Delineation of shallow seismic source zones usingK-means cluster analysis, with application to the Aegean region

Abstract

SUMMARY The selection of specific uniform seismic source zones for use in probabilistic seismic hazard analysis is often controversial. Recognizing that a consistent approach to source model development is not always possible, as the information available relating to geology and seismotectonics can vary from region to region, the K‐means algorithm for hierarchical cluster analysis can be used to partition regions based on observed seismicity. The Aegean [incorporating Greece, Albania, Former Yugoslav Republic of Macedonia (F.Y.R.O.M.), southern Bulgaria and western Turkey], with its varied seismotectonics and generally high seismicity, is used as an important area of seismicity in which to develop and demonstrate the application of K‐means. Two types of algorithm are considered. The first is a point‐source K‐means that can be used to partition a catalogue of earthquake hypocentres. The second is a novel line‐source development of the algorithm, appropriate in seismology as these are analogues for the traces of active faults, which is then applied to a catalogue of known fault ruptures in the Aegean. The common problems of the K‐means methodology are also addressed. Ensemble analyses are used to identify better choices of initial estimates for the cluster centres. A cluster quality index is used to identify the optimum number of clusters, and its robustness assessed when considering different subsets of the observed earthquake catalogue. An alternative approach is also implemented: Monte Carlo seismic hazard analysis is used to compare models with different numbers of clusters with the observed seismicity of the 20th century. Considerable variation is found in the optimum number of clusters identified either by the quality index or by stochastic seismic hazard analysis. Ultimately the K‐means partitions of seismicity are developed into source models and their representation of Aegean seismotectonics assessed. The result is that models containing between 20 and 30 clusters emerge as the most appropriate in capturing the spatial variation in hypocentral distribution and fault type in the Aegean.

Earthquake Clustering and Energy Release of the African–Arabian Rift System

“…This has been done for the 1800-km-long Sumatran fault (Burton and Hall, 2014) where the partitioned seismicity along the fault (or earthquake clusters), hence fault segments, complements the geological segmentation of Sieh and Natawidjaja (2000). Additionally, in regions where deformation occurs on more distributed fault networks, such as the Aegean, rather than on a single fault strand, the method successfully partitions seismicity in line with the structural segmentation (Weatherill and Burton, 2009). The procedure to partition seismicity into earthquake clusters is described in detail by Weatherill and Burton (2009) and Burton and Hall (2014), but the procedure is as follows.…”

“…The data were homogenized to the moment magnitude (M w ) scale using equations E Q -T A R G E T ; t e m p : i n t r a l i n k -; ; 5 5 ; 1 2 5 Scordilis, 2006). The K-means data-partitioning method of Hartigan (1975) has made it possible to partition such a regional seismicity into clusters of earthquakes and even into seismic zones (e.g., Greece; Weatherill and Burton, 2009). Indeed, it is possible to partition the seismicity on a long and specific fault into clusters of seismicity that are comparable to the geometry of a geological segmentation.…”

“…The K-means cluster analysis has already been successfully applied to the Aegean region (Weatherill and Burton, 2009), Pakistan (Rehman et al, 2013(Rehman et al, , 2017, and the Sumatran fault (Burton and Hall, 2014) where the optimum cluster model correlates well with known structural segmentation of the fault. Here, we apply it to the entire East African rift (EAR) system where there are variations in the extent and quality of structural observations throughout the region.…”

“…However, the characterization of seismic sources is often inconsistent, resulting in disagreement between zonation models of the same area. An emerging method of hazard zonation is to delineate source zones based on the hypocentral distribution of seismicity using a K-means cluster analysis (Weatherill and Burton, 2009;Burton and Hall, 2014). A clustering analysis of instrumentally observed and historical seismicity is particularly useful in areas where physical characteristics (i.e., structural observations) of seismic sources are absent.…”

Rushan earthquake swarm in eastern China and its indications of fluid-triggered rupture

“…Cluster analysis is a useful tool in seismology and has been widely applied to locate earthquakes (Frohlich and Davis 1990;Davis and Frohlich 1991;Dzwinel et al 2005;Shearer et al 2005;Weatherill and Burton 2009;Rehman et al 2014), split shear-waves (Teanby et al 2004) and estimate inversion error (Zheng et al 2015a). Godano et al (2013) demonstrated the presence of multiple earthquake clusters in research on an Italian swarm, while Lindenfeld et al (2012) recorded several similar clusters in a restricted area while researching fluidtriggered earthquake swarms in the East African Rift.…”