Applied mathematician honored for leading-edge work

A professor of applied mathematics at the University of Colorado has been named to the 2010 class of fellows of the Society for Industrial and Applied Mathematics.

CU Professor Stephen McCormick, one of 34 fellows who will be recognized during SIAM’s annual meeting in Pittsburgh, is being honored for “contributions to numerical partial differential equations, especially multigrid and first-order system least-square methods.”

In a statement, the society noted that each fellow had made significant contributions to the field of applied mathematics and computational science.

“Men and women from four continents and numerous disciplines, they exemplify SIAM’s vision of the central role mathematics plays in the advancement of science and technology,” said Douglas N. Arnold, president of the society, adding:

“They have created mathematical models for complex phenomena ranging from combustion to swarming.  They have designed algorithms that enabled the restoration of old movies and the design of new aircraft. They have developed mathematical tools for optimizing manufacturing processes, tracking epidemics, and quantifying the uncertainty in both. Along the way, they have developed elegant mathematics which strengthens the foundations on which future work will rely.”

“Steve is one of the top computational researchers in the world,” adds Jim Curry, professor and chair of CU’s applied mathematics department. “Further, he has helped to train a generation of computational scientist and engineers and was one of the original researchers to help develop the multi-scale and multi-level techniques known collectively as ‘multigrid methods.’”

McCormick explains that computational mathematics is an integral part of research and development in fields such as physics, engineering, biology, medicine and economics. Research in each discipline involves mathematical systems of equations that govern the state of physical systems.

These governing equations are important because their solutions allow researchers to simulate physical processes that would otherwise be too expensive, too time-consuming, too damaging or just impossible to observe in practice, McCormick notes.

For instance, modeling the equations of fluids, structures and electromechanics enables virtual surgery on a human heart. Because such equations are difficult to solve, computational mathematicians strive to develop better methods to numerically solve complex systems of equations accurately and in a reasonable amount of time.

McCormick’s work includes efforts to improve adaptive/bootstrap algebraic multigrid methods. Unlike other computational methods, which take more computing time with successive iterations until “numerical processing grinds almost to a halt,” the adaptive/bootstrap algebraic multigrid methods “have eliminated this bottleneck,” he notes.

That improvement enables particle simulation with much higher fidelity and more physical realism, he adds.

McCormick is author, co-author or editor several books, including “Multigrid Methods,” “A Multigrid Tutorial, Second Edition” and “Multilevel Adaptive Methods for Partial Differential Equations.”

McCormick is the second SIAM fellow in the Department of Applied Mathematics, Professor Tom Manteuffel was inducted last year.

Curry adds: “It is correct to say that Steve and his co-workers, and especially his applied mathematics department colleague Tom Manteuffel, also a SIAM fellow, have made seminal contributions in algorithm design, computational software implementation and the deep theory that are the foundations of multigrid methods.”

Finally, Curry says it is rumored that McCormick can be seen wearing shorts and a T-shirt “at least 312 days per year.”