Xinran Miao was staring at a map of sub-Saharan Africa, where 130,000 tiny dots marked the footprints of restoration projects—each one a promise to bring trees back to degraded land. What she and her team discovered, however, was that some of these projects were doing something unexpected: they were quietly protecting forests beyond their borders.
Across 27 countries, from Kenya to South Africa, researchers have uncovered the first large-scale evidence that tree restoration doesn’t just heal the land inside project zones—it can also shape what happens in the surrounding landscapes. This ripple effect, known as “leakage,” can be either positive or negative, and the difference hinges on how restoration is done.
The study, led by Miao at the University of Cambridge and supported by scientists at the University of Sheffield, analyzed nearly 6.8 million hectares (16.8 million acres) of restored land using decades of satellite data. They found that natural regeneration and active tree planting were linked to a drop in tree loss—up to 9.1 percentage points inside the sites and 8.9 percentage points in the areas within 10 kilometers (6.2 miles) outside. This positive spillover likely happens when communities are involved, land rights are clearer, and people gain fuelwood, jobs, or other benefits that reduce the need to cut down nearby forests.
But not all methods had this effect. Agroforestry—the practice of growing trees on farmland—was tied to more tree loss, both within the projects and in the surrounding areas. On average, tree loss rose by 7.9 percentage points inside agroforestry sites and 7.5 percentage points nearby. The reason? Adding trees to active farms can reduce crop yields by blocking sunlight or competing for water, pushing farmers to clear nearby forests to make up for lost production.
The results varied by country too. In Kenya and Tanzania, most restoration types helped reduce deforestation around project sites. But in Ethiopia and South Africa, some efforts were linked to increased tree loss nearby, highlighting how local conditions—like land ownership, farming pressures, and government support—shape outcomes.
These findings matter for global climate goals. With more than 350 million hectares (865 million acres) pledged for restoration by 2030, the study warns that current carbon markets and certification systems often ignore what happens beyond project boundaries. That means they might overvalue some projects while missing the true benefits of others.
"These findings provide the first large-scale empirical evidence that different restoration strategies can generate heavily contrasting positive and negative effects on deforestation rates in the wider landscape," said Dr. Oscar Morton of the University of Sheffield.
As the world leans into restoration as a climate solution, this study reminds us that success isn’t just about how many trees are planted. It’s about how, where, and with whom they’re grown.