A small dip, for the wrong reason
Global emissions from fossil fuels are projected to fall 0.4 percent in 2026 to 38.2 billion tonnes of CO2, according to the 21st Global Carbon Budget from the Global Carbon Project, presented during the COP31 talks in Antalya. The range of uncertainty runs from a 1.5 percent fall to a 0.6 percent rise.
The decline is not mainly a clean-energy story. Oil emissions are projected to drop 1.9 percent, which the authors describe as a likely short-term response to the energy shock caused by the Iran conflict and the closure of the Strait of Hormuz. Coal emissions rose 0.6 percent. When an energy crisis cuts oil demand, the atmosphere gets a small break, but the underlying trend has not changed.
Renewables are doing real work
There is a genuine positive. For the first time the report estimates avoided emissions from solar, wind and electric vehicles at about 0.5 billion tonnes of CO2 in 2026, offsetting roughly two thirds of the emissions growth caused by rising energy demand. In 2025, solar and wind accounted for about 95 percent of that avoided total.
The report also finds that 40 economies significantly cut fossil CO2 while growing their economies between 2016 and 2025, nearly twice the number in the previous decade. Global emissions grew about 0.3 percent a year over the past decade, against 1.7 percent in the one before.
Why the headline is still bad
Total CO2, which adds land-use change to fossil emissions, is projected to reach a record 43.3 billion tonnes in 2026, up from 42.8 billion last year, because deforestation emissions are expected to rise. The remaining carbon budget for a 50 percent chance of holding warming to 1.5C is now about 150 billion tonnes, less than four years at current rates. For 1.7C it is 510 billion tonnes, about 12 years, and for 2C about 1,035 billion tonnes, roughly 24 years.
The authors' conclusion is blunt: clean energy is growing fast enough to bend the curve but not yet fast enough to guarantee a durable decline. A single year of falling fossil emissions caused by a war is not the same as a peak.