A Unified Seismicity Catalog Development for Saudi Arabia: Multi-Network Fusion and Machine Learning-Based Anomaly Detection

This investigation concentrates on refining the accuracy of earthquake parameters as reported by various Saudi seismic networks, addressing the significant challenges arising from data discrepancies in earthquake location, depth, and magnitude estimations. The application of sophisticated machine learning techniques, particularly the Isolation Forest algorithm, has markedly enhanced the precision in the estimation of seismicity parameters by effectively identifying and eliminating outliers and discrepancies. A newly developed and refined seismicity catalog was employed to accurately determine key seismic parameters such as the magnitude of completeness (Mc), a-value, and b-value, thereby underlining their indispensable role in regional seismic hazard assessment. The research underscores the substantial impact of data inconsistencies on the evaluation of seismic hazards, thereby advocating for the advancement of research methodologies within the field of seismotectonics. The insights derived from this study significantly contribute to a more profound understanding of the seismotectonic processes in the region. These insights are crucial for the development of comprehensive seismic hazard assessments and the formulation of targeted earthquake preparedness strategies, thereby enhancing resilience against seismic risks in the region.