Climate Change and Computational Science

Climate change is a major challenge for the world, with enormous potential impact on our environment, economy and health. At the same time it offers many fascinating problems in geosciences, mathematics (nonlinear dynamics and chaos), physics and statistics. Central among them is the problem of turbulence (the last great unsolved problem of classical physics Richard Feynman, a Nobel Prize winner). Beginning with pioneering work of A. N. Kolmogorov (generally regarded as the greatest probabilist of the 20th century who also set off statistical fluid mechanics and the dynamical systems approach to turbulence) statistics has played a key role here. It provides powerful tools combining classical methods with those of computational statistics. The latter employs computers in the creation of new statistical methodology for typical climate studies situations where data sets are very large, high-dimensional, or nonhomogeneous. Current research involves capturing interactions between models describing climate processes (from simple nonlinear dynamical systems to high-end comprehensive climate models) and statistical analysis to quantify uncertainty and to predict surprises that climate has in store for us. Collaborations are envisaged with Climate and Extreme Weather (CLEW) research group in EAS and with Purdue Climate Change Research Center (PCCRC).

Relevant Courses

STAT 420 Introductory Time Series

STAT 520 Time Series and Applications

STAT 526 Advanced Statistical Methodology

EAS 520 Theory of Climate

EAS 591 Chaos and Predictability

Course Description:Chaotic dynamics is considered by many to be the most important discovery in the 20th century after relativity and quantum mechanics. It cuts across traditional subject boundaries in science, and it has had major impacts on many fields. A chaotic system is deterministic but it behaves in an apparently random manner. It turns out that such behavior is just as common in nature as periodic cycles. In the course, the mathematical concept of chaos is explained, with applications to nonlinear time series analysis, atmospheric dynamics, and atmospheric and climate predictability.