In the summer of 2022, the EGS Collab Project conducted a 2-month long multi-well hydraulic circulation test in the 4100L drift of the Sanford Underground Research Facility. This circulation test was conducted in a fracture network that had previously been hydraulically stimulated in the Yates Amphibolite crystalline rock formation hosting the second EGS Collab testbed. The cross-well injection and production of water was continuously conducted and measured using an instrumented network of downhole straddle packers which were operated through a flow and pressure-controlled hydraulic injection system. Onsite observers and an array of downhole geophysical monitoring systems recorded crucial data to supplement the hydraulic pressure and flow measurements from the injection control system. EGS Collab Team members executed a series of shut-in tests during the final days leading up to, and upon, the conclusion of the hydraulic circulation test. The purpose of this paper is to present the results and findings from the shut-in testing by describing 1) the primary motivations for conducting the tests; 2) the instrument configuration leading up to and through the shut-in tests; 3) the shut-in execution sequence and technical challenges; 4) data collected from the hydraulic injection control system, geophysical monitoring arrays, and on-site personnel on the 4100L; and, 5) interpretations and lessons/implications from this shut-in series. Shut-in tests are routinely conducted to assess fracture closure and evaluate in-situ stress parameters of the rock formation. In a multi-zone test bed, the pressure responses also provide information on connectivity of the injection zone to monitored intervals. By conducting shut-in testing after hydraulic stimulation and circulation, phenomena such as local stresses and near-wellbore interactions (i.e., skin effects) can also be examined. Analyses of the shut-in tests add crucial technical context specific to the EGS Collab experiment and have more broad implications to larger-scale applications in relevant conditions (e.g., commercial-scale EGS in crystalline rock). Straddle packers were deployed in four subhorizontal wells - one injector and three producers - at various depths. The two-month long injection was conducted at a rate of 3.4 liters per minute at a nearly steady injection pressure of over 30 MPa (approximately 4,400 psi). During that time, the intervals in the producing boreholes were open to atmospheric pressure. The production flow rates from short, targeted intervals, as well as the borehole sections above and below those intervals, were monitored continuously for outflow. Production rates in the wells ranged from 10’s to 100’s of mL per minute. Significant water production was also measured from multiple locations in the 4100L drift. The rates and locations of outflow varied over time. Shut-in testing was conducted in two stages. The first stage involved closing seven production zones (straddle packer intervals and bottom zones) while maintaining the 3.4 liter per minute rate in the injection well. This shut-in stage provided information on the pressures in the isolated production intervals. Once all production zones were shut-in, injection continued for another 24 hours before initiating a second shut-in stage in which injection was stopped and the injection interval was shut-in to monitor pressure decay. Pressure decay responses generally indicate that the stimulated fracture network had an average residual pressure of approximately 20 MPa, in agreement with minimum principal stress estimates obtained in a nearby borehole. Observed evolution of the outflow system in the wells and the drift indicates a dynamic, interconnected fracture network throughout the testbed. Geophysical monitoring data were recorded during the shut-in sequence, which provided additional insights on the timing and spatial distribution of flow propagation, changes in stress, and fracture relaxation.

Press the Back button in your browser, or search again.

Copyright 2023, Stanford Geothermal Program: Readers who download papers from this site should honor the copyright of the original authors and may not copy or distribute the work further without the permission of the original publisher.