AI that automatically detects methane plumes from orbit could be a powerful tool in combating climate change

The methane plume detection by the machine learning model. This particular plume was previously manually processed. The machine learning model can make this detection automatically. Image credit: Open source EMIT data (NASA) processed by Vít Růžička, illustration by Trillium Technologies.

For the first time, University of Oxford researchers, in partnership with Trillium Technologies’ NIO.space have developed a machine learning tool to automatically detect methane plumes on Earth from orbit using machine learning with hyperspectral data from the NASA’s new EMIT sensor. The AI can identify excessive ‘super emitters’ from hyperspectral satellites and could enable more effective action to reduce gas (GHG) emissions and achieve net-zero targets by allowing satellites to work together. The findings have been published today in the journal Nature Scientific Reports. 

Although net zero targets focus on carbon dioxide emissions, action on combating methane emissions remains a much faster path to slowing rising temperatures. Methane is 80 times as effective in trapping heat as CO 2 , however it has a much shorter atmospheric lifetime (around 7 to 12 years compared to centuries). Acting quickly to reduce methane emissions from anthropogenic sources would have an immediate impact on slowing global heating and improving air quality.