| Title | Using In-Situ Strain Measurements with the SIMFIP Tool to Test Different Proposed Methods of Estimating Stress from Shut-in Pressure Transients |
|---|---|
| Authors | Yves GUGLIELMI, Mark MCCLURE, Tim KNEAFSEY, Tom DOE, Pengcheng FU, Joe MORRIS, Hunter KNOX, Vince VERMEUL |
| Year | 2022 |
| Conference | Stanford Geothermal Workshop |
| Keywords | SIMFIP, DFIT, geomechanics, stress measurement |
| Abstract | The EGS Collab project is a mesoscale project performed at 1.25 and 1.5 km at the Sanford Underground Research Facility. A series of fracture injection tests and flow tests have been performed at two different locations within the mine. These tests involved cycles of injection at pressures above the minimum principal stress with fracture opening and propagation, followed by extended shut-in periods. The tests were instrumented with the SIMFIP tool, a double-packer probe with a high-resolution three-dimensional displacement borehole sensor. The tool allows direct observation of strain as the fractures open and close during the tests. These strain measurements can be correlated with pressure measurements to provide a high fidelity, direct measurement of the minimum principal stress. Typically, in practical applications, only pressure measurements are available. Different methods to estimate stress from shut-in pressure transients have been proposed in the literature, and sometimes they yield meaningfully different results. Thus, because of the difficulty of validating/testing these different proposed interpretation methods, stress measurement interpretations are sometimes ambiguous and/or debatable. The SIMFIP measurements provide an uncommon opportunity to test these proposed pressure transient methods against direct physical measurements. In this study, we compare the SIMFIP measurements against four extended shut-in pressure transients from the EGS Collab project. The shut-in transients were analyzed with two different techniques – the ‘tangent’ method and the ‘compliance’ method. In three of the four tests, the tangent method significantly underestimated the minimum principal stress. The compliance method was reasonably accurate in all four tests. |