H51M-1570
Low-frequency shear measurements on fractured samples to determine detectability of fractures at various stress conditions

Friday, 18 December 2015
Poster Hall (Moscone South)
Seth Saltiel, Brian P Bonner and Jonathan Blair Ajo Franklin, Lawrence Berkeley National Laboratory, Berkeley, CA, United States
Abstract:
In the subsurface energy community there is increasing interest in identifying and characterizing fractures (both natural and engineered) in reservoirs using field seismic techniques. The threshold of detection depends on fracture specific stiffness, acoustic impedance, and attenuation (both intrinsic and scattering) in the surrounding medium. The frequency of the probing seismic wave is therefore important. The simplest way to estimate detection conditions is by computing reflection and transmission coefficients and how they change with varying stiffness. In this setting we imagine that the stiffness is dominated by the asperity contact area, which is in turn a function of effective stress.

We have adapted a low-frequency (1-64 Hz) sub-resonance torsional shear system, capable of measuring shear modulus and attenuation, to explore the seismic signature of fractures in order to interpret field data from fractured reservoirs. Since the apparatus operates quasi-statically, we eliminate scattering effects. Our instrument is unique in its ability to measure at low normal stresses, simulating 'open' fractures in shallow or high fluid pressure reservoirs. We present calibration data that shows the accuracy of our instrument, and measurements from artificially fractured dry granite and rhyolite samples. By measuring the samples before and after fracturing under various normal stresses, we can separate the compliance of the rock from the compliance of the fracture. This method allows us to study well mated 'closed' fractures (stiffness about 40% of the intact shear modulus), until they are nearly open (<1%). With this data we calculate the reflectivity of the fracture in a field setting with varying effective stress, showing at what conditions we would expect to be able to detect fractures seismically.