To get an idea of how much carbon the Earth can store and how it changes over time, scientists would need to count a staggering number of trees and track their growth over time. Incredibly, the folks at NASA are now using supercomputers to do just that, via top-down imagery from space.
Scientists at NASA’s Goddard Space Flight Center in Maryland recently partnered with an international team of researchers to map trees using high-resolution satellite imagery – more than 1.8 billion trees found outside of forests, in a swath of more of half a million squares. miles.
The team used one of the world’s fastest supercomputers (Blue Waters at the University of Illinois) to perform “deep learning” analysis on terrain images of large sections of West Africa. They found that they could not only count trees that satellites hadn’t seen before, but they could begin to assess the carbon storage potential of those trees at the same time.
Much of the global effort to assess large numbers of trees has focused on well-forested regions. This is why the NASA team sought to focus on isolated trees in drylands and semi-arid regions in West Africa, to get a more complete picture.
“These dry areas are blank on the maps, they are basically masked because normal satellites just don’t see the trees,” said lead author Martin Brandt. in a sentence. “They see a forest, but if the tree is isolated, they cannot see it. We are now on our way to fill in these white dots on the maps. And that’s pretty exciting. “
To train the machine learning algorithms, Brandt, an assistant professor of geography at the University of Copenhagen, personally marked nearly 90,000 trees spanning different terrains, giving the software different shapes and shades to learn the difference. The team also trained their algorithms to recognize individual trees and small clusters in different types of terrain, ranging from savannas to deserts, and published their new study in Nature.
With the proper training in place, a job that may have taken trained eyes several years to complete took just a few weeks for artificial intelligence.
The team was able to map the crown diameter (the width of a tree seen from above) of 1.8 billion trees spanning an area of more than 500,000 square miles (1.3 square kilometers). They also compared variability in tree cover and density under different rainfall patterns, information the team plans to compare with upcoming tree height and biomass data to identify potential for carbon storage.
In the future, assessments of this type will more effectively track deforestation around the world for conservationists. The general data from one year will also be compared to subsequent years for scientists to assess whether or not conservation efforts are working.
Accurate automated tree counting should also promote the ability of owners to monetize the unused space they may have to plant new trees, to quantify the amount of carbon they store for carbon credits.
Ultimately, improving the ability of researchers to detect trees where they previously couldn’t with satellite imagery and to measure the carbon storage of those trees will allow climate scientists to make global measurements of carbon storage on earth. This will be a vital tool in a world where storing our excess carbon is increasingly crucial.
CLOCK a NASA video on the progress …
PLANT some positivity for your friends and family on social media …