Assessing the Effect of Velocity Model Accuracy on Microseismic Interpretation at the In Salah Carbon Capture and Storage Site

Abstract

Injection of carbon dioxide (CO2) to be stored at depth at the In Salah Carbon Capture and Storage (CCS) site began in 2004 with the subsequent installation of a pilot microseismic monitoring well in 2009. The project is one of only two industrial-scale storage projects to have been monitored for microseismicity. Such projects are vital to demonstrate the validity of CCS technology and the role of microseismic monitoring, a technology that could be used in real-time to regulate the geomechanical response of a site to CO2 injection. Substantial microseismicity (over 9000 events) was recorded by a single three-component geophone situated at 80m deep almost directly above one of the In Salah injection wells. The events occur in two main clusters with estimated locations of wellrecorded events within one of these clusters to be within 1km horizontal distance from the geophone and between 2.1km and 2.7km deep, at least 200m below the injection depth and CO2 storage interval. Errors in the depth range of event locations are investigated using modified velocity models, revealing that 10% slower velocities create uncertainties up to 450m in depth. Alternatively, 20% slower velocities in the shallow sub-surface or an anisotropic model have a similar effect. Independent of the absolute depth, there is no migration of event locations to shallower depths with time. Evidence from the analysis of shear-wave splitting delay times implies that, between 2009 and 2011, CO2 injection is opening pre-existing fractures that then close as pressure decreases, rather that creating new fractures. The estimated dominant fracture orientation is approximately NW-SE, in agreement with fracture orientations inferred from logging data, and the observed maximum moment magnitude, MW = 1.7, is also consistent with estimated pre-existing fracture dimensions at the injection depth. This work demonstrates the value of microseismic monitoring of CCS projects, even with a limited array, but an accurate velocity model is critical to allow reliable interpretation of the data. We recommend that microseismic monitoring is conducted prior to CO2 injection at future CCS sites to enable baseline and comparative studies. Real-time microseismic monitoring would help inform injection decision and contribute to the safe operation of a project. © 2013 The Authors. Published by Elsevier Ltd. Selection and peer-review under responsibility of GHGT.