The newest generation of global climate models is running hotter than earlier versions, with many models predicting stronger future warming than their predecessors.

It‘s a confusing trend, and scientists have been working to figure out why it‘s happening. A review paper out this week suggests that clouds—and the tiny particles that help them form in the atmosphere—have something to do with it.

For the past several years, researchers have been working on an international project known as the Coupled Model Intercomparison Project, or CMIP. Every few years or so, modelers around the world coordinate to develop a new generation of models for use in climate research, always more advanced than the generation that came before.

The newest suite, CMIP6, is still in progress. More than three dozen models have been submitted by teams all over the world, and dozens more are still expected to come in.

But scientists have noticed something surprising about the new suite. Many of the models suggest that a given level of future carbon emissions may result in stronger warming than previous models suggested.

The issue lies within a concept known as equilibrium climate sensitivity, or ECS. It‘s a metric that scientists often use to estimate the strength of future climate warming.

ECS refers to the amount of warming that scientists should expect if atmospheric carbon dioxide concentrations suddenly doubled their preindustrial levels. (For reference, atmospheric CO2 hovered around 280 parts per million prior to the Industrial Revolution. Today, they‘ve exceeded 400 ppm, and they‘re still climbing.)

Previous generations of climate models have typically predicted a range for ECS spanning anywhere from 1.5 to 4.5 degrees Celsius. CMIP6, on the other hand, includes more than dozen models so far with an ECS higher than 4.5 C.

Several have exceeded 5 C, with the highest reaching an unprecedented 5.6 C.

As E&E News reported in May, scientists still are debating whether the ECS in the new models is reasonable, or whether it may be too high (Climatewire, May 12). It‘s a question still under consideration.

At the same time, researchers have been speculating about what may be causing the higher sensitivity. Some of the new models have tweaked the way they represent clouds in the atmosphere, experts told E&E News. And clouds can have an outsize effect on the Earth‘s warming response.

The new review paper, published yesterday in the journal Science Advances, supports the idea.

Led by Gerald Meehl of the National Center for Atmospheric Research, the authors examined the new suite of climate models and compared them with earlier generations. They also reviewed a number of recent papers, published by the modeling teams themselves, that assessed some of the changes they‘d made to the newer models.

They were looking for clues about what‘s causing the increased sensitivity.

“Our goal was to look for any themes that were emerging, especially with the high-sensitivity models,” said Meehl in a statement. “The thing that came up again and again is that cloud feedbacks in general, and the interaction between clouds and tiny particles called aerosols in particular, seem to be contributing to higher sensitivity.”

The complexities of clouds

Clouds are notoriously difficult to simulate in climate models. In the first place, cloud formation is a highly complex phenomenon with a lot of small, moving parts.

Tiny particles in the air, called aerosols, have a huge influence on how quickly clouds form, how big they get, what type of clouds they turn out to be and how long they last in the atmosphere. That‘s on top of all the other weather-related factors that affect cloud formation, including air temperature, humidity and wind conditions.

Simulating these complicated physics takes a lot of computing power and requires models to operate at a very fine scale. That‘s hard enough. But it‘s extraordinarily difficult to do in global climate models, which are designed to simulate grand-scale climate processes across the entire world.

To compromise, climate models often contain simplified, built-in information about clouds and the way they form—a kind of shortcut that allows clouds to appear without requiring the models to actually recreate all the small-scale physical processes that influence their formation.

As scientists learn more about clouds and cloud physics, they‘ve been able to gradually improve the way clouds are represented in their models. That‘s important because clouds can have a huge effect on the climate system.

Clouds can either cool or warm their local climate, for instance, depending on how much water or ice they contain—characteristics often influenced by the types of aerosols that help the clouds form.

Sometimes clouds cool the Earth‘s surface by reflecting sunlight away, and sometimes they actually trap heat beneath them.

Many of the new climate models have made significant advancements in the way they represent clouds and aerosols. They might better depict the amount of liquid water versus the amount of ice that certain clouds contain. Or they might more accurately represent the way certain kinds of aerosols influence cloud formation.

But there seems to be some side effects, the new review paper notes. These improvements may be causing clouds to have a much bigger impact on the local climate in some models.

It‘s possible that in some cases these feedbacks are a little too strong. The clouds themselves might be more accurate, but the way they interact with the bigger climate system might still need to be tweaked. Scientists are debating whether that‘s the case and how realistic the increased sensitivity really is.

It‘s also likely that clouds and aerosols aren‘t the only factors contributing to the hotter models.

But the evidence so far suggests they‘re playing a key role. And that‘s information that can help improve the next generation of models, beyond CMIP6.

“Cloud-aerosol interactions are on the bleeding edge of our comprehension of how the climate system works, and it‘s a challenge to model what we don‘t understand,” Meehl said in a statement. “These modelers are pushing the boundaries of human understanding, and I am hopeful that this uncertainty will motivate new science.”

Reprinted from Climatewire with permission from E&E News. E&E provides daily coverage of essential energy and environmental news at