Topological Comparison Between the Stochastic and the Nearest‐Neighbor Earthquake Declustering Methods Through Network Analysis.

Earthquake clustering is a significant feature of seismic catalogs, both in time and space. Several methodologies for earthquake cluster identification have been proposed in the literature in order to characterize clustering properties and to analyze background seismicity. We consider two recent data‐driven declustering techniques, one based on nearest‐neighbor distance and the other on a stochastic point process. These two methods use different underlying assumptions and lead to different classifications of earthquakes into background events and clustered events. We investigated the classification similarities by exploiting graph representations of earthquake clusters and tools from network analysis. We found that the two declustering algorithms produce similar partitions of the earthquake catalog into background events and earthquake clusters, but they may differ in the identified topological structure of the clusters. Especially the clusters obtained from the stochastic method have a deeper complexity than the clusters from the nearest‐neighbor method. All of these similarities and differences can be robustly recognized and quantified by the outdegree centrality and closeness centrality measures from network analysis. Plain Language Summary: Clustering, in both space and time, is a widely recognized feature of seismicity. An adequate identification of earthquake clusters allows splitting seismicity into background and clustered events (e.g., aftershocks) and is an essential step in several studies, ranging from seismic hazard assessment to long‐ and short‐term earthquake forecasting. Also, the space‐time patterns of identified clusters may provide useful insights on the structural and dynamic tectonic features of a region. Among the several methods proposed so far to identify and characterize seismic clusters, we consider two recent data‐driven declustering techniques, one based on nearest‐neighbor distance and the other on a stochastic point process. These two methods use different underlying assumptions and may lead to different classifications of earthquakes into background events and clustered events. Therefore, this study aims to compare their performances, including clusters structure characterization, by exploiting tree graph representations and tools from network analysis. We found that (1) the two declustering algorithms produce similar partitions of the earthquake catalog; (2) they may differ in the internal structure outlined for individual clusters, with the nearest‐neighbor method usually providing simpler structures than the stochastic declustering method; and (3) these features can be robustly quantified by centrality measures widely used in network analysis. Key Points: Two recent data‐driven declustering methods are compared, one based on nearest‐neighbor distance and one on the ETAS modelSimilarities in classification and in earthquake clusters are investigated by tree graphs and tools from network analysisObtained clusters are consistent, though nearest‐neighbor method usually provides simpler structures than stochastic declustering method [ABSTRACT FROM AUTHOR]