用环境噪音被动监测地壳应力变化，提前发现地震前兆

Stress evolution governs material failure across scales, from microscopic fractures to large earthquakes, yet direct observation of its dynamics in natural systems has remained elusive. Laboratory experiments using active ultrasonic measurements have shown that seismic velocity and attenuation are sensitive to stress, but such monitoring has not previously been achievable remotely or passively. Here we introduce a stress-sensitive frequency-domain transform that enables passive monitoring of stress evolution using ambient seismic or acoustic noise. The method quantifies relative energy shifts between adjacent frequency bands, capturing subtle changes in wave-propagation properties linked to evolving shear and normal stress. Applied across scales, from laboratory stick-slip and slow-slip experiments to natural fault systems including the 2018 Kilauea collapse, Cascadia slow-slip episodes, and major earthquakes such as the 2011 Tohoku, 2010 Maule, 2002 Denali, and 2023 Turkey-Syria events, the transform consistently reveals distinctive precursory trajectories and stress-cycle patterns. These results demonstrate that stress evolution in elastic Earth materials can be remotely and passively monitored, bridging laboratory rock physics and large-scale seismology and offering a new foundation for real-time tracking of fault mechanics and earthquake preparation.