Distinguishing earthquakes and explosions

Our region has high level of mining and other technological activity. It makes the problem of distinguishing earthquakes from explosions very important for seismology. Large number of explosions of different kinds (open-pit, underground, underwater) registered by our stations make a good basis to study a fundamental question how do seismic recordings depend on sources and on waves propagation paths.

We uses several criteria to distinguish earthquakes from explosions. Most of them are not absolute. In other words, it is often impossible to state reliably a kind of seismic events by seismic recordings only.

We study questions concerned with events discrimination systematically. Our studies are concerned with:

• Study of mining activity. Clarification of operation modes of mines, kinds and structures of explosions, specific features of explosions of different kinds.
• Improving of seismic events locations. Seismic events occurred within mining areas can be identified as explosions with high likelihood.
• Checking the criterion of P and S body waves amplitudes ratio. It is considered that explosions due to spherical symmetry of their sources don't generate S-waves.
• Study of spectral properties of different events recordings, working out spectral criteria of events discrimination.
• Applying generalized envelopes (i.e., curves, which show average change of signals amplitudes and frequencies in time) for seismic events distinguishing.
• Registration of acoustic pulses from open-pit explosions.
• Study of difference between body wave (Mb) and surface wave (Ms) magnitudes.
• Study of specific features of recordings.

By the results of the studies we can make preliminary conclusions concerned with different approaches of events discrimination:

• S/P amplitudes ratio. High (greater than 3) value of the parameter can be argument in a favor that the event is earthquake. Small value of the parameter IS NOT A SERIOUS ARGUMENT THAT THE EVENT IS EXPLOSION.
• Depth of an event. This is not a reliable parameter. In most case it can be used as a discrimination criterion. Each case require a special analysis. Sometimes (event occurred near a seismic array and its array recording show extremely high values of apparent velocities) it may argue in a favor of natural origin of the event.
• Difference between magnitudes computed by body and surface waves.. It seems that the criterion works for strong enough events at regional distances (although overlap area does exist). To use the criteria it is necessary to have broadband recordings. Thus its usage is very restricted.
• Changing of spectrum in time. High values of a parameter (greater than 0.5) argue in a favor of technological origin of event with high likelihood.
• Comparison of generalized envelopes of seismic events. It seems to be a promising method of events classification and discrimination. For its usage it is necessary to have a data base of recordings of events of known nature.
• Existence of acoustic pulse. This is a reliable criterion of technological origin of an event.
• Specific features of recordings. Existence of features described here can reliable identify an event to be an ore fall in one of Khibiny mines.