A world without clouds? Climate scientists barely say | Science

By Paul VoosenFebruary 26, 2019, 10:40

Could the layers of gray clouds that hang down on the ocean suddenly disappear in a warming world? Yes, if you believe a study published yesterday in Nature Geoscience– and the amplifying media coverage of it. If atmospheric carbon dioxide (CO₂) Triple Levels – an unlikely but not implausible scenario considering past rates of human emissions – these stratocumulus clouds could disappear in a frightening feedback loop. Fewer cooling clouds would mean a warmer Earth, which in turn would mean fewer clouds, leading to an 8 ° C rise in global warming, an astonishing change that would change the world.

But many climate scientists who do research on clouds are pushing back the study, arguing that his analysis of a small corner of atmosphere does not apply to the entire globe. It's a "simple model" [that] basically has a button with two settings, "says Joel Norris, a cloud scientist at the Scripps Institution of Oceanography in San Diego, California. "But it is very likely that Earth has more pimples than that."

As sophisticated as they are, climate models have trouble managing clouds. Condensed moisture and turbulent air form clouds on a smaller scale than the models can simulate directly. They therefore use approximations for this behavior. To better understand the clouds, scientists have instead developed high-resolution vorticity simulations that recreate the life of small patches of the atmosphere, including the key physics of cloud formation that climate models can not. not manage directly.

Several years ago, a project comparing six advanced vorticity simulations examined how a 2 ° C temperature rise affected low clouds in the ocean. Two dynamics have emerged that have thinned the clouds, exacerbating warming. First, higher temperatures have allowed more dry air to penetrate thin clouds from above, to prevent them from thickening and to reflect more solar energy. Second, CO increase₂ levels trapped the heat near the tops of the clouds, preventing their cooling. Because such cooling causes turbulence that forms clouds, this effect could hinder the formation of clouds, thereby fueling warming. If emissions continued, it seemed plausible that these low clouds were melting.

The frustration caused by global mismanagement of clouds is one of the main reasons why Tapio Schneider, climate scientist at California Institute of Technology (Caltech) in Pasadena and lead author of the new study, began construction of a new climate model last year. Nicknamed the climate machine, he would use artificial intelligence to learn from Foucault simulations and satellite observations to improve cloud rendering. This consisted first of all in building, with his team, their own swirling simulation, which could interact dynamically or couple with the ocean, thus allowing the simulated clouds to warm up, and vice versa.

The new study, which uses this swirl simulation, shows the same comments that others had previously identified. But Schneider ran for much higher CO₂ concentrations that most had done. While levels reached 1200 parts per million – three times what they are today and a figure that could be reached in the next century if no effort is made to curb climate change – the layers of low clouds disappear quickly.

The results of the model themselves appear to be strong, even particularly innovative. However, many cloud scientists oppose the next step that Schneider has undertaken: extrapolating the results of its swirl simulation, which is just one place that seems to be subject to cloud loss, for each area with similar stratocumulus clouds. As a result, all these clouds disappeared almost at once, allowing more of the solar energy to be absorbed suddenly by the dark ocean. Bjorn Stevens, a climatologist at the Max Planck Institute of Meteorology in Hamburg, Germany, has a hard time believing that clouds and the ocean would bond in such a simple way. "This coupling is done in a way that does not give confidence in the result."

There is no doubt that these returns will be at stake. Previous work has shown, says Chris Bretherton, a cloud scientist at the University of Washington in Seattle. "But they would all occur at different times in different concentrations of CO₂ in different places. That would solve the problem. There would be no sudden tipping point where all the clouds would have disappeared. This would be done gradually, subject to the complex response of the ocean and the atmosphere. "That's where I do not agree with that," says Bretherton. "I think the tipping point is not right."

Indeed, the new model is so simple, it lacks elements such as the sound of the weather, that it can simulate only fast transitions, adds Stephen Klein, atmospheric scientist at Lawrence Livermore National Laboratory in California. "Because of these simplifications, I do not find the nature of the" turning point "of their work credible."

Schneider stands by his interpretation. "I looked for all the possible reasons to fool myself, but I did not find anything," he says. The main implication, he adds, is that climate models need to be better equipped to manage clouds. "We should not be content with trusting models to predict the future of the 22nd century. The models may not quite pick up other things. "

Bretherton reports that more models are being solved in the clouds. "Over the next few years, we will have global models that will do what [Schneider’s] more defensible. Bretherton is developing such a model himself, which is also based on Foucault simulations to feed his simulations. To his surprise, he added, the first passes seemed to suppress warming feedbacks from these clouds more than expected.

The Caltech climate model, meanwhile, will take a few more years to materialize. But it is no coincidence that Schneider started developing this model once. Two years ago, he witnessed a turbulence simulation that eliminated clouds.

It will be an interesting test to see if this trend extends to the climate machine he is developing, adds Matthew Huber, paleoclimatologist at Purdue University in West Lafayette, Indiana. The global model could capture this kind of dynamic, or show that the climate system as a whole dampens such "tippiness" at smaller scales of its system. "This is indeed the only reason to develop this new model, he predicts, to predict climate surprises."