A Big Tech-backed campaign to plant trees might have taken a wrong turn

Roughly half of the land targeted by a major tech-backed forest restoration campaign in Africa was never meant to be forest, according to a new analysis. Planting trees in the identified area could actually harm grasslands and savannas that may have been inadvertently mislabeled as “forests” in need of help, the report concludes.

The paper, published in the journal Science today, takes stock of AFR100, an initiative endorsed by 34 governments in Africa and that counts the Bezos Earth Fund and Meta among its major funders. The goal of AFR100, short for African Forest Landscape Restoration Initiative, is to restore at least 100 million hectares of land by 2030. AFR100 disputes the new analysis.

For forest restoration to be successful, the right kinds of trees need to be in the right places. It’s easy to mess that up, and this recent analysis aims to show just how big of a problem that could be across one continent. While it focuses on one initiative, the authors say it’s probably emblematic of major flaws across international conservation campaigns.

For forest restoration to be successful, the right kinds of trees need to be in the right places, which is easy to mess up

“We had suspected that this is a threat, but just the sheer extent of it was absolutely enormous,” says Catherine Parr, lead author of the paper and an ecologist at the University of Liverpool. “Some countries where there weren’t even any forest at all, planning on planting those trees and badging it as reforestation — that’s really quite a shock.”

Nearly one-fifth of the total area set aside for restoration — 25.9 million hectares — spreads across eight countries that naturally lack forest cover, Parr and co-authors from the University of Oxford and Utrecht University found in their analysis. That includes land in Burkina Faso, Chad, Lesotho, Mali, Namibia, Niger, Senegal, and Gambia. Eighteen countries in total have committed to “restore” an area greater than the amount of forest it should actually have, according to the analysis.

“The article is riddled with a lot of inaccuracies,” a spokesperson for AFR100 wrote to The Verge in an email. Gambia is not currently included in the initiative, according to AFR100, which would bring the figure attributable to AFR100 countries down to 21.9 million hectares. “Even if the Gambia is a member of the AFR100 that small country cannot pledge 4 million hectares,” Teko Nhlapho, communications officer for the African Union Development Agency that co-launched AFR100, said in the email.

To conduct their analysis, the researchers used publicly available information on the AFR100 website and a database of restoration projects maintained by the environmental news organization Mongabay. After looking up projects taking place in AFR100 countries, the researchers compared those locations to biome maps commonly used to identify what kinds of habitats are present. That’s how they came to the conclusion that many of the areas identified for restoration actually contain grasslands or savannas — not forests in need of more trees.

According to the analysis, around half of the total area committed for restoration in AFR100 countries is in savannas or grasslands, where planting trees could actually harm the local ecosystem. And since Parr suspects grassland and savanna cover is underestimated in biome maps, Parr says the figures in the analysis are actually quite conservative.

The authors argue that conservation groups need to change the way they identify land for restoration. Relying on measures of tree cover by satellite is one issue. Another is a standard commonly used by conservationists that defines forests as areas with at least 10 percent tree canopy cover. Parr says that process can wrongly categorize open areas with some trees, often savannas, as forest.

Lions forage at Serengeti National Park on February 11th, 2022. Photo by Xie Hao / Xinhua via Getty Images

The Verge also reached out to the World Resources Institute (WRI), a nonprofit organization named in the paper that uses the definition of forests as having 10 percent canopy cover and that maintains an atlas of areas it sees as ripe for restoration. WRI launched AFR100 alongside the African Union Development Agency, the World Bank, and Germany’s Federal Ministry for Economic Cooperation and Development back in 2015.

“AFR100 has made clear that native grasslands should not be converted into forests and this is reflected in its principles,” Sean Dewitt, director of the forest restoration initiative at WRI, said in an email.

Both WRI and AFR100 said in their responses that the authors of the article shouldn’t equate all restoration projects with reforestation. “It should be understood that the total area pledged to the AFR100 initiative is composed of both degraded forests and lands. So it will be a mistake to focus just on degraded forests,” Nhlapho said in an email.

WRI’s Dewitt says that a “vast majority” of restoration projects affiliated with AFR100 are actually agroforestry projects. “Agroforestry projects add trees to existing croplands to improve soil fertility, increase water retention, and reduce topsoil erosion,” he writes.

However, nearly 60 percent of agroforestry projects use non-native species, Parr’s analysis says. “A prime example of tree-centric approaches being misapplied is using agroforestry in non-forests as restoration,” Parr responded in an email. “We agree agroforestry comes with multiple social and economic benefits but increasing tree cover in non-forested systems is not ecological restoration.”

As tree planting campaigns have gotten more popular with brands and consumers conscious about their environmental impact, it has stoked conflicts like this over how effective these kinds of initiatives actually are.

The drama hasn’t deterred some major funders

A 2019 study published in Science on the potential trees have to fight climate change sparked a controversial World Economic Forum campaign to plant a trillion trees. Dozens of scientists published their own scathing critique of that research and the tree planting projects it spawned, saying the research inflated figures on how much potential tree-planting has to sequester planet-heating carbon. The chief scientific advisor for the trillion trees campaign has since left his post and was apparently “begging environmental ministers to stop planting so many trees” at a UN climate conference in December, Wired reported.

The drama hasn’t deterred some major funders. “Our partnership with AFR100 has helped us find and fund over 150 [locally led restoration] efforts, and we are enormously proud of the work they are doing,” Emily Averna, Bezos Earth Fund’s land restoration program officer, said in an emailed statement to The Verge. Meta didn’t respond to a request for comment.

Trees have become a powerful symbol for protecting the environment and stopping climate change, so much so that they could actually risk overshadowing other wildlife in need of conservation. In a literal sense, grasses wither in the shadow of trees. Their “woody encroachment” can crowd out savannas. “The lions, and the wildebeest, and the zebra of the Serengeti, they need those open grassland systems,” Parr says. “Trees are great, but the problem is if we get too many in the wrong place, then we’ve got problems.”