Attribute Analysis Research And Application In CO₂ Geological Storage Of Time-lapse Seismic Monitor Technology

At present, due to the serious impact, such as extreme climate, caused by the greenhouse effect in the worldwide, every country has placed the emission reduction on the agenda. And as the most potential emission reduction technology, carbon capture and storage ?CCS?, is undergoing, while in order to avoid CO₂ leakage via the faults or the cracks, we must monitor the underground situation of injected CO₂, and the 4D ?four dimensional? seismic is a very effective way to monitor it. Based on the 4D seismic data and logging data of Weyburn oilfield in Canada, we processed and interpreted the seismic data systematically.By the attribute correlation and synthetic seismogram verification,we eventually got the underground CO₂ distribution state by seismic attributes interpretation.In this paper, we analyzed the 4D seismic data repeatability, which includes the source coordinate deviation, receiver coordinate deviation and first break deviation, the base data of 1999 and the monitor data of 2001 and 2002. Then the 4D processes, which includes the 4D preprocessing and 4D cross equalization, respectively, are applied to the seismic data of 1999 and 2002. The 4D preprocessing includes 4D post-stack geometry macro,4D preprocessing macro,4D CDP intersect macro and 4D CDP time intersection macro. The 4D cross equalization includes 4D amplitude envelope XEQ macro,4D frequency balance macro,4D global shifts macro,4D correlation statics macro and 4D differential phase macro.After extracting 16 kinds of seismic attributes of the 4D processed data, which include amplitude attributes, frequency attributes, phase attributes and time attributes. We computed its repeatability estimate parameter, similarity?NRMS & REP, and selected out two kinds of attributes with high repeatability. Based on the chosen attributes, we optimized the attributes of the cap rocks and reservoirs, by 4D attribute distribution correlation. This method is to get the attribute distribution map after calibrating the cap and reservoir.After the correlation,the attribute difference distribution of cap we get is about 0,which can eliminate the non-fluid factor.Eventually the attribute difference distribution of reservoir can reflect the fluid change better.At the same time,we get the amplitude and velocity change before and after CO₂ injection by synthetic seismogram,which is based on the logging data. The amplitude and velocity change can verify the change we get from 4D seismic data process. Combination of correlation and synthetic seismogram predict the distribution state of injected CO₂ more efficient.