Monitoring results after 36 ktonnes of deep CO2 injection at the Aquistore CO2 storage site, Saskatchewan, Canada
Monitoring results after 36 ktonnes of deep CO2 injection at the Aquistore CO2 storage site, Saskatchewan, Canada
复制标题
加拿大萨斯喀彻温省 Aquistore 二氧化碳封存场深层注入 36 吨二氧化碳后的监测结果
DOI：
10.1016/j.egypro.2017.03.1546
复制
发表时间：
2017
期刊：
影响因子：
--
通讯作者：
C. Samson
中科院分区：
文献类型：
--
作者：
D. White;K. Harris;L. Roach;B. Roberts;K. Worth;A. Stork;C. Nixon;D. Schmitt;T. Daley;C. Samson
Aquistore二氧化碳储存点位于加拿大萨斯喀彻温省东南部。二氧化碳被注入前寒武纪基底上方∼3200米深的充满卤水的砂岩地层中。从2015年秋季开始，该工地实现了400-600吨/天的持续注入速度，截至2016年9月底，总共注入了88千吨。地震监测方法已经被用来追踪地下CO2羽流，并记录任何由注入引起的地震活动。目前正在使用两组正交短周期地震检波器、三台宽带地震仪和一组井下地震检波器进行被动地震监测。在二氧化碳注入的前17个月，没有检测到与注入有关的重大地震活动(MW>；-1)。2016年2月，在该工地进行了第一次注入后时间推移三维地震勘探(地面和VSP)。观测井采用2750m套管传输光缆，采用分布式声波传感系统采集VSP数据。流体流动模拟的三维地震模拟结合从现场数据获得的地震重复性估计表明，延时VSP应该能够在总共注入∼30kt后成像CO2羽流。此外，这项首次监测调查还测试了用稀疏永久阵列采集的地面地震数据在有限注入后探测或成像CO2羽流的能力。正在定期进行延时测井，以提供与二氧化碳饱和度变化相关的地震速度变化的现场测量。
The Aquistore CO2Storage Site is located in southeastern Saskatchewan, Canada. CO2is injected into a brine-filled sandstone formation at ∼3200 m depth immediately above the Precambrian basement. Sustained injection rates of 400-600 tonnes/day were achieved at the site starting in the fall of 2015 with a total of 88 ktonnes having been injected by the end of September, 2016. Seismic monitoring methods have been employed to track the subsurface CO2plume and to record any injection-induced seismicity. Passive seismic monitoring is being conducted using two orthogonal arrays of short-period geophones, 3 broadband seismographs, and an array of downhole geophones. No significant injection-related seismicity (Mw > -1) has been detected during the first 17 months of CO2injection. The first post-injection time-lapse 3D seismic surveys (surface and VSP) were conducted at the site in February, 2016. The VSP data were acquired with a distributed acoustic sensing system using a 2750 m casing-conveyed optical fibre cable in the observation well. 3D seismic modelling of fluid flow simulations in conjunction with seismic repeatability estimates obtained from field data indicate that the time-lapse VSP should be capable of imaging the CO2plume after a total injection of ∼30 ktonnes. In addition, this first monitor survey tests the ability of surface seismic data acquired with a sparse permanent array to detect or image the CO2plume after limited injection. Time-lapse logging is being conducted on a regular basis to provide in situ measurement of the change in seismic velocity associated with changes in CO2saturation.