How one climate model breaks the planet into a 10,242-cell
spherical geodesic grid. Source: Prabhat, LBNL.

In my QUEST radio story this week, we learn about how faster supercomputers will help scientists run climate simulations. One of the trickiest aspects of that is dealing with clouds. To find out why, I sat down with Bill Collins, head of Climate Science Department at Lawrence Berkeley National Lab.

How important are supercomputers to climate change science?

We understand the climate by making observations using satellites and ice sheets. But the only crystal ball we know about, short of a time machine, is the supercomputer.

We started with by running simple climate models on supercomputers that included simulating the weather, rainfall, and carbon dioxide. In the last 20 years, the complexity of models has vastly increased. They now include ocean dynamics, glaciers, sea ice and the exchange of carbon dioxide between the ocean and the land, known as the carbon cycle. All of that has required an immense increase in computing power.

Climate models today simulate the atmosphere and carbon cycle by breaking up the planet into a grid and running the calculations in those segments, right?

Right, in modern climate models, we simulate the weather every two to five minutes and then average that to see how the climate is going to change across that grid. We simulate the weather in segments that are 25 kilometers wide.

Our goal is model something the size of San Francisco County, which is about 10 kilometers wide. Once we get to that scale, we're going to be able to provide local projections of climate change. We're honing in, but we're not there yet. We need bigger computers to get there.

The other reason is we'd like a higher resolution is that we're having to make educated guesses about certain things, like clouds. And those educated guesses are a source of uncertainty. Cloud systems can be very large or very small. We don't know how they work at the large scale, but we do know how they work at the small scale. So the trick is to simulate them at the small scale.

What role do clouds play in the climate?

Clouds stabilize the climate. They reflect sunlight, so they act like a sun shield. But they also trap heat from the Earth. They both heat and cool, but their net effect is to cool the planet. So the question is, what happens if climate change makes the cloud cover decrease or increase? Understanding how clouds will be affected by climate change has become a critical question.

Where clouds form in the atmosphere makes all the difference. High clouds reflect sunlight, but they're mostly very efficient blankets. Clouds low in the atmosphere aren't very good blankets. They act as a big sunscreen, reflecting energy.

How do climate models today treat clouds?

Models today represent clouds throughout statistical methods over large areas. That models their effect, but not really how they work. And you don't want to assume how they work now is how they'll work in the future. We want to get to a level of physical modeling of clouds.

To do that, we need to be able to resolve them at a small scale. The current Intergovernmental Panel on Climate Change projections use a 50 kilometer grid, but that's still not good enough. The scale we need to get to is about 10km or so. So once supercomputers can get us there, we'll be on a much more solid footing to predict how clouds might be affected by climate change.

If we tried to run climate models at that resolution now, it would simply take too long. The rule of thumb is that we'd like to simulate the climate a thousand times faster than it happens. So simulating three years in a day is our rule of thumb. If we increase our resolution from 50 kilometers down to 10 kilometers, that increases the computation demand by a factor of 125. At that point, you're doing 9 days in a day. We can't afford to do that and make the kind of projections that policymakers need in the next century.

Climate model resolution of California. Source: LBNL.

What will we learn about California with better climate models?

Temperature changes are happening faster in the mountains than in the valley. So climate change in California is locally specific. A big questions is how much snowfall we'll get in the future. That's going to hinge on what the temperature is at the peaks of the Sierras. So knowing how fast the temperature change is going to happen at the peaks is going to make a big difference to our water supply.

Local climate predications are really important for state and local policymakers. How should building codes be changed? How will local areas adapt? We need accuracy at the state and local level to pull off that planning.

If you can resolve clouds better in the future, will that change overall projections about climate change?

I'd be shocked if they did. The physics of climate change is really basic. We're not going to get out of global warming. We know based on the projections that we've had in hand for the last 20 years that the time to act is now. The longer we wait, the harder the solutions are to avoid dangerous levels of climate change.

What better resolution of clouds is likely to give us is a better idea of changes in rainfall. That's really important to our water supply, our forests, and our crops. Higher resolution will also give us better predictions of climate change extremes, like when droughts happen or the impact of downpours on rivers and dams.

We want to know about climate change that goes bump in the night. We're concerned about abrupt climate change – the type that occurs quickly over a large region, like the melting of the permafrost. We're also worried about extreme climate change – intense, highly-localized changes like heat waves, hurricanes and tornadoes. Both of those are stressors on society and the environment. They've been difficult to simulate since we haven't had the computing power. But now, thanks to advances, we're getting there.

37.8077719 -122.2689661

John Shalf of Lawrence Berkeley National Lab stands inside the Hopper supercomputer.

Whether its laptops or cell phones, computers are getting smaller for most of us. But for many scientists, they’re getting larger. Supercomputers have become a critical tool for analyzing complex problems like climate change.

But as supercomputers grow, so does their energy appetite. Researchers are trying to solve that problem by using a smaller, more pervasive technology.

Supercomputers have improved at a break-neck speed, especially if you look back to the Cray-1. In 1976, this six-foot tall tower of wires was the most powerful supercomputer the world had ever seen. It was installed at Lawrence Livermore National Lab for fusion research.

“If you needed an icon for a supercomputer, you would use the Cray-1,” says Dag Spicer, senior curator at the Computer History Museum, where the computer is spending its retirement. “It blew people’s minds. It was so powerful, so fast.”

Of course, in today’s terms, “It’s roughly equivalent to a first generation iPhone from Apple,” says Spicer.

Listen to the QUEST radio story The Future of Supercomputers

The reason we don’t play Angry Birds on a supercomputer today is thanks to something called Moore’s Law.

“Moore’s law is a predication made by Intel cofounder Gordon Moore in 1965 that the number of transistors – that is the little switches that make up a computer – the number of transistors incorporated in a chip will double approximately every 12 months,” says Spicer. Moore later amended that timeline to every 18 months.

What that means is computer chips have gotten smaller and faster at an incredible rate over the last 40 years. Which leads us to a supercomputer known as Hopper.

Today's Supercomputers

“This is our new Cray XE6 supercomputing system,” says John Shalf, a computer scientist at Lawrence Berkeley National Lab. We’re standing next to row after row of tall black computer towers inside a building in downtown Oakland. The sound of the computer’s massive cooling system is deafening.

“You have to keep it cold or it’ll melt. We’ll have a puddle of chips on the bottom of the floor,” says Shalf.

Hopper is the eighth largest supercomputer in the world. And right now, it’s chewing on some complicated problems. “Number one here is particle accelerator design. We have fusion energy and then we also have laser plasma inertial fusion simulation,” says Shalf.

“Science has just really been revolutionized by the speed of computers,” says Kathy Yelick, associate director for computing sciences at Berkeley Lab. She says scientists use Hopper to simulate everything from black holes to climate models. There’s a special term to measure this supercomputer’s power: a petaflop.

“So how fast is that?” says Yelick. “Most people can do probably about one arithmetic operation per second if they’re pretty good.”

Now imagine asking a billion people on the planet to do one math problem per second. To get to Hopper’s speed, “we would need a million earths,” she says.

A million earths, each with a billion mathematicians – that’s how fast Hopper is. But it won’t be long before a faster model comes along. “Every four years we get a system that’s about 10 times larger than one we put in three or four years earlier” says Yelick.

According to Moore’s Law, those next generation supercomputers should be faster and more compact. But John Shalf says computer chips have hit a wall.

The End of Moore's Law?

“The problem is now we can’t make them go any faster. So we can cram more things on the chip, but if you make them go fast, it’s so hot they’ll melt.”

If chips themselves aren’t faster, supercomputers will simply have to add more and more of them to increase computing power. And that comes with a very big impact on the energy use.

Hopper uses around 3 megawatts of electricity – about as much as 2000 homes. But future supercomputers? “Projections say that at the end of the decade, we’d be at 100 megawatts if we continue,” says Shalf.

That’s enough power for a small city, about the size of Novato. The electricity bill alone would be roughly 100 million dollars a year.

“What that says is our current approach to doing supercomputing is dead end. And that we need to think of dramatically new ways to improve the efficiency of computing,” Shalf says.

That could be done with some very familiar technology. Cell phones have computer chips inside them, but not the same chips as desktop computers.

From Peter M. Kogge, "ExaScale Computing Study: Technology Challenges in Achieving Exascale Systems," Sept. 28, 2008

“For as long as they’ve existed, they’ve wanted a cell phone that would last longer, be less expensive,” says Shalf.

To do that, chips in cell phones have had to be smaller and more energy efficient. So Shalf says, why not build a supercomputer with chips that combine millions of these simple cell phone processors, specially designed for scientific jobs? In other words, use cell phone technology to make the world’s most powerful computers.

“We’re able to demonstrate an additional 80 times more energy efficiency than business as usual, and that gets us within striking distance of where we need to be to build a practical supercomputer,” he says.

Instead of a 100-megawatt supercomputer, it would be a three to ten megawatt computer. Whether or not it gets built depends on chipmakers like AMD and Intel, who would design the chips. But Shalf says a supercomputer with that power could make a big difference in climate change science.

“It enables policymakers to have the tools they need to make important decisions that have trillion dollar consequences. And that’s why you want to build a supercomputer that’s able to do this.”

Berkeley Lab hopes to use the supercomputer to better predict some of the trickier impacts of climate change – like changes in rainfall patterns, ice sheet melt and the effects of clouds.

37.8077719 -122.2689661

Electron microscope image of a hole embedded within a sheet of graphene. The corners of the green hexagons are carbon atoms which form graphene’s crystal structure. Image courtesy of the Zettl Research Group, Lawrence Berkeley National Laboratory and University of California at Berkeley.

Acquiring a sample of graphene is almost comically easy. Start with an ordinary piece of graphite, which is basically the same material that is used in pencil lead. Squeeze it between two pieces of Scotch tape and tear them apart. Repeat several times until pieces of the graphite have been cleaved into sheets no more than a single atom thick. Voila – graphene! Total cost of 1 pencil plus a roll of Scotch tape: about $3.

Simple as this process is, scientists did not even know that single sheets of graphene could exist until 2004. Now that we know that we can make graphene, it turns out that it has some amazing electrical properties and someday might even replace silicon as the most important component in computer circuitry. To that end, researchers in Alex Zettl’s group at Berkeley have endeavored recently to isolate suspended membranes of graphene for study and image them at Lawrence Berkeley Lab’s TEAM 0.5, the world’s most powerful transmission electron microscope (TEM). Results were published last spring by Çaglar Ö. Girit and others in the Science.

Two aspects of the Zettl group’s recent work have been particularly interesting. First, the TEAM 0.5 microscope not only has the ability to see individual atoms of graphene, but can also take pictures in close to real time. This means that Girit was able to see dynamics of graphene as they actually happened. Other types of microscopy (scanning tunneling microscopes, for example) can take several minutes to get a single picture.

Second, Girit and others centered their images at a hole within the graphene sheet. This allowed them to observe the dynamics that occur at the material’s edge. Such edges can have a notable effect on a graphene sheet’s electrical properties and thus understanding them and controlling them would be crucial in the design of any future technology.

Aside from technological applications, graphene is a theoretical physicist’s dream system because it beautifully combines the dynamics of relativistic particles from space such as neutrinos with the experimental accessibility of an easy system to make and manipulate here on Earth. Girit thinks that this is perhaps the single most exciting aspect of the system.

Only time will tell if graphene will have a long-term impact on society, but this would not be the first time a new discovery has transformed the Bay Area. In 1955 William Shockley moved to Mountain View, CA to found a new startup developing the silicon transistor. His company’s success was ultimately marred by Shockley’s own belligerent personality (“He understood everything except people,” Charles Townes once remarked), but the invention and the industry that grew up around it have revolutionized the region. The Santa Clara Valley’s old nickname, “the Valley of Heart’s Delight,” has long since been whisked away into a memory of a distant time and setting. Today most of us know it only as Silicon Valley. Our children may know the region as something entirely different.

37.8768 -122.251