The investigators form a "Mathematics and Climate Research Network." This is a framework for an intensive effort aimed at bringing to bear the full power of modern applied mathematics and statistics on the prediction and understanding of the Earth's climate. The investigators focus on three key themes: (1) Dynamics of Climate, (2) Climate Process Modeling, and (3) Data Analysis and Data Assimilation. Dynamics of Climate addresses critical climate processes and their interactions. Climate Process Modeling undertakes the modeling of climate components that have been underrepresented in extant climate models, such as the multi-scale material structure of sea ice. Data Analysis and Data Assimilation develops mathematical tools for analyzing climate data and assimilating them in current climate models.

The Research Network aims to be a national resource, with participants at thirteen U.S. universities. The investigators work together as a virtual community that holds regular weekly "webinars" and working meetings over the Internet. This multi-year effort is expected to help in defining a research area of "climate mathematics" and in educating a new generation of mathematical researchers to meet the scientific challenges associated with a changing climate.

This project is driven by the need to better understand the Earth's climate system. Climate is the result of many geophysical and chemical processes in the Earth's atmosphere, oceans and biosphere. These processes evolve in time over many scales, ranging from minutes to centuries, and interact in multiple ways, most often nonlinearly. Feedback mechanisms, many of which are poorly understood, further complicate the picture. Because there is only one Earth, climate cannot be studied by systematic experimentation; the only approach available to climate researchers is through computational experiments. These experiments are based on mathematical models, which must be simple enough not to exceed the capabilities of today's advanced computer architectures, while still incorporating the physical and chemical processes that are essential for realistic climate outcomes. The expertise of mathematical scientists in designing, assessing and interpreting these models is critical. This "Mathematics and Climate Research Network" helps engage the mathematical sciences community to address the mathematical and statistical issues of our changing climate. The Research Network takes full advantage of current information technology; communication and collaboration among the participants takes place mostly remotely over the Internet.