Use this URL to cite or link to this record in EThOS: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.702748
Title: Estimating and interpreting anisotropy and attenuation in large microseismic datasets
Author: Usher, Philip James
ISNI:       0000 0004 6059 0786
Awarding Body: University of Bristol
Current Institution: University of Bristol
Date of Award: 2016
Availability of Full Text:
Access from EThOS:
Abstract:
Fractures provide important fluid pathways in petroleum reservoirs. In impermeable reservoirs, like those of shale gas, fractures need to be hydraulically stimulated in order to produce from them. Stimulating such fractures causes microseismicity which can be monitored using passive seismic surveys. The goal of my thesis is to show how measurements of velocity and attenuation anisotropy from passive seismic monitoring can be used to understand reservoir properties, including fluid and fracture properties. I use three microseismic datasets: Cotton Valley, Delphi, and Rainbow from sandstone, silica rich shale and silica poor shale reservoirs, respectively. Automated shear-wave splitting measurements from the Delphi and Rainbow datasets show a dominant VTI symmetry, which is consistent with a shale reservoir. The Delphi dataset shows weaker anisotropy (18%) than Rainbow (40%) due to the amount of clay minerals in each formation. Best-fitting Thomsen's γ parameters are found (0.17 for the Delphi dataset and 0.50 for the Rainbow dataset), through a rock physics based inversion. Shear-wave splitting is measured manually on the Rainbow dataset for a cluster of events that are almost co-located in space but spread over two stages of injection. The manual measurements show higher amounts of anisotropy, but more interestingly the anisotropy drops from 60% to 20% between the two time periods. Modelled split shear-waves for the best fitting anisotropic model from the automated method, and one with an additional set of vertical fractures match this temporal change. A decrease in anisotropy is is interpreted as being caused by an increase in fracturing. Attenuation anisotropy is measured on the Cotton Valley dataset using the log-spectral ratio method to find ∆t* between two waveforms. A temporal increase in the slow S-wave attenuation (∆t* increase of 4 x 1O- 3s) is explained by an increase in fracturing. No significant change in P-wave attenuation is observed. A "squirt flow" model is used to show how changes in ∆t* can be interpreted as an increase in fracture density of up to 0.04. Measurements of velocity and attenuation anisotropy from three microseismic datasets show how seismic wave propagation can be used to analyse fracture properties. However, care must be 'taken as the degree of intrinsic anisotropy in shales can be highly variable, and in extreme cases can mask the effects of fracturing. Linked processing and modelling of attenuation and anisotropy helps to constrain the interpretation of reservoir fracturing.
Supervisor: Kendall, J.-Michael Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.702748  DOI: Not available
Share: