New AI method transforms global methane detection

NASAThe entrepreneurship challenge has inspired technological innovations such as GeoLab’s AI method for detecting methane emissions globally. This method improves Accuracy Monitoring greenhouse gases provides a quick way to address climate change impacts.

The NASA Science Mission Directorate (SMD) established the Entrepreneurship Challenge to identify innovative ideas and technologies from small business start-ups that have the potential to advance the agency’s science goals. GeoLab—a prize winner in a recent Entrepreneurship Challenge—has developed a way to use it Artificial intelligence To trace global methane emissions. Methane is a greenhouse gas that contributes significantly to global warming, and this promising new technology could provide data to help decision makers develop strategies to mitigate climate change.

Progress and participation in challenge events

SMD sponsored Entrepreneurship Challenge events in 2020, 2021 and 2023. Prize Money Awarded to Challenge Winners – Total Entrepreneur Challenge Prize Value $1M in 2023 SMD engages the venture capital community in Entrepreneurship Challenge events to leverage external funding sources to develop innovative technologies of interest to NASA. Many challenge winners have subsequently received funding from NASA and external sources (eg, other government agencies or the venture capital community) to further develop their technologies.

Each entrepreneur is challenged in specific focus areas such as mass spectrometry technology, quantum sensors, etc. Metamaterials– based sensor technologies and more. Central areas of the latest 2023 challenge include lunar surface payloads and climate science.

Case Study: Geolab’s Methane Detection Innovation

A recent Entrepreneurship Challenge success story includes 2023 Challenge winner Geolabe – founded in 2020 in Los Alamos, New Mexico by Dr. Claudia Hulbert and Dr. Bertrand Rood-Leduc. The Geolabe team developed a method that uses artificial intelligence (AI) to automatically detect methane emissions globally.

As global temperatures rise to record highs, pressure to curb greenhouse gas emissions has intensified. Controlling methane emissions is critical because methane is the second-largest contributor to global warming, and is estimated to account for one-third of global warming to date. Also, since methane stays in the atmosphere for a shorter time compared to CO₂Controlling methane emissions is widely regarded as one of the fastest ways to slow global warming.

However, monitoring methane emissions and quantifying them remains challenging due to limitations of existing detection methods. Methane plumes are invisible and odorless, so they are usually detected with special equipment such as infrared cameras. The difficulty of finding these leaks from space is comparable to finding a needle in a haystack. The leaks are distributed around the globe, and most methane plumes are relatively small, making them easy to miss in satellite data.

Advances in methane detection technology

Multispectral satellite imagery has emerged in recent years as a viable methane detection tool, enabling routine measurements of methane plumes globally every few days. However, in the case of methane, these measurements suffer from a very poor signal-to-noise ratio, which has so far only allowed the detection of very large emissions (2-3 tons/h) using manual methods.

The GeoLab team has developed a deep learning framework that automatically identifies methane signatures in existing open-source spectral satellite data and decomposes the signal from the noise. This AI method enables automatic detection of methane leaks of 200kg/hour and above, which account for 85% of methane emissions in well-studied, large oil and gas basins. Information obtained using this new technique can help inform efforts to mitigate methane emissions on Earth and automatically verify their effects. This Geolabe project featured Natural communication On May 14, 2024.