The 4D-time lapse seismic method allows to determine the changes occurring in a geological reservoir as a result of injection of gas, CO2, water or hydrocarbon production. This method involves acquisition, processing, and interpretation of repeated seismic surveys over the considered geological reservoir. The final result is a time-lapsed difference seismic dataset, obtained by comparing the repeated datasets (i.e. seismic data of survey 1 is subtracted from the data of survey 2). The difference should be close to zero except where the reservoir changes occurred. The paper presents the results obtained by a numerical modeling approach based on the construction of a detailed 3D geological model located in Italian Adriatic offshore. The 3D geological model consists of a sandy reservoir with salt water mineralization (Porto Garibaldi formation - Pliocene/Pleistocene) surmounted by a caprock layer of clay (Santerno clays - Pleistocene) and recent alluvial sediments. The 3D reservoir changes were generated by using the SIAM-Integrated System for Modeling Analysis developed by RSE and assuming the storage of supercritical CO2 into the reservoir at a rate of 1 million tons per year for a 30 years period according to the ENEL Porto Tolle coal power plant project. The offshore seismic baseline is an existing 2D stack lines available through the ViDEPI data base developed by the Ministry of Economic Development UNMIG. Several 2D seismic stack sections were synthetically created at different times (0, 15, 30, 60 year) and with different volumes of supercritical CO2 injected. The time-lapse difference was analyzed using different numerical simulators and with different comparison methods as SVD, covariance and instantaneous seismic attributes. The influence of random seismic noise with a 20% Gaussian distribution was also analyzed by generating appropriate seismic reflection stack sections. The obtained results seem to suggest that it is possible to highlight the time-lapsed difference in the area affected by the injection of CO2 but a real monitoring could not be effective in defining areas and volumes really interested by CO2 storage in the long-term. In fact, the results show that the method could be efficient in the short period (about 15 years) and this opens up a serious consideration about the definition of the monitoring objectives and about the relationship between costs and benefits in the implementation of this expansive prospecting method.
A numerical approach to test 4D seismic technology as a monitoring tool for deep geological reservoir
Caielli G;de Franco R;
2013
Abstract
The 4D-time lapse seismic method allows to determine the changes occurring in a geological reservoir as a result of injection of gas, CO2, water or hydrocarbon production. This method involves acquisition, processing, and interpretation of repeated seismic surveys over the considered geological reservoir. The final result is a time-lapsed difference seismic dataset, obtained by comparing the repeated datasets (i.e. seismic data of survey 1 is subtracted from the data of survey 2). The difference should be close to zero except where the reservoir changes occurred. The paper presents the results obtained by a numerical modeling approach based on the construction of a detailed 3D geological model located in Italian Adriatic offshore. The 3D geological model consists of a sandy reservoir with salt water mineralization (Porto Garibaldi formation - Pliocene/Pleistocene) surmounted by a caprock layer of clay (Santerno clays - Pleistocene) and recent alluvial sediments. The 3D reservoir changes were generated by using the SIAM-Integrated System for Modeling Analysis developed by RSE and assuming the storage of supercritical CO2 into the reservoir at a rate of 1 million tons per year for a 30 years period according to the ENEL Porto Tolle coal power plant project. The offshore seismic baseline is an existing 2D stack lines available through the ViDEPI data base developed by the Ministry of Economic Development UNMIG. Several 2D seismic stack sections were synthetically created at different times (0, 15, 30, 60 year) and with different volumes of supercritical CO2 injected. The time-lapse difference was analyzed using different numerical simulators and with different comparison methods as SVD, covariance and instantaneous seismic attributes. The influence of random seismic noise with a 20% Gaussian distribution was also analyzed by generating appropriate seismic reflection stack sections. The obtained results seem to suggest that it is possible to highlight the time-lapsed difference in the area affected by the injection of CO2 but a real monitoring could not be effective in defining areas and volumes really interested by CO2 storage in the long-term. In fact, the results show that the method could be efficient in the short period (about 15 years) and this opens up a serious consideration about the definition of the monitoring objectives and about the relationship between costs and benefits in the implementation of this expansive prospecting method.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
