A study of induced seismicity: From micro- to macro-scale

点击翻译按钮获取中文摘要

The number of earthquakes in North America has increased dramatically over the past decade, where many of these earthquakes are large enough to be felt by humans at the surface and damage has been caused in some cases. These events have been termed induced seismicity and result from slip along a pre-existing fault, which is caused by fluid-induced stress changes along that fault. The observed increase of induced seismicity has primarily been linked to wastewater disposal; however, induced seismicity has also been linked both domestically and abroad to hydraulic fracturing and enhanced geothermal systems (EGS). Conversely, as combating global climate change remains a central political issue in the 21st century, the race is on to find suitable renewable energy alternatives of which EGS is one. As a result, injection-induced seismicity and its impact on human populations and the environment will continue to be a central issue in the years to come. To better regulate these industries and minimize negative societal and environmental impacts, fundamental research is required to understand better the key processes that lead to induced earthquakes and how they unfold in the natural environment. This is the overarching objective of the proposed project. The long-term goal of this research program is to improve our understanding of fluid-induced seismicity through novel laboratory experiments, the use and development of data analysis techniques and numerical modelling, and to close the gap between laboratory experimentation and geophysical observations. A series of novel fluid injection experiments on natural and saw-cut faults will be conducted to study the fault reactivation processes where measurements of acoustic emission (AE), ultrasonic velocity, fault displacement, stress, fluid pressure, and injection rate will be collected. AE sensors will be calibrated which will enable comparisons with field-scale seismicity, thereby establishing a link between micro and macro scales. Using numerical modelling, laboratory experiments will be simulated and the relevant physical parameters extracted allowing for laboratory observations to be scaled up using numerical modelling. Lastly, data-driven machine learning techniques will be used to study continuous AE waveform data in the period leading up to an induced slip event. A purely data-driven approach will eliminate any preconceptions about how the data ought to be analyzed and may deliver new insights helping to shed light on the process of fluid-induced fault reactivation. This research program will lead to an improved understanding of critical processes, allowing the resource industry to be better regulated and human and environmental impacts mitigated. The result will be new economic growth opportunities in these resource and green tech industries. Additionally, they will provide insight to inform public policy related to lowering the potential of these earthquakes and improving public health and safety.