Monitoring seismic velocity changes in the Dongtan Coal Mine using ambient noise correlation

This study analyzed ambient seismic noise using the MSNoise package to monitor temporal changes in the underground seismic wave velocity in Mining Area 6 of the Dongtan Coal Mine in China. The data was recorded continuously over 76 days by 7 three-component stations and 10 single-component microseismic stations deployed in Dongtan Coal Mine, with station spacing ranging from 0.1 km to approximately 3 km. Using the causal and non-causal components of the Z-component cross-correlation function, along with moving-window cross-spectrum analysis and cumulative calculations with a 5-day window overlay, stable seismic velocity changes were obtained in the frequency band of 0.1 to 2 Hz. We found a correlation between the timing of average velocity changes and seismic events caused by underground mining processes. In particular, when the relative seismic velocity increased by 0.23%, larger energy minequakes typically occurred. This study shows that ambient noise correlation has great potential for predicting minequakes, guiding pressure-relief production, and providing warnings about the impact of overburden pressure.