in seismic hazard models or observational data.

Main Methods:

• Compared PSHA predictions (peak ground acceleration/velocity) with historically observed earthquake shaking intensities.
• Analyzed discrepancies using numerical simulations across diverse tectonic settings.
• Investigated the impact of conversion equations between instrumental and qualitative shaking measures.

Main Results:

• Recent PSHA consistently overpredicted historically observed earthquake shaking intensities.
• Numerical simulations indicated that observed shaking is equally likely to be above or below PSHA predictions.
• Discrepancies are partly attributed to conversion equations linking instrumental measures and qualitative shaking reports.

Conclusions:

• A potential systematic bias exists in current seismic hazard models, observational data, or both.
• Further research is needed on relating qualitative historical shaking reports to instrumental seismic measures.
• Improving the conversion of qualitative observations is essential for accurate seismic hazard assessment.