Cross-Validation of Spatial Elongation as a Global Indicator of M≥7 Earthquake Potential: Eigenvalue Analysis of Seismicity Covariance Matrices

Read full text
Authors: Marcelo Hollander
Cited by: 3

Abstract

We present a cross-validated evaluation of spatial elongation as an indicator of tectonic zones capable of generating M≥7 earthquakes. The method computes the eigenvalue ratio of the spatial covariance matrix of seismicity within a zone, comparing recent (12-month) to historical (5-year) organization. Applied to 910 deduplicated M≥7 targets globally (1960–2026), the method achieves a 94.5%–95.2% detection rate at threshold 1.1, with a 57.1% false positive rate in pure control zones (10,000 Monte Carlo samples). Three independent implementations (Ryzen: 95.1%, Xeon: 95.2%, Antigravity AI: 94.5%) confirm methodological reproducibility and robustness. Detection rate improves from 76.2% (1960s) to 99.0% (2020s) with catalog completeness. A novel azimuth rotation indicator (mean 6.9° before rupture) complements the elongation score. We conclude that spatial elongation identifies active tectonic infrastructure rather than isolated precursors, with direct applications to seismic hazard zoning, megathrust monitoring, and operational risk prioritization. Includes full LaTeX source, high-resolution figures, 10,000-sample Monte Carlo ROC verification dataset, and cryptographic SHA256 manifest.

Fields of Study & Subjects

earthquake and tectonic studies Geophysics and Sensor Technology High-pressure geophysics and materials

Publication details

Journal / source
Zenodo (CERN European Organization for Nuclear Research)
Publication date
14 September 2026
Document type
Preprint