Summary
This article presents the results of a multidisciplinary, four-dimensional (4D) (time-lapse), three-component (3C) (multicomponent) seismic study of a CO2 injection project in vacuum field, New Mexico. The ability to sense bulk rock/fluid properties with 4D, 3C seismology enables characterization of the most important transport property of a reservoir, namely, permeability. Because of the high volume resolution of the 4D, 3C seismology, we can monitor the sweep efficiency of a production process to see if reserves are bypassed by channeling around lower permeability parts of the reservoir and the rate at which the channeling occurs. In doing so, we can change production processes to sweep the reservoir more efficiently.
Introduction
Improving reservoir performance and enhancing hydrocarbon recovery while reducing environmental impact are critical to the future of the petroleum industry. To do this, it must be possible to characterize reservoir parameters including fluid properties and their changes with time, i.e., dynamic reservoir characterization. The objectives of our research arerepeated acquisition of a three-dimensional (3D), three-component (3C) seismic survey;demonstrate the ability of 3D, 3C, and four-dimensional (4D), 3C seismology to detect and monitor rock/fluid property change associated with a production process;incorporate geological, petrophysical, petroleum engineering, and other geophysical studies;refine the reservoir model and recommend procedures for scaling up from a pilot injection program to partial field flood to achieve maximum sweep efficiency and minimize bypassed reservoir zones;link bulk rock/fluid property variation monitored by time-lapse multicomponent (4D, 3C) seismic surveying to dynamic attributes of the reservoir including permeability, fluids, and flow characterization.
Three-dimensional, 3C seismology involves seismic data acquisition in three orientations at each receiver location—two orthogonal horizontal and one vertical. When three source components are used, nine times the amount of data of a conventional P-wave 3D survey can be recorded. Horizontal components of source and receiver displacements enable shear- (S-) wave recording; this is a powerful complement to vertical P-wave recording.
Three-dimensional, 3C seismic surveys provide significantly more information about the rock/fluid properties of a reservoir than can be achieved from conventional P-wave seismic surveys alone. By combining P- and S-wave recording, the seismic ability to determine rock/fluid property changes in the subsurface is increased. Seismic wave propagation includes travel time, reflectivity, and the effects of anisotropy and attenuation. In-situ stress orientation and relative magnitudes can be derived from seismic anisotropy. Rock/fluid properties, including lithology and porosity, may be obtained from comparative travel times or velocities of P and S waves. Other rock/fluid properties, including permeability, may be determined from comparative P and S anisotropy, travel time, reflectivity, and attenuation measurements. By combining P- and S-wave recording, seismic wave propagation characteristics can be transformed into reservoir parameters.
Introducing time as the "fourth dimension," new time-lapse (4D), 3C seismology is a tool to monitor production processes and to determine reservoir property variations under changing conditions. Using 4D, 3C seismic monitoring as an integral part of dynamic reservoir characterization, refinements can be made to production processes to improve reservoir hydrocarbon recovery. VP/VS ratios for both the fast S1 shear component and slow S2 shear component may provide a tool for separating bulk rock changes due to fluid property variations from bulk rock changes due to effective stress variations. Changes in shear wave anisotropy may reflect varying concentrations of open fractures and low aspect ratio pore structure in both a spatial and temporal sense across the reservoir. The permeability of a formation, or the connectivity of the pore space, will be the target in 4D, 3C seismology.
Refinements made to reservoir characterization techniques and their applications, now extending into the fourth dimension, are an important new area of research. Benefits of this research will include improved reservoir characterization and correlative increased hydrocarbon recovery and reduction in operating costs through improved reservoir management.
Geologic Setting
Since early Permian time, the general evolution of the portion of the Permian Basin which includes vacuum field is that of a progressively shallowing-upward carbonate platform. Aggrading and prograding cycles represent, respectively, the results of high stand and still stand sea levels. At the shelf edge these platform carbonates typically grade into reef-type deposits such as the Abo, Goat Seep, and Capitan formations. The San Andres is an exception; no reef-like rocks have been detected. Beyond the shelf edge, in the Delaware basin, clastic rocks, especially siliciclastics, were deposited during a lowstand sea level. Vacuum field is located on a large anticlinal structure that plunges slightly to the east-northeast.
The San Andres and Grayburg formations correspond to the rim of a broad carbonate shelf province to the north and northwest, northwest shelf, and of a deeper intracratonic basin, Delaware basin, on the southeast and east.1 The overall area including the Midland basin, northern and eastern shelves, and central basin platform are part of a major restricted intracratonic basin which existed during Permian time. West Texas and southeast New Mexico were in the low latitudes throughout the late Paleozoic period, making them an ideal location for carbonate sedimentation. As a consequence of this tropical environment, broad carbonate shelves were established on the margins of the Delaware and Midland basins.2
Unsupervised machine learning for time-lapse seismic studies and reservoir monitoring