When Autumn Leaves Begin to Fall: As the Climate Warms, Leaves on Some Trees are Dying Earlier

The finding counters scientists’ previous assumptions and indicates a reduction in the amount of carbon deciduous forests can remove from the atmosphere.

Tourists are seen playing with falling maple leaves at a Mughal garden during the autumn season in Kashmire on Nov. 12, 2020.
Tourists are seen playing with falling maple leaves at a Mughal garden during the autumn season in Kashmire on Nov. 12, 2020. Credit: Idrees Abbas/SOPA Images/LightRocket via Getty Images

Share this article

When the last shimmering autumn leaves drift to the ground, it doesn’t just mark a seasonal change. It’s also a turning point in the global carbon cycle, as forests and other plants start to emit carbon dioxide instead of soaking it up.

New research shows that, as the planet warms, deciduous trees in temperate European forests are losing their leaves earlier, and that could reduce the amount of CO2 forests will remove from the atmosphere in the decades ahead.

Against previous expectations, leaves are likely to fall earlier in the autumn,” said Constantin Zohner, a climate biologist with the Crowther Lab at ETH Zürich who was a co-author of the new study on leaf senescence, published in the journal Science on Thursday.


We deliver climate news to your inbox like nobody else. Every day or once a week, our original stories and digest of the web's top headlines deliver the full story, for free.


For decades, scientists have projected that increasing temperatures and CO2 would cause trees to shed their leaves later in the year and lengthen the overall growing season, which would help slow the rate of global warming. But the new study “reverses our expectations” that global warming would delay leaf die-off by two to three weeks, Zohner said. 

Accurately measuring how carbon dioxide cycles through forests is critical for predicting how fast and how much Earth will warm, and for understanding how forests can help in the fight to limit the buildup of greenhouse gases. The findings “substantially lower our expectations of the extent to which longer growing seasons will increase seasonal carbon uptake in forests,” the scientists wrote in their study.

The potential implications of a changing growing season are large, said Mathias Neumann, a forest researcher at the University of Natural Resources and Life Sciences in Vienna, Austria, who was not involved in the study.

“The study suggests that longer growing seasons are associated with lower annual photosynthesis, which is a surprising outcome and contradicts common sense,” Neumann said. Until now, available data from remote sensing and ground observations suggested that a longer growing season on a warming planet would boost carbon uptake by forests, he added.

But instead of slowing the autumn leaf die-off by the expected two to three weeks, global warming could hasten it by three to six days by the end of the century, despite projections for warmer autumns. That’s because trees don’t just respond to light and temperatures. 

The timing of the leaf die-off is also tied to how much trees engage in photosynthesis during the entire growing season. 

Trees convert and store carbon by using sunlight to make nutrients from CO2 and water. More CO2 and warmer weather speeds up tree metabolism, and when the trees have converted and stored enough carbon for the annual growing cycle, the leaves become an energy drain, so they are shed, the study found.

Accurately extrapolating the findings from the European forest study to the global scale requires more observations and modeling. But Zohner estimated that the change in timing could reduce annual carbon uptake by trees in temperate forests by about one gigaton per year from previous estimates. 

Watching Every Leaf

A new record of tree and leaf observations from 1948 through 2015 underpins the  findings. In some cases, there were reports on autumn leaf senescence from individual trees for that entire time, enabling scientists to see the timing change, Zohner said. He added that the research also highlighted the importance of citizen scientists, who regularly report early leaves and blossoms in the spring, and leaf changes in the fall.

Some of the previous difficulty in establishing the timing of autumn leaf die-off stems from the fact that it’s more difficult to observe than when leaves emerge in the spring.

Autumn senescence is a bit messier, Zohner said. “Hobby biologists wouldn’t necessarily know, did this leaf just die? Or is it really a sign of the tree shutting down photosynthesis at the end of the growing season?”

He continued, “The models didn’t work for senescence. We could predict spring with much greater confidence. In the spring, as soon as the leaf emerges, photosynthesis starts. People just assumed that autumn would be hard to predict.”

When the researchers combined the detailed record of leaf observations with the results of lab experiments tracking photosynthesis, and then plugged that data into climate models, the trend toward earlier leaf die-off became clear.

“We never really expected to have this drastic effect,” Zohner  said.

Neumann said the significance of the new research will depend on whether the results withstand scrutiny from subsequent studies, and how the results could be scaled up from the narrow species selection to other regions and evergreen species. Those are  “particularly important in cold climates, where climate change is expected to lead to considerable changes,” he said.

“Eastern North America, eastern China and Japan features ecosystems that are comparable,” where the findings could be used to assess carbon cycle impacts on a broader scale, Neumann said. Extrapolating the findings will be most useful in other parts of the world where similar trees are common, he added.

Temperate forests hold about 10 percent of terrestrial carbon, more than the boreal forests of northern latitudes, even though they cover a smaller area. 

Tree researchers know that spring has been coming earlier, but “fall has always been like, spin the wheel and see what you get,” said Christy Rollinson, a forest ecologist with the Morton Arboretum in Illinois, who wrote an accompanying Perspective piece on the new study in the same issue of the journal. “At the arboretum, the biggest question I get is, when are the leaves going to change, and it has been very, very difficult to forecast.”

The new research can help explain the changes in the progression of fall color, which is an important economic factor in many U.S. regions, she said. And, she added, the lab experiments used in the study also show how and why that timing is changing, which is more useful than just showing a correlation between rising CO2, warmer temperatures and changes in fall color timing.

Zohner added that the new study can also help assess other climate impacts of changing forests.

“We need to know the growing season length to measure the energy budget of the Earth,” he said. For example, when trees are bare, it changes how much incoming solar radiation is trapped near the surface, because leaves are shinier than bare branches and open ground, so a full canopy of leaves reflects some heat upward, allowing it to escape the atmosphere again. When the leaves are gone, it also gives understory plants more light, which can change their growth pattern, he added.

But perhaps the most important calculation is how much carbon trees can take out of the atmosphere in the race to slow global warming. And the new findings show we can’t rely on existing forests to dig us out of our carbon hole.

“If we want to increase forests as a way to capture carbon, we need to plant new trees, or restore forests,” he said. “Trees want to help us, but there are limits.”