Google and NASA's Jet Propulsion Laboratory have developed an artificial intelligence that tracks global methane emissions from space. The model is called MAPL-EMIT and was presented in the scientific journal Proceedings of the National Academy of Sciences.
The tool uses data from the EMIT spectrometer installed aboard the International Space Station. It automates the detection and localization of methane sources, including cases of overlapping plumes.
The model was trained on 3,6 million physically simulated methane plumes. This allowed it to better handle complex landscapes and to identify leaks in conditions where manual analysis was previously required.
According to test results, MAPL-EMIT detects 50% more methane leaks than human experts. In addition, the system identified more than 23 thousand hidden plumes around the world, including 24 of the 25 largest landfill emission sources.
Municipal solid waste landfills are among the top three anthropogenic sources of methane in the United States. Automating the search for such sites helps accelerate targeted mitigation — the targeted reduction of emissions.
The collaboration continues Google's work on combating super-pollutants. The company previously entered into a partnership with AMP for waste processing and the generation of carbon credits.
How will this affect global efforts to reduce methane emissions in the coming years?
