Characterization of in-Situ Stress and Permeability in Fractured Reservoirs

Abstract

During the past six months we have adapted our 3-D elastic, anisotropic finite difference code by implementing the rotated staggered grid (RSG) method to more accurately represent large contrasts of elastic moduli between the fractures and surrounding formation, and applying the perfectly matched layer (PML) absorbing boundary condition to minimize boundary reflections. Two approaches for estimating fracture spacing from scattered seismic energy were developed. The first relates notches in the amplitude spectra of the scattered wavefield to the dominant fracture spacing that caused the scattering. The second uses conventional FK filtering to isolate the backscattered signals and then recovers an estimate of the fracture spacing from the dominant wavelength of those signals. Both methods were tested on synthetic data and then applied to the Emilio field data. The spectral notch method estimated the Emilio fracture spacing to be about 30 to 40 m, while the FK method found fracture spacing of about 48 to 53 m. We continue to work on two field data sets from fractured carbonate reservoirs provided by our industry sponsors-the offshore Emilio Field data (provided by ENI-AGIP), and an onshore reservoir from the Middle East (provided by Shell). Calibration data in the form of well logs and previous fracture studies are available for both data sets. In previous reports we showed the spatial distribution fractures in the Emilio Field based on our calculated scattering index values. To improve these results we performed a map migration of all the scattering indices. The results of this migration process show a very strong correlation between the spatial distribution and orientation of our estimated fracture distribution and the fault system in the field. We observe that the scattering index clusters tend to congregate around the fault zones, particularly near multiple faults and at fault tips. We have also processed a swath of data from the second data set (the onshore carbonate field). FMI data are available from a number of wells for comparison to our seismic scattering analysis results. The agreement is very good, providing confidence that these methods can be applied to land seismic data that do not have the ideal azimuthal coverage. I. Executive SummaryThe purpose of this project is to develop and implement large-scale numerical models to quantify the effects of fracture parameter variations on seismic reflection signals and in-situ stress variations on flexural modes in boreholes. These models will be used as the basis of data analysis and inversion routines for estimating the heterogeneous fracture distribution in fractured reservoirs from seismic and borehole field data.Fracture property distributions estimated from seismic data will be used to estimate the permeability tensor in the reservoir for input to reservoir simulators.During the past six months we have adapted our 3-D elastic, anisotropic finite difference code in two ways. First, we implemented the rotated staggered grid (RSG) method to more ac...