Grant: Alliance Grants

Many studies have been dedicated to understanding the occurrence of SCVF and GM issues in Western Canada, including the quantification of GHG emissions from these sources. Such studies align closely with current federal and provincial policies that focus on enhanced screening techniques through routine monitoring and improved detection technologies. Current efforts and state-of-the-art methodologies, however, are primarily reactive and insufficiently mature-aiming to detect and quantify fugitive gas emissions after well integrity failures have occurred through the use of advanced monitoring technologies (e.g., optical gas monitoring, mobile leak detection platforms, and satellite-based platforms), followed by remediation based on identifying the source and depth of the problem. This research seeks to initiate a paradigm shift from a reactive to a proactive approach for managing well integrity issues. Our goal is to develop advanced predictive models using machine learning that can forecast potential well integrity failures. This shift represents a new direction in the field, aiming to eliminate GHG emissions at the source rather than simply mitigating them after failures have been detected and releases have occurred.