POINT OF VIEW: Fusion of climate models and geology adds insight

July 9, 2007
Geologic knowledge can provide vital information for scientists modeling the climate, said Eric J. Barron, dean of the Jackson School of Geosciences at the University of Texas, Austin.

Geologic knowledge can provide vital information for scientists modeling the climate, said Eric J. Barron, dean of the Jackson School of Geosciences at the University of Texas, Austin.

Climate models create mathematical representations of atmospheric patterns and ocean movements. The current focus is on models that relate temperature to concentrations of carbon dioxide in the atmosphere.

Scientists are developing increasingly sophisticated models that explain and project climate change.

For oil and gas exploration, climate modeling is a concept that has come and gone and now come again, Barron said. Scientists report great strides since 20-25 years ago in their ability to generate simulations that accurately examine specific time periods and basins.

“The geologic record is a window on how the earth can change,” Barron said of the emerging climate-change science. “I like to say that climate modeling brings the discipline of forecasting to the geologic record.”

Eric J. Barron
Click here to enlarge image

He serves on an Association of American Petroleum Geologists special committee on climate changes. At the AAPG April convention in Long Beach, Calif., he presented a paper entitled “Is It Time for a Rebirth in the Geologic Applications of Climate Models?”

The committee is evaluating AAPG’s policy statement on climate change, which Barron calls “a fast-moving field in some ways.” The committee addresses whether AAPG should renew its statement to reflect growing public concern about climate change. “We’re also looking for the opportunity to provide more education for AAPG members on climate change topics.”

Climate projections have informed decision-makers worldwide about water resource management, ecosystems maintenance, and alternative energy development. Despite an inherent uncertainty in climate models, the science is gaining corporate attention as executives across all types of industries try to address global warming issues.

“It used to be that many of the automakers and oil and gas companies had mixed views about the issue of global warming,” Barron said. “But I think the debate is being more broadly engaged than it ever has been. In my opinion, when this country starts talking about energy issues and global warming, it’s almost as if you are for or against-you can do it this way but not that way. I hope we are coming to realize that there are all sorts of degrees, issues, and responses. We can actually address them quite intelligently if we set out to do so.”

An advocate of integrating intelligent energy policy with climate policy, Barron notes that industrialized countries are better able to address environmental issues than many developing countries are. “It’s very clear that the countries that have the most problems associated with any kind of environmental change are those countries that do not have economies that enable them to address issues. So I think robust economies are much more capable of addressing these issues. It’s hard to separate energy from economic well-being.”

Climate models

Climate modeling stems from science that started in the 1920s to predict weather. But climate modeling differs greatly from weather forecasting. Weather forecasters are concerned about near-term, localized phenomena while climate modelers are concerned with long-term, large-scale features and variables that respond to external pressures, called forcings. The models involve land, oceans, ice, interactive atmospheric aerosols, atmospheric chemistry, and the carbon cycle.

Climate models are very complex. Hundreds of independent teams having diverse expertise are comparing models primarily based on simulations performed in 2004-05 for the United Nations Intergovernmental Panel on Climate Change.

Barron said climate models can be applied to geological information in two ways. One involves improving knowledge of climate change so scientists can better project the future, and the other involves new ways of studying Earth’s history.

In the typical model study of future climates, scientists study the period from 1895 to 2100. For the first 100 years, researchers have information to compare against their models, but they lack this information for the next century. Various models get very different results for the next century. The question confronting climate modelers is whether they can believe models based on simulating 100 years of history in which the changes were relatively modest.

“A great deal of change has a lot of significance for humans, while a little change has much less significance for humans,” Barron said. “One application here is the geologic record is a wonderful way to explore the versatility and capability of climate models. How well can you predict the past? The one thing geology has is abundant examples of climate change. So can we replicate those changes? If we can then we certainly are adding insight.”

The second application is exploration of past time periods. In the study of Earth’s history, researchers start with a set of observations. These can be derived from cores or outcrops containing pollen or isotopes that document past conditions.

“You constantly try to build a bigger picture from a limited number of observations,” Barron said. “You come up with this bigger picture, and you test it by trying to gather some more observations. A really good model gives you the potential to say, ‘OK, I know what the physical geography of the continent was, and I know a lot of the conditions. What environment do I predict for a time period in the past?’ Then all those observations become an opportunity to prove whether you did it right or wrong.”

Climate modelers try to make world reconstruction projections that are independent of key observations, Barron said. If the observations fit the predictions, then scientists have more confidence when they look at geographic areas lacking many observations.

Barron said: “It’s a way to fill in a lot of the holes if you can prove your predictions work well. If you attempt to reconstruct a past set of conditions, you tend to think rather simply about particular variables: Was it hot or cold? Was it dry or wet? Whereas what you are looking at could be one wet season and one dry season....

“The type of information that you derive could turn out to provide much more insight as well. Not only are you filling in the gaps but you are starting to tie what you see in the geologic record with a more defined set of physical variables. For instance, consider predicting severe weather in the past. Storms are capable of moving the sediments and altering the record.”

Tsunamis and hurricanes inundate coastal regions, depositing sandy sediment across broad areas. Scientists believe identifying a sandy bed in the geologic record as a tsunami or hurricane deposit is crucial to accurately identifying a recurrent interval, which in turn could produce a statistical measurement regarding the probability of future tsunamis or hurricanes.

The variables in a climate model also are crucial to the model’s accuracy. For instance, scientists are examining whether greenhouse gases (GHGs) influenced past climates. Ice cores document past GHG levels. When building climate models, scientists could not get the models to simulate past climate change unless they considered changes in CO2 levels. Although scientists are not sure why the CO2 levels changed, research indicates the changes contribute to observed warming. Climate scientists examine the cause-effect relationship between CO2 levels and warming.

Climate models project additional warming on a global scale. About 30 major centers for atmospheric or ocean research develop climate models.

Universities involved in climate modeling in Earth history are the University of California, Santa Cruz; the University of New Mexico; the University of Michigan; Purdue University; and Penn State.

Barron said the list of universities involved is growing, and he believes climate models will continue to improve.

“If you have confidence that these models can predict the past, it starts to give you more confidence that they can predict the future. There are definite flaws. There are definitely things in Earth history that we are having trouble predicting,” Barron said. “There are cases where you can safely say we don’t understand this problem well enough. There are other cases fitting exactly as predicted, and that’s where you start to believe you have something to say.”

Career highlights

Eric J. Barron joined the University of Texas at Austin as dean of the Jackson School of Geosciences in 2006. He succeeded William L. Fisher. Barron’s research interests are climatology, numerical modeling, and Earth history.

Previously, he was dean of the College of Earth and Mineral Sciences at Penn State. Before that he directed Pennsylvania State University’s College of Earth and Mineral Sciences Earth System Science Center and was a geosciences professor.


Barron has a BS degree in geology from Florida State University and a MS degree in oceanography and a Ph.D in oceanography from the University of Miami.


Barron belongs to the American Geophysical Union and the American Association for the Advancement of Science. He has chaired several national research boards, including the Board on Atmospheric Sciences and Climate of the National Academies (1997-2003). He is chairman of the Consortium for Ocean Leadership, a combination of the former Consortium for Oceanographic Research and Education with the Joint Oceanographic Research Institutions.