Researchers mapped out 1.8 billion individual tree canopies across millions of kilometers of West Africa’s Sahel and Sahara regions. It’s the first survey ever to map out trees in detail over such a vast area of land.
How did they map out so many trees? A team analyzed a massive database of satellite images via artificial intelligence (AI). They used neural networks that can recognize trees based on their colors and shapes.
First, lead author Martin Brandt conducted the laborious process of identifying and labeling almost 90,000 trees himself. The AI system was then trained by introducing it to satellite images of these manually traced trees. These images informed the computer what a tree looked like, and as a result, the system could select individual canopies from the thousands of pictures in the database.
In an overview of the research, published in Nature on October 14, scientists at New Mexico State University claim “it will soon be possible, with certain limitations, to map the location and size of every tree worldwide.” Such a task would have taken humans millions of years to complete without the AI system.
In 2015, there were approximately three trillion trees on Earth. This was then scaled up from data collected in forests worldwide. However, that number underestimates tree cover in drier areas like the Sahel and Sahara deserts, as there wasn’t much data from these regions.
When the team found out how many trees were growing in dry areas, they were shocked. It’s crucial for ecology to understand how much vegetation is found in deserts.
Jesse Meyer, a NASA programmer who worked on the research, said:
For preservation, restoration, climate change, and so on, data like this are very important to establish a baseline. In a year or two or ten, the study could be repeated to see if efforts to revitalize and reduce deforestation are effective or not.
In addition to measuring deforestation, the satellites could also determine how much carbon is stored in deserts, which is a factor that’s currently not included when scientists model climate change.
This survey is promising, but it’s still far from knowing whether the new data will impact our understanding of climate change.
