Automatic passive seismic data processing with no prior information: the contribution of Surface Wave Tomography

Abstract

Passive Seismic Tomography (PST) has been successfully applied for years in solving difficult exploration and production problems. Using local microearthquakes as seismic sources and dense microseismic networks as receivers, this methodology exploits significant numbers of continuous records that have to be thoroughly examined, in order to extract all useful information. In such applications, the greatest possible number of microseismic events is required in order to increase our knowledge of the subsurface. Although processing and analysis of microseismic data has been traditionally carried out manually by expert analysts, automation of these procedures has become mandatory. The vast volume of continuous data to be processed, as well as the subjective factor that is unavoidably introduced by the involvement of different analysts in the phase picking process, leads to the need for automation of the specific procedures (Leontarakis et al., 2015). Regardless of the way that processing and analysis of passive seismic data is carried out, the choice of the appropriate initial 1D velocity model is of primary importance, especially in the case of the PST method, where the sources (local earthquakes) are unknown and hypocentres should be accurately defined. The 1D velocity model, as well as the mean Vp/Vs ratio, are important prerequisites for a reliable initial event location.