NASA Discovers 1.8 Billion Trees In The Sahara Desert Using Artificial Intelligence

• NASA discovered these trees scattered in the West African Sahara and Sahel regions, which are typically dry.
• The experts used a deep-learning algorithm equipped with very high-resolution satellite imageryto identify trees.
• The new AI model can automatically identify and map trees over large areas in just a few hours.

When one thinks of the Sahara Desert, it’s usually smooth stretches of land that seems to go into infinity, with very little clumps of trees in sight. But a study from NASA says different – not only are there trees in the Sahara, there are BILLIONS of them.

With the help of powerful supercomputers and machine learning algorithms, NASA discovered these trees scattered in the West African Sahara and Sahel regions. These areas are typically dry.

So when scientists from NASA’s Goddard Space Flight Center, along with a team of collaborators, found billions of trees in these arid parts of the world, it was something they totally didn’t expect.

These experts used a deep-learning algorithm equipped with very high-resolution satellite imagery (0.5 m per pixel) to identify trees and currently, there are now more than 1.8 billion of them that have been seen across an area of more than 500,000 square miles.

This technological breakthrough allows experts to get data faster and efficiently. Normally, it would take months to map non-forest trees at this level of detail.

The coding of the training data took more than a year, which involves marking almost 90,000 individual trees manually across a variety of terrain. This allows the deep learning tool to understand which shadowns and shapes signify the presence of trees in an area.

But once the AI model is trained, it can automatically identify and map trees over large areas in just a few hours. The data is useful for researchers, conservationists, and policymakers in their environmental conservation work.

“We were very surprised to see that quite a few trees actually grow in the Sahara Desert, because up until now, most people thought that virtually none existed. We counted hundreds of millions of trees in the desert alone. Doing so wouldn’t have been possible without this technology. Indeed, I think it marks the beginning of a new scientific era,” explains Martin Brandt, an assistant professor of geography at the University of Copenhagen and the study’s lead author.

“In one kilometer of terrain, say it’s a desert, many times there are no trees, but the program wants to find a tree. It will find a stone and think it’s a tree. Further south, it will find houses that look like trees.”

“It sounds easy, you’d think – there’s a tree, why shouldn’t the model know it’s a tree? But the challenges come with this level of detail. The more detail there is, the more challenges come.”

The next step for the researchers is to map over a larger area in Africa and establish a global database of all out-of-forest trees.