At first glance, the seemingly barren expanses of the Sahel and Sahara deserts are characterized by little greenery, but detailed satellite images combined with deep computer training reveal a different picture.
In fact, about 1.8 billion trees are dotting parts of West Africa’s Sahara and Sahel deserts and the so-called sub-wetland, a previously unknown grace that overturns previous assumptions about such habitats, the researchers said.
“We were very surprised that a lot (so many) trees were growing in the Sahara Desert,” lead author Martin Brand told AFP.
“There are certainly huge areas without trees, but there are still areas with high tree density, and even between the sand dunes there are some trees growing here and there,”
The study provides researchers and conservationists with data that can help guide efforts to combat deforestation and more accurately measure onshore carbon storage.
“For conservation, recovery, climate change, and so on, data like this is very important for baseline,” said Jesse Meyer, a programmer at NASA’s Goddard Space Flight Center who worked on the study.
“In a year or two or ten, the study could be repeated … to see if efforts to revive and reduce deforestation are effective or not,” he said in a press release from NASA.
Finding and counting the trees was not an easy task. In areas with many trees thick lumps of growth appear relatively clearly in satellite images, even at low resolutions, and are easily distinguished from bare lands.
But where they are more common, satellite imagery may be too low resolution to select individual trees or even small groups.
Higher resolution images are already available, but even then the problems remain: counting individual trees, especially over vast areas, is an almost impossible task.
Brand and his team came up with a solution, pairing satellite imagery at very high resolutions with deep training – essentially training a computer program to do the work for them.
But that didn’t mean they could just sit and wait for the results.
Before the in-depth training program could work, it had to be trained, a cumbersome process in which Brand individually counted and labeled nearly 90,000 trees. It took him a year.
“The level of detail is very high and the model needs to know what all kinds of different trees look like in different landscapes,” he said.
“I did not accept misclassification and additional training when I saw misclassified trees.”
Determination of baseline level of protection
The effort was worth it, he said, allowing what would take years of work for millions to be calculated in hours.
“Other research is based on estimates and extrapolations, here we see and count each tree directly, this is the first wall-to-wall assessment.”
The study, published Wednesday in the journal Nature, covers an area of 1.3 million square kilometers (about 500,000 square miles) and includes analysis of more than 11,000 images.
The technique suggests that “it will soon be possible, with certain limitations, to map the location and size of every tree in the world,” wrote Niall P. Hanan and Julius Anchang of the New Mexico Department of Plant and Environmental Sciences in a review of research.
And accurate information on vegetation in deserts and other arid areas is “fundamental to our understanding of global ecology, biogeography and the biogeochemical cycles of carbon, water and other nutrients,” they wrote in a review commissioned by Nature.
Better information could help determine how much carbon is stored in those facilities that are not typically included in climate models, Brand said.
But it is too early to say whether an accurate count of the life of this tree will affect how we understand climate change and its acceleration, he added.
He hopes to now use the technique elsewhere to map more previously hidden trees within 65 million square kilometers (25 million square miles) of the world’s arid regions.
© Agence France-Presse