Iowa State professor explains climate models and their uses

Since the Industrial Revolution, humans have been emitting greenhouse gases into the atmosphere through sectors like transportation, electricity production, industry and agriculture. Those gases — including carbon dioxide, methane and nitrous oxide — trap heat and make the Earth grow warmer than it naturally would be.

As a result, the climate is changing, sending rippling impacts worldwide, as detailed in the newly released Fifth National Climate Assessment, which was mandated by Congress.

How do we know all this? For the most part, we have climate models to thank.

William Gutowski is a professor of meteorology at Iowa State University who has worked with climate models for decades. He has contributed to several Intergovernmental Panel on Climate Change reports, National Academy of Sciences reports about climate change and climate reports from former President George W. Bush’s administration. He’s also involved with the World Climate Research Programme, which coordinates regional climate modeling around the world.

Gutowski chatted with The Gazette about climate modeling and what people should know about it. His answers are edited for brevity and clarity.

  Buy Photo

Q: By definition, what is a climate model?

A: A climate model is a computer simulation that tries to simulate how weather evolves over many months, years, decades and centuries. It includes all sorts of environmental processes and interactions that go on in our planet. It's obviously very complex, but the attempt is to be as comprehensive as possible.

I want to emphasize the word “climate.” There's the day-to-day weather, which meteorologists try to predict using very similar models. But when we talk about simulating climate, we're not talking about the day-to-day weather. We’re making a climate projection. (The National Oceanic and Atmospheric Administration defines “climate” as is the long-term pattern of temperature and precipitation averages and extremes at a location.)

Q: How do climate models work?

A: We start these models with climate conditions at some point in history, often the middle of the 19th century when greenhouse gases started to grow in our atmosphere. Then, you let that historical data simulate weather behavior into the 21st century.

From there, you can treat the simulations as an experimental lab to map future climate trends. You can input known changes in, say, greenhouse gases and watch how an increase affects our climate. Or you can run simulations where you freeze the greenhouse gases at some level and say, “OK, what's the difference now?”

We can get the overall picture of how climate is established and evolving. We may not be able to tell you exactly what's going to happen in Cedar Rapids compared to Ames. But we can tell you the general pattern of what summers, winters, springs and falls are going to be like.

Q: What are the different types of climate models?

A: Global climate models try to simulate everything, including oceans, land cover, rainfall patterns, volcanic activity, ice sheets and biological interactions. They're the most comprehensive models — and the most challenging to evaluate. But people have been evaluating them for decades.

Then, there are regional climate models that cover just a portion of the planet. They have finer detail and might include behaviors that global models can't include. Some can go down to the size of about 5 miles.

There are simpler models that complement the most comprehensive models by focusing on certain processes within the complex models. For example, the simplest model tells you the average temperature of the planet. You look at what’s heating the planet, which is primarily coming from the sun, and what's cooling the planet, which is primarily what we give off back to space. Then, you can start to make things gradually more complicated by building in additional dimensions, like latitude and elevation.

Q: What are the limitations of climate modeling?

A: One limitation is the observations that help test our models. There are some things that we’ve observed very well for a long time, like temperature and precipitation. But there are other things that we don’t observe very well, like water evaporating back into the atmosphere. In a global sense, I think we're largely capturing it, but we're not going to get all the exact details.

We also like to repeat the same simulation multiple times. There’s random variability that goes on in our climate system, and we try to take that into account by running our models multiple times. But it can result in small differences in our simulations. Those help tell us what’s a true climate change signal versus natural variability.

The amount of computing power we have also limits the scales we can resolve in models.

Q: How reliable are climate models?

A: There are general principles that govern how the climate system evolves. The starting point is the fundamental laws of physics. We have methods for working with those laws to replicate what’s going on in the Earth's climate system. And these models have been tested over and over and over and over. I can't emphasize that enough. It's not like we just threw this together and said, “Hey, this is what's happening.”

We understand the basic processes that control our climate. We understand what the patterns of climate change are from these models, and we understand how they compare with what is observed. Any evaluation of how are our climate models are performing would basically show they're capturing the overall behavior and they're telling us what is going on. It is so clearly tied to the increasing greenhouse gases in the atmosphere.

No climate model is perfectly accurate. But through our many decades of assessing their behavior, we know that they do a very good job in telling us what factors are most important in how our climate evolves.

Q: What are key findings we’ve learned from climate models so far?

A: For where we live, winters are getting warmer, which is consistent with observed trends in the past 20 to 40 years. They will have less snow, maybe more rainfall.

Iowa summers are governed by moisture coming up from the Gulf of Mexico. In the past 20 to 30 years, that’s actually kept summer temperatures from warming up as much. We have cloud cover blocking sunlight, and we have water evaporating from the surface because there's more rainfall occurring.

Despite all that, as we go 30 to 40 years into the future, there's still an expectation that things are going to get warmer here in the summertime. Tied in with that, we’ll tend to have more dry periods — like right now — and then we'll get punctuated by periods of very heavy rainfall. When it rains, it pours. And when it's not pouring, it's going to tend to be dry. That's a simplification. It's not like it's going to be like that all the time. But there's going to be that tendency in that direction.

Brittney J. Miller is the Energy & Environment reporter for The Gazette and a corps member with Report for America, a national service program that places journalists in local newsrooms to report on under-covered issues.

Comments: (319) 398-8370; brittney.miller@thegazette.com