Climate science needs more mathematicians and physicists. So say prominent climatologists who are trying to spark enthusiasm for their field in budding researchers who might otherwise choose astrophysics or cosmology. Talented physical scientists are needed to help resolve mysteries that are crucial to modelling the climate — and, potentially, saving the planet — the group says, such as the ways in which clouds are formed.

There is a misconception that the major challenges in physical climate science are settled. “That’s absolutely not true,” says Sandrine Bony, a climate researcher at the Laboratory of Dynamic Meteorology in Paris. “In fact, essential physical aspects of climate change are poorly understood.”

To attract physics and mathematics students to the speciality, Bony and her collaborators have presented some of the field’s grand challenges in magazines such as Physics Today, and are organizing summer schools for students from an array of scientific backgrounds.

Last week in Nature Geoscience, Bony’s team outlined four of the field’s deepest questions, including how clouds and climate interact and how the position of tropical rain belts and mid-latitude storm tracks might change in a warming world. The questions are best tackled, says Bony, by creating more realistic climate simulations — an approach that she hopes will appeal to physicists.

The perception that climate science is ‘solved’ is an inadvertent result of pressure on climatologists to convey a simple message to the public — for instance, that all dry regions will get dryer and all wet regions wetter in a warming climate, says Piers Forster, a climate modeller at the University of Leeds, UK. That has made the science “sound somewhat dull”, he says.

“We too quickly turn to the policy implications of our work and forget the basic science,” adds Bjorn Stevens, a director at the Max Planck Institute for Meteorology in Hamburg, Germany, and a co-author of the Nature Geoscience paper. Although climate scientists agree on the basics — for example, climate change is primarily the result of human activity — large uncertainties persist in ‘climate sensitivity’, the increase in average global temperature caused by a given rise in the concentration of carbon dioxide.

As Bony and co-authors argue, understanding how the warming climate might affect cloud cover, which influences the amount of sunlight reflected back into space and thus Earth’s energy cycle, is key to addressing these uncertainties. A major weakness of current climate models is their limited ability to simulate the convection by which humid air is lifted into the atmosphere and which drives cloud formation and rainfall. In some instances, the models cannot even agree on whether the future will bring more rain or less.

Building better cloud-resolving models requires enormous computer power, as well as people who have a deep understanding of climate physics combined with skills in numerical modelling. But the number of scientists involved in developing computer algorithms for improved climate models is tiny, says Christian Jakob, an atmosphere researcher at Monash University in Clayton, Australia.

Physicists agree that climate science is not a big attractor of physics students. “Very few, and rarely the best, choose to do a master thesis in climatology,” says Thierry Fichefet, a physicist and climate modeller at the French-speaking Catholic University of Leuven in Belgium. “Talented physicists commonly go into more glamorous fields such as astronomy, cosmology or particle physics.” According to the American Institute of Physics in College Park, Maryland, 49 PhDs were awarded in atmospheric chemistry and climatology in the United States in 2013, compared with 303 for astronomy and almost 2,000 each for physics and mathematics.

Many physicists applaud Bony’s effort to raise interest in climate science, but whether physics students will heed her call remains to be seen. “We offer courses in climate science and our students do recognize the importance of the field,” says Paul Linden, a fluid-dynamics researcher at the University of Cambridge, UK. However, he says, classical subjects with a long history such as cosmology, are just more attractive, particularly at his university. “Most physics students would rather study with someone like Stephen Hawking, who is a member of our faculty.”