Global warming caused by humans might be closer to a crucial climate threshold than current estimates suggest. A study1 of Antarctic ice cores argues that, in 2023, human-driven warming reached 1.49 °C above pre-industrial levels.

In 2015, nearly all countries adopted the Paris climate agreement, a legally binding treaty that states that global temperatures should be kept to less than 1.5 °C above pre-industrial levels to reduce the impacts of climate change.

The Intergovernmental Panel on Climate Change (IPCC) suggests that human-induced warming is currently close to 1.31 °C. However, the IPCC does not track warming in individual years; instead, it compares temperature averages calculated over decades, so its figures trail behind current temperatures. And it uses the average temperature between 1850 and 1900 as its ‘pre-industrial’ baseline.

But carbon dioxide levels and temperatures were increasing long before 1850, so the standard 1850–1900 baseline doesn’t capture the full picture. This suggests that current approaches for estimating global temperature changes — which rely on climate models and statistical methods — might underestimate the increase in human-driven warming, says study co-author Andrew Jarvis, a climate scientist at Lancaster University, UK.

Frozen record

To address this problem, Jarvis and his colleague Piers Forster, a climate scientist at the University of Leeds, UK, turned to Antarctic ice-core data that record atmospheric CO2 levels over the past 2,000 years. The researchers used this data to establish a pre-industrial baseline from AD 13 to 1700, a period when CO2 levels were around 280 parts per million and relatively stable. They combined this with global temperature data from 1850 to 2023 to calculate the amount of warming from the baseline in individual years of the twenty-first century.

Their analysis showed that by 2023, CO2 levels had increased by 142 parts per million above the pre-1700 baseline, indicating that human-induced warming had reached 1.49 °C. When Jarvis and Forster switched their baseline to the standard 1850–1900 time frame, they calculated a temperature increase of 1.31 °C. This suggests the more recent baseline doesn't fully capture pre-industrial warming that was already occurring before 1850. Other methods for calculating global temperature change, such as long-lived marine sponges, suggest warming has even passed 1.50 °C.

And they found that temperature-change estimates using their pre-1700s baseline were around 30% more certain than those made with standard methods.

The method is “important and useful” because it offers a relatively quick and simple way to calculate near-real-time estimates of human-induced warming without relying on models or statistical methods that are prone to uncertainties, says Richard Betts, a climate scientist at the University of Exeter, UK. “We need immediate information to inform urgent responses,” he says.