A machine learning tool developed by researchers at the U.K.'s University of Oxford and Trillium Technologies' NIO.space uses data from hyperspectral satellites to automatically detect methane plumes from space.
The tool could be used to identify "super emitters" of methane and help cut greenhouse gas emissions.
The model was trained with 167,825 hyperspectral tiles, which of each represent a 1.64 sq. km. (about .6 sq. mi.) area, from NASA's AVIRIS aerial sensor over the Four Corners area of the U.S.
The researchers found the model was more than 81% accurate in detecting large methane plumes, and 21.5% more accurate than the most accurate approach used previously.
The annotated dataset and code have been made available on GitHub.
From University of Oxford (U.K.)
View Full Article
Abstracts Copyright © 2023 SmithBucklin, Washington, D.C., USA
No entries found