Monitoring Velocity Changes Caused By Underground Coal Mining Using Seismic Noise RAFAl CZARNY, 1 HENRYK MARCAK, 1 NORI NAKATA, 2 ZENON PILECKI, 1 and ZBIGNIEW ISAKOW 3 Abstract—We use passive seismic interferometry to monitor temporal variations of seismic wave velocities at the area of underground coal mining named Jas-Mos in Poland. Ambient noise data were recorded continuously for 42 days by two three-com- ponent broadband seismometers deployed at the ground surface. The sensors are about 2.8 km apart, and we measure the temporal velocity changes between them using cross-correlation techniques. Using causal and acausal parts of nine-component cross-correlation functions (CCFs) with a stretching technique, we obtain seismic velocity changes in the frequency band between 0.6 and 1.2 Hz. The nine-component CCFs are useful to stabilize estimation of velocity changes. We discover correlation between average velocity changes and seismic events induced by mining. Especially after an event occurred between the stations, the velocity decreased about 0.4 %. Based on this study, we conclude that we can monitor the changes of seismic velocities, which are related to stiffness, effective stress, and other mechanical properties at subsurface, caused by mining activities even with a few stations. Key words: Monitoring, scattering, coda waves, coal mine, induced seismicity. 1. Introduction Two techniques of coal exploitation are often used: open-pit mining for shallower coal seams (up to about 300 m) and underground mining for deeper seams. In Poland, underground mining is more com- mon, and people use longwall and room-and-pillar systems for mining. The first one consists of a long wall of coal in a single slice. The latter one refers to cutting a network of rooms into the seam and leaving behind pillars to support the roof. Both methods dramatically change stress–strain conditions at and around the mines, and hence the mining changes seismic velocities up to hundreds of meters above the exploitation (HOEK and BROWN 1980;BRADY and BROWN 1993). Dangers of mining especially come from regions of stress concentration, where strong seismicity may be induced (DUBIN ´ SKI and MUTKE 1996). Therefore, obtaining information on temporal elastic variations near the exploitation area is impor- tant. To prevent damages in mine and protect mine crews, some active (DUBIN ´ SKI and DWORAK 1989; SZREDER et al. 2008;HE et al. 2011) and passive (ZUBEREK and CHODYN 1989;LURKA 2008;HOSSEINI et al. 2012) seismic methods are used to measure elastic moduli. Furthermore, almost all rock burst prone mines in Poland have seismometer networks to observe seismic activity during exploitation. These data are useful for seismic hazard assessment by studying distributions of seismic events (GIBOWICZ and KIJKO 1994;LASOCKI and ORLECKA-SIKORA 2008;LES ´ - NIAK and ISAKOW 2009). All these methods have one main drawback; these active and passive sources are not temporally continuous because of natural, practi- cal and/or economic reasons. To fill this temporal gap, we use continuous records of ambient seismic noise. The ambient noise cross-correlation technique (WAPENAAR et al. 2010a, b) has been used to monitor temporal velocity changes due to pressure changes in volcanic calderas (SENS-SCHO ¨ NFELDER and WEGLER 2006;BRENGUIER et al. 2008a, b), strong earthquakes (WEGLER and SENS-SCHO ¨ NFELDER 2007; BRENGUIER et al. 2008a, b;NAKATA and SNIEDER 2012), slow slips (RIVET et al. 2011) and landslides (MAINSANT et al. 2012). One can also monitor velocities and hence stiffness of civil structures with the ambient noise technique (NAKATA and SNIEDER 2014). According to 1 The Mineral and Energy Economy Research Institute of the Polish Academy of Sciences, Wybickiego 7, 31-261 Krakow, Poland. E-mail: czarny@min-pan.krakow.pl; marcak@agh.edu.pl; pilecki@min-pan.krakow.pl 2 Department of Geophysics, Stanford University, 397 Panama Mall, Stanford, CA 94305, USA. E-mail: nnakata@stanford.edu 3 Institute of Innovative Technologies EMAG, Leopolda 31, 30-189 Katowice, Poland. E-mail: zbigniew.isakow@ibemag.pl Pure Appl. Geophys. 173 (2016), 1907–1916 Ó 2016 Springer International Publishing DOI 10.1007/s00024-015-1234-3 Pure and Applied Geophysics