Skip to Content

A powerful new model could make global warming estimates less vague

One of the biggest sources of climate uncertainty is how clouds will behave. Caltech physicist Tapio Schneider is trying to give us some answers.

Photo of Tapio Schneider
Photo of Tapio SchneiderRyan Young

The severity and speed of climate change will depend on the quantity of greenhouse gases we emit into the sky, but also on how sensitive the climate is to those gases.

One uncertainty is how clouds will respond as the atmosphere heats up. Tapio Schneider, a climate scientist at Caltech, and his colleagues from Caltech, the Naval Postgraduate School, JPL, and MIT are building a climate model that will use machine learning, powerful computing, and petabytes of data to help resolve some of the unknowns around how, why, and where clouds form, produce precipitation, or dissipate. The goal: to cut the uncertainty in current predictions of carbon dioxide’s impact on the planet in half.

Science journalist Mallory Pickett sat down with Schneider to find out how his research will do this, and why it matters.

How much uncertainty is there in current climate models?

There is a measurement called “climate sensitivity.” It’s the global mean surface temperature increase that you get after doubling CO2 concentrations and letting the system equilibrate. With current climate models, the climate sensitivity for doubling CO2 ranges somewhere between two degrees [Celsius] warming up to five degrees warming.

What are the implications?

Take the two-degree target of the Paris agreement. We’ve had about one degree of warming already, so it’s one more degree to go. How much more CO2 can we put into the atmosphere before we have warmed Earth another degree?

For a model that has a climate sensitivity of around two degrees, you can get to CO2 concentrations of close to 600 parts per million. We’re at 410 parts per million, so even if we continue emitting a lot, we won’t reach 600 before 2060 or so. In a model that has a climate sensitivity closer to five degrees, [one more degree requires] about 480 ppm, so that’s only about 70 to go. That’s something we’ll reach in the next two decades or so.

Why the uncertainty?

The single biggest contributor is uncertainties about clouds, and specifically about low clouds in the tropics. Low clouds over tropical oceans reflect sunlight because they are white, and this cools the Earth. We don’t know if we’ll get more or fewer of them as it warms, and that’s the key uncertainty in climate predictions.

One other important piece is how much carbon is being taken out by the biosphere. Right now only about half the carbon that humans emit ends up in the atmosphere. The rest is taken up by oceans and the land biosphere, and we don’t quite know where it goes.

Photo of Tapio Schneider
Ryan Young

If there’s so much uncertainty, do we really even know that things will get bad with a lot of CO2?

When you put more CO2 or other greenhouse gases in the atmosphere, they absorb thermal radiation. What happens if you put more of these greenhouse gases in the atmosphere is that everything else being equal, you ought to warm the surface. The physics of that’s completely clear, undisputed by any serious scientist.

Where the uncertainties come in is to say, well, how much warmer will it get? What happens to these little clouds? They reflect a lot of sunlight. If we get more of them and more sunlight will be reflected, it will be less warming. If you get fewer of them, you have an amplifying feedback effect where you get more warming.

Are things worse than predicted?

I think the evidence in recent years—for example, from studies looking at cloud variations over the past decades—points more toward higher climate sensitivity.

The goal is to cut some of these uncertainties in half. How will you do that?

We want to use the data we have available: terabytes per day coming down from satellites. We want to assimilate those data into the model. That’s a computationally challenging task, but it’s just doable now and will give you a model that simulates a present climate better.

If climate sensitivity is on the high side, how worried should we be?

I would be very worried. It would mean higher heat extremes, especially in summer. It means more extreme precipitation in places like the [US] Northeast. It is possible the climate sensitivity is on the high end of what models predict, and if that’s right then we’ll experience severe impact in our lifetime, certainly in our children’s lifetimes. Simply put, the higher the climate sensitivity, the more worried we should be.

Keep Reading

Most Popular

Large language models can do jaw-dropping things. But nobody knows exactly why.

And that's a problem. Figuring it out is one of the biggest scientific puzzles of our time and a crucial step towards controlling more powerful future models.

OpenAI teases an amazing new generative video model called Sora

The firm is sharing Sora with a small group of safety testers but the rest of us will have to wait to learn more.

Google’s Gemini is now in everything. Here’s how you can try it out.

Gmail, Docs, and more will now come with Gemini baked in. But Europeans will have to wait before they can download the app.

This baby with a head camera helped teach an AI how kids learn language

A neural network trained on the experiences of a single young child managed to learn one of the core components of language: how to match words to the objects they represent.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at customer-service@technologyreview.com with a list of newsletters you’d like to receive.