In the heart of Illinois, a groundbreaking approach to monitoring carbon dioxide storage is taking shape, and it’s not just about keeping an eye on the subsurface—it’s about revolutionizing how we understand and interact with it. Hanchen Wang, a researcher at the Los Alamos National Laboratory in New Mexico, has developed a novel deep learning method called DeFault, designed to enhance passive seismic monitoring for carbon capture, utilization, and storage (CCUS) projects.
The stakes are high in the world of CCUS. As industries grapple with the urgent need to reduce greenhouse gas emissions, the success of these projects hinges on a comprehensive understanding of subsurface geology and geomechanics. Passive seismic monitoring, which involves detecting and locating seismic events naturally occurring in the subsurface, offers vital insights into these structures and the migration pathways of fluids like CO2. However, the process is fraught with challenges, including the need for high-quality data and advanced computational methods.
Enter DeFault. This innovative approach leverages data domain-adaptation, allowing researchers to train a neural network using labeled synthetic data and then apply it directly to field data. “DeFault automatically clusters passive seismic sources based on their recording time and spatial locations, and subsequently delineates faults and fractures accordingly,” Wang explains. This automation not only saves time but also enhances the accuracy of seismic event relocation and fault detection.
The efficacy of DeFault was demonstrated in a field case study involving CO2 injection-related microseismic data from the Decatur, Illinois area. The results were promising, with DeFault accurately and efficiently relocating passive seismic events and identifying faults. This capability is crucial for monitoring potential damage induced by seismicity, ensuring the safety of CCUS projects.
The implications of this research are significant for the energy sector. As CCUS technologies become increasingly vital in the fight against climate change, tools like DeFault can play a pivotal role in their deployment. By refining subsurface characterization methods, DeFault and similar machine learning approaches can enhance the safety, efficiency, and reliability of CCUS projects, ultimately contributing to the reduction of greenhouse gas emissions.
Wang’s work, published in the journal “Earth and Space Science,” underscores the potential of machine learning to transform our understanding of the subsurface. As the energy sector continues to evolve, the integration of advanced technologies like DeFault will be key to navigating the complexities of CCUS and ensuring a sustainable energy future.
In the words of Wang, “Our approach highlights the potential of deep learning in passive seismic monitoring, emphasizing its role in ensuring CCUS project safety.” As the world looks to innovative solutions to combat climate change, DeFault stands as a testament to the power of technology and human ingenuity in shaping a greener future.