Our planet reveals its sensitive side

FORGET the “pause” in global warming. The climate is highly sensitive to greenhouse gases after all, and global temperatures could still shoot up this century. Some recent analyses had suggested that the climate does not respond much to carbon dioxide, but three new studies say this hopeful scenario is unlikely, and Earth reacts strongly to CO2. “All the lines of evidence now point to the high end,” says Drew Shindell of the NASA Goddard Institute for Space Studies in  New York. “They’re all telling us the exact same thing.” Climate sensitivity is a measure of how much Earth will heat up, on various timescales, if CO2 levels in the atmosphere double. The latest Intergovernmental Panel on Climate Change report says  the medium-term warming  will be between 1.5 °C and 4.5 °C. Studies of past climates usually point to the high end of this range, as do climate models. But it has been hard to reconcile this with the slow warming over the past decade. Most climate models projected more warming than observed, suggesting they are too sensitive to CO2. What’s more, the slowdown meant that simple estimates of climate sensitivity based on the recent temperature record were lower. But new analyses can explain these discrepancies. In February, Shindell and colleagues showed that models inflated the warming because they were fed the wrong data. The modellers had assumed that solar output, volcanic activity and aerosol pollution would stay at 1990s levels. But solar output fell, while volcanic activity and aerosol pollution rose. All that had a cooling effect. Shindell reran the models with the correct data, and most of the mismatch vanished (Nature Geoscience, doi.org/rsh). Now Shindell has found a  flaw in simple estimates of climate sensitivity that use  recent temperature data. These underestimate the global cooling effects of aerosol pollution, so  also underestimate sensitivity. That means the short-term sensitivity is between 1.3 °C and 3.2 °C – most likely 1.7 °C (Nature Climate Change, doi.org/rtm). It also explains why a muchtrumpeted study from 2013, from a team that included Shindell, used the temperature record and concluded that the short-term sensitivity to CO2 was just 1.3 °C (Nature Geoscience, doi.org/mj7). A high short-term sensitivity suggests that the medium-term sensitivity, which builds on the initial warming, is also high. That’s not the only evidence for high sensitivity. Steven Sherwood of the University of New South Wales in Sydney, Australia, and his colleagues looked at why different models give different estimates  of sensitivity. They found that the way clouds behave in the models explains half of the differences. Models that simulate clouds best also predict higher medium-term sensitivity (Nature, doi.org/rsj). “Sensitivity is not down at the low end,” Shindell says. “We can’t take any solace from warming being slower of late.” That means we must cut our emissions fast, he says, or the planet could warm 6 °C by 2100.