We propose a unified geophysical hypothesis treating Earth's solid inner core as a triaxial gyroscope subject to the Dzhanibekov effect (intermediate-axis instability). Gyroscopic precession generates a slowly sweeping stress arc across the mantle-crust system, selectively triggering earthquakes at mechanically weak zones while transmitting energy to antipodal points through electromagnetic diffusion when encountering stable crust. The model integrates four independent observational correlations: (1) antipodal earthquake pairing statistics (22% vs. 12–15% random baseline), with specific axes (Chile–Sichuan, Sumatra–Ecuador) showing elevated rates; (2) the precise antipodal alignment of Kamchatka (52. 6°N, 160°E) — Earth's most seismically active zone — with the South Atlantic Anomaly core (52. 6°S, 20°W), showing cross-correlation r ≈ 0. 65 with 20–40 year lag; (3) inner core differential rotation periodicity (~70 years) correlating with Length-of-Day variations and decadal M8+ earthquake frequency (r ≈ 0. 55) ; (4) the four major geomagnetic polarity chrons of the past 5. 2 Ma fitting prime-number multiples of a ~100 ka base unit. Version 1. 0 extends the framework with: (a) Euler equation simulations revealing four dynamical regimes of inner core flip behavior, identifying Earth as operating in the "dissipative + periodically forced" regime; (b) deterministic chaos analysis demonstrating that reversal timing contains irreducible randomness near the separatrix despite fully deterministic physics; (c) an obliquity–inner core coupling hypothesis constrained by angular momentum conservation (Icore/Iₘantle ≈ 0. 0875) ; (d) a coupled ball-in-shell numerical experiment proving that an internal gyroscope can spontaneously generate obliquity in an outer shell through viscous coupling, with an optimal coupling strength regime. The model yields falsifiable predictions: SAA entering a transitional regime ~2090–2110 CE; elevated M6. 5+ seismicity near the 2025 Myanmar earthquake's antipodal point (21°S, 84°W) within 2026–2035; and detectable non-planetary periodic components in obliquity acceleration residuals. All simulation code and data sources are documented for reproducibility.
Zhendong Wang (Sun,) studied this question.