With the help of machine learning algorithms and artificial intelligence, scientists were able to determine the approximate amount of carbon found in African trees. This will provide a better understanding of the carbon cycle and climate change.
Using artificial intelligence and satellite imagery, scientists were able to find out how much carbon is in trees in arid regions of Africa. Earlier it was impossible to collect accurate information, but modern technology has helped scientists. According to SciTechDaily, this information is valuable for predicting future climate changes.
Focus.Technologies own telegraph channel. Subscribe to not miss the latest and most interesting news from the world of science!
Using satellite imagery and artificial intelligence, a NASA team mapped nearly 10 billion trees in Africa’s arid regions to estimate the amount of carbon outside the continent’s rainforests.
The researchers found that the semi-arid regions of Africa actually have more trees than previously thought, but store less carbon than the simulations suggested. In total, about 0.84 petagrams of carbon are found in the arid regions of Africa. 1 petagram contains 1 billion metric tons.
Such an estimate is important for predicting climate change because it has a direct impact on how long trees and plants store carbon. Grasses and shrubs store very little carbon due to their short lifespan, unlike trees.
Outside of the African rainforest, Africa’s landscape is generally scattered from dry grasslands with single trees to savannas and wet areas with many scattered trees. Therefore, scientists’ estimates are complex, and the amount of carbon in studies is often underestimated or, conversely, overestimated.
“Our team has collected and analyzed carbon data down to the level of individual trees in vast semi-arid regions of Africa and elsewhere – something that has only been done on small local scales before,” said project lead scientist and scientist Compton Tucker. Goddard (NASA) at the Space Flight Center in Greenbelt, Maryland.
Carbon is constantly circulating between land, atmosphere and ocean. Trees remove it from the atmosphere during the process of photosynthesis and store it inside themselves. Because of this feature, it is quite common to propose an increase in tree cover as a way to offset increased carbon emissions.
In the study, the team used artificial intelligence to analyze more than 326,000 satellite images. Martin Brandt of the University of Copenhagen collected AI training data from around 90,000 individual trees, and his colleague Ankit Kariyaa adapted the neural network to detect individual trees in satellite images.
Researchers defined a “tree” as something with a green crown and a shadow near it. They used this definition to teach the Blue Waters supercomputer at the University of Illinois to count trees. The team compared the results to human measurements in the field: the computers were 96.5% accurate in measuring the tree canopy area.
Using information about the crown region and allometry, which studies the variation of individual characteristics of living things with size, scientists can determine the amount of carbon in the leaves, stems and roots of trees. A team of researchers analyzed 30 tree species to measure leaf, tree and root mass. Thus, they established a statistical dependence on the area of the tree crowns.
Focus He also talked about the desperate steps scientists want to take to save Earth.
Source: Focus
Ashley Fitzgerald is an accomplished journalist in the field of technology. She currently works as a writer at 24 news breaker. With a deep understanding of the latest technology developments, Ashley’s writing provides readers with insightful analysis and unique perspectives on the industry.