As the effort to transition away from fossil fuels gains momentum, scientists have turned to deep learning models to assist in the hunt for new sources of low-carbon energy. In this pursuit, researchers have successfully developed an algorithm that can scan the Earth’s surface for indications of subsurface reservoirs of naturally occurring free hydrogen. By leveraging this advanced technology, they hope to uncover ovoids or semicircular depressions (SCDs) that are often associated with “gold hydrogen” deposits. In recent years, these circular patterns have been discovered in several countries worldwide, revealing their prevalence to be greater than previously thought. The integration of artificial intelligence (AI) has proven to be instrumental in mapping out potential subsurface hydrogen reservoirs, providing a baseline for further investigation and potentially reshaping the global energy landscape.
Identifying SCDs: The Role of AI
Lead authors of two recent papers, Sam Herreid and Saurabh Kaushik, both postdoctoral scholars at The Ohio State University’s Byrd Polar and Climate Research Center, have successfully combined their deep learning model with global satellite imagery data to identify SCDs. Their algorithm was trained using a compilation of known SCD locations, enabling it to search for these features worldwide. By analyzing remote sensing data and studying geomorphic and spectral patterns, they were able to pinpoint areas most likely to be associated with SCDs related to geologic hydrogen. This groundbreaking research was presented at the annual meeting of the American Geophysical Union, highlighting AI’s unique ability to uncover surface expressions of potential subsurface hydrogen reservoirs.
Hydrogen as an Attractive Energy Source
Hydrogen possesses enormous potential as a clean and efficient energy source. Unlike other renewable energies such as wind or solar, hydrogen can be stored and transported, making it highly versatile. When burned, hydrogen only produces water as a byproduct, making it an environmentally friendly alternative. Governments worldwide are increasingly investing in cleaner energy alternatives, generating significant interest in natural hydrogen. Access to this low-carbon energy resource, especially “gold hydrogen” continuously produced within the Earth’s crust, could potentially revolutionize the global energy landscape.
Despite hydrogen’s promising attributes, locating hydrogen reservoirs poses significant challenges. These deposits are believed to occur in different geologies and locations than traditional oil or gas reservoirs, making them difficult to pinpoint. However, AI is aiding in the identification and mapping process, allowing researchers to investigate every potential SCD. Satellite data helps narrow down areas of interest, but distinguishing real hydrogen deposits from other circular-looking land features, such as lakes or golf courses, remains a challenge. As countries around the world race to expedite the search for hydrogen sources, innovative exploration methods are being developed to overcome these challenges.
The Role of AI in Mitigating the Climate Crisis
The integration of AI in the search for natural hydrogen not only contributes to the development of clean energy but also plays a proactive role in mitigating the climate crisis. Sam Herreid, a glaciologist who previously used AI to model the impact of climate change on glaciers, emphasizes the excitement of being involved in a project that is rapidly evolving. Europe is already exploring ways to leverage their stores of gold hydrogen, while in the United States, initiatives like the Inflation Reduction Act aim to expand the clean energy production industry. However, successful integration of natural hydrogen reservoirs as a reliable energy source will require a few more years of research and development.
Moving forward, researchers must focus on deepening our understanding of hydrogen systems. The primary challenge lies in the discovery of more SCDs and a thorough investigation of their formation process. By expanding the knowledge base and uncovering the geological and environmental factors contributing to the creation of hydrogen reservoirs, scientists can overcome the existing challenges. This understanding will be crucial in making natural hydrogen a viable and widespread energy source.
The utilization of AI in the search for natural hydrogen reservoirs marks an important milestone in the transition to cleaner energy sources. Through the development of deep learning models and the analysis of satellite imagery, scientists are making significant progress in mapping out potential subsurface hydrogen deposits. While challenges remain, the exploration of natural hydrogen holds the key to a sustainable and low-carbon energy future. Continued research and innovation in this field will play a pivotal role in mitigating climate change and reshaping the global energy landscape.