The 2023 SSJ Fall Meeting

Presentation information

Room B

Regular session » S14. Earthquake prediction and forecast

[S14] PM-2

Wed. Nov 1, 2023 3:15 PM - 4:15 PM Room B (F201)

chairperson:Masajiro Imoto (National Research Institute for Earth Science and Disaster Resilience)

4:00 PM - 4:15 PM

[S14-04] No Evidence of Magnitude Dependence in Earthquake Triggering

*Giuseppe Petrillo1, Jiancang Zhuang1 (1. Institute of Statistical Mathematics (ISM))

The question of whether earthquakes can be predicted is a complex one, and the answer is likely to depend on a variety of factors. One important factor is the presence of correlations between seismic magnitudes. Some researchers have reported evidence of such correlations, while others have argued that they are caused by incomplete data. We study the magnitude correlations for a synthetic seismic catalogue produced with a 2-layer OFC model, which is able to produce a realistic earthquake statistics. We also propose a direct correlation analysis of a machine learning high-resolution catalogue for the Amatrice-Norcia (2016) sequence in Italy. By using stochastic declustering technique, we are able to calculate correlations weighting the results based on the probability of the two events being correlated. After establishing the completeness magnitude of the catalog and performing statistical analysis on correlated pairs, we can check whether the magnitude correlation hypothesis can be rejected with an high level of confidence.
Our results suggest that there is no evidence of magnitude correlation in either the synthetic or real catalogs. This is in contrast to some previous studies, which have reported evidence of such correlations, but they are consistent with the results of other studies that have found no evidence of magnitude correlation.
This study has important implications for the development of seismic forecasting models. If there is no evidence of magnitude correlation, then traditional ETAS models, which do not account for such correlations, may be sufficient for seismic forecasting. However, further research is needed to confirm our results and to understand the reasons for the apparent discrepancy between our results and those of previous studies.