SCIENCE

A new project funded by the National Science Foundation links researchers at leading U.S. universities to study the mathematical problems related to climate change research.

The University of North Carolina at Chapel Hill leads the Mathematics and Climate Change Network, with the Renaissance Computing Institute (RENCI) providing logistical support and cyber tools to support the creation of a virtual organization spanning the U.S. The NSF will provide $1 million a year for five years to support the network.

"The math community is not being properly involved in climate change research," said Chris Jones, mathematics professor at UNC Chapel Hill and principal investigator for the project. "But the fact is, we have only one Earth, so experiments must be done using computer models."

The Mathematics and Climate Change Network includes faculty members, post-docs and students at 13 institutions: University of North Carolina at Chapel Hill, University of North Carolina at Asheville and RENCI at UNC Asheville, Arizona State University, Bowdoin College, Cal Poly San Luis Obispo, New York University, Northwestern University, University of California at Berkeley, University of Chicago, University of Minnesota, University of Utah, University of Vermont and University of Washington.

Representatives of the member institutions are in Chapel Hill Sept. 23 and 24 for a kickoff meeting at RENCI headquarters.

The network's mathematicians will work closely with climate scientists at research centers such as the National Center for Atmospheric Research, the National Climatic Data Center, Los Alamos National Laboratory and Oak Ridge National Laboratory.

Jones called the network "a virtual climate mathematics center with national reach" that will conduct research under three themes: data reconstruction and analysis, climate process models and the dynamics of climate.

Data problems to be tackled include optimizing existing climate models so they more accurately describe climate processes. That work involves assimilating data from many disparate sources, including historical data, and developing mathematical methods for interpreting historical data so it can be used in mathematical models. Incorporating accurate empirical data into climate models will help to develop more accurate models of future climatic conditions, said Jones.

Climate process modeling involves using mathematical formulas to understand microstructures in natural systems. Sea ice, for example, is an intricate, honeycombed structure and understanding its physical properties can help answer crucial questions about its stability and how rapidly it will melt.

Mathematicians studying dynamics will look at historical changes in climate including sudden, dramatic changes, such as the "little ice age" of the 16th – 19th centuries.  Sudden, abrupt changes are common in the world of mathematics, said Jones, and studying disruptive climate events will help scientists understand the tipping points that trigger these changes.

Most of the NSF funding will support researchers at participating universities and travel costs for visits with climate scientists and for biannual, face-to-face meetings. During most of the year, the network will operate like a university center or department, holding regular working group meetings over the Internet instead of in a conference room. Over time, Jones hopes the project will make mathematics as integral to climate research as it is to the physics and biology research communities.

"Our charge as mathematicians is not so much to go out and solve the climate change problem, but to develop the mathematical ideas and tools that will be crucial to climate scientists in their work to understand and predict climate changes," he said.