Full waveform inversion (FWI) is a useful, and powerful tool for finding accurate es- timates of subsurface properties. However, when applied to conventional land data, the quality of the inversion can suffer from nonideal acquisition. FWI is most successful when we have densely sampled, wide aperture data, with a large bandwidth. Seismic data ac- quired with standard three component (3C) geophones typically lacks the low frequencies and dense sampling required for successful inversions. Recent advances, in the use of distributed acoustic sensors (DAS), may hold the key to remediation of these problems. Distributed acoustic sensors employ a continuous optical fibre, offering tighter spatial sam- pling at a greatly reduced cost, especially in the borehole environment. It has also been shown in laboratory experiments, that DAS fibres can recover significantly lower frequen- cies than standard 3C geophones. The trade-off, however, is that DAS fibres only sense strain along their tangent and therefore only provide one wavefield component, at a lower signal-to-noise ratio, limiting our ability to invert for elastic parameters. Taken together, both datasets share complementary aspects that should benefit FWI.
In September of 2018, CREWES in partnership with the Containment and Monitoring Institute (CaMI), acquired a large 3D walkaway-walkaround VSP dataset into both straight fibre, helical fibre, and 3C geophones in our geophysics well. Future work will focus on utilization of this dataset to develop an FWI formulation that leverages the complementary aspects of the DAS and geophone data. This paper is concerned with exploring the mod- elling of these complementary aspects as they relate to FWI, using the CaMI field site as our model.
View full article as PDF (3.84 Mb)