The global push to plant trees has become one of the most visible responses to climate change. Governments, corporations, and nonprofits have pledged to put billions of saplings into the ground, often with ambitious timelines and impressive-sounding numbers. But new research is forcing a more careful conversation about where those trees should go and where they should not.
A study published this year analyzed the ecological outcomes of tree planting across different biomes and found that tropical moist broadleaf forests stand apart as the gold standard. In these environments, planted trees store more carbon per hectare, support greater species diversity, and integrate more successfully into existing ecosystems than in any other biome studied. The warm, wet conditions accelerate growth. The dense canopy structure provides habitat layers from the forest floor to the treetops. A sapling planted in a tropical rainforest is joining a system that already knows how to use it.
When Planting Backfires
The same study found that planting trees in grasslands and savannas can produce the opposite of the intended effect. These open ecosystems evolved without dense tree cover. Their grasses, soils, and fire regimes are adapted to sunlight and periodic burning. Introducing trees disrupts these processes, shading out native grasses, altering soil chemistry, and suppressing the fires that many grassland species depend on for survival.
In parts of the Brazilian Cerrado and the East African savannas, tree planting initiatives have replaced biodiverse grassland with monoculture plantations of eucalyptus or pine. These plantations absorb carbon, but they support a fraction of the species that the original grassland hosted. The net ecological result is a loss, dressed up as a gain.
This creates a friction between climate goals and biodiversity goals that policymakers have been slow to acknowledge. Carbon accounting treats a tree as a tree regardless of where it stands. A eucalyptus plantation in a former savanna registers the same carbon credit as a native hardwood in a regenerating rainforest. The metric fails to capture the ecological difference between the two, and that failure has real consequences on the ground.
Respecting Open Ecosystems
Grasslands and savannas are not empty spaces waiting to be filled. They are functioning ecosystems with their own carbon sequestration pathways. Grasses store carbon primarily in their root systems, which can extend several meters below the surface. This belowground carbon is more stable and more resistant to fire than the aboveground carbon stored in tree trunks. Peatlands, another open ecosystem frequently targeted for afforestation, hold enormous carbon reserves in their waterlogged soils, reserves that tree planting can actually release by draining the landscape.
The distinction matters for policy. Restoration programs that respect the ecological identity of each landscape will deliver better outcomes for both carbon and biodiversity than programs that apply a single solution everywhere. A sapling in a rainforest is a continuation. A sapling in a grassland is an intrusion.
The researchers behind the study advocate for biome-specific restoration targets that prioritize native species and natural ecosystem processes over raw tree counts. In tropical forests, aggressive planting and natural regeneration can work in tandem. In savannas, the better investment may be protecting existing grassland from conversion rather than converting it further.
Not all earth is meant to hold a forest. Recognizing that fact is not a retreat from climate ambition. It is a refinement of it, one that treats the complexity of the living world as something to work with rather than something to override.