Microseismic monitoring of a CO2 sequestration project
October 2, 2008 - 10:31am — dana
The Midwest Regional Carbon Sequestration Partnership (MRCSP) is one of seven nationwide partnerships formed by the U.S. Department of Energy (DOE) to help develop the technology, infrastructure, and regulations to implement large-scale CO2 sequestration in different regions and geologic formations within the U.S. The MRCSP encompasses eight states (Indiana, Kentucky, Maryland, Michigan, New York, Ohio, Pennsylvania and West Virginia), and is formed from a variety of stakeholders including research, energy industry, government and non-government organizations. As a member of the MRCSP, ESG has been actively involved in deploying its ResMap™ real-time microseismic reservoir monitoring technologies, to assist in the monitoring, measuring and verification (MMV) of carbon sequestration programs. In this case a temporary microseismic monitoring system was deployed to assist in the monitoring of a CO2 injection project in Northern Michigan. The purposes of this injection project was to conduct a variety of focused field tests of sequestration technologies, including microseismic monitoring, to determine their viability for being unrolled in wide-scale sequestration projects across the region. Challenge The objective of this particular project was to inject 10,000 metric tons of CO2 into a target interval containing the porous portions of the Bois Blanc and all of the Bass Islands Dolomite formation, at a depth of 3,190-3,515 feet. The actual target interval is located in the Bass Islands Dolomite at 3442-3515 ft. The injection site is located in the northern portion of the Michigan Basin, a large mature sedimentary basin that covers most of Lower Michigan. The primary seal is the Amherstburg formation, which acts as a caprock layer, overlying the Bois Blanc and Bass Islands Dolomite. The injected CO2 was transported from a nearby natural gas processing plant. The plant removes the CO2 found in the natural gas produced from the Antrim shales, compressed it, and sent it via pipeline to the injection site. Prior to this injection project the CO2 from the plant was either vented or used for EOR floods in the Niagaran reefs. With the project duration set for 30-60 days, the average injection rate was planned for 100-300 metric tons per day, with a maximum rate of 600 tons per day. Bottom-hole fracture pressure and corresponding maximum surface fracture pressures were used as limits for injection. The MRCSP wanted to ensure they had a means of maintaining control of the geology of the region at all injection rates. In particular they needed to make sure they were not injecting at rates that would fracture the formation or breach the capping Amherstburg layer. Assessment of the acoustical emissions created by the injection process was also needed to refine the geomechanical model. ESG Solution ESG designed and deployed a temporary ResMap™ solution for monitoring the CO2 injection. Data was acquired using two 8-level dual 3-component sensor arrays, spaced 15 meters apart and deployed down two monitoring wells on retrievable bow spring clamps. Proper design and deployment of these downhole arrays resulted in excellent coupling to the formation for maximum event detection.
Fig. 1: The bow spring sensor array being deployed
The sensor arrays were also designed to be used for 3D VSP shoots. ESG’s Paladin Data Acquisition Units were used to digitize and record the data, and GPS and fiber optic communications were used between the well sites to ensure accurate time stamping. Figure 2 and 3 illustrate the schematics of the sensor arrays, including their location error and ranges of detectability. Fig. 2: Sensor array schematic showing location error
Three string shots and four perforation shots were used to orient the sensors, providing P-wave dominant events of a known location that allowed for the calculation of the orientation of the sensors in the observation wells. An initial velocity model was developed from the monopole and dipole sonic logs provided by the client. In order to correct for the variances in the velocity model, ESG performed its own proprietary Particle Swarm Optimization technique on the orientation shots to refine the velocity model and enhance the event location accuracy 26 microseismic events were recorded during the injection period, of particular importance was some microseismic activity captured by both arrays that occurred following an increase in the gas injection rates. This activity was located within the upper region of the Bois Blanc, immediately below the Amherstburg formation’s capping layer. Because the activity was detected immediately following a period of increased injection pressure, it is likely the injection activated a previously existing structure within the formation or that the increased injection pressure itself caused a fracture within the formation. The majority of the events recorded lined up in a certain orientation, indicating there may be some form of intersection with a regional structure in the subsurface. Outcome The microseismic activity detected near the capping shale gave the operator a clear indication that their injection was influencing the formation, either through fault activation by the creation of new fractures. Incorporating this information into their operating practices can help determine optimal injection rates as well as provide operators with direct feedback of injection compliance.
Deployment on bow spring clamps provided a cost-effective tradeoff between higher cost conventional wire line deployment, and permanently cementing the sensors in the well.
Such knowledge will be useful when conducting future injections. ESG continues to work with the MRCSP on a variety of CO2 sequestration projects, and is the main provider of passive seismic monitoring instrumentation and services for several large scale CO2 EOR operations, including the one in Weyburn, SK.
Join our team! ESG is currently hiring. Please go to our careers section to see a listing of current job postings.
|
