HR: 1340h
AN: GC13A-1219    [Abstracts]
TI: Can We Estimate Injected Carbon Dioxide Prior to the Repeat Survey in 4D Seismic Monitoring Scheme?
AU: * Sakai, A
EM: akios@japex.co.jp
AF: Japan Petroluem Exploration Co.,Ltd.(Japex), 2-2-20 Higashi-shinagawa,Shinagawa, Tokyo, 140-0002 Japan
AB: To mitigate global climate change, the geologic sequestration by injecting carbon dioxide in the aquifer and others is one of the most promising scenarios. Monitoring is required to verify the long-term safe storage of carbon dioxide in the subsurface. As evidenced in the oil industry, monitoring by time-lapse 3D seismic survey is the most effective to spatially detect fluid movements and change of pore pressure. We have conducted 3D seismic survey onshore Japan surrounding RITE/METI Iwanohara carbon dioxide injection test site. Target aquifer zone is at 1100m deep in the Pleistocene layer with 60m thick and most permeable zone is approx. 12m thick. Baseline 3D seismic survey was conducted in July-August 2003 and a monitor 3D seismic survey was in July-August 2005 by vibrating source with 10-120Hz sweep frequency band. Prior to the monitor survey, we evaluated seismic data with integrating wireline logging data. As target carbon dioxide injection layer is thin, high-resolution seismic data is required to estimate potential spreading of injected carbon dioxide. To increase seismic resolution, spectrally enhancing method was in use. The procedure is smoothing number of seismic spectral amplitude, computing well log spectrum, and constructing matching filter between seismic and well spectrum. Then it was applied to the whole seismic traces after evaluating test traces. Synthetic seismograms from logging data were computed with extracting optimal wavelets. Fitting between spectrally enhanced seismic traces and synthetic seismograms was excellent even for deviated monitor wells. Acoustic impedance was estimated by inversion of these 3D seismic traces. In analyzing logging data of sonic, density, CMR, and others, the elastic wave velocity was reconstructed by rock physics approach after estimating compositions. Based on models, velocity changes by carbon dioxide injection was evaluated. The correlation of acoustic impedance with porosity and logarithmic permeability was good and relying on this relation and geological constraints with inversion techniques, porosity and permeability was estimated in 3D volume. If the carbon dioxide movement was solely controlled by permeability, estimated permeability volume might predict the time-lapse seismic data prior to a repeat survey. We compare the estimate with the actual 4D changes and discuss related variations.
DE: 0910 Data processing
DE: 0915 Downhole methods
DE: 5102 Acoustic properties
DE: 5114 Permeability and porosity
SC: Global Climate Change [GC]
MN: Fall Meeting 2005