On the Kola peninsula one of the strongest seismic events (MS 4.2) in recent years occurred on 17 August 1999. The event was located in the Lovozero Massif close to the city of Revda and the Umbozero mining site. The Revda event was recorded with a temporary broadband network in Finnmark (Masi region), northern Norway. Based on high-quality three-component seismic data from this network we analyzed the event by full waveform modeling using a frequency-wavenumber method. We systematically varied the source depth, the size and orientation of the fault plane, and the crustal velocity model and compared the resulting synthetic seismograms with the observed data. The spectra could be matched quite well with synthetics for a circular fault plane of radius 1.6 km, an effective stress drop of 4.5 MPa, and a seismic moment of 6.0 × 1015 N m. The source depth was estimated to be about 5 km. The fault-plane solution is mainly characterized by a reverse focal mechanism with strike 240°, dip 60°, and rake 70°, consistent with first motions observed at 28 Fennoscandian stations. The distinct dispersion of the observed data could be explained by introducing a relatively strong shear-wave velocity gradient in the uppermost crust.
We think that the Revda event may have been influenced, possibly even triggered, by the presence of the mine. The Lovozero Massif has a distinct block structure with a northeast-southwest-oriented fault system and a conjugated system in the north-west-southeast direction associated and consistent with the regional north-northwest compressional tectonic stress field. Long-term mining can redistribute stress and initiate ruptures along favorably oriented faults, and to this end our fault-plane solution is in agreement with an event on the major Alluaiv fault of the northeast-southwest-oriented system. Local inspections revealed that a rockburst occurred in a part of the mine that was abandoned 10-20 years ago. This rockburst has also a reverse fault mechanism, but with a shear plane striking perpendicular to our solution. We therefore consider the observed rockburst as a secondary effect.