Abstract
Fluid flow through low permeability rocks is largely controlled by fractures. As a result, the remote detection and characterisation of fractures is critical for projects involving storage or extraction of fluids. A better understanding of how different fracture characteristics affect fracture specific stiffness and in turn, seismic wave propagation (velocity and attenuation), could aid the interpretation of fractures in seismic data. In this thesis, I investigate the effect of fracture aperture, angle, roughness and infill material on wave propagation by 1) interpreting ultrasonic data from isotropic compression tests conducted by Skurtveit et al. (2020) on core samples with and without fractures; and 2) running 2D numerical models, calibrated against the experimental results, in COMSOL Multiphysics. Additionally, I relate the ultrasonic data to flow test results from Skurtveit et al. (2020) to investigate how seismic measurements can be related to fracture permeability. Experimental results show that P-waves propagating across thinner fractures with more contact points arrive faster, with larger amplitudes and higher central frequencies. Closure of fractures under increasing stress can be identified in the results by increasing P-wave velocity, whereas changes in first-arrival amplitude and frequency are ambiguous. The experimental results also highlight that increases in P-wave velocity can be related to decreases in fracture aperture and permeability. Modelling results show that 1) increasing fracture aperture leads to linear increases in arrival time delay and non-linear decreases in arrival amplitude, 2) increasing the roughness of fracture boundaries decreased arrival amplitudes by 10%, and 3) fracture infill material affects estimates of P-wave velocity and fracture specific stiffness depending upon the stiffness of the material. Furthermore, the distribution of mineral precipitates within the fracture, e.g. as linings or mineral bridges, is also shown to affect the arrival time, amplitude and frequency content of transmitted P-waves. Knowledge from this project could be applied to aid the remote characterisation of fractures and lead to safer and more efficient subsurface operations.