This study is motivated by the tectonic complexity of the Bali-NTB region, which results in a heterogeneous seismic distribution and active stress dynamics. The objective is to analyze the spatiotemporal evolution of the b-value using the Ogata-Katsura maximum likelihood estimation (OK1993), combined with Voronoi tessellation as an adaptive spatial partition and the Bayesian Information Criterion (BIC) for ensemble model optimization. Spatial analysis prior to the 2018 event revealed a contradiction in the form of high b-value (>1.0) in Lombok. However, the spatiotemporal analysis successfully explains the condition by identifying a zone of low b-value (<1.0) that developed following the 2007 intermediate sized earthquake through the 2018 mainshock distinct from the 2004 local anomaly that did not progress into a major earthquake. This indicates that the persistence of low b-values is a key indicator of high stress accumulation. Following the 2018 earthquake, the high stress zone expanded across Lombok and West Nusa Tenggara (NTB). This study emphasizes the importance of evaluating statistical reliability, such as N(b) and Median Absolute Deviation (MAD), to avoid misinterpretations due to data limitations, thereby providing a more reliable framework for regional seismic hazard assessment.
Copyrights © 2026