Researchers at ETH Zurich have developed a world map that for the first time uses machine learning to derive vegetation heights from satellite images in high resolution.
by Stéphanie Hegelbach
Using an artificial neural network, researchers at ETH Zurich have created the first high-resolution global vegetation height map for 2020 from satellite images. This map could provide key information for fighting climate change and species extinction, as well as for sustainable regional development planning.
Last year marked the beginning of the UN Decade on Ecosystem Restoration. This initiative is aimed at halting the degradation of ecosystems by 2030, preventing it going forward and, if possible, remedying the damage that has already been done. Delivering on these kinds of projects calls for accurate foundations, such as surveys and maps of the existing vegetation.
In an interview, Ralph Dubayah, the Principal Investigator of NASA’s Global Ecosystem Dynamics Investigation (GEDI) mission, explains: “We simply do not know how tall trees are globally. […] We need good global maps of where trees are. Because whenever we cut down trees, we release carbon into the atmosphere, and we don’t know how much carbon we are releasing.”
Analyzing and preparing precisely this kind of environmental data is what the EcoVision Lab in the ETH Zurich Department of Civil, Environmental and Geomatic Engineering specializes in. Founded by ETH Zurich Professor Konrad Schindler and University of Zurich Professor Jan Dirk Wegner in 2017, this lab is where researchers are developing machine learning algorithms that enable automatic analysis of large-scale environmental data. One of those researchers is Nico Lang. In his doctoral thesis, he developed an approach—based on neural networks—for deriving vegetation height from optical satellite images. Using this approach, he was able to create the first vegetation height map that covers the entire Earth: the Global Canopy Height Map.
Continue reading… “Neural network can read tree heights from satellite images”