Earthquake Source Processes in Deep South
African Gold Mines- The NELSAM Project

This unique and exciting project, the Natural Earthquake Laboratory in South African Mines (NELSAM), provides an opportunity to directly access and monitor typically inaccessible, seismogenic faults. Due to mining-induced stresses, hundreds of earthquakes occur each day, making earthquake hazard an ever-present reality. The NELSAM data set, with a wide range of well-recorded deformation frequencies, is helping to bridge the gap between natural earthquakes and laboratory instabilities and to develop a consistent theoretical framework to describe seismicity of all sizes.

Besides myself there are many participants in the NELSAM project, most notably Ze'ev Reches (U. Oklahoma), Mal Johnston (USGS, bottom right picture), Art McGarr (USGS, bottom left picture), Dave Lockner (USGS), Vincent Heesakkers (U. Oklahoma), Tom Jordan (USC), Mark Zoback (Stanford), Yehuda Ben-Zion (USC), and many others.

Minimum Magnitude Earthquakes
		One goal of the NELSAM project is to investigate whether a minimum magnitude earthquake exists in the gold mines. Using twelve years of data collected by Integrated Seismic System International, we found that it is detection capabilities, rather than a physical process that limits the observed minimum earthquake size. Currently the mine's earthquake catalog is complete to magnitude -1.3, but with our array of NELSAM seismometers we have observed many smaller earthquakes, as small as magnitude -3.0 so far. As this project has only been recording reliable data for a short time, we still have much work to do to determine if we can detect a physically controlled minimum magnitude earthquake.
Earthquake Energy Budget:
		Determining the energy budget of an earthquake, including the radiated seismic energy, the fracture energy, heat generated, and surface area created have been a long-held goal of earthquake scientists.  In TauTona Mine, we are closer to reaching this goal than in most environments.  We record the seismic energy, which like for larger earthquakes, is only a small fraction of the total energy release (1-4%). And in exceptional cases, such as this magnitude 2.1 earthquake that Ze’ev Reches identified underground, we can directly observe the fault zone, including the fault gouge generated during sliding and the slip during the earthquake.  We have thermisters in place, but not yet recording, so hopefully we will soon have yet another constraint on earthquake energies.
Seismic Moment Tensors:
		Art McGarr (USGS) and I conducted a field expedition to the South African gold mines where we deployed a temporary seismic array using portable broadband accelerometers.  During our nine-day experiment we recorded many earthquakes in the size range -1.0 ≤ MW ≤ 3.4.  With these data we have been able to verify the recording parameters for our permanently installed seismometers, and also use the lower frequency data to calculate moment tensors for mining-induced seismicity.  We found that the larger events are likely to involve an implosive component and the smaller earthquakes often show explosive components to their focal mechanisms.
Scaling of Earthquake Source Processes: A central goal of the NELSAM project is to provide insight into whether the rupture process of small and large earthquakes is the same.  We have only just started computing the apparent stress and other source parameters of the mining induced earthquakes and comparing our results with those from larger natural events to observe any scaling with seismic moment.  Initial results from a small sampling of mining-induced earthquakes suggest that apparent stress maybe independent of seismic moment.