Climate models have underestimated Earth’s sensitivity to CO2 changes, study finds
By Jim Shelton
April 7, 2016
A Yale University study says global climate models have significantly underestimated how much the Earth’s surface temperature will rise if greenhouse gas emissions continue to increase as expected.Yale scientists looked at a number of global climate projections and found that they misjudged the ratio of ice crystals and super-cooled water droplets in “mixed-phase” clouds — resulting in a significant under-reporting of climate sensitivity. The findings appear April 7 in the journal Science.Equilibrium climate sensitivity is a measure used to estimate how Earth’s surface temperature ultimately responds to changes in atmospheric carbon dioxide (CO2). Specifically, it reflects how much the Earth’s average surface temperature would rise if CO2 doubled its preindustrial level. In 2013, the Intergovernmental Panel on Climate Change (IPCC) estimated climate sensitivity to be within a range of 2 to 4.7 degrees Celsius.The Yale team’s estimate is much higher: between 5 and 5.3 degrees Celsius. Such an increase could have dramatic implications for climate change worldwide, note the scientists.“It goes to everything from sea level rise to more frequent and extreme droughts and floods,” said Ivy Tan, a Yale graduate student and lead author of the study.Trude Storelvmo, a Yale assistant professor of geology and geophysics, led the research and is a co-author of the study. The other co-author is Mark Zelinka of Lawrence Livermore National Laboratory’s Program for Climate Model Diagnosis and Intercomparison.A key part of the research has to do with the makeup of mixed-phase clouds, which consist of water vapor, liquid droplets, and ice particles, in the upper atmosphere. A larger amount of ice in those clouds leads to a lower climate sensitivity — something known as a negative climate feedback mechanism. The more ice you have in the upper atmosphere, the less warming there will be on the Earth’s surface.“We saw that all of the models started with far too much ice,” said Storelvmo, an assistant professor of geology and geophysics. “When we ran our own simulations, which were designed to better match what we found in satellite observations, we came up with more warming.”Storelvmo’s lab at Yale has spent several years studying climate feedback mechanisms associated with clouds. Little has been known about such mechanisms until fairly recently, she explained, which is why earlier models were not more precise.“The overestimate of ice in mixed-phase clouds relative to the observations is something that many climate modelers are starting to realize,” Tan said.The researchers also stressed that correcting the ice-water ratio in global models is critical, leading up to the IPCC’s next assessment report, expected in 2020.Support for the research came from the NASA Earth and Space Science Fellowship Program, the National Science Foundation, and the U.S. Department of Energy.