According to recent studies by the National Petroleum Council (NPC) and the United States Department of Energy (DOE), significant gas (203 trillion cubic feet) and oil (325 billion barrels, 64% of the total original oil in place) remains trapped in existing, previously-discovered reservoirs. The resource is left behind because the earth is not simple, continuous, or homogeneous. The principal barrier to extracting bypassed oil and gas in existing reservoirs is an inadequate description of the geologic heterogeneities between wells. Three-dimensional seismic data collected from the surface of the earth provide a low-resolution view of the reservoir over a wide area, not always adequate to describe the structure and porosity/permeability variations at a scale that controls flow. Well logs and cores describe the reservoir with high resolution at the wellbore, but offer little information about reservoir properties between the wells. Crossed seismic profiling places a seismic source in a wellbore and receivers in a nearby wellbore to provide high-resolution images and formation property information between the wells. Traveltimes of the energy directly transmitted from source to receiver are used in traveltime tomography to provide a two-dimensional map of the formation velocity in the interwell region. Traveltimes and amplitudes of reflections generated at rock interfaces are used in reflection imaging to produce a detailed image of the reservoir structure. Because seismic waves with frequencies as high as a few kilohertz are typically recorded in crosswell surveys, the resolution of crosswell imaging has been demonstrated to be 10 to 100 times that of surface seismic data with vertical resolution of 5 to 10 feet or better. In current traveltime tomography methods, earth models are often described by rectangular cells (pixels), with a constant velocity within each cell. Although this formulation is appropriate for medical tomography, when applied to the crosswell configuration it leads to tomograms that suffer from artifacts. These artifacts are caused by the fact that the number of pixels is much larger than the number of measurements. Geologic interpretation of tomograms contaminated by artifacts becomes ambiguous. To address this practical difficulty, algorithms will be developed in Phase I that constrain seismic traveltime tomography by taking into account the generally highly layered structure of the earth. In Phase II the algorithms will be generalized to assure compatibility with software commonly used in the gas and oil industry. Anticipated Results/Potential Commercial Applications as described by the awardee: The result anticipated is a more geologically correct two-dimensional velocity estimate between wells, suitable for both information property estimation and velocity estimation in reflection imaging. The result should be a prototype algorithm available to commercialize within a crosswell processing package or for use in crosswell processing services.
