Application of a Multivariate-Bayesian Approach to Evaluating Geological Risk at Play Level, Nanpu Sag, Eastern China
Geological risk evaluation is one of the important components in petroleum resource assessment. In this paper, a multivariate-Bayesian approach was applied to the Nanpu Sag of the Bohai Bay Basin, eastern China, to evaluate the geological risk of the structural play in the Oligocene Dongying Formation (Ed1). Geological data from 343 exploratory wells and seismic interpretation available at the end of 2004 were used to map the geological risk of petroleum occurrence, employing the multivariate-Bayesian approach. After the evaluation, seventeen wells were drilled in 2005 and sixteen penetrated the Ed1 and tested for oil. Post drilling analysis shows that eight of the nine wells drilled in areas with predicted high probability of hydrocarbon occurrence (＞50％) encountered commercial oil with high flow rates and thick net pay, and one tested oil with a low flow rate. The other seven wells were drilled in areas with relatively low probability (＜50％), of which three tested dry and four recovered oil with relatively low flow rates. The predicted high potential and low risk areas in southeastern Nanpu Sag were confirmed by these sixteen exploratory wells with an 81％ success rate. Comparison of post-drilling results with the pre-drilling prediction suggests that the multivariate-Bayesian approach can help visualize geological risk, thus improving exploration success by optimizing the drilling strategy.
Suyun Hu Qiulin Guo Zhuoheng Chen Shiyun Mi Hongbing Xie
China University of Petroleum, Beijing 102249, China;Research Institute of Petroleum Exploration & D Research Institute of Petroleum Exploration & Development, PetroChina, Beijing 100083, China;Institu Geological Survey of Canada, Calgary, AB T2L 2A7, Canada Research Institute of Petroleum Exploration & Development, PetroChina, Beijing 100083, China
国际会议
The 12th Conference of the International Association for Mathematical Geology(第12届国际数学地质大会)
北京
英文
641-644
2007-08-26（万方平台首次上网日期，不代表论文的发表时间）