Climate Models Overestimate Forest Carbon Sink by 30% as Warming Slows Tree Growth

The Earth's forests have long served as a vital shock absorber for human emissions, quietly pulling roughly 27 percent of our carbon dioxide out of the atmosphere each year. But the mathematical models used to predict how much carbon these forests will store in the future contain a critical biological blind spot.[1]

According to a new study published today in Geophysical Research Letters, widely used climate land models may overestimate the future carbon storage capacity of forests by as much as 30 percent. The discrepancy stems from a fundamental misunderstanding of how trees respond to hotter, drier air.[1]

For decades, the algorithms powering global climate projections have operated on a relatively simple assumption: if a tree is photosynthesizing, it is growing. Photosynthesis is the process by which leaves draw in carbon dioxide and sunlight to create sugars. Modellers assumed that as long as this process continued, the tree was actively converting that carbon into wood, locking it away for decades or centuries.[4]

However, forest ecologists have increasingly warned that photosynthesis and physical growth are two entirely separate mechanisms that can become "decoupled" under stress. The new Cornell University study, led by postdoctoral researcher Brendan Clark, finally quantifies exactly how much this decoupling skews our climate forecasts.[2][4]

The biological culprit is a mechanism known as turgor pressure—the internal water pressure within a tree's cells. In order for a tree to physically grow, its cells must divide and expand. This cellular expansion requires high turgor pressure, which acts like water inflating a balloon to stretch the cell walls.[1]

When the surrounding air becomes hotter and drier, trees lose moisture through their leaves. To compensate, the water pressure inside their cells drops. Without sufficient turgor pressure, cell division grinds to a halt. The tree stops building new wood, even though its leaves may still be actively photosynthesizing and absorbing carbon dioxide.[1][4]

"The tree may be photosynthesizing, but it's not growing," Clark noted in the study's release. Because current land surface models fail to account for this pressure drop, they systematically assume that all the carbon being absorbed by the leaves is being permanently stored in the trunk.[2]

To measure the impact of this error, Clark's team analyzed eight years of high-resolution growth records from Swiss forests, tracking both broadleaf and coniferous species. They built a statistical model based on the actual biological responses of the trees and compared it against the projections of a widely used open-source land surface model.[1]

The results were stark. The standard climate model overestimated the actual physical growth of broadleaf trees by a factor of two. For coniferous trees, the model overshot reality by a factor of three. The widest gaps between the model's optimistic predictions and the trees' actual growth occurred in regions forecast to experience the most significant increases in heat and atmospheric dryness.[1]

This physiological bottleneck helps explain a growing body of field observations that have puzzled scientists. Just last week, a separate study in Science Advances revealed that oak trees in temperate climates continue to photosynthesize for months after their seasonal wood production has completely shut down.[2]

If the carbon isn't becoming wood, where does it go? Ecologists note that trees redirect this absorbed carbon toward short-term survival mechanisms. It may be used for basic cellular maintenance, stored temporarily as starches, or pushed into the soil through root systems. Unlike dense wood, which traps carbon for generations, these short-term sinks release carbon back into the atmosphere much faster.[2][4]

The implications for global climate policy are profound. Humanity currently relies on the terrestrial biosphere to offset more than a quarter of fossil fuel emissions. If the land sink is 30 percent weaker than projected, atmospheric carbon dioxide concentrations will rise faster than the models currently anticipate, accelerating global warming.

The weakening of the forest sink is not just a future projection; it is already happening. Data from the World Resources Institute indicates that the global net forest carbon sink hit a two-decade low in 2023, driven by extreme heat, drought, and unprecedented wildfires.

Furthermore, researchers have identified a "growth-lifespan tradeoff" in global forests, where trees that grow rapidly in carbon-rich environments tend to die younger, releasing their stored carbon back into the atmosphere sooner than expected.[3][4]

The Cornell team is now working to translate their biological findings into open-source code that can be integrated directly into global climate models. By correcting the algorithms to account for turgor pressure, scientists hope to provide policymakers with a much more accurate—if sobering—picture of the Earth's future carbon balance.[1]

Ultimately, the evidence points to a difficult reality for climate strategy: nature-based solutions, such as mass tree-planting campaigns, cannot absorb as much of our pollution as the math previously suggested. As the planet warms, the trees themselves are losing their capacity to bail us out, placing even greater urgency on the need to halt fossil fuel emissions at the source.[3][4]