To understand how much carbon the earth can store, and how it changes over time, scientists must calculate an obscure number of trees, and find out how they grow over time. Surprisingly, NASA people are now using the supercomputer to do exactly that – through top-down imaging 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 images. Miles.
The team used one of the world’s fastest supercomputers (Blue Waters at the University of Illinois) to analyze “deep learning” on terrain images from large parts of West Africa. They found that they could not only count the trees that had failed to see the satellites before, but at the same time begin to assess the tree’s carbon storage capacity.
Most of the world’s efforts to evaluate a large number of trees have focused well on forest-regions. This is why the NASA team sought to focus on the dry lands and isolated trees in the semi-arid regions of West Africa – for the full picture.
Related: For the first time ever, scientists have identified how many trees to plant and where to plant them to prevent a climate crisis.
“These arid areas are white on the map – they are originally hidden because normal satellites just don’t see the trees,” lead author Martin Brand said in a statement. “They see the forest, but if the tree is separated they cannot see it. We are now on our way to filling these white spots on the map. And it’s absolutely thrilling. “
To train machine-learning algorithms, Brandt, an assistant professor of geography at the University of Copenhagen, individually identifies about 90,000 trees spread across different terrain – giving the software different shapes and shadows to distinguish these differences. The team trained their algorithms to recognize both individual trees and small groups in different regions, ranging from savannas to deserts, and published their new study Nature.
Being in the right training location, a job that took several years of trained eyes to complete, took only a few weeks for artificial intelligence.
Check out: Scientists use recycled sewage water to grow 500 acres of forest in the middle of the Egyptian desert
The team was able to map the crown diameter (width of trees seen from above) of 1.8 billion trees spread over an area of more than 500,000 square miles (1.3 sq km). They also compare variability in tree cover and density under different precipitation patterns – which data plan to compare with the height and biomass data of the next tree to identify the potential for carbon storage.
In the future, such assessments will more effectively track deforestation around the world for conservationists. Global scientists also compare one-year overhead data to see if conservation efforts are working.
Famous: The search engine that planted trees with every search has just planted its 100-million trees
Even in accurate, automated tree calculations, landowners should advance their ability to monetize unused space for planting new trees – to determine how much carbon they store for carbon credits.
Ultimately, researchers will be able to see trees where they can improve the ability to do satellite images before – and measure the carbon storage of those trees – eventually allowing meteorologists to make global measurements of carbon storage on Earth. This will be an important tool in a world where storing our excess carbon is becoming even more crucial.
See NASA video about progress
Plant some positivity for your friends and family on social media …
