The role of clouds in our changing climate
Bruce Wielicki uses satellites to understand “tricky” nature of clouds.
Bruce Wielicki of NASA’s Langley Research Center in Hampton, Virgina is the Principal Investigator for CERES, a project now using four instruments on two different Earth–orbiting satellites to monitor how clouds affect our climate. Earth & Sky’s Abby Frank spoke to Dr. Wielicki in 2005.
Frank: What is the CERES project?
Wielicki: It’s a project to measure all the energy coming off the planet. Instead of perspiring like your body does, Earth emits at thermal infrared wavelengths, which are much longer than your eye sees. It’s the sort of radiation you feel coming off a hot fire. That radiation has to escape out to space in the same amount that we receive it from the sun. But there are huge variations over the planet as to where it comes out. These variations depend on a range of things, for example, the temperature, how much cloudiness there is, and so on. So we keep track of that outgoing infrared radiation all over the globe.
We currently have four satellite instruments on two different spacecraft. We pull all of that data together, sampling every part of the Earth about four times a day. And that let’s us keep track of what the Earth’s doing seasonally, and year to year, and to try to watch how climate change is happening.
Frank: How do clouds help regulate the temperature on Earth?
Wielicki: Clouds have a lot of effects. That’s what makes them so tricky. For one thing, clouds refelect sunlight back to space, which cools the Earth. You know that yourself from the times you’re out on a hot beach in the summer, and a cloud comes between you and the sun. It suddenly gets a little cooler. That’s less of the sun’s energy coming down to the surface of the Earth.
But clouds don’t just reflect energy. If Earth were a billiard ball in space, it would be able to radiate right from the hot temperatures of the surface directly to space. But the Earth has an atmosphere, and that atmosphere doesn’t let all of the radiation escape. A lot of it gets reabsorbed and then reradiated at the much colder temperatures up in the atmosphere. If you’re up in a jet, for example, and you ever see the monitor they put up there on the TV that tells you what the temperature outside is, it’s extremely cold. The atmosphere drops in temperature as it goes up in altitude. What’s happening is that radiation is being absorbed in the atmosphere.
Clouds do the same thing. Clouds block that radiation even more effectively than greenhouse gases like carbon dioxide and water vapor. So clouds act both to cool by reflecting incoming solar radiation, and they act to warm by blocking radiation from escaping. And it’s the balance between those two big effects that creates the net effect on the Earth.
Right now, clouds’ net effect is cooling. The Earth would be a lot hotter place if we didn’t have clouds.
Frank: What effect has CERES data had on climate models? What role do clouds play in climate change?
Wielicki: What we don’t know is what role clouds will play in future climate change. We know on average their role is cooling. We’re pretty sure they’ll still be cooling 100 years from now. But if they’re less cooling or more cooling, it can be either a positive or negative feedback on the climate change as we do things like add greenhouse gases or add aerosols.
What CERES is trying to do is to make the climate models much more accurate. These big climate models we put on super computers to try to predict what the clouds are going to do in a future, warmer climate are struggling. For example, if you look out your window, you see these puffy, cumulus, up–and–down drafts in the atmosphere of maybe a kilometer scale. But a typical, fast supercomputer running a current generation climate model is probably a hundred times coarser than that in resoluton. There’s maybe 100 kilometers, or 200 kilometers, between the data points it keeps track of.
What CERES tries to do is tell these models what the clouds are really doing. Then the models are trying to get more and more accurate at simulating what those clouds are doing. So, it’s kind of an iterative process that we and the climate modelers are using to improve the accuracy with which we compare possible future climates.
Frank: What are some of the findings from CERES?
Wielicki: Of the findings so far, I would say the biggest thing is that we have the first data where we can actually distinguish between the amounts of energy coming in and out of the planet by individual cloud type. So, for example, we can now accurately distinguish between cumulus clouds versus big stratus decks versus deep convective, cumulonimbus clouds. Now we’re trying to give the climate models much tighter hoops to jump through, to prove that they really know how to handle clouds. So, that’s been the first big breakthrough.
The other things we’ve been seeng are some of the changes that are going on in the Earth’s energy budget year to year. For example, the total amount of heat the oceans are storing as they warm, because they are warming, varies from year to year. And it looks like it’s actually tied in with how the clouds are changing. So, we have an ocean–cloud link now that we’re at the very beginning stages of trying to puzzle out. So, that’s one of the other exciting things I think that we’ve seen so far in the data.
Frank: In climate models, how is current cloud data used to predict future climate change?
Wielicki: That’s a good question. You can’t take the data directly to try to predict the change. What you do instead is you try to take the models to predict current day data, for example, the last 4 or 5 years when we had CERES data. For example, we ask the models to do things like, when we go from El Ni?os to La Ni?as, do you get the right cloud changes? We also use the models to show what the seasonal cycle is. When we go from winter to summer to spring, do we get the right cloud changes? You know clouds change dramatically from the storm systems in the winter to the convection in the summer.
So, right now what we’re doing is constantly stressing the models’ accuracy to prove whether they’re reaching accurate enough levels to predict climate out into the future. It’s kind of a big grey area as to how exactly accurately they have to get the clouds to be good enough to predict climate 100 years out in the future. We’re still struggling with that one.
Frank: How have school children been involved in validating CERES data?
Wielicki: That’s been a lot of fun. Lin Chambers in our group started that, and I have to admit I was even skeptical of it in the beginning. What she really set up was a system where the kids can get online, find out when the satellite’s coming over their school, and then go out and do cloud observations. They look at sky cover, look at the type of clouds they see, and in that process learn something about the physics of clouds and what things they’re looking at. And then they put their data on the web. We bring it down here to Langley; and this is done from all over the world. We have schools in about, I think it’s about 60 countries and maybe 1,500 schools. And then we actually take our analysis of the cloud heights, fractions, ice versus water cloud type in the same region, at the same time, and actually do comparisons with the kids.
One of the things they’re extremely good at is seeing things like a totally clear sky or really thin cirrus. The satellite may have a very difficult time seeing versus a complex background, for example, surface trees and vegetation and changing agricultural patterns. So, the kids actually are helping us validate some of our data.
And of course they’re taking their data just as we come over. A normal meteorological site may take it within two or three hours, and of course if you’re used to looking at the sky, clouds change pretty rapidly. So, there’s been a real advantage in having them do it as we actually come over.
Frank: How large is the CERES project? How many scientists are involved?
Wielicki: The science team itself, we have roughly about 40 scientists involved. There are a lot of support people, who are sort of masters level computer programmers. The job we have to put this data together is we basically pull together instrument data from about up to 11 instruments and 7 spacecraft, not only our own data, we need data from other things as well, as well as global weather data. So we actually have about a half a million lines of computer code to pull together for the production and verify it all at climate accuracy. So, there’s also a lot of validation building and development work going on in the team as well.
Frank: What is your goal for the project? What result would you ultimately like to see from your data collection?
Wielicki: What I’d ultimately love to see is we reach the point where the climate models are predicting clouds that agree with the data collected via satellite. We’d like to be able to prove as a communtiy that we’ve reached the accuracy that we need to predict climate out 50 or 100 years from now. That’ll really be the ultimate goal of all of this: to prove to ourselves scientifically that we really know how to predict clouds.
Frank: How accurate are the climate models we have today? How much variation is there between models?
Wielicki: If you ask these models to predict how much warming in the year 2100, there is still about a factor of 3 to 4 uncertainty, and even that is just the differences in models. We can’t yet prove the real world lies within the range of those models. But what we can show unequivocally is that indeed clouds are one of their big uncertainties.
Another uncertainty is what the deep ocean is doing in terms of how much carbon is being sucked up there – or in the biosphere.
So, I think in terms of where we go in the future, clouds will be one of the big things we try to figure out, so that we can narrow uncertainties in our future predictions.
Frank: Is there anything you would like to add about CERES?
Wielicki: I think the thing to keep in mind is that, although the climate models involve uncertainties, that doesn’t change the fact that things are going to be warming. The physics of that future warming is pretty solid. It’s more a matter of the amplitude and timing that we’re arguing about now, in order to figure out what’s going to happen and how we have to react to it.
Frank: Is climate change going to be as bad as everyone is saying? Can clouds save us?
Wielicki: For example in 2100 the range you can get for what the warming is going to be globally can be anywhere from 2 or 3 degrees centigrade up to 3 or 4 times that. It’s still a very big range and the impact level is pretty sizable with both scenarios.
I think the wildcard in all of this is that the climate system is unlikely to change as a nice, simple, linear, flat line. That’s just not a likely scenario. A much more likely scenario is, given the complexity of the ice sheets, ocean, atmosphere interaction, that it will go in lurches and bounds. We may get some abrupt changes that will suddenly pop it into different modes. And then it will sit there for awhile. And then it will pop into a different mode again. I think those will be the bigger challenges.
